| Page Range | 66128-66340 | |
| FR Document | 2015-26607 |
[Federal Register Volume 80, Number 208 (Wednesday, October 28, 2015)] [Rules and Regulations] [Pages 66128-66340] From the Federal Register Online [www.thefederalregister.org] [FR Doc No: 2015-26607] [[Page 66127]] Vol. 80 Wednesday, No. 208 October 28, 2015 Part II Bureau of Consumer Financial Protection ----------------------------------------------------------------------- 12 CFR Part 1003 Home Mortgage Disclosure (Regulation C); Final Rule Federal Register / Vol. 80 , No. 208 / Wednesday, October 28, 2015 / Rules and Regulations [[Page 66128]] ----------------------------------------------------------------------- BUREAU OF CONSUMER FINANCIAL PROTECTION 12 CFR Part 1003 [Docket No. CFPB-2014-0019] RIN 3170-AA10 Home Mortgage Disclosure (Regulation C) AGENCY: Bureau of Consumer Financial Protection. ACTION: Final rule; official interpretations. ----------------------------------------------------------------------- SUMMARY: The Bureau of Consumer Financial Protection is amending Regulation C to implement amendments to the Home Mortgage Disclosure Act made by section 1094 of the Dodd-Frank Wall Street Reform and Consumer Protection Act (Dodd-Frank Act). Consistent with section 1094 of the Dodd-Frank Act, the Bureau is adding several new reporting requirements and clarifying several existing requirements. The Bureau is also modifying the institutional and transactional coverage of Regulation C. The final rule also provides extensive guidance regarding compliance with both the existing and new requirements. DATES: This rule is effective on January 1, 2018, except that the amendment to Sec. 1003.2 in amendatory instruction 3 is effective on January 1, 2017; the amendments to Sec. 1003.5 in amendatory instruction 8, the amendments to Sec. 1003.6 in amendatory instruction 10, the amendments to appendix A to part 1003 in amendatory instruction 12, and the amendments to supplement I to part 1003 in amendatory instruction 16 are effective on January 1, 2019; and the amendments to Sec. 1003.5 in amendatory instruction 9 are effective on January 1, 2020. See part VI for more information. FOR FURTHER INFORMATION CONTACT: Jaydee DiGiovanni, David Jacobs, Terry J. Randall, or James Wylie, Counsels; or Elena Grigera Babinecz, Courtney Jean, Joan Kayagil, Thomas J. Kearney, or Laura Stack, Senior Counsels, Office of Regulations, at (202) 435-7700. SUPPLEMENTARY INFORMATION: I. Summary of the Final Rule Regulation C implements the Home Mortgage Disclosure Act (HMDA), which was amended by the Dodd-Frank Wall Street Reform and Consumer Protection Act (Dodd-Frank Act). On July 24, 2014, the Bureau issued a proposed rule to amend Regulation C, which was published in the Federal Register on August 29, 2014 (the 2014 HMDA Proposal or the proposal).\1\ The Bureau is publishing final amendments to Regulation C modifying the types of institutions and transactions subject to the regulation, the types of data that institutions are required to collect, and the processes for reporting and disclosing the required data. --------------------------------------------------------------------------- \1\ 79 FR 51731 (Aug. 29, 2014). See also Press Release, U.S. Bureau of Consumer Fin. Prot., CFPB Proposes Rule to Improve Information About Access to Credit in the Mortgage Market (July 24, 2014), available at http://www.consumerfinance.gov/newsroom/cfpb-proposes-rule-to-improve-information-about-access-to-credit-in-the-mortgage-market/. --------------------------------------------------------------------------- A. Modifications to Institutional and Transactional Coverage The Bureau is modifying Regulation C's institutional and transactional coverage to better achieve HMDA's purposes in light of current market conditions and to reduce unnecessary burden on financial institutions. The Bureau is adopting uniform loan-volume thresholds for depository and nondepository institutions. The loan-volume thresholds require an institution that originated at least 25 closed-end mortgage loans or at least 100 open-end lines of credit in each of the two preceding calendar years to report HMDA data, provided that the institution meets all of the other criteria for institutional coverage. The final rule also includes a separate test to ensure that covered institutions that meet only the 25 closed-end mortgage loan threshold are not required to report their open-end lending, and that covered institutions that meet only the 100 open-end line of credit threshold are not required to report their closed-end lending. In addition, the final rule retains the current institutional coverage criteria for depository institutions, which require reporting by depository institutions that satisfy an asset-size threshold, have a branch or home office in an Metropolitan Statistical Area (MSA) on the preceding December 31, satisfy the current federally related test, and originated at least one first-lien home purchase loan or refinancing secured by a one- to four-unit dwelling in the previous calendar year. For nondepository institutions, the final rule replaces the current loan-volume or -amount test with the loan-volume thresholds discussed above, and removes the current asset-size or loan-volume threshold, but retains the current criterion that the institution have a branch or home office in an MSA on the preceding December 31. The Bureau also is modifying the types of transactions subject to Regulation C. The final rule adopts a dwelling-secured standard for all loans or lines of credit that are for personal, family, or household purposes. Thus, most consumer-purpose transactions, including closed- end home-equity loans, home-equity lines of credit, and reverse mortgages, are subject to the regulation. Most commercial-purpose transactions (i.e., loans or lines of credit not for personal, family, or household purposes) are subject to the regulation only if they are for the purpose of home purchase, home improvement, or refinancing. The final rule excludes from coverage home improvement loans that are not secured by a dwelling (i.e., home improvement loans that are unsecured or that are secured by some other type of collateral) and all agricultural-purpose loans and lines of credit. B. Modifications to Reportable Data Requirements The final rule amends several of Regulation C's currently required data points to clarify the requirements and make the data more useful. To streamline the regulation, the final rule removes appendix A; all of the substantive requirements contained in appendix A have been moved, with some modifications, to the regulation text or commentary. The final rule also adopts several new data points, many of which were added by the Dodd-Frank Act, and some of which were added pursuant to the Bureau's discretionary authority to carry out the purposes of HMDA. The final rule does not adopt some of the new or amended data points set forth in the 2014 HMDA Proposal, such as the proposed requirements to report qualified mortgage status or the initial draw on an open-end line of credit. The data points required to be reported under the final rule can be grouped into four broad categories:Information about applicants, borrowers, and the underwriting process, such as age, credit score, debt-to-income ratio, and automated underwriting system results. Information about the property securing the loan, such as construction method, property value, and additional information about manufactured and multifamily housing. Information about the features of the loan, such as additional pricing information, loan term, interest rate, introductory rate period, non-amortizing features, and the type of loan. Certain unique identifiers, such as a universal loan identifier, property address, loan originator identifier, and a [[Page 66129]] legal entity identifier for the financial institution.\2\ --------------------------------------------------------------------------- \2\ All of the data points required by the final rule are discussed in detail below in the section-by-section analysis of Sec. 1003.4(a). --------------------------------------------------------------------------- The final rule also amends the current requirements related to the collection of ethnicity, race, and sex of applicants and borrowers. The final rule requires financial institutions to report whether ethnicity, race, or sex information was collected on the basis of visual observation or surname when an application is taken in person and the applicant does not provide the information. For transactions where ethnicity and race information is provided by the applicant or borrower, the final rule requires financial institutions to permit applicants and borrowers to self-identify using disaggregated ethnic and racial categories. However, when race and ethnicity data is completed by the financial institution, the final rule retains the current requirements, requiring financial institutions to provide only aggregated ethnic or racial data. C. Modifications to Disclosure and Reporting Requirements The final rule retains the current requirement that financial institutions submit their HMDA data to the appropriate Federal agency by March 1 following the calendar year for which the data are collected. The final rule imposes a new requirement that financial institutions that report large volumes of HMDA data for a calendar year also submit their data for the first three quarters of the following calendar year to the appropriate Federal agency on a quarterly basis. However, the final rule removes the current requirements that a financial institution provide to the public its disclosure statement and its loan/application register, modified to protect applicant and borrower privacy, and instead requires financial institutions to provide a notice to members of the public seeking these data that the information is available on the Bureau's Web site. II. Background A. HMDA and Regulation C For nearly 40 years, HMDA has provided the public with information about mortgage lending activity within communities throughout the nation. Public officials use the information available through HMDA to develop and allocate housing and community development investments, to respond to market failures when necessary, and to monitor whether financial institutions may be engaging in discriminatory lending practices. The data are used by the mortgage industry to inform business practices, and by local communities to ensure that lenders are serving the needs of individual neighborhoods. To maintain the data's usefulness, HMDA and Regulation C have been updated and expanded over time in response to the changing needs of homeowners and evolution in the mortgage market. This part II.A provides an abbreviated discussion of the detailed background information presented in the proposal, which the Bureau considered and relied on in preparing this final rule.\3\ --------------------------------------------------------------------------- \3\ See 79 FR 51731, 51734-39 (Aug. 29, 2014). --------------------------------------------------------------------------- The Statute and Current Regulation The Home Mortgage Disclosure Act (HMDA), 12 U.S.C. 2801 et seq., requires certain depository institutions and for-profit nondepository institutions to collect, report, and disclose data about originations and purchases of mortgage loans, as well as mortgage loan applications that do not result in originations (for example, applications that are denied or withdrawn). As originally adopted, HMDA identifies its purposes as providing the public and public officials with information to help determine whether financial institutions are serving the housing needs of the communities in which they are located, and to assist public officials in their determination of the distribution of public sector investments in a manner designed to improve the private investment environment.\4\ Congress later expanded HMDA to, among other things, require financial institutions to report racial characteristics, gender, and income information on applicants and borrowers.\5\ In light of these amendments, the Board of Governors of the Federal Reserve System (Board) subsequently recognized a third HMDA purpose of identifying possible discriminatory lending patterns and enforcing antidiscrimination statutes, which now appears with HMDA's other purposes in Regulation C.\6\ --------------------------------------------------------------------------- \4\ HMDA section 302(b), 12 U.S.C. 2801(b); see also 12 CFR 1003.1(b)(1)(i)-(ii). \5\ Financial Institutions Reform, Recovery, and Enforcement Act of 1989, Public Law 101-73, section 1211 (``Fair lending oversight and enforcement'' section), 103 Stat. 183, 524-26 (1989). \6\ 54 FR 51356, 51357 (Dec. 15, 1989), codified at 12 CFR 1003.1(b)(1). --------------------------------------------------------------------------- In 2010, Congress enacted the Dodd-Frank Act, which amended HMDA and also transferred HMDA rulemaking authority and other functions from the Board to the Bureau.\7\ Among other changes, the Dodd-Frank Act expands the scope of information relating to mortgage applications and loans that must be compiled, maintained, and reported under HMDA. New data points include the age of loan applicants and mortgagors, information relating to the points and fees payable at origination, the difference between the annual percentage rate (APR) associated with the loan and a benchmark rate or rates for all loans, the term of any prepayment penalty, the value of real property to be pledged as collateral, the term of the loan and of any introductory interest rate for the loan, the presence of contract terms allowing non-amortizing payments, the origination channel, and the credit scores of applicants and mortgagors.\8\ The Dodd-Frank Act also authorizes the Bureau to require, ``as [it] may determine to be appropriate,'' a unique identifier that identifies the loan originator, a universal loan identifier, and the parcel number that corresponds to the real property pledged or proposed to be pledged as collateral for the mortgage loan.\9\ The Dodd-Frank Act also provides the Bureau with the authority to require ``such other information as the Bureau may require.'' \10\ --------------------------------------------------------------------------- \7\ Public Law 111-203, 124 Stat. 1376, 1980, 2035-38, 2097-101 (2010). Also, in 2010, the Board conducted public hearings on potential revisions to Regulation C. The Board's hearings are discussed below. \8\ Dodd-Frank Act section 1094(3), amending HMDA section 304(b), 12 U.S.C. 2803(b). \9\ Id. \10\ Id. --------------------------------------------------------------------------- The Bureau's Regulation C, 12 CFR part 1003, implements HMDA. Regulation C currently requires depository institutions (i.e., banks, savings associations, and credit unions) and for-profit nondepository mortgage lending institutions to submit and publicly disclose certain HMDA data if they meet criteria set forth in the rule. Whether a depository institution is required to report and publicly disclose data depends on its asset size, the location of its home and branch offices, the extent to which it engages in residential mortgage lending, and the extent to which the institution or its loans are federally related. Whether a for-profit nondepository mortgage lending institution is required to report and publicly disclose data depends on its size, the location of its home and branch offices, including the extent of its business in MSAs, and the extent to which it engages in residential mortgage lending. Covered financial institutions are required to report originations and purchases of mortgage loans (home purchase and refinancing) and home improvement loans, as well as loan [[Page 66130]] applications that do not result in originations. The information reported under Regulation C currently includes, among other items: application date; loan or application type, purpose, and amount; property location and type; race, ethnicity, sex, and annual income of the loan applicant; action taken on the loan application (approved, denied, withdrawn, etc.), and date of that action; whether the loan is subject to the Home Ownership and Equity Protection Act of 1994 (HOEPA); lien status (first lien, subordinate lien, or unsecured); and certain loan price information. Depository financial institutions report HMDA data to their supervisory agencies, while nondepository financial institutions report HMDA data to the U.S. Department of Housing and Urban Development (HUD). Financial institutions report their data on an application-by- application basis using a register format referred to as the loan/ application register. Institutions must make their loan/application registers available to the public, with certain fields redacted to preserve applicants' and borrowers' privacy. At present, the Federal Financial Institutions Examination Council (FFIEC),\11\ on behalf of the supervisory agencies, compiles the reported data and prepares an individual disclosure statement for each institution and aggregate reports for all covered institutions in each metropolitan area. These disclosure statements and reports are available to the public. On behalf of the agencies, the FFIEC also annually releases a loan-level dataset containing all reported HMDA data for the preceding calendar year with certain fields redacted to protect the privacy of applicants and borrowers. --------------------------------------------------------------------------- \11\ The FFIEC is a formal interagency body empowered to prescribe uniform principles, standards, and report forms for the Federal examination of financial institutions by the Bureau, the Board, the Federal Deposit Insurance Corporation (FDIC), the National Credit Union Administration (NCUA), and the Office of the Comptroller of the Currency (OCC), and to make recommendations to promote uniformity in the supervision of financial institutions. In 2006, the State Liaison Committee was added to the Council as a voting member. --------------------------------------------------------------------------- Overview of HMDA's Purposes and Evolution In the decades that followed World War II, the standard of living declined sharply in many U.S. cities as people migrated to the suburbs. A significant cause of this decline was the gradual deterioration of the urban housing supply. Although Congress took several steps to address this problem, by the 1970s it was clear that inadequate private investment and a lack of access to credit was contributing to an ongoing cycle of decline in urban neighborhoods. However, Congress lacked adequate data to determine the extent and severity of these market failures. To create transparency in the mortgage market Congress enacted HMDA in 1975, which the Board implemented by promulgating Regulation C in 1976. As originally enacted, HMDA applied to certain depository institutions that were located in standard metropolitan statistical areas, and required the disclosure of a limited amount of data regarding home improvement and residential mortgage loans.\12\ --------------------------------------------------------------------------- \12\ See 79 FR 51731, 51735-36 (Aug. 29, 2014). --------------------------------------------------------------------------- HMDA substantially improved the public's ability to determine whether financial institutions were serving the needs of their communities, but during the 1980s several events occurred that illustrated the need to improve and expand the HMDA data. A series of investigative reports and studies revealed that discrimination against certain applicants and borrowers was common during the mortgage lending process. Concerns over this discrimination, coupled with the need to respond to the savings and loan crisis of the late 1980s, led Congress to amend HMDA significantly in 1988 and 1989. These amendments, among other things, expanded the coverage of depository and nondepository institutions, required transaction-level disclosure of applications and loans, and added new reporting requirements regarding the applicant's or borrower's race, gender, and income. These amendments dramatically improved the public's understanding of how mortgage lending decisions affected both communities and individual applicants and borrowers.\13\ --------------------------------------------------------------------------- \13\ See 79 FR 51731, 51736-37 (Aug. 29, 2014). --------------------------------------------------------------------------- The mortgage market evolved and became more complex during the 1990s, particularly with respect to the expansion of the secondary market and the growth of the subprime market. Faced with concerns about potential predatory and discriminatory practices in the subprime market, community groups and others began to call for new amendments to HMDA to provide increased visibility into market practices. The Board addressed some of these concerns by amending Regulation C in 2002. However, as delinquencies, foreclosures, and other harmful effects of subprime lending unfolded, it became apparent that communities throughout the nation lacked sufficient information to understand the magnitude of the risk to which they were exposed. Community groups, local, State, and Federal officials relied on the HMDA data to identify at-risk neighborhoods and to develop foreclosure relief and homeownership stabilization programs. However, the limited data provided presented several challenges for those who attempted to create effective and responsive relief programs. As discussed above, Congress added several new reporting requirements, but left the Bureau to determine which additional information is necessary. Many argue that more publicly available information is needed to help inform communities of lending practices that affect local economies and may endanger neighborhood stability. The Bureau believes that the HMDA data must be updated to address the informational shortcomings exposed by the financial crisis and to meet the needs of homeowners, potential homeowners, and neighborhoods throughout the nation.\14\ --------------------------------------------------------------------------- \14\ See 79 FR 51731, 51737-39 (Aug. 29, 2014). --------------------------------------------------------------------------- B. Applicant and Borrower Privacy In its proposal, the Bureau set forth the approach it proposed to take to protect applicant and borrower privacy in light of HMDA's purposes. It proposed the use of a balancing test to determine whether and how HMDA data should be modified prior to its disclosure to the public in order to protect applicant and borrower privacy while also fulfilling the disclosure purposes of the statute.\15\ For the reasons described below, the Bureau is adopting the balancing test described in the proposal. The Bureau will provide at a later date a process for the public to provide input on the application of the balancing test to determine the HMDA data to be publicly disclosed. --------------------------------------------------------------------------- \15\ 79 FR 51731, 51742 (Aug. 29, 2014). --------------------------------------------------------------------------- HMDA's purposes are to provide the public and public officials with sufficient information to enable them to determine whether institutions are serving the housing needs of the communities and neighborhoods in which they are located, to assist public officials in distributing public sector investments in a manner designed to improve the private investment environment, and to assist in identifying possible discriminatory lending patterns and enforcing antidiscrimination statutes. Today, HMDA data are the primary source of information for regulators, researchers, economists, industry, and advocates analyzing the mortgage market both for HMDA's purposes and for general market monitoring. Developing appropriate protections for applicant and borrower [[Page 66131]] privacy in light of HMDA's purposes is a significant priority for the Bureau. The Bureau is mindful that privacy concerns may arise both when financial institutions compile and report HMDA data to their regulators and when the data are disclosed to the public. Compiling and Reporting of HMDA Data Financial institutions collect various types of information from consumers in the course of processing loan applications. To promote HMDA's goals, HMDA and Regulation C require financial institutions to compile and report to regulators some of this information and other information obtained or generated concerning the application or loan. As discussed above, the Dodd-Frank Act both expanded the scope of information that financial institutions must compile and report and authorized the Bureau to require financial institutions to compile and report additional data. The Bureau carefully considered the potential risks to applicant and borrower privacy associated with compiling and reporting data in developing the proposal and adopting this final rule. Neither consumer advocate commenters nor the privacy advocate that submitted a comment identified concerns about applicant and borrower privacy associated with the compilation and reporting of data to regulators under the proposal. However, the Bureau received many comments from industry arguing that the compilation and reporting of certain data under the proposal created significant and unjustified risks to applicant and borrower privacy. These comments focused on concerns relating to the potential identifiability and sensitivity of the data to be compiled and reported. Most commenters expressed concerns about potential harms to applicants and borrowers if the data compiled and reported under the proposal were subject to unauthorized access. A few commenters also expressed concerns about potential legal liability and costs to financial institutions associated with the compilation and reporting of the proposed data. Many industry commenters argued that the proposed requirement to report the postal address of the property securing the covered loan or, in the case of an application, proposed to secure the covered loan \16\ would allow data users to easily link all reported data to an individual applicant or borrower. Some commenters also suggested that proposed data fields other than postal address could allow individual applicants and borrowers to be identified in the reported HMDA data. Many industry commenters asserted that some of the proposed data fields, if tied to an individual, would reveal sensitive information about the applicant or borrower.\17\ --------------------------------------------------------------------------- \16\ Proposed Sec. 1003.4(a)(9)(i). \17\ Some commenters suggested that the Bureau require certain data to be reported in ranges, rather than exact values, to mitigate privacy concerns. Comments received concerning particular data points are addressed in the applicable section-by-section analysis below. --------------------------------------------------------------------------- Some industry commenters expressed general concern about government collection of information that may be linkable to individuals, but most commenters expressed specific concerns about potential harms to applicants and borrowers in the event of unauthorized access to the HMDA data maintained by the agencies. Commenters asserted that the proposal increased both the potential harm a breach of the HMDA data at the Bureau or another agency could cause affected applicants and borrowers as well as the risk that such a data breach would occur. Many comments stated that the proposed HMDA data could be used to target applicants and borrowers with marketing for harmful financial products and to commit identity theft and other fraud. Several commenters stated that data breaches at corporations and government agencies have become common and suggested that the proposed HMDA data are sufficiently valuable to identity thieves and others that agency systems maintaining the data would be subject to hacks and other attacks aiming to access the data. A few commenters expressed concern that the HMDA data would be vulnerable to unauthorized access during transmission from financial institutions to their regulators. Several industry commenters expressed particular concern with the Bureau's information security practices and suggested that HMDA data held by the Bureau would be at heightened risk of breach. A few of these commenters urged the Bureau to publish the details of its information security practices and procedures in order to address these concerns. Some industry commenters questioned the benefit of some of the proposed data in light of HMDA's purposes. Several commenters argued that, in light of the potential risks to applicant and borrower privacy presented by the compilation and reporting of the some of the proposed data, any benefits of such compilation and reporting were not justified. In addition, a few commenters expressed concern that compiling and reporting the proposed data would create legal risks for financial institutions and would impose related costs. A few comments suggested that a financial institution would face regulatory or legal liability if an agency suffered a breach that compromised the financial institution's HMDA data. One comment suggested that reporting the proposed data would expose financial institutions to liability under the Right to Financial Privacy Act (RFPA) \18\ and a few other commenters suggested that doing so would violate the Gramm-Leach-Bliley Act (GLBA)\19\. Several national trade associations argued that compiling and reporting the proposed data would require financial institutions to strengthen significantly their information security programs and would also increase costs associated with compensating customers in the event of a financial institution's data breach. --------------------------------------------------------------------------- \18\ 12 U.S.C. 3401 et seq. \19\ 15 U.S.C. 6801 et seq. --------------------------------------------------------------------------- The Bureau has analyzed these industry comments carefully and has determined that any risks to applicant and borrower privacy created by the compilation and reporting of the data required under the final rule are justified by the benefits of the data in light of HMDA's purposes.\20\ The Bureau takes seriously the concerns raised about the security of reported HMDA data maintained at the agencies. The Bureau has addressed or is actively addressing each of the recommendations made in the Government Accountability Office (GAO) report cited by some industry commenters as a basis for concern that the Bureau's information security practices are insufficient to protect HMDA data.\21\ The GAO report [[Page 66132]] recognized the many steps that the Bureau has taken to ensure the privacy and security of the data it collects; indeed, the report's recommendations focused primarily on formalizing and documenting the privacy and information security practices the Bureau already had in place at the time the report was issued. The Bureau takes strong measures to mitigate and address any risks to the security of sensitive data it receives, consistent with the guidance and standards set for Federal information security programs,\22\ and is committed to protecting the privacy and information security of the HMDA data it receives from financial institutions. As discussed in its proposal,\23\ the Bureau is developing improvements to the HMDA data submission process, including, for example, further advancing encryption if necessary to protect data reported under the final rule. --------------------------------------------------------------------------- \20\ Several industry commenters asserted that, under the Bureau's proposal, none of the proposed new data points would be made available to the public, or would be made available only in aggregate form, and that this was evidence of the limited value of the proposed data in light of HMDA's purposes. These commenters misunderstood the proposal. The Bureau proposed that the data financial institutions would disclose on their modified loan/ application registers would be limited to the currently disclosed data, see proposed Sec. 1003.5(c), but stated that it would apply a balancing test to determine whether and how the HMDA data should be modified prior to its public release by the agencies in their annual loan-level data release, see 79 FR 51731, 51742, 51816 (Aug. 29, 2014). Based on its analysis to date, the Bureau believes that some of the proposed new data points may create privacy concerns sufficient to warrant some degree of modification, including redaction, before public disclosure, but it has determined that all of the data required to be compiled and reported under the final rule significantly advance HMDA's purposes. \21\ U.S. Gov't Accountability Office, GAO-14-758, Consumer Financial Protection Bureau: Some Privacy and Security Procedures for Data Collections Should Continue Being Enhanced (2014), available at http://www.gao.gov/assets/670/666000.pdf. In this report, the GAO examined the Bureau's authority to receive consumer financial information as well as steps taken to implement the privacy and information security protections to address risks associated with the receipt of such information. The report contained eleven recommendations directed to the Bureau. \22\ The Bureau's information security program is aligned with the requirements of the Federal Information Security Management Act of 2002 (FISMA). Like other Federal information security programs, the policies and principles that form the CFPB information security program are based on guidance and standards provided by the National Institute of Standards and Technology (NIST). The Bureau declines to publish details of its information security safeguards, as suggested by some industry commenters, because such disclosure would pose a significant security risk. \23\ 79 FR 51731, 51741 (Aug. 29, 2014). --------------------------------------------------------------------------- The Bureau does not believe a financial institution could be held legally liable for the exposure of data due to a breach at a government agency or for reporting data to a government agency if the institution was legally required to provide the data to the agency and did so in accordance with other applicable law. The comments raising this concern provided no evidence or analysis concerning how such liability might be created. Contrary to a few commenters' suggestions, reporting data as required under the final rule would not create liability for a financial institution under the RFPA or cause the financial institution to violate the GLBA, as both of these laws permit financial institutions to disclose information as required by Federal law or regulation.\24\ Finally, in light of the significant amounts of highly sensitive, personally identifiable information concerning customers that financial institutions collect and maintain in the course of conducting their business regardless of HMDA and Regulation C, the Bureau does not believe the requirement to compile and report some of these data pursuant to the final rule will meaningfully increase financial institutions' information security needs or the amounts required for victim compensation in the event of a financial institution's security breach. The industry commenters that made these arguments offered no detail or evidence of such needs or costs. It is the Bureau's understanding that substantially all of the new data to be compiled under the final rule are either data that HMDA reporters compile for reasons other than HMDA or Regulation C or are calculations that derive from such data, and must be retained by a financial institution to comply with other applicable laws. --------------------------------------------------------------------------- \24\ See 12 U.S.C. 3413(d) (providing an exception to the RFPA's general prohibition on disclosure to the Federal government for financial records or information ``required to be reported in accordance with any Federal statute or rule promulgated thereunder''); 15 U.S.C. 6802(e)(8), 12 CFR 1016.15(a)(7)(i) (providing an exception to GLBA's general prohibition on disclosing nonpublic personal information to a nonaffiliated third party absent notice and an opportunity to opt-out of such disclosure where the disclosure is ``to comply with Federal, State, or local laws, rules, and other applicable legal requirements.''). --------------------------------------------------------------------------- Disclosures of HMDA Data As discussed in part II.A above, HMDA is a disclosure statute. To fulfill HMDA's purposes, the types of data a financial institution is required to compile and report under HMDA and Regulation C have been expanded since the statute's enactment in 1975, and the formats in which HMDA data have been disclosed to the public also have evolved. At present, HMDA and Regulation C require data to be made available to the public in both aggregate and loan-level formats. First, each financial institution must make its ``modified'' loan/application register available to the public, with three fields deleted to protect applicant and borrower privacy.\25\ Each financial institution must also make available to the public a disclosure statement prepared by the FFIEC that shows the financial institution's HMDA data in aggregate form.\26\ In addition, the FFIEC makes available to the public disclosure statements for each financial institution \27\ as well as aggregate reports for each MSA and metropolitan division (MD) showing lending patterns by certain property and applicant characteristics.\28\ Since 1991, on behalf of the agencies receiving HMDA data, the FFIEC also has released annually a loan-level dataset containing all reported HMDA data for the preceding calendar year (the agencies' release). To reduce the possibility that data users could identify particular applicants or borrowers in these data, the same three fields that are deleted from the modified loan/application register are deleted from the agencies' release.\29\ --------------------------------------------------------------------------- \25\ Section 1003.5(c); HMDA section 304(j)(2)(B). Section 1003.5(c) requires that, before making its loan/application register available to the public, a financial institution must delete three fields to protect applicant and borrower privacy: Application or loan number, the date that the application was received, and the date action was taken. \26\ Section 1003.5(b); HMDA section 304(k). \27\ Section 1003.5(f); HMDA section 304(f). \28\ Section 1003.5(f); HMDA section 310. \29\ The agencies first released loan-level HMDA data in October 1991. In announcing that the loan-level data submitted to the agencies on the loan/application register would be made available to the public, the FFIEC noted that ``[a]n unedited form of the data would contain information that could be used to identify individual loan applicants'' and that the data would be edited prior to public release to remove the application identification number, the date of application, and the date of final action. 55 FR 27886, 27888 (July 6, 1990). --------------------------------------------------------------------------- Changes to financial institutions' disclosure obligations under the final rule. The Bureau's proposal addressed both of the disclosures financial institutions must make to the public under current Regulation C. First, the Bureau proposed to allow a financial institution to meet its obligation to make its disclosure statement available to the public by making available a notice that clearly conveys that the disclosure statement may be obtained on the FFIEC Web site and that includes the FFIEC's Web site address.\30\ Second, it proposed to require that the modified loan/application register a financial institution must make available show only the data fields that currently are released on the modified loan/application register.\31\ The Bureau explained that the new data points adopted under the final rule would be disclosed in the agencies' release, modified as appropriate to protect applicant and borrower privacy.\32\ These proposals aimed to reduce burden on financial institutions associated with their disclosure of HMDA data and allow for the appropriate protection of applicant and borrower privacy in HMDA data disclosed by shifting much of the responsibility for making HMDA data available to the public to the agencies. --------------------------------------------------------------------------- \30\ Proposed Sec. 1003.5(b)(2). \31\ Proposed Sec. 1003.5(c). \32\ 79 FR 51731, 51742-43, 51816 (Aug. 29, 2014). --------------------------------------------------------------------------- The Bureau received several comments on the proposed provisions relating to financial institutions' disclosure obligations. As discussed below in the applicable section-by-section analysis, after consideration of [[Page 66133]] these comments and further analysis, the Bureau has decided to finalize proposed Sec. 1003.5(b)(2) concerning the disclosure statement with minor modifications. The Bureau is not finalizing Sec. 1003.5(c) concerning the modified loan/application register as proposed and instead is aligning Sec. 1003.5(c) with Sec. 1003.5(b)(2) by adopting a requirement that a financial institution make available to the public a notice that clearly conveys that the institution's modified loan/ application register may be obtained on the Bureau's Web site. Thus, under the final rule, the disclosure of HMDA data is shifted entirely to the agencies; financial institutions will no longer be required to provide their HMDA data directly to the public, but only a notice advising members of the public seeking their data of where it may be obtained online. Use of a balancing test to determine data to be publicly disclosed. The Dodd-Frank Act amendments to HMDA added new section 304(h)(1)(E), which directs the Bureau to develop regulations, in consultation with the other agencies, that ``modify or require modification of itemized information, for the purpose of protecting the privacy interests of the mortgage applicants or mortgagors, that is or will be available to the public.'' Section 304(h)(3)(B), also added by the Dodd-Frank Act, directs the Bureau to ``prescribe standards for any modification under paragraph (1)(E) to effectuate the purposes of [HMDA], in light of the privacy interests of mortgage applicants or mortgagors. Where necessary to protect the privacy interests of mortgage applicants or mortgagors, the Bureau shall provide for the disclosure of information . . . in aggregate or other reasonably modified form, in order to effectuate the purposes of [HMDA].'' \33\ --------------------------------------------------------------------------- \33\ Section 304(h)(3)(A) provides that a modification under section 304(h)(1)(E) shall apply to information concerning ``(i) credit score data . . . in a manner that is consistent with the purpose described in paragraph (1)(E); and (ii) age or any other category of data described in paragraph (5) or (6) of subsection (b), as the Bureau determines to be necessary to satisfy the purpose described in paragraph (1)(E), and in a manner consistent with that purpose.'' --------------------------------------------------------------------------- The Bureau explained in its proposal that it interprets HMDA, as amended by the Dodd-Frank Act, to call for the use of a balancing test to determine whether and how HMDA data should be modified prior to its disclosure to the public in order to protect applicant and borrower privacy while also fulfilling HMDA's public disclosure purposes.\34\ Using the balancing test to evaluate particular HMDA data points, individually and in combination, and various options for providing access to HMDA data, the Bureau proposed to balance the importance of releasing the data to accomplish HMDA's public disclosure purposes against the potential harm to an applicant or borrower's privacy interest that may result from the release of the data without modification. The proposal explained that modifications the Bureau may consider where warranted include various disclosure limitation techniques, such as techniques aimed at masking the precise value of data points,\35\ aggregation, redaction, use restrictions, and query- based systems. HMDA's public disclosure purposes might also be furthered by implementing a restricted access program.\36\ The Bureau explained that it interpreted HMDA, as amended by the Dodd-Frank Act, to require that public HMDA data be modified when the release of the unmodified data creates risks to applicant and borrower privacy interests that are not justified by the benefits of such release to the public in light of the statutory purposes. The Bureau also sought comment on its view that, considering the public disclosure of HMDA data as a whole, applicant and borrower privacy interests arise under the balancing test only where the disclosure of HMDA data may both substantially facilitate the identification of an applicant or borrower in the data and disclose information about the applicant or borrower that is not otherwise public and may be harmful or sensitive. The proposal explained that the Bureau's analysis of the proposed HMDA data under the balancing test was ongoing and included data fields currently disclosed on the modified loan/application register and in the agencies' release. The Bureau stated that it would provide at a later date a process for the public to provide input on the application of the balancing test to determine the HMDA data to be publicly disclosed. --------------------------------------------------------------------------- \34\ Section 1022(c)(8) of the Dodd-Frank Act provides that, ``[i]n collecting information from any person, publicly releasing information held by the Bureau, or requiring covered persons to publicly report information, the Bureau shall take steps to ensure that'' certain information is not ``made public under this title.'' The Bureau interprets ``under this title'' to not include data made public pursuant to HMDA and Regulation C. \35\ Binning and suppression are examples of commonly-used data masking techniques. Binning, sometimes known as recoding or interval recoding, provides only a range for certain fields. Binning allows data to be shown clustered into ranges rather than as precise values. \36\ A restricted access program could allow ``trusted researchers'' access to privacy-sensitive information that is unavailable to the public, for research purposes. --------------------------------------------------------------------------- The Bureau received very few comments concerning the proposed balancing test itself, most of which supported the balancing test. One industry commenter stated that the balancing test was too narrow, but its comment concerned the types of available information the Bureau should consider in analyzing the potential risks of re-identification and harm to applicants and borrowers presented by the public disclosure of HMDA data, and the types of potential harmful uses of HMDA data, rather than the balancing test itself. The Bureau received many comments from consumer advocates, researchers, industry, and a privacy advocate concerning the application of the balancing test to the current and proposed HMDA data. These comments concerned (i) the benefits of public disclosure of the data, (ii) the potential risks to applicant and borrower privacy created by such disclosure, and (iii) modifications and data access and use restrictions the Bureau might consider to protect applicant and borrower privacy where warranted. Many comments, especially from consumer advocates and researchers, identified the benefits of public disclosure of the current and proposed HMDA data. These commenters noted that public disclosure is the fundamental purpose of the Act and argued that public availability of HMDA data: Allows the public to supplement limited government resources to enforce fair lending and other laws and otherwise accomplish the goals of the Act; mitigates the impact of regulator capture or inattention to illegal practices and troublesome trends; and reduces information asymmetry between industry and the public concerning the residential mortgage market. Several comments raised concerns about potential risks to applicant and borrower privacy created by the disclosure of HMDA data. Similar to comments received concerning such potential risks associated with the compilation and reporting of HMDA data, these comments addressed sources of data that could be combined with HMDA data to identify applicants and borrowers in the HMDA data. Several comments also suggested that the Bureau consider how HMDA data may be combined with other available data to harm consumers. Many comments, especially from industry, raised concerns about a variety of specific proposed data points as well as potential harmful uses to which data disclosed to the public may be put, including fraud, identity theft, and [[Page 66134]] targeted marketing of harmful financial products. Finally, several comments concerned data access and use restrictions that the Bureau could consider. Some consumer advocate and researcher comments offered suggestions and recommendations concerning a restricted access program. Several industry comments expressed concerns about the implementation of a restricted access program, however, including concerns that it may create opportunities for data leakage and unauthorized access to the HMDA data. A privacy advocate commenter urged the Bureau to restrict the uses of HMDA data to certain defined purposes, similar to the approach taken with respect to consumer reports under the Fair Credit Reporting Act.\37\ --------------------------------------------------------------------------- \37\ 15 U.S.C. 1681 et seq. --------------------------------------------------------------------------- The Bureau has determined that its interpretation of HMDA to call for the use of the balancing test described above is reasonable and best effectuates the purposes of the statute. The Bureau interprets HMDA, as amended by the Dodd-Frank Act, to require that public HMDA data be modified when the release of the unmodified data creates risks to applicant and borrower privacy interests that are not justified by the benefits of such release to the public in light of the statutory purposes. In such circumstances, the need to protect the privacy interests of mortgage applicants or mortgagors requires that the itemized information be modified. Considering the public disclosure of HMDA data as a whole, applicant and borrower privacy interests arise under the balancing test only where the disclosure of HMDA data may both substantially facilitate the identification of an applicant or borrower in the data and disclose information about the applicant or borrower that is not otherwise public and may be harmful or sensitive. Thus, disclosure of an unmodified individual data point or field may create a risk to applicant or borrower privacy interests if such disclosure would either substantially facilitate the identification of an applicant or borrower or disclose information about an applicant or borrower that is not otherwise public and that may be harmful or sensitive. This interpretation implements HMDA sections 304(h)(1)(E) and 304(h)(3)(B) because it prescribes standards for requiring modification of itemized information, for the purpose of protecting the privacy interests of mortgage applicants and borrowers, that is or will be available to the public. In applying the balancing test, the Bureau will carefully consider all comments received concerning the benefits of disclosure of HMDA data, the risks to applicant and borrower privacy created by such disclosure, and options for data use and access restrictions. However, the Bureau believes that it will be most helpful in applying the balancing test to provide an additional process through which all stakeholders can provide additional comment now that the data to be compiled and reported are finalized. Accordingly, the Bureau intends to provide a process for the public to provide input on the application of the balancing test to determine the HMDA data to be publicly disclosed. The Bureau received some comments suggesting that disclosure of certain HMDA data could reveal confidential business information. As these comments do not concern applicant and borrower privacy, they are addressed in the appropriate section-by-section analyses below. III. Summary of the Rulemaking Process This final rule is the product of several years of research and analysis. In 2010, when the Board had rulemaking authority over HMDA, the Board conducted a series of public hearings that elicited feedback on improvements to Regulation C. After the rulemaking authority for HMDA was transferred to the Bureau, the Bureau conducted additional outreach by soliciting feedback in Federal Register notices, by meeting with community groups, financial institutions, trade associations, and other Federal agencies, and by convening a Small Business Review Panel. To prepare this final rule, the Bureau considered, among other things, the comments presented to the Board during its public hearings, feedback provided to the Bureau prior to the issuance of its proposal, including information provided during the Small Business Review Panel, interagency consultations, and feedback provided in response to the proposed rule. A. Pre-Proposal Outreach In 2010, the Board convened public hearings on potential revisions to Regulation C (the Board's 2010 Hearings).\38\ The Board began the reassessment of HMDA in the aftermath of the financial crisis, as Congress was considering the legislation that later became the Dodd- Frank Act. Participants addressed whether the Board should require reporting from additional types of institutions, whether certain types of institutions should be exempt from reporting, and whether any other changes should be made to the rules for determining which types of institutions must report data. For example, representatives from Federal agencies, lenders, and consumer advocates urged the Board to adopt a consistent minimum loan threshold across all types of institutions, including banks, savings associations, credit unions, and nondepository institutions.\39\ In particular, industry representatives noted the limited value derived from data reported by lower-volume depository institutions.\40\ Industry and community advocate representatives also asserted that loan volume, rather than asset size, should trigger reporting, particularly for nondepository lenders because they tend to have a different capital structure than banks, savings associations, and credit unions.\41\ Participants also urged the Board to expand coverage of nondepository institutions.\42\ In addition, participants commented that the coverage scheme for nondepository institutions was too complex and should be simplified.\43\ --------------------------------------------------------------------------- \38\ See 75 FR 35030 (June 21, 2010). \39\ Transcript of Fed. Reserve Board Public Hearing on Potential Revisions to the Home Mortgage Disclosure Act, Washington DC (Sept. 24, 2010) [hereinafter Washington Hearing], (remarks of Faith Schwartz, Senior Advisory, HOPE Now Alliance) (``I think everyone should have the burden of reporting that has any meaningful originations out there. * * *''), http://www.federalreserve.gov/communitydev/files/full_transcript_board_20100924.pdf ; id. (remarks of Josh Silver, Vice President of Research and Policy, National Community Reinvestment Coalition) (``[I]n terms of your threshold, it is very confusing because you have depository institutions that have different thresholds and nondepository institutions . . . I suggested just make it the same for everybody. If you make more than [50 reportable loans under HMDA], you disclose.. . . So that's a threshold I would propose across the board for nondepository institutions and depository institutions.''). \40\ See, e.g., Transcript of Fed. Reserve Board Public Hearing on Potential Revisions to the Home Mortgage Disclosure Act, Atlanta, Georgia (July 15, 2010) [hereinafter Atlanta Hearing], http://www.federalreserve.gov/communitydev/files/full_transcript_atlanta_20100715.pdf. \41\ See, e.g., id. (remarks of Faith Anderson, Vice President and General Counsel, American Airlines Federal Credit Union) (``[A]n exemption from HMDA reporting should be based on the volume of mortgage loans that are given. Exemptions should not be based on the asset size of a financial institution.''). \42\ See, e.g., Transcript of Fed. Reserve Board Public Hearing on Potential Revisions to the Home Mortgage Disclosure Act, San Francisco, California (Aug. 5, 2010) [hereinafter San Francisco Hearing], http://www.federalreserve.gov/communitydev/files/full_transcript_sf_20100805.pdf; Washington Hearing, supra note 39; Atlanta Hearing, supra note 40. \43\ See, e.g., Washington Hearing, supra note 39. --------------------------------------------------------------------------- The Board solicited feedback on ways to improve the quality and usefulness of HMDA data, including whether any data elements should be added, modified, or deleted. Participants provided [[Page 66135]] suggestions about ways to improve the utility of HMDA data. Participants discussed modifications to the data fields currently collected in Regulation C that may clarify reporting requirements and improve the usefulness of HMDA data. For example, participants urged the Board to augment the information collected concerning multifamily properties\44\ and manufactured housing \45\ and to expand the reporting of rate spread to all originations.\46\ Participants also urged the Board to clarify specific reporting requirements, such as how to report modular homes \47\ and conditional approvals.\48\ Participants discussed the reluctance of applicants to provide demographic information, such as race and ethnicity, and the challenges financial institutions face in collecting the information.\49\ --------------------------------------------------------------------------- \44\ See, e.g., San Francisco Hearing, supra note 42; Washington Hearing, supra note 39. \45\ See, e.g., id. \46\ See, e.g., Atlanta Hearing, supra note 40; Transcript of Fed. Reserve Board Public Hearing on Potential Revisions to the Home Mortgage Disclosure Act, Chicago, Illinois (Sept. 16, 2010) [hereinafter Chicago Hearing], http://www.federalreserve.gov/communitydev/files/full_transcript_chicago_20100916.pdf; id. (remarks of Professor Jim Campen, University of Massachusetts). \47\ See, e.g., Atlanta Hearing, supra note 40. \48\ See, e.g., Washington Hearing, supra note 39. \49\ See, e.g., Atlanta Hearing, supra note 40; San Francisco Hearing, supra note 42; Chicago Hearing, supra note 46. --------------------------------------------------------------------------- In addition, participants commented on data fields that could be added to the data collected under HMDA to improve its utility. For example, participants suggested collecting information regarding points and fees, including prepayment penalties,\50\ information concerning the relationship of the loan amount to the value of the property securing the loan,\51\ and information concerning whether an application was submitted through a mortgage broker.\52\ --------------------------------------------------------------------------- \50\ See, e.g., San Francisco Hearing, supra note 42; Chicago Hearing, supra note 46. \51\ See, e.g., Atlanta Hearing, supra note 40; San Francisco Hearing, supra note 42; Chicago Hearing, supra note 46; Washington Hearing, supra note 39. \52\ See, e.g., Chicago Hearing, supra note 46. --------------------------------------------------------------------------- In developing the proposal to amend Regulation C, the Bureau, through outreach and meetings with stakeholders, built on the feedback received during the Board's 2010 HMDA hearings. The Bureau conducted meetings in-person and through conference calls. In addition, the Bureau solicited feedback through correspondence and Federal Register notices.\53\ --------------------------------------------------------------------------- \53\ 76 FR 31222 (May 31, 2011); 76 FR 43570 (Jul. 21, 2011); 76 FR 75825 (Dec. 5, 2011); 76 FR 78465 (Dec. 19, 2011). --------------------------------------------------------------------------- In 2011, the Bureau issued a proposed rule seeking feedback on regulations inherited from other agencies (2011 Streamlining Proposal).\54\ While the Bureau sought general feedback on opportunities to streamline inherited regulations, the Bureau also solicited specific feedback on whether a small number of refinancings should not trigger Regulation C coverage.\55\ The Bureau received comments from consumer advocates, fair housing advocates, financial institutions, State bank supervisory organizations, and national industry trade associations. Comments addressed issues ranging from reporting thresholds and data reporting exemptions to clarifying certain definitions and reporting issues.\56\ --------------------------------------------------------------------------- \54\ 76 FR 75825 (Dec. 5, 2011). \55\ The Bureau noted in the 2011 Streamlining Proposal that a depository institution that did not ordinarily originate home purchase loans, but that occasionally refinanced a home purchase loan to accommodate a customer, would be required to report under Regulation C. 76 FR 75825, 75828 (Dec. 5, 2011). \56\ The Bureau's 2014 HMDA proposal provides a more detailed description of the comments received. See 79 FR 51731, 51744 (Aug. 29, 2014). --------------------------------------------------------------------------- On December 19, 2011, the Bureau published an interim final rule establishing Regulation C in 12 CFR part 1003, implementing the assumption of HMDA authority from the Board (the Bureau's 2011 Regulation C Restatement).\57\ The Bureau's 2011 Regulation C Restatement substantially duplicated the Board's Regulation C and made only non-substantive, technical, formatting, and stylistic changes. As part of the Bureau's 2011 Regulation C Restatement, the Bureau solicited comment on any outdated, unduly burdensome, or unnecessary technical issues and provisions.\58\ Commenters generally suggested aligning Regulation C definitions with other regulations, providing a tolerance for enforcement actions based on low error rates, and establishing a loan-volume threshold. Commenters also raised other issues, some of which the Bureau discussed in the proposal and which are also discussed in the section-by-section analysis below. --------------------------------------------------------------------------- \57\ 76 FR 78465 (Dec. 19, 2011). \58\ Id. --------------------------------------------------------------------------- The Bureau met with a few groups to better understand existing and emerging data standards and whether Regulation C could be aligned with those standards. The Bureau met with staff from Mortgage Industry Standards Maintenance Organization (MISMO) \59\ and the GSEs \60\ regarding the MISMO dataset and the ULDD \61\, respectively. The Bureau also met with community, regional, and national banks to understand their HMDA compliance processes and obtain feedback on areas for improvement, and with consumer and fair housing advocates as well as industry trade associations to understand their concerns with the HMDA data and Regulation C. --------------------------------------------------------------------------- \59\ MISMO is the federally registered service mark of the Mortgage Industry Standards Maintenance Organization, a wholly-owned subsidiary of the Mortgage Bankers Association. \60\ Government-sponsored enterprises, specifically Federal National Mortgage Association (Fannie Mae) and Federal Home Loan Mortgage Corporation (Freddie Mac). \61\ The Uniform Loan Delivery Dataset is a common set of data elements required by Fannie Mae and Freddie Mac. --------------------------------------------------------------------------- B. Small Business Review Panel In February 2014, the Bureau convened a Small Business Review Panel (Panel) with the Chief Counsel for Advocacy of the Small Business Administration (SBA) and the Administrator of the Office of Information and Regulatory Affairs with the Office of Management and Budget (OMB).\62\ As part of this process, the Bureau prepared an outline of the proposals then under consideration and the alternatives considered (Small Business Review Panel Outline), which the Bureau posted on its Web site for review by the small financial institutions participating in the panel process, as well as the general public.\63\ --------------------------------------------------------------------------- \62\ The Small Business Regulatory Enforcement Fairness Act of 1996 (SBREFA), as amended by section 1100G(a) of the Dodd-Frank Act, requires the Bureau to convene a Small Business Review Panel before proposing a rule that may have significant economic impact on a substantial number of small entities. See Public Law 104-121, tit. II, 110 Stat. 847, 857 (1996) as amended by Public Law 110-28, and Public Law 111-203, section 1100G (2010). \63\ Press Release, CFPB Takes Steps to Improve Information About Access to Credit in the Mortgage Market (Feb. 7, 2014), http://www.consumerfinance.gov/newsroom/cfpb-takes-steps-to-improve-information-about-access-to-credit-in-the-mortgage-market/. The Bureau also gathered feedback on the Small Business Review Panel Outline from other stakeholders and members of the public, and from the Bureau's Consumer Advisory Board and Community Bank Advisory Council. --------------------------------------------------------------------------- Prior to formally convening, the Panel participated in teleconferences with small groups of the small entity representatives to introduce the materials and to obtain feedback. The Panel conducted a full-day outreach meeting with the small entity representatives in March 2014 in Washington, DC. The Panel gathered information from the small entity representatives and made findings and recommendations regarding the potential compliance costs and other impacts of the proposed rule on those entities. Those findings and [[Page 66136]] recommendations are set forth in the Panel's report (Small Business Review Panel Report), which will be made part of the administrative record in this rulemaking.\64\ The Bureau carefully considered the findings and recommendations in preparing the proposal and this final rule. --------------------------------------------------------------------------- \64\ See Final Report of the Small Business Review Panel on the CFPB's Proposals Under Consideration for the Home Mortgage Disclosure Act (HMDA) Rulemaking (April 24, 2014), http://files.consumerfinance.gov/f/201407_cfpb_report_hmda_sbrefa.pdf. --------------------------------------------------------------------------- C. The Bureau's Proposal In July 2014, the Bureau published on its Web site for public comment a proposed rule regarding Regulation C to implement section 1094 of the Dodd-Frank Act, which amended HMDA to improve the utility of the HMDA data and revise Federal agency rulemaking and enforcement authorities. The proposal was published in the Federal Register in August 2014.\65\ The Bureau proposed modifications to the institutional coverage and transactional coverage in light of market conditions, to reduce burden on financial institutions, and to address gaps in the HMDA data regarding certain segments of the housing market. The proposed modification to institutional coverage would have simplified the coverage criteria for depository and nondepository institutions with a uniform threshold of 25 loans. Under the proposal, depository and nondepository institutions that originated 25 covered loans, excluding open-end lines of credit, in the previous calendar year would be required to report HMDA data so long as all the other reporting criteria were met. The proposed modification to transactional coverage would have expanded the types of transactions subject to Regulation C. Under the proposal, financial institutions would be required to report all closed-end loans, open-end lines of credit, and reverse mortgages secured by dwellings, which would have relieved financial institutions from the requirement to ascertain an applicant's intended purpose for a dwelling-secured loan to determine if the loan was reportable under HMDA. --------------------------------------------------------------------------- \65\ 79 FR 51731 (Aug. 29, 2014). See part II.A for a discussion of section 1094 of the Dodd-Frank Act. --------------------------------------------------------------------------- The Bureau also proposed modifications to reportable data requirements. First, the Bureau proposed to align many HMDA data requirements with the MISMO data standards for residential mortgages. Second, the Bureau proposed to modify existing data points already established under Regulation C as well as add new data points to the reporting requirements. Some of these data points were specifically identified by the Dodd-Frank Act and others were proposed pursuant to the Bureau's discretionary rulemaking authority to carry out the purposes of HMDA by addressing data gaps. The following four categories of new or modified data points were proposed by the Bureau: Information about applicants, borrowers, and the underwriting process, such as age, credit score, debt-to-income ratio, reasons for denial if the application was denied, the application channel, and automated underwriting system results. Information about the property securing the loan, such as construction method, property value, lien priority, the number of individual dwelling units in the property, and additional information about manufactured and multifamily housing. Information about the features of the loan, such as additional pricing information, loan term, interest rate, introductory rate period, non-amortizing features, and the type of loan. Certain unique identifiers, such as a universal loan identifier, property address, loan originator identifier, and a legal entity identifier for the financial institution. In addition, the Bureau proposed modifications to the disclosure and reporting requirements and clarifications to the regulation. Under the proposal, financial institutions that report large volumes of HMDA data would be required to submit their data to the appropriate agency on a quarterly basis rather than an annual basis. The Bureau noted its belief that quarterly reporting would reduce reporting errors and improve the quality of HMDA data, allow regulators to use the data in a more timely and effective manner, and could facilitate an earlier release of annual HMDA data to the public. The Bureau also proposed to allow HMDA reporters to make their disclosure statements available by referring members of the public that request a disclosure statement to a publicly available Web site, which would facilitate public access to the HMDA data and minimize the burden on HMDA reporters. The Bureau also proposed clarifications to Regulation C to address issues that are unclear or confusing. These proposed clarifications included guidance on types of residential structures that are considered dwellings; the treatment of manufactured and modular homes and multiple properties; preapproval programs and temporary financing; how to report a transaction that involved multiple financial institutions; reporting the action taken on an application; and reporting the type of purchaser for a covered loan. D. Feedback Provided to the Bureau The Bureau received approximately 400 comments on the HMDA proposal during the comment period from, among others, consumer advocacy groups; national, State, and regional industry trade associations; banks, community banks, credit unions, software providers, housing counselors; Federal agencies, including the Office of Advocacy of the Small Business Administration (SBA); and individual consumers and academics. In addition, the Bureau also considered other information, including ex parte communications.\66\ Materials on the record are publicly available at http://www.regulations.gov. This information is discussed below in the section-by-section analysis and subsequent parts of the notice, as applicable. The Bureau considered the comments and ex parte communications, modified the proposal in certain respects, and adopts the final rule as described below in the section-by-section analysis. --------------------------------------------------------------------------- \66\ CFPB Bulletin 11-3, CFPB Policy on Ex Parte Presentations in Rulemaking Proceedings (2011), available at http://files.consumerfinance.gov/f/2011/08/Bulletin_20110819_ExPartePresentationsRulemakingProceedings.pdf. --------------------------------------------------------------------------- IV. Legal Authority The Bureau is issuing this final rule pursuant to its authority under the Dodd-Frank Act and HMDA. Section 1061 of the Dodd-Frank Act transferred to the Bureau the ``consumer financial protection functions'' previously vested in certain other Federal agencies, including the Board.\67\ The term ``consumer financial protection function'' is defined to include ``all authority to prescribe rules or issue orders or guidelines pursuant to any Federal consumer financial law, including performing appropriate functions to promulgate and review such rules, orders, and guidelines.'' \68\ Section 1022(b)(1) of the Dodd-Frank Act authorizes the Bureau's Director to prescribe rules ``as may be necessary or appropriate to enable the Bureau to administer and carry out the purposes and objectives of the Federal consumer financial laws, and to prevent evasions thereof.'' \69\ Both HMDA and title X of the Dodd-Frank Act are Federal [[Page 66137]] consumer financial laws.\70\ Accordingly, the Bureau has authority to issue regulations to administer HMDA. --------------------------------------------------------------------------- \67\ 12 U.S.C. 5581. Section 1094 of the Dodd-Frank Act also replaced the term ``Board'' with ``Bureau'' in most places in HMDA. 12 U.S.C. 2803 et seq. \68\ 12 U.S.C. 5581(a)(1)(A). \69\ 12 U.S.C. 5512(b)(1). \70\ Dodd-Frank Act section 1002(14), 12 U.S.C. 5481(14) (defining ``Federal consumer financial law'' to include the ``enumerated consumer laws'' and the provisions of title X of the Dodd-Frank Act); Dodd-Frank Act section 1002(12), 12 U.S.C. 5481(12) (defining ``enumerated consumer laws'' to include HMDA). --------------------------------------------------------------------------- HMDA section 305(a) broadly authorizes the Bureau to prescribe such regulations as may be necessary to carry out HMDA's purposes.\71\ These regulations can include ``classifications, differentiations, or other provisions, and may provide for such adjustments and exceptions for any class of transactions, as in the judgment of the Bureau are necessary and proper to effectuate the purposes of [HMDA], and prevent circumvention or evasion thereof, or to facilitate compliance therewith.'' \72\ --------------------------------------------------------------------------- \71\ 12 U.S.C. 2804(a). \72\ Id. --------------------------------------------------------------------------- A number of HMDA provisions specify that covered institutions must compile and make their HMDA data publicly available ``in accordance with regulations of the Bureau'' and ``in such formats as the Bureau may require.'' \73\ HMDA section 304(j)(1) authorizes the Bureau to issue regulations to define the loan application register information that HMDA reporters must make available to the public upon request and to specify the form required for such disclosures.\74\ HMDA section 304(j)(2)(B) provides that ``[t]he Bureau shall require, by regulation, such deletions as the Bureau may determine to be appropriate to protect--(i) any privacy interest of any applicant . . .; and (ii) a depository institution from liability under any Federal or State privacy law.'' \75\ HMDA section 304(j)(7) also directs the Bureau to make every effort in prescribing regulations under the subsection to minimize the costs incurred by a depository institution in complying with the subsection and regulations.\76\ --------------------------------------------------------------------------- \73\ See, e.g., HMDA section 304(a)(1), (j)(2)(A), (j)(3), (m)(2), 12 U.S.C. 2803(a)(1), (j)(2)(A), (j)(3), (m)(2); see also HMDA section 304(b)(6)(I), 12 U.S.C. 2803(b)(6)(I) (requiring covered institutions to use ``such form as the Bureau may prescribe'' in reporting credit scores of mortgage applicants and mortgagors). HMDA section 304(k)(1) also requires depository institutions covered by HMDA to make disclosure statements available ``[i]n accordance with procedures established by the Bureau pursuant to this section.'' 12 U.S.C. 2803(k)(1). \74\ 12 U.S.C. 2803(j)(1). \75\ 12 U.S.C. 2803(j)(2)(B). \76\ 12 U.S.C. 2803(j)(7). --------------------------------------------------------------------------- HMDA section 304(e) directs the Bureau to prescribe a standard format for HMDA disclosures required under HMDA section 304.\77\ As amended by the Dodd-Frank Act, HMDA section 304(h)(1) requires HMDA data to be submitted to the Bureau or to the appropriate agency for the reporting financial institution ``in accordance with rules prescribed by the Bureau.'' \78\ HMDA section 304(h)(1) also directs the Bureau, in consultation with other appropriate agencies, to develop regulations after notice and comment that: --------------------------------------------------------------------------- \77\ 12 U.S.C. 2803(e). \78\ 12 U.S.C. 2803(h)(1); see also HMDA section 304(n), 12 U.S.C. 2803(n) (discussing submission to the Bureau or the appropriate agency ``in accordance with regulations prescribed by the Bureau''). For purposes of HMDA section 304(h), HMDA section 304(h)(2) defines the appropriate agencies for different categories of financial institutions. The agencies are the Federal banking agencies, the FDIC, the NCUA, and the Secretary of HUD. 12 U.S.C. 2803(h)(2). --------------------------------------------------------------------------- (A) Prescribe the format for such disclosures, the method for submission of the data to the appropriate agency, and the procedures for disclosing the information to the public; (B) require the collection of data required to be disclosed under [HMDA section 304(b)] with respect to loans sold by each institution reporting under this title; (C) require disclosure of the class of the purchaser of such loans; (D) permit any reporting institution to submit in writing to the Bureau or to the appropriate agency such additional data or explanations as it deems relevant to the decision to originate or purchase mortgage loans; and (E) modify or require modification of itemized information, for the purpose of protecting the privacy interests of the mortgage applicants or mortgagors, that is or will be available to the public.\79\ --------------------------------------------------------------------------- \79\ 12 U.S.C. 2803(h)(1). The Dodd-Frank Act also added new HMDA section 304(h)(3), which directs the Bureau to prescribe standards for any modification pursuant to HMDA section 304(h)(1)(E), to effectuate HMDA's purposes, in light of the privacy interests of mortgage applicants or mortgagors. 12 U.S.C. 2803(h)(1)(E), 2803(h)(3). HMDA also authorizes the Bureau to issue regulations relating to the timing of HMDA disclosures.\80\ --------------------------------------------------------------------------- \80\ HMDA section 304(l)(2)(A), 12 U.S.C. 2803(l)(2)(A) (setting maximum disclosure periods except as provided under other HMDA subsections and regulations prescribed by the Bureau); HMDA section 304(n), 12 U.S.C. 2803(n). --------------------------------------------------------------------------- As amended by the Dodd-Frank Act, HMDA section 304 requires itemization of specified categories of information and ``such other information as the Bureau may require.'' \81\ Specifically, HMDA section 304(b)(5)(D) requires reporting of ``such other information as the Bureau may require'' for mortgage loans, and section 304(b)(6)(J) requires reporting of ``such other information as the Bureau may require'' for mortgage loans and applications. HMDA section 304 also identifies certain data points that are to be included in the itemization ``as the Bureau may determine to be appropriate.'' \82\ It provides that age and other categories of data shall be modified prior to release ``as the Bureau determines to be necessary'' to satisfy the statutory purpose of protecting the privacy interests of the mortgage applicants or mortgagors.\83\ --------------------------------------------------------------------------- \81\ HMDA section 304(b)(5)(D), (b)(6)(J), 12 U.S.C. 2803(b)(5)(D), (b)(6)(J). \82\ HMDA section 304(b)(6)(F), (G), (H), 12 U.S.C. 2803(b)(6)(F), (G), (H). \83\ HMDA section 304(h)(3)(A)(ii), 12 U.S.C. 2803(h)(3)(A)(ii). --------------------------------------------------------------------------- The Dodd-Frank Act amendments to HMDA also authorize the Bureau's Director to develop or assist in the improvement of methods of matching addresses and census tracts to facilitate HMDA compliance by depository institutions in as economical a manner as possible.\84\ The Bureau, in consultation with the Secretary of HUD, may also exempt for-profit mortgage-lending institutions that are comparable within their respective industries to a bank, savings association, or credit union that has total assets of $10,000,000 or less.\85\ --------------------------------------------------------------------------- \84\ HMDA section 307(a), 12 U.S.C. 2806(a) (authorizing the Bureau's Director to utilize, contract with, act through, or compensate any person or agency to carry out this subsection). \85\ HMDA section 309(a), 12 U.S.C. 2808(a). --------------------------------------------------------------------------- In preparing this final rule, the Bureau has considered the changes below in light of its legal authority under HMDA and the Dodd-Frank Act. The Bureau has determined that each of the changes addressed below is consistent with the purposes of HMDA and is authorized by one or more of the sources of statutory authority identified in this part. V. Section-by-Section Analysis Section 1003.1 Authority, Purpose, and Scope 1(c) Scope As summarized in part I, the Bureau proposed to revise the provisions of Regulation C that determine which financial institutions and transactions are covered by the regulation. The Bureau also proposed to reorganize the regulation to reduce burden. The Bureau proposed to revise Sec. 1003.1(c) and its accompanying commentary to reflect both the proposed substantive changes to Regulation C's institutional and transactional coverage and the proposed reorganization of the regulation. The Bureau did not receive any comments addressing proposed Sec. 1003.1(c).\86\ As [[Page 66138]] discussed in the section-by-section analyses of Sec. 1003.2(d), (e), (g), and (o) and of Sec. 1003.3, the final rule in some cases revises the Bureau's proposed changes to institutional and transactional coverage. However, none of those changes affect the technical revisions that the Bureau proposed for Sec. 1003.1(c). The Bureau thus is finalizing Sec. 1003.1(c) largely as proposed, with several non- substantive revisions for clarity. --------------------------------------------------------------------------- \86\ The Bureau received a large number of comments about the proposed revisions to Regulation C's transactional and institutional coverage. Those comments are addressed in the section-by-section analyses of Sec. 1003.2(d), (e), (g), (o) and of Sec. 1003.3(c)(10). --------------------------------------------------------------------------- Section 1003.2 Definitions Section 1003.2 of Regulation C sets forth definitions that are used in the regulation. As discussed below, the Bureau proposed both substantive revisions to several definitions and technical revisions to Sec. 1003.2 to enumerate the terms defined therein. The Bureau addresses comments concerning its proposed substantive revisions below. The Bureau received no comments opposing its proposal to enumerate the terms in Sec. 1003.2, and the final rule sets forth enumerations for all such terms. The Bureau believes that this technical revision will facilitate compliance with Regulation C by making defined terms easier to locate and cross-reference in the regulation, commentary, and the procedures published by the Bureau. 2(a) Act Section 1003.2 of Regulation C sets forth a definition for the term ``act.'' The Bureau is adopting a technical amendment to add a paragraph designation for this definition. No substantive change is intended. 2(b) Application 2(b)(1) In General Section 1003.2 currently defines an application as an oral or written request for a home purchase loan, a home improvement loan, or a refinancing that is made in accordance with the procedures used by a financial institution for the type of credit requested. The Bureau proposed to make technical corrections and minor wording changes to conform the definition of application to the proposed changes in transactional coverage. In addition, the Bureau proposed to make technical and minor wording changes to the applicable commentary. For the reasons discussed below, the Bureau is adopting Sec. 1003.2(b)(1) and the associated commentary as proposed. Commenters generally addressed aspects of the definition of application that differ from other regulations or challenges in applying the definition in multifamily and commercial lending. The Bureau received several comments urging that the Regulation C definition of application should be aligned with the definition used in Regulation Z Sec. 1026.2(a)(3)(ii) to simplify compliance across regulations. As the Bureau noted in the proposed rule, the Bureau did not propose to align the definitions because they serve different purposes.\87\ The definition of application in Regulation Z Sec. 1026.2(a)(3)(ii) establishes a clear rule for triggering when disclosures must be provided. In contrast, the definition for Regulation C is closely related to Regulation B and serves HMDA's fair lending purposes by requiring information about the disposition of credit requests received by financial institutions that do not lead to originations.\88\ Therefore it is important for the Regulation C definition of application to be based on the procedures used by the financial institution for the type of credit requested rather than the defined elements of the definition in Regulation Z Sec. 1026.2(a)(3)(ii) under which creditors may be sequencing and structuring their information collection processes in various different ways.\89\ --------------------------------------------------------------------------- \87\ 79 FR 51731, 51746 (Aug. 29, 2014). \88\ 12 CFR 1003.2, comment Application-1; 12 CFR 1002.2(f). \89\ 78 FR 79730, 79767 (Dec. 31, 2013). --------------------------------------------------------------------------- Some comments argued that the definition of application would be difficult to comply with for multifamily loans, which generally involve a more fluid application process. They also argued that the Bureau should exclude ``pitch book requests'' from the definition of application. Pitch book requests are preliminary investment packages related to multifamily residential structures requesting specific loans terms. The Bureau has considered the comments but believes that changes to the proposed definition of application related to multifamily loans are not warranted. Because the definition of application in Regulation C is closely related to the Regulation B definition of application and Regulation B applies to business credit, including multifamily lending,\90\ the Bureau believes that the flexible definition of application as proposed and the commentary in Regulation B and Regulation C provide adequate guidance for multifamily lending. The Bureau is also concerned that an exception for pitch book requests may be difficult to adopt because financial institutions may have different definitions of pitch book request or procedures for handlings them. The Bureau is not adopting an exclusion specific to pitch book requests, and believes that the existing commentary regarding the definition of application and prequalifications is appropriate.\91\ Whether pitch book requests would be considered applications under Regulation C would depend on how the specific financial institution treated such requests under its application process for covered loans secured by multifamily residential structures under the definition of application in Regulation C. As discussed below, the Bureau is also excluding covered loans secured by multifamily dwellings from the definition of a preapproval program, which may address some of the commenters' concerns. After considering the comments, the Bureau is finalizing Sec. 1003.2(b)(1) and comments 2(b)-1 and 2(b)-2 as proposed. --------------------------------------------------------------------------- \90\ 12 CFR 1002.2(j), comment 2(j)-1. \91\ See existing comment Application-2, final comment 2(b)-2. --------------------------------------------------------------------------- 2(b)(2) Preapproval Programs Regulation C incorporates certain requests under preapproval programs into the definition of application under Sec. 1003.2. Such programs are only covered if they involve a comprehensive analysis of the creditworthiness of the applicant and include a written commitment for up to a specific amount, subject only to certain limited conditions. The Bureau proposed to make technical and clarifying wording changes to the definition of a preapproval program under Sec. 1003.2(b)(2) and the applicable commentary to add language adapted from additional FAQs regarding preapproval programs that had been provided by the FFIEC.\92\ For the reasons discussed below, the Bureau is finalizing Sec. 1003.2(b)(2) with modifications to exclude certain types of covered loans from the definition. --------------------------------------------------------------------------- \92\ 79 FR 51731, 51747 (Aug. 29, 2014). --------------------------------------------------------------------------- Several commenters addressed the Bureau's proposed definition of preapproval programs. Some commenters questioned the value of preapproval reporting or argued that preapproval reporting discourages financial institutions from offering preapproval programs. However, the Bureau is not excluding preapproval requests from Regulation C in this final rule because this information is valuable for fair lending purposes, as it provides visibility into how applicants are treated in an early stage of the lending [[Page 66139]] process.\93\ The statute requires lenders to report action taken on applications,\94\ and the Bureau believes that requests for preapproval as defined in the proposal and final rule represent credit applications. The Bureau does not believe that Regulation C's coverage of preapproval programs has discouraged offering of preapproval programs, and it concludes that any discouragement would be justified by the benefits of reporting. The reporting requirement is limited only to preapproval programs that meet certain conditions. Additionally, the Bureau is finalizing changes to comment 2(b)-3 that specify that programs described as preapproval programs that do not meet the definition in Sec. 1003.2(b)(2) are not preapproval programs for purposes of HMDA reporting. --------------------------------------------------------------------------- \93\ 67 FR 7222, 7224 (Feb. 15, 2002); 79 FR 51731, 51747 (Aug. 29, 2014). \94\ HMDA section 303(4). --------------------------------------------------------------------------- Some commenters requested clarification about occasional preapprovals and some argued for a broader and more flexible definition of preapproval programs. The Bureau is not adopting a broader or more flexible definition of preapproval programs because it believes that limiting the scope of the definition allows for comparison of similar programs across institutions, where a broader definition could expand reportable transactions, lead to new compliance issues, and make preapproval data less comparable across institutions. The Bureau continues to believe that a financial institution that does not have a preapproval program and only occasionally considers preapproval requests on an ad hoc basis need not report those transactions and believes that proposed comment 2(b)-3 addresses the commenters' concerns. It provides, in part, that a financial institution need not treat ad hoc requests as part of a preapproval program for purposes of Regulation C. The Bureau is therefore finalizing comment 2(b)-3 as proposed. After considering the comments and conducting additional analysis, the Bureau is finalizing Sec. 1003.2(b)(2) generally as proposed, with minor revisions to exclude home purchase loans that will be open-end lines of credit, reverse mortgages, or secured by multifamily dwellings. Some loans secured by multifamily dwellings have been previously reported in HMDA under preapproval programs. The definition of a home purchase loan could include these types of loans. The definition of preapproval programs in current Regulation C and adopted by the final rule is primarily focused on programs associated with closed-end home purchase loans for one- to four-unit dwellings. The Bureau believes it is appropriate to categorically exclude loans secured by multifamily dwellings, open-end lines of credit, and reverse mortgages from the definition of preapproval programs in order to facilitate consistent reporting and analysis of preapprovals by limiting the definition to closed-end home purchase loans for one- to four-unit dwellings. 2(c) Branch Office Section 1003.2 currently provides a definition of branch office, which includes separate definitions for branches of (1) banks, savings associations, and credit unions and (2) for-profit mortgage-lending institutions (other than banks, savings associations, and credit unions). The Bureau proposed technical and nonsubstantive modifications to the definition of branch office. The Bureau received no comments on proposed Sec. 1003.2(c) or proposed comments 2(c)-2 and -3. The Bureau is adopting Sec. 1003.2(c) and comments 2(c)-2 and -3, renumbered as comment 2(c)(1)-2 and comment 2(c)(2)-1, with technical modifications. The Bureau is also republishing comment (Branch Office)-1, renumbered as comment 2(c)(1)-1. 2(d) Closed-End Mortgage Loan Under existing Regulation C, financial institutions must report information about applications for, and originations of, closed-end loans made for one of three purposes: Home improvement, home purchase, or refinancing.\95\ Closed-end home purchase loans and refinancings must be reported if they are dwelling-secured.\96\ Closed-end home improvement loans must be reported whether or not they are dwelling- secured. --------------------------------------------------------------------------- \95\ Reverse mortgages currently are subject to these same criteria for reporting; thus, a closed-end reverse mortgage currently must be reported if it is for one of Regulation C's three purposes. \96\ Regulation C defines ``dwelling'' broadly to include single-family homes, rental properties, multifamily residential structures (e.g., apartment buildings), manufactured homes, and vacation homes. See the section-by-section analysis of Sec. 1003.2(f) and related commentary. --------------------------------------------------------------------------- As discussed in the section-by-section analysis of Sec. 1003.2(e) (``covered loan''), the Bureau proposed to adjust Regulation C's transactional coverage to require financial institutions to report all dwelling-secured loans (and applications), instead of reporting only those loans and applications for the purpose of home improvement, home purchase, or refinancing.\97\ To facilitate this shift in transactional coverage, the Bureau proposed to define the term ``closed-end mortgage loan'' in Regulation C. Proposed Sec. 1003.2(d) provided that a closed-end mortgage loan was a dwelling-secured debt obligation that was not an open-end line of credit under Sec. 1003.2(o), a reverse mortgage under Sec. 1003.2(q), or an excluded transaction under Sec. 1003.3(c). The Bureau did not propose commentary to accompany proposed Sec. 1003.2(d) but solicited feedback about whether commentary would be helpful. --------------------------------------------------------------------------- \97\ As discussed in the section-by-section analysis of Sec. 1003.2(o) and (q), the proposal applied the same dwelling-secured test to open-end lines of credit and reverse mortgages, the two other categories of ``covered loans'' in proposed Sec. 1003.2(e). --------------------------------------------------------------------------- The proposal to remove Regulation C's current purpose-based reporting approach for closed-end mortgage loans in some cases broadened, and in some cases limited, the closed-end loans that would be reported under the regulation. For example, the proposal provided for reporting of all closed-end home-equity loans and all closed-end, dwelling-secured commercial-purpose loans. At the same time, the proposal eliminated the requirement to report home improvement loans not secured by a dwelling. As discussed in the section-by-section analysis of Sec. 1003.2(e), the Bureau is finalizing the proposed shift to dwelling-secured transactional coverage for consumer-purpose transactions and is retaining Regulation C's traditional purpose test for commercial- purpose transactions. The Bureau believes that the shift serves HMDA's purposes, will improve HMDA data, and will simplify transactional reporting requirements. Accordingly, the Bureau is finalizing Sec. 1003.2(d) largely as proposed, but with technical revisions for clarity, to define the universe of closed-end mortgage loans that must be reported under Regulation C unless otherwise excluded under Sec. 1003.3(c). The Bureau also is finalizing commentary to Sec. 1003.2(d) to address questions that commenters raised about the scope of the closed-end mortgage loan definition. Relatively few commenters specifically addressed the benefits and burdens of reporting all dwelling-secured, consumer-purpose, closed-end mortgage loans.\98\ Consumer advocacy [[Page 66140]] group commenters supported the proposal to cover all such loans, and industry stakeholders expressed mixed views. A number of consumer advocacy group commenters also requested that the Bureau clarify in the final rule whether particular categories of transactions are included under the closed-end mortgage loan definition. --------------------------------------------------------------------------- \98\ As discussed in the section-by-section analysis of Sec. 1003.2(e), nearly all commenters addressed in some fashion the Bureau's proposal to shift Regulation C's transactional coverage test from a purpose-based test to a collateral-based test. However, most commenters focused either on the benefits and burdens of the shift overall, or on the specific benefits and burdens of reporting all open-end lines of credit, all reverse mortgages, or all dwelling-secured, commercial-purpose mortgage loans and lines of credit. Those comments are addressed in the section-by-section analyses of Sec. 1003.2(e), (o), (q), and Sec. 1003.3(c)(10), respectively. The section-by-section analysis of Sec. 1003.2(d) focuses on the comments that specifically addressed the proposal to cover all consumer-purpose, closed-end home-equity loans. --------------------------------------------------------------------------- Coverage of Dwelling-Secured, Consumer-Purpose, Closed-End Mortgage Loans A large number of consumer advocacy group and community development commenters supported having information about all closed-end home- equity loans. They stated that having information about all such loans would be valuable in assessing whether neighborhoods that the consumer groups serve, especially those that are low- and moderate-income, are receiving the full range of credit that they need and would be appropriate to ensure an adequate understanding of the mortgage market. A small group of industry commenters supported the proposed shift to dwelling-secured coverage to the extent that it meant reporting all dwelling-secured, closed-end, consumer-purpose loans. Some of these commenters argued that reporting all such loans would be less burdensome than discerning whether each loan was for a reportable purpose.\99\ Others asserted that dwelling-secured coverage would eliminate the possibility that exists under current Regulation C of erroneously gathering race, gender, and ethnicity data for consumer- purpose loans that later are determined not to be reportable. One industry commenter supported dwelling-secured coverage only for closed- end, consumer-purpose loans secured by one- to four-unit dwellings, arguing that these transactions are the most common, are similar in their underwriting and in their risks to consumers, and have hit the economy hardest when they default en masse. Other industry commenters agreed that the shift to dwelling-secured coverage for closed-end, consumer-purpose loans was appropriate and would serve HMDA's purposes, would simplify reporting, would improve data for HMDA users, and would better align Regulation C's coverage with Regulations X and Z.\100\ --------------------------------------------------------------------------- \99\ One commenter provided a specific example. The commenter stated that, when a borrower owns a home outright but takes out a dwelling-secured debt consolidation loan, the loan is recorded as a refinancing in the lender's loan origination system and on the GSE's standard loan application form. However, the loan currently is not reported under Regulation C because it does not meet the purpose- based test. Therefore, an employee later must remove the transaction manually from the institution's HMDA report. If all dwelling- secured, consumer-purpose, closed-end loans were covered, the transaction would be reported and the extra, manual step of removing the transaction would be unnecessary. \100\ Regulation X implements the Real Estate Settlement Procedures Act (RESPA), 12 U.S.C. 2601 et seq. Regulation Z implements the Truth in Lending Act (TILA), 15 U.S.C. 1601 et seq. --------------------------------------------------------------------------- As discussed in the section-by-section analysis of Sec. 1003.2(e), a majority of industry commenters opposed the proposed shift to dwelling-secured coverage, and some of those commenters specifically objected to reporting data about all closed-end home-equity loans. Some argued that the Bureau should maintain current coverage; a few argued that closed-end home-equity loans should be excluded from coverage altogether. The commenters argued that funds obtained through home- equity loans could be used for any purpose. If a transaction's funds were not used for home purchase, home improvement, or refinancing purposes, commenters asserted, then having data about that transaction would not serve HMDA's purpose of ensuring that financial institutions are meeting the housing needs of their communities. One commenter argued that concerns about home-equity lending's role in the financial crisis no longer justified covering all home-equity loans, because the Bureau's ability-to-repay and qualified mortgage rules have addressed any issues with such lending.\101\ A few commenters also objected that such reporting would increase loan volume or argued that compiling data about all closed-end home-equity loans would be onerous, would require costly systems upgrades, or would distort HMDA data because loans would be reported even if their funds were not used for housing-related purposes. --------------------------------------------------------------------------- \101\ See the Bureau's Ability-to-Repay and Qualified Mortgage Standards rule (2013 ATR Final Rule), 78 FR 6408 (Jan. 30, 2013). --------------------------------------------------------------------------- As discussed in the proposal, the Bureau believes that covering all dwelling-secured, consumer-purpose, closed-end mortgage loans will provide useful data that will serve HMDA's purposes by providing additional information about closed-end home-equity loans, which research indicates were a significant factor leading up to the financial crisis,\102\ and which impeded some borrowers' ability to receive assistance through foreclosure relief programs during and after the crisis.\103\ The Bureau also believes, as some industry commenters observed, that covering all such transactions will simplify the regulation and ease compliance burden. The Bureau thus is adopting proposed Sec. 1003.2(d) largely as proposed, but with several revisions for clarity, as discussed below. --------------------------------------------------------------------------- \102\ See 79 FR 51731, 51747-48 (Aug. 29, 2014) (citing Atif Mian & Amir Sufi, House Prices, Home Equity-Based Borrowing, and the U.S. Household Leverage Crisis, 101 Am. Econ. Rev. 2132, 2154 (Aug. 2011) and Donghoon Lee et al., Fed. Reserve Bank of New York, Staff Report No. 569, A New Look at Second Liens, at 11 (Aug. 2012)). \103\ See id. (citing Vicki Been et al., Furman Ctr. for Real Estate & Urban Policy, Essay: Sticky Seconds--The Problems Second Liens Pose to the Resolution of Distressed Mortgages, at 13-18 (Aug. 2012)). --------------------------------------------------------------------------- Clarifications to the Closed-End Mortgage Loan Definition General. The Bureau is making two clarifying changes to Sec. 1003.2(d) and is adding comment 2(d)-1 to provide general guidance about the definition of closed-end mortgage loan. First, proposed Sec. 1003.2(d) provided that a closed-end mortgage loan was a dwelling- secured debt obligation that was not an open-end line of credit under Sec. 1003.2(o), a reverse mortgage under Sec. 1003.2(q), or an excluded transaction under Sec. 1003.3(c). To align with lending practices, to streamline the definitions of closed-end mortgage loan and open-end line of credit, and to streamline the reverse mortgage flag in final Sec. 1003.4(a)(36), the final rule eliminates the mutual exclusivity between closed-end mortgage loans and reverse mortgages.\104\ Second, the final rule eliminates the proposed language that provided that an excluded transaction under Sec. 1003.3(c) was not a closed-end mortgage loan. The Bureau is making this change to avoid circularity with final Sec. 1003.3(c), which incorporates for clarity the defined terms ``closed-end mortgage loan'' and ``open-end line of credit'' into the descriptions of excluded transactions. Final Sec. 1003.2(d) thus provides that a closed-end mortgage loan is a dwelling-secured extension of credit that is not an open-end line of credit under Sec. 1003.2(o). Comment 2(d)-1 provides an example of a loan that is not a closed-end mortgage loan because it is not dwelling- secured. --------------------------------------------------------------------------- \104\ As discussed in the section-by-section analysis of Sec. 1003.2(q), under the final rule a reverse mortgage thus may be either a closed-end mortgage loan or an open-end line of credit, as appropriate. --------------------------------------------------------------------------- Extension of credit and loan modifications. As proposed, Sec. 1003.2(d) [[Page 66141]] generally provided that a closed-end mortgage loan was a dwelling- secured ``debt obligation.'' Many consumer advocacy group commenters asked the Bureau to clarify the scope of transactions covered under the term ``debt obligation.'' In particular, a large number of consumer advocacy group commenters asked the Bureau to require reporting of all loan modifications.\105\ The commenters argued that financial institutions' performance in modifying loans is and will continue to be a major factor in determining whether they are meeting local housing needs, particularly the needs of communities that have been devastated by the mortgage crisis. The commenters also argued that financial institutions' loan modification performance will be a major factor in determining whether they are complying with fair housing and fair lending laws. Specifically, commenters cited several studies showing that, since the mortgage crisis, borrowers of color, or borrowers who live in communities of color or in low-to-moderate income communities, have received less favorable loss mitigation outcomes than white borrowers. Commenters stated that many millions of loan modifications have been made since the mortgage crisis, and millions more will be made in the coming years. Commenters argued that the need for data about loan modifications is compelling given the volume of transactions, the identified fair lending concerns, and the lack of other publicly available data about them. --------------------------------------------------------------------------- \105\ These comments related to loan workout modifications. Several commenters also addressed coverage of loan consolidation, extension, and modification agreements. Those comments are discussed separately, below. --------------------------------------------------------------------------- As several of these commenters noted, however, loan modifications currently are not reported because they are not ``originations'' under existing Regulation C. Indeed, since its adoption, Regulation C has required reporting only of applications, originations, and purchases, and the proposal did not seek to change this. While there is a need for publicly available data about loan modifications, the final rule does not require reporting of loan modifications. Covering all loan modifications would be a complex undertaking and would constitute a major revision of Regulation C. However, the Bureau has no information about the burdens to financial institutions of reporting loan modifications under Regulation C, and the Bureau neither has proposed, nor has received feedback about, how existing data points would need to be modified, or whether additional data points would be required, to accommodate reporting of loan modifications. After considering the comments, the Bureau is adopting Sec. 1003.2(d) to provide that a ``closed-end mortgage loan'' is a dwelling- secured ``extension of credit'' that is not an open-end line of credit under Sec. 1003.2(o). Comment 2(d)-2 provides guidance about ``extension of credit.'' First, comment 2(d)-2 provides an example of a transaction that is not a closed-end mortgage loan because no credit is extended. Comment 2(d)-2 also explains that, for purposes of Regulation C, an ``extension of credit'' refers to the granting of credit pursuant to a new debt obligation. If a transaction modifies, renews, extends, or amends the terms of an existing debt obligation without satisfying and replacing the original debt obligation with a new debt obligation, the transaction generally is not an extension of credit under Regulation C. The Bureau understands that it is interpreting the phrase ``extension of credit'' differently in Sec. 1003.2(d) than in Regulation B, 12 CFR part 1002, which implements the Equal Credit Opportunity Act (ECOA).\106\ Regulation B defines ``extension of credit'' under Sec. 1002.2(q) to include the granting of credit in any form, including the renewal of credit and the continuance of existing credit in some circumstances. As discussed above, the Bureau generally is interpreting the phrase ``extension of credit'' in Sec. 1003.2(d) to refer at this time only to the granting of credit pursuant to a new debt obligation. The Bureau may in the future revisit whether it is appropriate to require loan modifications to be reported under Regulation C. --------------------------------------------------------------------------- \106\ 15 U.S.C. 1691 et seq. --------------------------------------------------------------------------- Exceptions to ``extension of credit'' rule. As discussed below, comments 2(d)-2.i and .ii provide two narrow exceptions to the general rule that an ``extension of credit'' under the final rule occurs only when a new debt obligation is created. One exception addresses assumptions, which Regulation C historically has covered. The second addresses transactions completed pursuant to New York consolidation, extension, and modification agreements (New York CEMAs). As discussed below, the Bureau believes that both assumptions and transactions completed pursuant to New York CEMAs represent situations where a new debt obligation is created in substance, if not in form, and that the benefits of requiring such transactions to be reported justify the burdens. Assumptions. The final rule adds new comment 2(d)-2.i to address Regulation C's coverage of assumptions. Under existing comment 1(c)-9, assumptions are reportable transactions. Existing comment 1(c)-9 provides that assumptions occur when an institution enters into a written agreement accepting a new borrower as the obligor on an existing obligation. Existing comment 1(c)-9 also provides that assumptions are reportable as home purchase loans. The Bureau proposed to move existing comment 1(c)-9 to the commentary to the definition of home purchase loan, and the Bureau is finalizing that comment, with certain modifications, as comment 2(j)-5. See the section-by-section analysis of Sec. 1003.2(j). Consistent with the final rule's continued coverage of assumptions, the Bureau is adding comment 2(d)-2.i to the definition of closed-end mortgage loan to clarify that an assumption is an ``extension of credit'' under Regulation C even though the new borrower assumes an existing debt obligation. When the Board first clarified Regulation C's application to assumptions, it stated that, when an institution expressly agrees in writing with a new party to accept that party as the obligor on an existing home purchase loan, the transaction should be treated as a new home purchase loan.\107\ The Bureau agrees and final comment 2(d)-2.i thus provides that assumptions are considered ``extensions of credit'' even if the new borrower assumes an existing debt obligation. --------------------------------------------------------------------------- \107\ See 53 FR 31683, 31685 (Aug. 19, 1988). --------------------------------------------------------------------------- Comment 2(d)-2.i also addresses successor-in-interest transactions. A successor-in-interest transaction is a transaction in which an individual first succeeds the prior owner as the property owner and afterward seeks to take on the debt secured by the property. One industry association recommended that the Bureau exclude successor-in- interest transactions from Regulation C's definition of assumption. The comment noted that the Bureau recently published interpretive guidance under Regulation Z stating that successor-in-interest transactions are not assumptions under that regulation because the successor already owns the property when the debt is assumed.\108\ The comment argued that successor-in-interest transactions should be treated the same under Regulations C and Z. --------------------------------------------------------------------------- \108\ See 79 FR 41631 (July 17, 2014). --------------------------------------------------------------------------- [[Page 66142]] The Bureau is clarifying in comment 2(d)-2.i that successor-in- interest transactions are assumptions under Regulation C. The Bureau's interpretive guidance providing that successor-in-interest transactions are not assumptions under Regulation Z relies on Regulation Z's existing definition of assumption in Sec. 1026.2(a)(24), which provides that the new transaction must be a residential mortgage transaction, i.e., a transaction to finance the acquisition or initial construction of the dwelling being financed. Successor-in-interest transactions do not fit Regulation Z's definition because no dwelling is being acquired or constructed.\109\ In contrast, Regulation C's definition of assumption requires only that a new borrower be accepted as the obligor on an existing obligation. Successor-in-interest transactions fit Regulation C's definition.\110\ --------------------------------------------------------------------------- \109\ See id. at 41633 (``Although [successor-in-interest] transactions are commonly referred to as assumptions, they are not assumptions under Sec. 1026.20(b) because the transaction is not a residential mortgage transaction as to the successor.'') \110\ Consistent with Regulation Z's interpretive guidance, however, final comment 2(j)-5 provides that successor-in-interest transactions are not home purchase loans under Sec. 1003.2(j). --------------------------------------------------------------------------- Moreover, when the Bureau issued its Regulation Z interpretive guidance, it was concerned that subjecting successor-in-interest transactions to an ability-to-repay analysis could decrease the frequency of such transactions, which could harm successors inheriting homes after, for example, a family member's death. The Bureau does not believe that similar concerns apply to requiring such transactions to be reported under Regulation C. On the contrary, the Bureau believes that collecting information about successor-in-interest transactions under Regulation C will help to monitor for discrimination in such transactions. Comment 2(d)-2.i thus specifies that successor-in- interest transactions are assumptions under Regulation C. Like assumptions generally, successor-in-interest transactions represent an exception to the general rule that an ``extension of credit'' requires a new debt obligation. As noted, the Bureau believes that assumptions, including successor-in-interest transactions, represent new debt obligations in substance, if not in form, and should be reported as such. Consolidation, Extension, and Modification Agreements Several consumer advocacy group commenters stated that it was unclear whether the proposal covered transactions completed pursuant to modification, extension, and consolidation agreements (MECAs) or consolidation, extension, and modification agreements (CEMAs). They asked the Bureau to specify that MECAs/CEMAs are reportable transactions. As noted below, Regulation C's commentary at one time specified that MECAs/CEMAs were not reportable as refinancings, and this guidance currently exists in an FFIEC FAQ. Some uncertainty has remained, however, about the reportability of MECAs/CEMAs used for home purchase or home improvement purposes. For the reasons discussed below, the final rule clarifies that CEMAs completed pursuant to section 255 of the New York Tax Law are covered loans. Other MECA/CEMA transactions are not covered loans under the final rule. New York CEMAs are loans secured by dwellings located in New York State. They generally are used in place of traditional refinancings, either to amend a transaction's interest rate or loan term, or to permit a borrower to take cash out. However, unlike in traditional refinancings, the existing debt obligation is not ``satisfied and replaced.'' Instead, the existing obligation is consolidated into a new loan, either by the same or a different lender, and either with or without new funds being added to the existing loan balance. Under New York State law, if no new money is added during the transaction, there is no ``new'' mortgage, and the borrower avoids paying the mortgage recording taxes that would have been imposed if a traditional refinancing had been used and the original obligation had been satisfied and replaced. If new money is part of the consolidated loan, the borrower pays mortgage recording taxes only on the new money.\111\ While generally used in place of traditional refinancings, New York CEMAs also can be used for home purchases (i.e., to complete an assumption), where the seller and buyer agree that the buyer will assume the seller's outstanding principal balance, and that balance is consolidated with a new loan to the borrower for the remainder of the purchase price. --------------------------------------------------------------------------- \111\ See N.Y. Tax Law Sec. 255 (Consol. 2015). --------------------------------------------------------------------------- A number of consumer advocacy group commenters stated that the Bureau should include MECAs/CEMAs, particularly New York CEMAs, as reportable transactions under the dwelling-secured coverage scheme. These commenters stated that New York CEMAs very often are used in lieu of traditional refinance loans, especially for larger-dollar, multifamily apartment building loans, which are central to maintaining the stock of private affordable housing complexes. The commenters argued that, without New York CEMA data, it is difficult or impossible to know where and how much credit banks are extending for such residential buildings, and whether the credit is extended on equitable terms. The commenters noted that CEMAs optionally are reported under the Community Reinvestment Act (CRA) but that CRA reporting provides less data to the public or to policymakers than if the transactions were HMDA-reportable. These commenters also stated that HMDA reporters historically have experienced confusion about whether to report MECAs/CEMAs. Under Regulation C's traditional loan-purpose coverage scheme, the Board declined to extend coverage to MECAs/CEMAs, because the Board found that the transactions did not meet the definition of a refinancing (because the existing debt obligation was not satisfied and replaced). The Board determined that maintaining a bright-line ``satisfies and replaces'' rule for refinancings was preferable to revising the definition to a ``functional equivalent'' test that would cover MECAs/ CEMAs but that also would introduce uncertainty about whether other types of transactions should be reported as refinancings.\112\ Because the Board's guidance concerning MECAs/CEMAs was limited to refinancings, however, it appears that at least some financial institutions have reported MECAs/CEMAs as home improvement loans when the transactions involved new money for home improvement purposes, or as home purchase loans when the transactions were the functional equivalent of traditional assumptions. --------------------------------------------------------------------------- \112\ See 59 FR 63698, 63702 (Dec. 9, 1994); 65 FR 78656 (Dec. 15, 2000); 67 FR 7222, 7227 (Feb. 15, 2002). In 1995, the Board adopted commentary to clarify that MECAs/CEMAs were not reportable as refinancings. 60 FR 63393 (Dec. 11, 1995). This commentary later was dropped from Regulation C inadvertently, but it was retained in an FFIEC FAQ. --------------------------------------------------------------------------- The various consumer advocacy group commenters that addressed MECAs/CEMAs asserted that the proposal did not resolve the uncertainty that has existed about whether to report these transactions. Proposed Sec. 1003.2(d) provided that all closed-end, dwelling-secured ``debt obligations'' were reportable transactions, and ``debt obligations'' arguably would include MECAs/CEMAs. At the same time, however, the proposal retained Regulation C's existing definition of ``refinancing,'' which arguably would continue to exclude MECAs/CEMAs from coverage or would make it unclear [[Page 66143]] how such transactions should be reported. The Bureau concludes that having data about New York CEMAs, in particular, will improve HMDA data. These transactions are used regularly in New York in place of traditional refinancings and sometimes in place of traditional home purchase loans. New York CEMAs are used not only for multifamily dwellings, but also for single-family transactions in high-cost areas like New York City. While it is difficult to identify precisely how often New York CEMAs are used, industry professionals familiar with the New York CEMA market believe that the transactions are used on a daily basis in New York State and represent a significant percentage of the refinancings that occur in the State. Requiring reporting of New York CEMAs will improve HMDA data and also will resolve lingering confusion about how Regulation C applies to them. Finally, the change is consistent with the shift to dwelling-secured coverage for most transactions.\113\ --------------------------------------------------------------------------- \113\ The Bureau understands that MECAs/CEMAs may be used in States other than New York. However, based on the comments received and the Bureau's own research, it appears that CEMAs are particularly common in New York State. As noted elsewhere in this section-by-section analysis, the Bureau understands that, by requiring reporting of New York CEMAs, it is departing from the Board's historical guidance on this topic. The Bureau believes that such a departure is warranted based on the apparent frequency with which such transactions are used. Like the Board, however, the Bureau believes that the benefits of modifying the overall ``satisfies and replaces'' standard for refinancings to capture MECAs/CEMAs do not justify the burdens of such a change. Therefore, the Bureau is incorporating New York CEMAs into the final rule by referencing the specific provision of the New York Tax Code that permits them. If the Bureau becomes aware of CEMAs/MECAs being completed in significant numbers in other States, the Bureau may evaluate whether it would be practicable to require them to be reported in a similar manner. --------------------------------------------------------------------------- Like assumptions, New York CEMAs represent an exception to the general rule that an ``extension of credit'' requires a new debt obligation. However, the Bureau believes that New York CEMAs represent new debt obligations in substance, if not in form, and should be reported as such. The Bureau acknowledges that, by requiring reporting of New York CEMAs, it is departing from the Board's historical guidance that such transactions need not be reported. The Bureau believes that the benefits of this departure justify the burdens both for the reasons discussed above and because the Bureau is defining the scope of transactions to be reported narrowly to encompass only those transactions that fall within the scope of New York Tax Law section 255.\114\ The Bureau believes that limiting the scope of reportable MECAs/CEMAs to those covered by New York Tax Law section 255 will permit New York CEMAs to be reported while avoiding the confusion that, as the Board worried, could result from departing from a bright-line ``satisfies and replaces'' rule for the definition of refinancings generally. --------------------------------------------------------------------------- \114\ Under the final rule, MECAs/CEMAs completed in States other than New York are not reported, regardless of whether they are used for home purchase, home improvement, or refinancing purposes, and regardless of whether new money is extended as part of the transaction. --------------------------------------------------------------------------- After considering the comments received, the Bureau is adopting new comment 2(d)-2.ii, specifying that a transaction completed pursuant to a New York CEMA and classified as a supplemental mortgage under N.Y. Tax Law Sec. 255, such that the borrower owed reduced or no mortgage recording taxes, is an extension of credit under Sec. 1003.2(d). To avoid any implication that other types of loan modifications or extensions must be reported, the commentary language is narrowly tailored to require reporting only of transactions completed pursuant to this specific provision of New York law. See the section-by-section analysis of Sec. 1003.2(i), (j), and (p) for details about whether a New York CEMA is a home improvement loan, a home purchase loan, or a refinancing. 2(e) Covered Loan HMDA requires financial institutions to collect and report information about ``mortgage loans,'' which HMDA section 303(2) defines as loans secured by residential real property or home improvement loans. When the Board adopted Regulation C, it implemented this requirement by mandating that financial institutions report information about applications and closed-end loans made for one of three purposes: Home improvement, home purchase, or refinancing.\115\ As noted, under existing Regulation C, closed-end home purchase loans and refinancings must be reported if they are dwelling-secured, and closed-end home improvement loans must be reported whether or not they are dwelling- secured.\116\ For transactions that meet one of the three purposes, reporting of closed-end loans is mandatory and reporting of home-equity lines of credit is optional.\117\ Under existing Regulation C, reverse mortgages are subject to these same criteria for reporting: A closed- end reverse mortgage must be reported if it is for one of the three purposes; a reverse mortgage that is an open-end line of credit is optionally reported. --------------------------------------------------------------------------- \115\ 41 FR 23931, 23932 (June 14, 1976). \116\ See the section-by-section analysis of Sec. 1003.2(d), (f), (i). \117\ Specifically, under existing Sec. 1003.4(c)(3), financial institutions optionally may report home-equity lines of credit made in whole or in part for the purpose of home improvement or home purchase. --------------------------------------------------------------------------- To simplify Regulation C's transactional coverage test and to expand the types of transactions reported, the Bureau proposed to require financial institutions to report applications for, and originations and purchases of, all dwelling-secured loans and lines of credit. The Bureau also proposed to add the defined term ``covered loan'' in Sec. 1003.2(e). The term referred to all transactions reportable under the proposed dwelling-secured coverage scheme: Closed- end mortgage loans under proposed Sec. 1003.2(d), open-end lines of credit under proposed Sec. 1003.2(o), and reverse mortgages under proposed Sec. 1003.2(q). The term provided a shorthand phrase that HMDA reporters and data users could use to refer to any transaction reportable under Regulation C. For the reasons discussed below, the Bureau is finalizing in Sec. 1003.2(e) the defined term ``covered loan'' and the shift to dwelling-secured coverage largely as proposed for consumer-purpose loans and lines of credit. The Bureau is retaining Regulation C's existing purpose-based test for commercial-purpose loans and lines of credit. Only a few commenters specifically addressed the Bureau's proposal to add the defined term ``covered loan'' to Regulation C to refer to all covered transactions, and the commenters generally favored the proposal. They believed that having a standard shorthand for all covered transactions would facilitate compliance. The Bureau is finalizing Sec. 1003.2(e) ``covered loan'' to define the universe of transactions covered under Regulation C. A large number of commenters addressed the proposed shift from purpose-based to collateral-based transactional coverage, with consumer advocacy group commenters supporting the shift and industry commenters expressing mixed views.\118\ Some consumer advocacy groups stated that having information about all loans secured by residential property would [[Page 66144]] improve the usefulness and quality of HMDA data. Others stated that having data about all such loans would be valuable in assessing whether financial institutions are providing the neighborhoods that the consumer advocacy groups serve with the full range of credit the neighborhoods need. One consumer advocacy commenter asserted that financial institutions should report any transaction that could result in a borrower losing his or her home. Another stated that removing the subjectivity from determining whether to report a loan would ease burden for financial institutions, and that having information about more loans would improve HMDA's usefulness. The commenter noted that consumer mortgage lending products evolve rapidly, and there is no principled reason to require reporting of some but not others. --------------------------------------------------------------------------- \118\ This section-by-section analysis provides a high-level discussion of comments concerning the proposed shift to dwelling- secured coverage. See the section-by-section analyses of Sec. 1003.2(d), (i), (o), (q) and of Sec. 1003.3(c)(10) for specific comments concerning closed-end mortgage loans, home improvement loans, open-end lines of credit, reverse mortgages, and commercial- purpose transactions, respectively. --------------------------------------------------------------------------- Industry commenters and a group of State regulators expressed mixed views about the proposed shift to dwelling-secured coverage. A small number of industry commenters supported the proposal unconditionally because they believed that it would ease burden. These commenters, who generally were smaller financial institutions and compliance consultants, stated that deciding which loans meet the current purpose test is confusing. They stated that a simplified transactional coverage test would stop the erroneous over-reporting of loans that has occurred despite financial institutions' best efforts,\119\ and that the benefits of a streamlined test justified the burdens of more reporting. One industry commenter appreciated the fact that HMDA would provide a more comprehensive view of mortgage transactions across the country. A group of State regulators supported dwelling-secured coverage for consumer-purpose transactions only. --------------------------------------------------------------------------- \119\ These commenters seemed to be concerned about erroneously classifying consumer-purpose transactions as HMDA-reportable and, in turn, unnecessarily collecting race, sex, and ethnicity data from applicants and borrowers. --------------------------------------------------------------------------- The majority of industry commenters that addressed transactional coverage opposed the proposed shift to dwelling-secured coverage, supported it only for consumer-purpose transactions or for closed-end mortgage loans, or supported it only to the extent that it would eliminate reporting of home improvement loans not secured by a dwelling. Numerous industry commenters generally objected to the overall compliance burdens and costs of reporting additional transactions, particularly in light of the Bureau's proposal simultaneously to expand the data reported about each transaction and to lower (for some institutions) the institutional coverage threshold.\120\ One government agency commenter expressed concern that the revisions to transactional coverage would burden small financial institutions and urged the Bureau not to adopt the proposed changes. Some industry commenters generally asserted that their reportable transaction volume would increase significantly,\121\ that they would not be able to comply without hiring additional staff, and that compliance costs would be passed to consumers. Others generally argued that the Bureau should keep Regulation C's existing purpose-based coverage because it serves HMDA's purposes better than a collateral- based scheme. Most industry commenters that opposed the proposed shift, however, specifically objected to the burdens of reporting all home- equity lines of credit and all dwelling-secured commercial-purpose loans and lines of credit. --------------------------------------------------------------------------- \120\ A number of commenters argued that, in light of the Bureau's proposal to expand transactional coverage, the Bureau should modify its institutional coverage threshold proposal. Those comments are discussed in the section-by-section analysis of Sec. 1003.2(g). \121\ Many commenters discussed the overall increase in reporting from a shift to dwelling-secured coverage. Others estimated only the increase from particular categories of transactions, such as home-equity lines of credit or commercial- purpose transactions. Those estimates are discussed in the section- by-section analyses of Sec. 1003.2(o) and of Sec. 1003.3(c)(10). --------------------------------------------------------------------------- As explained in the section-by-section analyses of Sec. 1003.2(d) and (o), the Bureau is finalizing the shift to dwelling-secured coverage for closed- and open-end consumer-purpose transactions, with some modifications to ease burden for open-end reporting. After considering the comments received, and as discussed fully in the section-by-section analyses of those sections, the Bureau believes that the benefits of expanded reporting justify the burdens. As discussed in the section-by-section of Sec. 1003.3(c)(10), however, the Bureau is maintaining Regulation C's existing purpose-based coverage test for commercial-purpose transactions. 2(f) Dwelling The Bureau proposed to revise the definition of dwelling in Sec. 1003.2 by moving the geographic location requirement currently in the definition of dwelling to Sec. 1003.1(c), to add additional examples of dwellings to the definition and commentary, and to revise the commentary to exclude certain structures from the definition of dwelling. A few commenters supported the proposed changes to the definition of dwelling, while others argued that certain types of structures should be included or excluded from the definition. For the reasons discussed below, the Bureau is finalizing Sec. 1003.2(f) with minor technical revisions to the definition and with additional revisions to the commentary discussed in detail below. The definition is revised to clarify that multifamily residential structures include complexes and manufactured home communities. Some commenters argued that second homes and investment properties should no longer be covered by Regulation C and that only primary residences should be reported because second homes and investment properties do not relate to housing needs in the same way that primary residences do. HMDA section 303(2) defines a mortgage loan, in part, as one secured by ``residential real property'' and HMDA section 304(b)(2) requires collection of information regarding ``mortgagors who did not, at the time of execution of the mortgage, intend to reside in the property securing the mortgage loan.'' The Bureau believes that second homes as well as investment properties are within the scope of information required by HMDA and should continue to be covered by Regulation C. The Bureau is therefore finalizing comment 2(f)-1 generally as proposed, with certain material from proposed comment 2(f)-1 incorporated into comment 2(f)-2 as discussed below. Some commenters argued that all multifamily properties should be excluded from Regulation C. The Bureau believes that multifamily residential structures should continue to be included within Regulation C because they provide for housing needs and because, as the Bureau noted in the proposal, HMDA data highlight the importance of multifamily lending to the recovering housing finance market and to consumers.\122\ --------------------------------------------------------------------------- \122\ 79 FR 51731, 51800 (Aug. 29, 2014); San Francisco Hearing, supra note 42. --------------------------------------------------------------------------- Many commenters addressed multifamily loan reporting in more specific ways. Some commenters supported the proposal's coverage of manufactured home community loans and other aspects related to multifamily lending. Others requested guidance on reporting multifamily transactions. Some commenters argued that certain types of multifamily lending should be excluded from Regulation C. The Bureau is adopting new comment 2(f)-2 dealing specifically with multifamily residential structures and communities, [[Page 66145]] which incorporates certain material from proposed comment 2(f)-1 and additional material in response to comments. The Bureau believes that providing a specific comment relating to multifamily residential structures will facilitate compliance by providing guidance on when loans related to multifamily dwellings would be considered loans secured by a dwelling for purposes of Regulation C. The comment provides that a manufactured home community is a dwelling for purposes of Regulation C regardless of whether any individual manufactured homes also secure the loan. The comment also provides examples of loans related to certain multifamily structures that would nevertheless not be secured by a dwelling for purposes of Regulation C, and would therefore not be reportable, such as loans secured only by an assignment of rents or dues or only by common areas and not individual dwelling units. The Bureau is adopting new comment 2(f)-3 relating to exclusions from the definition of dwelling (incorporating material from proposed comment 2(f)-2) and clarifying that recreational vehicle parks are excluded from the definition of dwelling for purposes of Regulation C. Several commenters agreed with the proposed exclusions for recreational vehicles, houseboats, mobile homes constructed prior to June 15, 1976 (pre-1976 mobile homes),\123\ and other types of structures. --------------------------------------------------------------------------- \123\ The HUD standards for manufactured homes do not cover mobile homes constructed before June 15, 1976. 24 CFR 3282.8(a). 79 FR 51731, 51749 (Aug. 29, 2014). --------------------------------------------------------------------------- Regarding the exclusion of recreational vehicles, the Bureau agrees with the commenters that supported the proposed clarification that recreational vehicles are not dwellings for purposes of Regulation C, regardless of whether they are used as residences. As noted in the proposal, the Bureau believes that making this exclusion explicit will provide more clarity on what structures qualify as dwellings and reduce burden on financial institutions. The Bureau also believes it will improve the consistency of reported HMDA data. Clarifying that recreational vehicle parks are excluded from the definition of dwelling for purposes of Regulation C is consistent with the exclusion of recreational vehicles. The Bureau believes that, as discussed above, while manufactured home communities should be included in the definition of dwelling for purposes of Regulation C, including recreational vehicle parks would not be appropriate given that they are not frequently intended as long-term housing. Some commenters stated that the proposed exclusion of pre-1976 mobile homes would create compliance problems because the financial institution could mistakenly collect race, ethnicity, and sex information before knowing whether the home was a manufactured home and therefore violate Regulation B. The Bureau believes that this concern is unlikely to result in ECOA violations because Regulation B would still require collection of demographic information on some pre-1976 mobile home lending.\124\ Other commenters argued that pre-1976 mobile home lending should be reported under Regulation C because of consumer protection and housing needs concerns related to this type of housing. The Bureau does not believe this concern justifies the additional burden of requiring financial institutions to report these loans and identify them distinctly from manufactured home loans, especially given that the amount of lending secured by this type of collateral will continue to decrease as time passes. Therefore, the Bureau is finalizing the exclusion of pre-1976 mobile homes as part of comment 2(f)-3. Clarifying that recreational vehicle parks are excluded from the definition of dwelling for purposes of Regulation C is consistent with the exclusion of recreational vehicles.\125\ The Bureau believes that, as discussed above, while manufactured home communities should be included in the definition of dwelling for purposes of Regulation C, including recreational vehicle parks would not be appropriate given that they are not frequently intended as long term housing. --------------------------------------------------------------------------- \124\ 12 CFR 1002.13(a)(2). \125\ As discussed in the proposal, the final rule's definition of dwelling would differ from Regulation Z's definition of dwelling with respect to some recreational vehicles, because Regulation Z treats recreational vehicles used as residences as dwellings. 12 CFR part 1026, comment 2(a)(19)-2. 79 FR 51731, 51749 (Aug. 29, 2014). --------------------------------------------------------------------------- The Bureau proposed a special rule for mixed-use properties that contained five or more individual dwelling units. The Bureau proposed that such a property always be considered to have a primary residential use and therefore report a covered loan secured by it. A few commenters supported the proposal to report all residential structures with five or more individual dwelling units, but most commenters who addressed mixed-use property argued that this was overbroad and that the current primary use rules should apply to multifamily residential structures as well. The Bureau is revising comment 2(f)-3 relating to mixed-use properties and finalizing it as comment 2(f)-4 by removing the sentence requiring that financial institutions always treat residential structures with five or more individual dwelling units as having a primary residential purpose. Requiring financial institutions to report mixed-use multifamily properties in all circumstances would result in reporting of multifamily properties with relatively small housing components and large commercial components. Data users could not differentiate between those properties and multifamily properties with larger housing components, which would decrease the data's usefulness. Thus retaining the existing discretion for financial institutions to determine the primary use for multifamily properties is appropriate. The Bureau is adopting new comment 2(f)-5 relating to properties with medical and service components. Some commenters requested guidance on when properties such as retirement homes, assisted living, and nursing homes should be reported under Regulation C. Other commenters requested exclusions for all properties that provide any service or medical care component. The Bureau does not believe it is appropriate to exclude all such properties. Information about loans secured by properties that provide long-term housing and that are not transitory or primarily medical in nature provides valuable information on how financial institutions are serving the housing needs of their communities. The comment provides that properties that provide long- term housing with related services are reportable under Regulation C, while properties that provide medical care are not, consistent with the exclusion of hospitals in comment 2(f)-3. The comment also clarifies that such properties are reportable when they combine long-term housing and related services with a medical care component. The comment will facilitate compliance by expanding on earlier guidance provided by the Board.\126\ Section 1003.2(f) is being adopted to implement, in part, the definition of ``mortgage loan'' in HMDA section 303(2). That term would be implemented through other terms in Regulation C as well, including the definitions of ``closed-end mortgage loan'' and ``covered loan.'' In combination with other relevant provisions in Regulation C, the Bureau [[Page 66146]] believes that the proposed definition of ``dwelling'' is a reasonable interpretation of the definition in that provision. Section 1003.2(f) is also adopted pursuant to the Bureau's authority under section 305(a) of HMDA. Pursuant to section 305(a) of HMDA, the Bureau believes that this proposed definition is necessary and proper to effectuate the purposes of HMDA. The definition will serve HMDA's purpose of providing information to help determine whether financial institutions are serving the housing needs of their communities by providing information about various types of housing that are financed by financial institutions. The definition will facilitate compliance with HMDA requirements by providing clarity regarding what transactions must be reported for purposes of Regulation C. --------------------------------------------------------------------------- \126\ 60 FR 63393, 63395 (Dec. 11, 1995). Fed. Reserve Bank of St. Louis, CRA/HMDA Reporter, Census 2000 and CRA/HMDA Data Collection, (Sept. 2000), available at http://www.ffiec.gov/hmda/pdf/00news.pdf. --------------------------------------------------------------------------- 2(g) Financial Institution Regulation C requires institutions that meet the definition of financial institution to collect and report HMDA data. HMDA and current Regulation C establish different coverage criteria for depository institutions (banks, savings associations, and credit unions) than for nondepository institutions (for-profit mortgage-lending institutions other than banks, savings associations, or credit unions).\127\ Under the current definition, depository institutions that originate one first-lien home purchase loan or refinancing secured by a one- to four- unit dwelling and that meet other criteria for ``financial institution'' must collect and report HMDA data, while certain nondepository institutions that originate many more mortgage loans annually do not have to collect and report HMDA data. --------------------------------------------------------------------------- \127\ HMDA sections 303(3) and 309(a); Regulation C Sec. 1003.2 (definition of financial institution). --------------------------------------------------------------------------- The Bureau proposed to adjust Regulation C's institutional coverage to adopt a uniform loan-volume threshold of 25 loans applicable to all financial institutions. Under the proposal, depository institutions and nondepository institutions that meet all of the other criteria for a ``financial institution'' would be required to report HMDA data if they originated at least 25 covered loans, excluding open-end lines of credit, in the preceding calendar year. For the reasons discussed below, the Bureau is finalizing changes to Regulation C's institutional coverage and adopting uniform loan- volume thresholds for depository and nondepository institutions. The loan-volume thresholds require an institution that originated at least 25 closed-end mortgage loans or at least 100 open-end lines of credit in each of the two preceding calendar years to report HMDA data, provided that the institution meets all of the other criteria for institutional coverage. The final rule's changes to institutional coverage will provide several important benefits. First, the coverage test will improve the availability of data concerning the practices of nondepository institutions. The expanded coverage of nondepository institutions will ensure more equal visibility into the practices of nondepository institutions and depository institutions. With expanded HMDA data about nondepository lending, the public and public officials will be better able to protect consumers because historically, some riskier lending practices, such as those that led to the financial crisis, have emerged from the nondepository market sector.\128\ --------------------------------------------------------------------------- \128\ See the section-by-section analysis of Sec. 1003.2(g)(ii) below for complete discussion. --------------------------------------------------------------------------- Second, a significant number of lower-volume depository institutions will no longer be required to report HMDA data under the revised coverage test, which will eliminate those institutions' compliance costs. At the same time, the coverage test will preserve sufficient data for analyzing mortgage lending at the national, local, and institutional levels. Third, the coverage test, by considering both an institution's closed-end and open-end origination volumes, will support the goal of increasing visibility into open-end dwelling-secured lending. This change to institutional coverage, along with the change to transactional coverage discussed in the section-by-section analysis of Sec. 1003.2(o), will improve the public and public officials' ability to understand whether, and how, financial institutions are using open- end lines of credit to serve the housing needs of their communities. Incorporating open-end lending into the institutional coverage test will not require financial institutions that originate a small number of closed-end mortgage loans or open-end lines of credit to report those loans. As discussed below in the section-by-section analysis of Sec. 1003.3(c)(11) and (12), the final rule also includes transactional thresholds. The transactional thresholds ensure that financial institutions that meet only the 25 closed-end mortgage loan threshold are not required to report their open-end lending, and that financial institutions that meet only the 100 open-end line of credit threshold are not required to report their closed-end lending. Finally, by considering two years of lending for coverage, the final rule will provide stability in reporting obligations for institutions. Accordingly, a financial institution that does not meet the loan-volume thresholds established in the final rule and that has an unexpected and unusually high loan-origination volume in one year will not be required to report HMDA data unless it maintains that level of lending for two consecutive years. The specific changes to the definition of financial institution applicable to nondepository institutions and depository institutions are discussed below separately. The Bureau also proposed technical modifications to the commentary to the definition of financial institution. The Bureau received no comments on the proposed comments 2(g)-1 or -3 through -6, and is finalizing the commentary as proposed and with technical modifications to conform to definition of financial institution included in the final rule. The Bureau is also renumbering proposed comments 2(g)-3 through - 6 as comments 2(g)-4 through -7. The Bureau is also adopting new comment 2(g)-3 to address how to determine whether an institution satisfies the definition of financial institution after a merger or acquisition. For ease of publication, the Bureau is reserving comment 2(g)-2, which sets forth the asset-size adjustment for depository financial institutions for each calendar year. The Bureau updates comment 2(g)-2 annually to make the adjustments to the level of the asset-size exemption for depository financial institutions required by HMDA section 309(b). The reserved comment will be replaced when the asset- size adjustment for the 2018 calendar year is published. 2(g)(1) Depository Financial Institutions HMDA extends reporting responsibilities to depository institutions (banks, savings associations, and credit unions) that satisfy certain location, asset-size, and federally related requirements.\129\ Regulation C implements HMDA's coverage criteria in the definition of financial institution in Sec. 1003.2. Under the current definition of financial institution in Sec. 1003.2, a bank, savings association, or credit union meets the definition of financial institution if it satisfies all of the following criteria: (1) On the preceding December 31, it had assets of at least $44 million; \130\ (2) on the preceding December 31, it had a home or branch office in a Metropolitan Statistical Area (MSA); (3) during the previous calendar [[Page 66147]] year, it originated at least one home purchase loan or refinancing of a home purchase loan secured by a first-lien on a one- to four-unit dwelling; and (4) the institution is federally insured or regulated, or the mortgage loan referred to in item (3) was insured, guaranteed, or supplemented by a Federal agency or intended for sale to the Federal National Mortgage Association or the Federal Home Loan Mortgage Corporation.\131\ --------------------------------------------------------------------------- \129\ 12 U.S.C. 2802(3). \130\ Comment Financial institution-2 to Sec. 1003.2. \131\ Section 1003.2(financial institution)(1). --------------------------------------------------------------------------- Proposed Sec. 1003.2(g)(1) modified the definition of financial institution by defining a new term, depository financial institution, and adding a loan-volume threshold to the coverage criteria for depository institutions. The proposed loan-volume threshold would require reporting only by depository institutions that met the current criteria in Sec. 1003.2 and that originated at least 25 covered loans, excluding open-end lines of credit, in the preceding calendar year. The Bureau received a large number of comments on proposed Sec. 1003.2(g)(1). Industry commenters generally supported eliminating the requirement to report from low-volume depository institutions, but urged the Bureau to exclude more institutions from the requirement to report HMDA data. Consumer advocate commenters generally opposed decreasing Regulation C's depository institution coverage. The Bureau is adopting Sec. 1003.2(g)(1), which defines depository financial institution, to include banks, savings associations, and credit unions, that meet the current criteria to be considered a financial institution,\132\ and originated at least 25 closed-end mortgage loans or 100 open-end lines of credit in each of the two preceding calendar years. The Bureau is finalizing the proposed exclusion of depository institutions that originate fewer than 25 closed-end mortgage loans. In addition, the final rule also requires lenders that meet the other criteria and that originate at least 100 open-end lines of credit to report HMDA data, even if those institutions did not originate at least 25 closed-end mortgage loans. The final rule includes a two-year look-back period for the loan-volume threshold. Each of these aspects of the final rule is discussed below separately. --------------------------------------------------------------------------- \132\ Under Sec. 1003.2, a bank, savings association, or credit union meets the definition of financial institution if it satisfies all of the following criteria: (1) On the preceding December 31, it had assets in excess of the asset threshold established and published annually by the Bureau for coverage by the Act; (2) on the preceding December 31, it had a home or branch office in a MSA; (3) during the previous calendar year, it originated at least one home purchase loan or refinancing of a home purchase loan secured by a first-lien on a one- to four-unit dwelling; and (4) the institution is federally insured or regulated, or the mortgage loan referred to in item (3) was insured, guaranteed, or supplemented by a Federal agency or intended for sale to the Federal National Mortgage Association or the Federal Home Loan Mortgage Corporation. --------------------------------------------------------------------------- Loan-Volume Threshold for Closed-End Mortgage Loans The Bureau received many comments on proposed Sec. 1003.2(g)(1). Industry commenters generally supported adopting a loan-volume threshold that would eliminate reporting by low-volume depository institutions,\133\ but urged the Bureau to adopt a much higher loan- volume threshold that would exempt more depository institutions from reporting. Industry commenters stated that low-volume depository institutions lack resources and sophistication and that their data have limited value. Industry commenters argued that a higher loan-volume threshold would not impact the availability of data for analysis at the national level or the ability to analyze lending at an institutional level. The commenters also advocated a consistent approach between the loan-volume threshold in Regulation C and the small creditor and small servicer definitions in the Bureau's title XIV Rules.\134\ --------------------------------------------------------------------------- \133\ Participants in the Board's 2010 Hearings also urged the Board to eliminate reporting by lower-volume depository institutions. See, e.g., Atlanta Hearing, supra note 40, (remarks of Phil Greer, Senior Vice President of Loan Administration, State Employees Credit Union) (noting that the burden of reporting only one loan would be low, but that the data reported would not provide ``meaningful information''). \134\ 12 CFR 1026.35(b)(2)(iii)(B) (describing small creditor thresholds); 12 CFR 1026.41(e)(4) (defining small servicer). --------------------------------------------------------------------------- On the other hand, several community advocate commenters expressed strong opposition to decreasing Regulation C's coverage of depository institutions. Most noted that the depository institutions that would be excluded are currently reporting, and therefore are accustomed to reporting. Many also highlighted the importance of the data reported by the depository institutions that would be excluded at the community level, especially in rural and underserved areas or to low- and moderate-income (LMI) individuals and minorities. Commenters provided examples of reports and programs that rely on HMDA data at the census tract level. Other community advocate commenters expressed support for the proposed loan-volume threshold, but noted concerns about the loss of data that may result if the Bureau adopted a loan-volume threshold greater than 25 loans. They highlighted concerns about the loss of data on particular types of transactions, such as applications submitted by African Americans, loans related to multifamily properties, and loans related to manufactured housing. The Bureau believes that Regulation C's institutional coverage criteria should balance the burden on financial institutions with the value of the data reported. Depository institutions that are currently reporting should not bear the burden of reporting under Regulation C if their data are of limited value in the HMDA data set. At the same time, Regulation C's institutional coverage criteria should not impair HMDA's ability to achieve its purposes. Higher closed-end mortgage loan-volume thresholds, as suggested by industry, might not significantly impact the value of HMDA data for analysis at the national level. For example, it is possible to maintain reporting of a significant percentage of the national mortgage market with a closed-end mortgage loan-volume threshold higher than 25 loans annually. In addition, it may also be true that data reported by some institutions that satisfy the proposed 25-loan-volume threshold may not be as useful for statistical analysis as data reported by institutions with much higher loan volumes. However, the higher closed-end mortgage loan-volume thresholds suggested by industry commenters would have a material negative impact on the availability of data about patterns and trends at the local level. Data about local communities is essential to achieve HMDA's three purposes, which are to provide the public and public officials with sufficient information: (1) To determine whether institutions are meeting their obligations to serve the housing needs of the communities in which they are located; (2) to identify communities in need of targeted public and private investment; and (3) to assist in identifying discriminatory lending patterns and enforcing antidiscrimination statutes.\135\ Public officials, community advocates, and researchers rely on HMDA data to analyze access to credit at the neighborhood level and to target programs to assist underserved communities and consumers. --------------------------------------------------------------------------- \135\ Section 1003.1(b). --------------------------------------------------------------------------- Local and State officials have used HMDA data to identify and target relief to localities impacted by high-cost lending or discrimination. For example, policy makers in Lowell, Massachusetts identified a need for homebuyer counseling and education in Lowell, based on HMDA data, which showed a high percentage of high-cost loans [[Page 66148]] compared to surrounding communities.\136\ Similarly, in 2008 the City of Albuquerque used HMDA data to characterize neighborhoods as ``stable,'' ``prone to gentrification,'' or ``prone to disinvestment'' for purposes of determining the most effective use of housing grants.\137\ As another example, Antioch, California, monitors HMDA data, reviews it when selecting financial institutions for contracts and participation in local programs, and supports home purchase programs targeted to households purchasing homes in census tracts with low origination rates.\138\ In addition, the City of Flint Michigan, in collaboration with the Center for Community Progress, used HMDA data to identify neighborhoods in Flint to target for a blight eradication program.\139\ Similarly, HMDA data helped bring to light discriminatory lending patterns in Chicago neighborhoods, resulting in a large discriminatory lending settlement.\140\ Researchers and consumer advocates also analyze HMDA data at the census tract level to identify patterns of discrimination at the national level.\141\ --------------------------------------------------------------------------- \136\ See City of Lawrence, Massachusetts, HUD Consolidated Plan 2010-2015, at 68 (2010), available at http://www.cityoflawrence.com/Data/Sites/1/documents/cd/Lawrence_Consolidated_Plan_Final.pdf. \137\ See City of Albuquerque, Five Year Consolidated Plan and Workforce Housing Plan, at 100 (2008), available at http://www.cabq.gov/family/documents/ConsolidatedWorkforceHousingPlan20082012final.pdf. \138\ See City of Antioch, California, Fiscal Year 2012-2013 Action Plan, at 29 (2012), available at http://www.ci.antioch.ca.us/CitySvcs/CDBGdocs/Action%20Plan%20FY12-13.pdf. \139\ Luke Telander, Flint's Framework for the Future, Ctr. for Cmty. Progress, Cmty. Progress Blog (July 1, 2014), http://www.communityprogress.net/blog/flints-framework-future. \140\ See, e.g., Yana Kunichoff, Lisa Madigan Credits Reporter with Initiating Largest Discriminatory Lending Settlements in U.S. History, Chicago Muckrakers Blog (June 14, 2013, 2:53 p.m.), http://www.chicagonow.com/chicago-muckrakers/2013/06/lisa-madigan-credits-reporter-with-initiating-largest-discriminatory-lending-settlements-in-u-s-history/ (``During our ongoing litigation . . . the Chicago Reporter study looking at the HMDA data for the City of Chicago came out. . . . It was such a startling statistic that I said . . . we have to investigate, we have to find out if this is true. . . . We did an analysis of that data that substantiated what the Reporter had already found. . . . [W]e ultimately resolved those two lawsuits. They are the largest fair-lending settlements in our nation's history.''). \141\ See, e.g., California Reinvestment Coalition, et al, Paying More for the American Dream VI: Racial Disparities in FHA/VA Lending (2012), available at http://www.woodstockinst.org/sites/default/files/attachments/payingmoreVI_multistate_july2012_0.pdf; Samantha Friedman & Gregory D. Squires, Does the Community Reinvestment Act Help Minorities Access Traditionally Inaccessible Neighborhoods?, 52 Social Problems 209 (2005). --------------------------------------------------------------------------- Any loan-volume threshold will affect individual markets differently, depending on the extent to which individual markets are served by smaller creditors and the market share of those creditors. The Bureau believes that a 25-closed-end mortgage loan-volume threshold would impact the robustness of the data that would remain available only in a relatively small number of markets. For example, only about 45 census tracts would lose over 20 percent of currently reported data if a 25 closed-end mortgage loan-volume threshold is used to trigger reporting.\142\ In contrast, the higher closed-end mortgage loan-volume thresholds requested by industry commenters would have a negative impact on data about more communities and consumers. For example, at a closed-end mortgage loan-volume threshold set at 100, the number of census tracts that would lose 20 percent of reported data would increase from about 45 tracts to about 385 tracts, almost eight times more than the number with a threshold set at 25 closed-end mortgage loans.\143\ The number of affected lower-middle income tracts would increase from about 20 tracts to about 145 tracts, an increase of over six times over the number at the 25-loan level.\144\ The Bureau believes that the loss of data in communities at closed-end mortgage loan-volume thresholds higher than 25 would substantially impede the public's and public officials' ability to understand access to credit in their communities. --------------------------------------------------------------------------- \142\ As discussed in part VII below, the Bureau derived these estimates using 2013 HMDA data. \143\ Id. \144\ As discussed in part VII below, the Bureau prepared these estimates using 2013 HMDA data and 2012 Community Reinvestment Act data. --------------------------------------------------------------------------- In addition, the Bureau does not believe that it should set the closed-end mortgage loan-volume threshold at the levels in the small creditor and small servicer definitions in the Bureau's title XIV rules.\145\ While the Bureau's title XIV rules and Regulation C may apply to some of the same institutions and transactions, Regulation C and the Bureau's title XIV rules have different objectives. HMDA aims to provide specific data to the public and public officials. For example, HMDA aims to provide sufficient information to the public and public officials to identify whether the housing needs of their communities are being served by the existing financial institutions. In contrast, the title XIV rule thresholds are designed to balance consumer protection and compliance burden in the context of very specific lending practices. As discussed above, an institutional coverage threshold at the levels of the small creditor and small servicer thresholds, which include thresholds of 2,000 and 5,000 loans, respectively,\146\ would undermine both the utility of HMDA data for analysis at the local level and the benefits that HMDA provides to communities. --------------------------------------------------------------------------- \145\ 12 CFR 1026.35(b)(2)(iii)(B) (describing small creditor thresholds); 12 CFR 1026.41(e)(4) (defining small servicer). \146\ Id. The Bureau recently increased the small creditor threshold to 2,000 applicable loans annually. See 80 FR 59943 (Oct. 2, 2015). --------------------------------------------------------------------------- Finally, the Bureau believes that eliminating the requirement to report by institutions that originated fewer than 25 closed-end mortgage loans annually would meaningfully reduce burden. As discussed in part VII below, the proposed loan-volume threshold would relieve about 22 percent of depository institutions that are currently reporting of the obligation to report HMDA data on closed-end mortgage loans. For the reasons discussed above, the Bureau is adopting a loan- volume threshold for depository institutions that will require reporting by depository institutions that originate at least 25 closed- end mortgage loans annually and meet the other applicable criteria in Sec. 1003.2(g)(1). The Bureau, as discussed below in part VI, believes that the 25 closed-end loan-volume threshold for depository institutions should go into effect on January 1, 2017, one year earlier than the effective date for most of the remaining rule. To effectuate this earlier effective date, the Bureau is amending the definition of ``financial institution'' in Sec. 1003.2. Loan-Volume Threshold for Open-End Lines of Credit The loan-volume threshold provided in proposed Sec. 1003.2(g)(1)(v) excluded open-end lines of credit from the loans that would count toward the threshold.\147\ The Bureau solicited feedback on what types of loans should count toward the proposed loan-volume threshold and, in particular, whether open-end lines of credit should count toward the proposed loan-volume threshold. The final rule incorporates an institution's origination of open-end lines of credit into HMDA's institutional coverage criteria. Under the final rule, a financial institution will be required to [[Page 66149]] report HMDA data on open-end lines of credit if it meets the other applicable criteria and originated at least 100 open-end lines of credit in each of the two preceding calendar years.\148\ --------------------------------------------------------------------------- \147\ Under the proposed loan-volume threshold, the definition of open-end line of credit did not include open-end reverse mortgages. As a result, neither open-end nor closed-end reverse mortgages were excluded from the proposed loan-volume threshold. The definitions of closed-end mortgage loan and open-end line of credit included in the final rule include closed-end and open-end reverse mortgages, respectively, as discussed in the section-by-section analysis of Sec. 1003.2(d) and (o). \148\ Under the final rule, all open-end transactions, whether traditional, reverse, or a combination of the two, count toward the open-end loan-volume threshold. --------------------------------------------------------------------------- Relatively few commenters provided feedback on this issue. Some industry commenters stated that they supported the proposed exclusion of open-end lines of credit from the loans that count toward the loan- volume threshold. These commenters also suggested excluding other types of loans from the loans that count toward the threshold, including commercial loans, home-equity loans, and reverse mortgages. On the other hand, some industry commenters and a community advocate commenter stated that open-end lines of credit should count toward the loan- volume threshold. They explained that this would prevent institutions from steering consumers to open-end lines of credit to avoid being required to report HMDA data. The Bureau is not finalizing the proposed exclusion of open-end lines of credit from Regulation C's institutional coverage criteria for the reasons discussed below. As noted above, the Bureau believes that Regulation C's institutional coverage criteria should balance the burden on financial institutions with the value of the data reported. Depository institutions that are currently reporting should not bear the burden of reporting under Regulation C if their data are of limited value in the HMDA data set. At the same time, Regulation C's institutional coverage criteria should support HMDA's purposes. The Bureau has determined that the exclusion of open-end lines of credit from Regulation C's institutional coverage criteria would not appropriately balance those considerations. As discussed in the section-by-section analysis of Sec. 1003.2(o), the Bureau is finalizing the proposed expansion of the transactions reported in HMDA to include dwelling-secured, consumer-purpose open-end lines of credit, unless an exclusion applies.\149\ Data about such transactions are not currently publicly available and, as discussed in the section-by-analysis of Sec. 1003.2(o), the Bureau believes that having data about them will improve the understanding of how financial institutions are serving the housing needs of their communities and assist in the distribution of public sector investments. Like closed- end home-equity loans and refinancings, both of which are subject to broad coverage under the final rule, dwelling-secured credit lines may be used for home purchase, home improvement, and other purposes. Regardless of how they are used, they liquefy equity that borrowers have built up in their homes, which often are their most important assets. Borrowers who take out dwelling-secured credit lines increase their risk of losing their homes to foreclosure when property values decline, and in fact, the expansion of open-end line of credit originations in the mid-2000s contributed to the foreclosure crises that many communities experienced in the late 2000s.\150\ Had open-end line of credit data been reported in HMDA, the public and public officials could have had a much earlier warning and a better understanding of potential risks, and public and private mortgage relief programs could have better assisted distressed borrowers in the aftermath of the crisis. As discussed in the section-by-section analysis of Sec. 1003.2(o), dwelling-secured open-end lending is again on the rise now that the mortgage market has begun to recover from the crisis. The Bureau believes that it is important to improve visibility into this key segment of the mortgage market for all of the reasons discussed here and in the section-by-section analysis of Sec. 1003.2(o). --------------------------------------------------------------------------- \149\ Under the final rule, dwelling-secured, commercial-purpose open-end lines of credit will be covered loans only if they are for home purchase, home improvement, or refinancing purposes. See the section-by-section analysis of Sec. 1003.3(c)(10). \150\ See 79 FR 51731, 51757 (Aug. 29, 2014). --------------------------------------------------------------------------- By excluding open-end lines of credit from the loan-volume threshold, the proposed coverage test would not support that goal. Under the proposed institutional coverage test, institutions that originate large numbers of open-end lines of credit, but fewer than 25 closed-end mortgage loans, would not be required to report HMDA data on any of their loans. The proposed test may, therefore, exclude institutions with significant open-end lending, whose data may provide valuable insights into the open-end dwelling-secured market. The proposed test may also create an incentive for institutions to change their business practices to avoid reporting open-end data (e.g., by transferring all open-end lending to a separate subsidiary). This result would undermine the goals articulated in the section-by-section analysis of Sec. 1003.2(o) to increase visibility into open-end dwelling-secured lending. In addition to possibly excluding high volume open-end lenders, the proposed test may also burden some institutions with low open-end origination volumes with the requirement to report data concerning their open-end lending. The proposed institutional coverage test would require institutions with sufficient closed-end--but very little open- end--mortgage lending to incur costs to begin open-end reporting. As discussed in the section-by-section analysis of Sec. 1003.2(o) below, commencing reporting of open-end lines of credit, unlike continuing to report closed-end mortgage loans, represents a new, and in some cases significant, compliance burden. The proposal would have imposed these costs on small institutions with limited open-end lending, where the benefits of reporting the data do not justify the costs of reporting. In light of these considerations and those discussed in the section-by-section analysis of Sec. 1003.2(o), the Bureau concludes that only institutions that originate at least 100 open-end lines of credit in each of the two preceding calendar years should report HMDA data concerning open-end lines of credit. Accordingly, the Bureau is adopting a separate, open-end loan-volume threshold to determine whether an institution satisfies the definition of financial institution. The Bureau is also adopting transactional coverage thresholds, discussed below in the section-by-section analysis of Sec. 1003.3(c)(11) and (12). The institutional and transactional coverage thresholds are designed to operate in tandem. Under these thresholds, a financial institution will report closed-end mortgage loans only if it satisfies the closed-end mortgage threshold and will report open-end lines of credit only if it satisfies the separate open-end line credit threshold. The Bureau believes that adopting a 100-open-end line of credit threshold will avoid imposing the burden of establishing open-end reporting on many small institutions with low open-end lending volumes. Specifically, the Bureau estimates that almost 3,400 predominately smaller-sized institutions, that would have been required to begin open-end reporting under the proposal will not be required to report open-end data under the final rule.\151\ At the same time, the final rule [[Page 66150]] will improve the availability of data concerning open-end dwelling- secured lending by collecting data from a sufficient array of institutions and about a sufficient array of transactions. The Bureau estimates that nearly 90 percent of all open-end line of credit originations will be reported under the final rule.\152\ This change to institutional coverage, along with the finalization of mandatory reporting of all consumer-purpose open-end lines of credit, will improve the public and public officials' ability to monitor and understand all sources of dwelling-secured lending and the risks posed to consumers and communities by those loans. --------------------------------------------------------------------------- \151\ As the Bureau discussed in the proposal, due to the lack of available data concerning open-end lending, the Bureau has faced challenges in analyzing the impact on HMDA's institutional and transactional coverage of including open-end lines of credit. See 79 FR 51731, 51754 (Aug. 29, 2014). Although it solicited information that would assist it in making these estimates, see id., commenters did not provide responsive data. After careful analysis, the Bureau has developed rough estimates of home-equity line of credit origination volumes by institutions using 2013 HMDA data, 2013 Reports of Condition and Income (Call Report) data, and the Bureau's Consumer Credit Panel data. Given the scarcity of certain underlying data, these estimates rely on a number of assumptions. Nonetheless, for the reasons given above, including supporting increased visibility into the open-end line of credit market and reducing compliance burden for many institutions, the Bureau believes HMDA's purposes are best effectuated by adopting an open-end line of credit threshold. Part VII below discusses these estimates in more detail. \152\ See part VII. --------------------------------------------------------------------------- For those reasons, the Bureau is modifying Regulation C's definition of depository financial institution by adopting an open-end loan-volume threshold. Under the revised definition, an institution satisfies the definition of a depository financial institution if it meets the other applicable criteria and either originated at least 25 closed-end mortgage loans or 100 open-end lines of credit in each of the two preceding calendar years. Two-Year Look-Back Period The proposed loan-volume threshold provided in proposed Sec. 1003.2(g)(1)(v) considered only a financial institution's lending activity during the previous calendar year. The Bureau solicited feedback on whether to structure the loan-volume threshold over a multiyear period to provide greater certainty about the reporting requirements. Many industry commenters, including small entity representatives, urged the Bureau to include a multiyear look-back period in the loan-volume threshold. The Bureau believes that a two-year look-back period is advisable to eliminate uncertainty surrounding reporting responsibilities. Under the final rule, a financial institution that does not meet the loan- volume thresholds established in the final rule and that experiences an unusual and unexpected high origination-volume in one year will not be required to begin HMDA reporting unless and until the higher origination-volume continues for a second year in a row. A first-time HMDA reporter must undertake significant one-time costs that include operational changes, such as staff training, information technology changes, and document retention policies. Therefore, the Bureau believes that it is appropriate to develop a two-year look-back period for HMDA reporting to provide more stability around reporting responsibilities. Regulations that implement the Community Reinvestment Act provide similar look-back periods to determine coverage.\153\ --------------------------------------------------------------------------- \153\ See, e.g., 12 CFR 345.12(u)(1). --------------------------------------------------------------------------- Therefore, the Bureau is finalizing the loan-volume threshold included in Sec. 1003.2(g)(1)(v) and (2)(ii) with modifications to include a two-year look-back period. Sections 1003.2(g)(1)(v) and (2)(ii) provide that, assuming the other criteria are satisfied, an institution qualifies as a depository financial institution or a nondepository financial institution if the institution meets the applicable loan-volume threshold in each of the two preceding calendar years. Multifamily-Only Depository Institutions Under Regulation C, loans related to multifamily dwellings (multifamily mortgage loans) do not factor into the coverage criteria applicable to depository institutions. A depository institution that does not originate at least one home purchase loan or refinancing of a home purchase loan, secured by a first lien on a one- to four-unit dwelling in the preceding calendar year is not required to report HMDA data.\154\ The Bureau did not propose to eliminate the current loan activity test included in the coverage criteria for depository institutions. The proposal also did not solicit feedback on this aspect of the current coverage criteria or on other aspects of depository institutions' current coverage criteria. --------------------------------------------------------------------------- \154\ See 12 CFR 1003.2 (definition of financial institution). When HMDA was enacted, the term ``federally related mortgage loan'' was defined in the Real Estate Settlement Procedures Act (RESPA) to include a loan secured by real property secured by a first lien on a one- to four-family dwelling and that meets other federally related tests. See Public Law 93-533, section 3164, 88 Stat. 1724 (1974). --------------------------------------------------------------------------- Many community advocate commenters nonetheless urged the Bureau to expand depository institution coverage to require reporting by depository institutions that originate multifamily mortgage loans, but do not originate first-lien one- to four-unit home purchase loans or refinancings, and that meet the other coverage criteria. They argued that the current formulation makes it more difficult to understand availability of credit for multifamily dwellings. No industry commenters addressed this issue. The Bureau is not adopting the commenters' suggestion at this time. The Bureau recognizes that this prong of HMDA's depository institution coverage test may exclude certain depository institutions and their loans from HMDA data. However, the Bureau estimates that this provision excludes a very small number of depository institutions and loans, fewer than 20 institutions and about 200 covered loans under the final rule.\155\ --------------------------------------------------------------------------- \155\ The Bureau developed this estimate using 2013 Call Report data. --------------------------------------------------------------------------- The Bureau adopts Sec. 1003.2(g)(1) pursuant to its authority under section 305(a) of HMDA to provide for such adjustments and exceptions for any class of transactions that the Bureau judges are necessary and proper to effectuate the purposes of HMDA. Pursuant to section 305(a) of HMDA, for the reasons given above, the Bureau finds that this proposed exception is necessary and proper to effectuate the purposes of HMDA. By reducing burden on financial institutions and establishing a consistent loan-volume test applicable to all financial institutions, the Bureau finds that the proposed provision will facilitate compliance with HMDA's requirements. 2(g)(2) Nondepository Financial Institutions HMDA extends reporting responsibilities to certain nondepository institutions, defined as any person engaged for profit in the business of mortgage lending other than a bank, savings association, or credit union.\156\ HMDA section 309(a) also authorizes the Bureau to adopt an exemption for covered nondepository institutions that are comparable within their respective industries to banks, savings associations, and credit unions with $10 million or less in assets in the previous fiscal year.\157\ --------------------------------------------------------------------------- \156\ See generally HMDA sections 303(5) (defining ``other lending institutions''), 303(3)(B) (including other lending institutions in the definition of depository institution), and 304(a) (requiring depository institutions to collect, report, and disclose certain data if the institution has a home or branch office located in an MSA), 12 U.S.C. 2802(5), 2802(3), 2803(a). \157\ See HMDA section 309(a), 12 U.S.C. 2808(a). --------------------------------------------------------------------------- Under the current definition of financial institution in Sec. 1003.2, a nondepository institution is a financial institution if it meets three criteria. First, the institution satisfies the following loan-volume or amount test: In the preceding calendar year, the [[Page 66151]] institution originated home purchase loans, including refinancings of home purchase loans, that equaled either at least 10 percent of its loan-origination volume, measured in dollars, or at least $25 million.\158\ Second, on the preceding December 31, the institution had a home or branch office in an MSA.\159\ Third, the institution meets one of the following two criteria: (a) On the preceding December 31, the institution had total assets of more than $10 million, counting the assets of any parent corporation; or (b) in the preceding calendar year, the institution originated at least 100 home purchase loans, including refinancings of home purchase loans.\160\ --------------------------------------------------------------------------- \158\ The Board adopted the 10 percent loan-volume test in 1989 to implement the 1989 FIRREA amendments, which extended HMDA's reporting requirements to institutions ``engaged for profit in the business of mortgage lending.'' See 54 FR 51356, 51358-59 (Dec. 15, 1989). In 2002, the Board modified the test and added the $25 million loan-volume test to require reporting by additional nondepository institutions. See 67 FR 7222, 7224 (Feb. 15, 2002). \159\ Under Sec. 1003.2 (definition of branch office), a nondepository institution has a branch office in an MSA if it originated, received applications for, or purchased five or more covered loans in that MSA in the preceding calendar year. \160\ In 1989, the $10 million asset test, derived from section 309, applied to both depository and nondepository institutions. See 54 FR 51356, 51359 (Dec. 15, 1989). Because the 1989 amendments failed to cover as many nondepository institutions as Congress had intended, in 1991, Congress amended the asset test in HMDA section 309 to apply only to depository institutions, and it granted the Board discretion to exempt comparable nondepository institutions. See Public Law 102-242, section 224 (1991). Pursuant to that authority, the Board added the 100 loan-volume test for nondepository institutions in 1992. See 57 FR 56963, 56964-65 (Dec. 2, 1992). --------------------------------------------------------------------------- The Bureau proposed to modify the coverage criteria for nondepository institutions by replacing the current loan-volume or amount test with the same loan-volume threshold that the Bureau proposed for depository institutions. Proposed Sec. 1003.2(g)(2) defined a new term, nondepository financial institution, and provided that an institution that is not a bank, saving association, or credit union is required to report HMDA data if it had a home or branch office in an MSA on the preceding December 31 and it originated at least 25 covered loans, excluding open-end lines of credit, in the preceding calendar year. For the reasons discussed below, the Bureau is adopting Sec. 1003.2(g)(2), which revises the coverage criteria applicable to nondepository institutions. Under the final rule, a nondepository institution is a nondepository financial institution and required to report HMDA data if it has a home or a branch office in an MSA and if it originated at least 25 closed-end mortgage loans in each of the two preceding calendar years or 100 open-end lines of credit in each of the two preceding calendar years. Loan-Volume Threshold Most of the industry comments on this issue opposed the proposed expansion of nondepository institution coverage. These commenters explained that the proposed expansion would add only a small amount of additional data. Commenters also raised concerns about the burden on the nondepository institutions that would be newly covered. Some commenters suggested excluding more nondepository institutions from HMDA's institutional coverage, rather than expanding coverage of nondepository institutions, by adopting a loan-volume threshold higher than 100 closed-end mortgage loans annually, such as one set at origination of 250 closed-end mortgage loans annually. On the other hand, several consumer advocate commenters and a few industry commenters expressed support for the proposed expansion of nondepository institution coverage, arguing that nondepository institutions, like depository institutions, should be held accountable for their lending practices. The Bureau believes, as stated in the proposal, that it is important to increase visibility into nondepository institutions' practices due to their history of riskier lending practices, including their role in the financial crisis, and the lack of available data about lower-volume nondepository institutions' mortgage lending practices. Therefore, the Bureau is adopting Sec. 1003.2(g)(2), which requires reporting if the institution meets the location test and originated at least 25 closed-end mortgage loans in each of the two preceding calendar years or 100 open-end lines of credit in each of the two preceding calendar years. The Bureau estimates that the final rule will require HMDA reporting by as much as 40 percent more nondepository institutions than are currently reporting.\161\ --------------------------------------------------------------------------- \161\ As discussed in part VII below, the Bureau developed this estimate using 2012 HMDA data and NMSLR data. --------------------------------------------------------------------------- The expansion of nondepository institution reporting will address the longstanding need for additional monitoring of the mortgage lending practices of nondepository institutions. During the years leading up to the financial crisis, many stakeholders called for increased monitoring of nondepository institution activity in the mortgage market. Concerns about nondepository institution involvement in the subprime market motivated the Board to expand nondepository institution coverage in 2002.\162\ In 2007, the GAO also identified risks associated with the lending practices of nondepository institutions, which were not subject to regular Federal examination at the time.\163\ GAO found that 21 of the 25 largest originators of subprime and Alt-A loans in 2006 were nondepository institutions and that those 21 nondepository institutions had originated over 80 percent in dollar volume of the subprime and Alt-A loans originated in 2006.\164\ GAO concluded that nondepository institutions ``may tend to originate lower-quality loans.'' \165\ In 2009, GAO found that nondepository institutions that reported HMDA data had a higher incidence of potential fair lending problems than depository institutions that reported HMDA data.\166\ GAO also suggested that the loan products and marketing practices of those nondepository institutions may have presented greater risks for applicants and borrowers.\167\ --------------------------------------------------------------------------- \162\ See 65 FR 78656, 78657 (Dec. 15, 2000) (proposing changes to coverage of nondepository institutions); 67 FR 7222, 7224-25 (Feb. 15, 2002) (finalizing changes to coverage of nondepository institutions). \163\ See U.S. Gov't Accountability Office, GAO-08-78R, Briefing to the House of Representatives Committee on Fin. Services, Information on Recent Default and Foreclosure Trends for Home Mortgages and Associated Economic and Market Dev., at 54 (2007), available at http://www.gao.gov/assets/100/95215.pdf. \164\ Id. \165\ Id. \166\ See U.S. Gov't Accountability Office, GAO-09-704, Fair Lending: Data Limitations and the Fragmented U.S. Financial Regulatory Structure Challenge Federal Oversight and Enforcement Efforts at 28-29 (2009) (``[I]ndependent lenders and nonbank subsidiaries of holding companies are more likely than depository institutions to engage in mortgage pricing discrimination.''), available at http://www.gao.gov/new.items/d09704.pdf. \167\ Id. at 29-30. See also GAO-08-78R at 54. --------------------------------------------------------------------------- In the aftermath of the financial crisis, Congress also expressed concerns about the lending practices of nondepository institutions generally and called for greater oversight of those institutions.\168\ In the Dodd-Frank Act, Congress granted Federal supervisory authority to the Bureau over a broad range of mortgage-related nondepository [[Page 66152]] institutions because it was concerned about nondepository institutions' practices generally and believed that the lack of Federal supervision of those institutions had contributed to the financial crisis.\169\ In addition, officials that participated in the Financial Crisis Inquiry Commission hearings in 2010 noted that practices that originated in the nondepository institution mortgage sector, such as lax underwriting standards and loan products with potential payment shock, created competitive pressures on depository institutions to follow the same practices, which may have contributed to the broader financial crisis.\170\ During the Board's 2010 Hearings, community advocates and Federal agencies specifically urged expansion of HMDA's institutional coverage to include lower-volume nondepository institutions. They stated that Regulation C's existing institutional coverage framework prevented them from effectively monitoring the practices of nondepository institutions.\171\ --------------------------------------------------------------------------- \168\ See, e.g., House Consideration of HR 4173, 155 Cong. Rec. H14430 (daily ed. Dec. 9, 2009) (statement of Cong. Ellison), ``One of the most important causes of the financial crisis, as I mentioned, is the utter failure of consumer protection. The most abusive and predatory lenders were not federally regulated, were not regulated at all in some cases, while regulation was overly lax for banks and other institutions that were covered.''); U.S. Gov't Accountability Office, GAO-09-704, Fair Lending: Data Limitations and the Fragmented U.S. Financial Regulatory Structure Challenge Federal Oversight and Enforcement Efforts at 28-29 (2009), available at http://www.gao.gov/new.items/d09704.pdf. \169\ See Dodd-Frank Act section 1024. \170\ See Official Transcript of First Public Hearing of the Financial Crisis Inquiry Commission at 97-98 (Jan. 10, 2010), (remarks of Sheila C. Bair, Chairman, Federal Deposit Insurance Corporation, and Mary L. Schapiro, Chairman, U.S. Securities and Exchange Commission), available at http://fcic-static.law.stanford.edu/cdn_media/fcic-testimony/2010-0114-Transcript.pdf. \171\ See, e.g., San Francisco Hearing, supra note 42; Washington Hearing, supra note 39 (remarks of Faith Schwartz, Senior Advisor, HOPE NOW Alliance) (urging reporting by all institutions that have ``any meaningful originations''); id. (remarks of Allison Brown, Acting Assistant Director, Division of Financial Practices, Federal Trade Commission) (urging expanded reporting by nondepository institutions ``to ensure that all nondepository institutions that made significant numbers of mortgage decisions report these essential data, providing the government and the public an accurate, timely picture of mortgage lending activity''); id. (remarks of Michael Bylsma, Director for Community and Consumer Law, Office of the Comptroller of the Currency) (urging the Board to ``review whether its rule-making authority'' would permit it to expand HMDA coverage to additional institutions); Atlanta Hearing, supra note 40. --------------------------------------------------------------------------- Despite these calls for increased monitoring of nondepository institutions, currently there are less publicly available data about nondepository institutions' mortgage lending practices than about those of depository institutions. Currently, under Regulation C, lower-volume depository institutions may be required to report even if they originated only one mortgage loan in the preceding calendar year, but lower-volume nondepository institutions may not be required to report unless they originated 100 applicable loans in the preceding calendar year.\172\ In addition, outside of HMDA, there are less publicly available data about nondepository institutions than about depository institutions. Depository institutions, even those that do not report HMDA data, report detailed financial information at the bank level to the Federal Deposit Insurance Corporation (FDIC) or to the National Credit Union Association (NCUA), much of which is publicly available.\173\ Nondepository institutions, on the other hand, report some data to the Nationwide Mortgage Licensing System and Registry (NMLSR), but detailed financial information and data on mortgage applications and originations are not publicly available.\174\ --------------------------------------------------------------------------- \172\ Banks, savings associations, and credit unions are required to report if they originate at least one home purchase or refinancing of a home purchase loan secured by a first lien on a one- to four-family dwelling and if they meet the other criteria in the definition of financial institution. See Section 1003.2 (definition of financial institution). \173\ Every national bank, State member bank, and insured nonmember bank is required by its primary Federal regulator to file consolidated Reports of Condition and Income, also known as Call Report data, for each quarter as of the close of business on the last day of each calendar quarter (the report date). The specific reporting requirements depend upon the size of the bank and whether it has any foreign offices. See, e.g., FDIC, Call and Thrift Financial Reports, http://www2.fdic.gov/call_tfr_rpts/. Credit unions are also required to report Call Report data to NCUA. See, e.g., NCUA, 53000 Call Report Quarterly Data, http://www.ncua.gov/DataApps/QCallRptData/Pages/default.aspx. \174\ NMLSR is a national registry of nondepository institutions. Nondepository institutions report information about mortgage loan originators, mortgage loan originations, the number and dollar amount of loans brokered, and HOEPA originations. --------------------------------------------------------------------------- The final rule addresses this information gap by including the same loan-volume threshold for nondepository institutions as for depository institutions. The expanded coverage of nondepository institutions will provide more data to the public and public officials for analyzing whether lower-volume nondepository institutions are serving the housing needs of their communities. In addition, with the expanded coverage, the public and public officials will be better able to understand access to and sources of credit in particular communities, such as a higher concentration of risky loan products in a given community, and to identify the emergence of new loan products or underwriting practices. In addition, the final rule will provide more data to help the public and public officials in understanding whether a lower-volume nondepository institution's practices pose potential fair lending risks. The final rule also considers origination of open-end lines of credit in the institutional coverage test for nondepository institutions. The Bureau believes that this revision is necessary to achieve greater visibility into all extensions of credit secured by a dwelling, as discussed above in the section-by-section analysis of Sec. 1003.2(g)(1). In addition, for the reasons discussed above in the section-by-section analysis of Sec. 1003.2(g)(1), the final rule also incorporates a two-year look-back period for nondepository institution coverage. Asset-Size or Loan-Volume Threshold The current coverage criteria for nondepository institutions include an asset-size or loan-volume threshold.\175\ This test is satisfied both by institutions that meet a certain asset-size threshold and by those with smaller asset sizes that have a higher loan- volume.\176\ The Bureau proposed to eliminate the asset-size or loan- volume threshold for nondepository institutions currently included in Regulation C because, for the reasons discussed above, the Bureau believes it is important to increase visibility into the practices of nondepository institutions. A few industry commenters objected to the proposal's elimination of the asset-size portion of the asset-size or loan-volume threshold for nondepository institutions. The Bureau believes that the current asset-size or loan-volume threshold is no longer necessary, because the Bureau is adopting the 25 closed-end mortgage loan-volume threshold and 100 open-end line of credit threshold discussed in this section. Under the final rule, nondepository institutions will be required to report if they originated 25 closed-end mortgage loans or 100 open-end lines of credit in each of the two preceding calendar years. An institution's asset- size will no longer trigger reporting (i.e., nondepository institutions with assets greater than $10 million that originated fewer than 25 closed-end mortgage loans or fewer than 100 open-end lines of credit in each of the two preceding calendar years will not be required to report HMDA data). In addition, at this time and in light of the coverage criteria being finalized, the Bureau does not believe the asset-size exemption is necessary. The Bureau believes that it is appropriate to exercise its discretion under HMDA section 309(a) to eliminate the exemption of certain nondepository institutions based on their asset- size.\177\ --------------------------------------------------------------------------- \175\ Section 1003.2 (financial institution) (2). \176\ Section 1003.2 (financial institution) (2)(B)(iii). \177\ The Bureau consulted with HUD as part of the interagency consultation process for this rulemaking. --------------------------------------------------------------------------- [[Page 66153]] Loan-Amount or Loan-Volume Threshold No commenters discussed the proposed new implementation of HMDA sections 303(3)(B) and 303(5), which require persons other than banks, savings associations, and credit unions that are ``engaged for profit in the business of mortgage lending'' to report HMDA data. As the Bureau stated in the proposal, the Bureau interprets these provisions, as the Board also did, to evince the intent to exclude from coverage institutions that make a relatively small volume of mortgage loans.\178\ In light of more recent activities of nondepository institutions discussed above, the Bureau believes that Regulation C's current coverage test for nondepository institutions inappropriately excludes certain persons that are engaged for profit in the business of mortgage lending. The Bureau estimates that financial institutions that reported 25 loans in HMDA for the 2012 calendar year originated an average of approximately $5,359,000 in covered loans annually. Given this level of mortgage activity, and consistent with the policy reasons discussed above, the Bureau interprets ``engaged for profit in the business of mortgage lending'' to include nondepository institutions that originated at least 25 closed-end mortgage loans or 100 open-end lines of credit in each of the two preceding calendar years. Due to the questions raised about potential risks posed to applicants and borrowers by nondepository institutions and the lack of other publicly available data sources about nondepository institutions, the Bureau believes that requiring additional nondepository institutions to report HMDA data will better effectuate HMDA's purposes. --------------------------------------------------------------------------- \178\ See 54 FR 51356, 51358-59 (Dec. 15, 1989). --------------------------------------------------------------------------- 2(h) Home-Equity Line of Credit Regulation C currently defines ``home-equity line of credit'' as an open-end credit plan secured by a dwelling as defined in Regulation Z (Truth in Lending), 12 CFR part 1026. The Bureau did not propose to change this definition. Existing Sec. 1003.4(c)(3), in turn, provides that financial institutions optionally may report home-equity lines of credit made in whole or in part for home improvement or home purchase purposes. As discussed in the section-by-section analysis of Sec. 1003.2(e) and (o), the Bureau proposed to expand Regulation C's transactional coverage to require reporting of all home-equity lines of credit. As part of the shift to dwelling-secured coverage, the Bureau proposed a separate definition for ``open-end lines of credit'' in Sec. 1003.2(o), to reflect the proposed coverage of both consumer- and commercial-purpose lines of credit. As proposed, Sec. 1003.2(o) generally defined an open-end line of credit as a dwelling-secured transaction that was an open-end credit plan under Regulation Z Sec. 1026.2(a)(20), but without regard to whether the transaction: (1) Was for personal, family, or household purposes; (2) was extended by a creditor; or (3) was extended to a consumer. In other words, the proposal defined ``open-end line of credit'' broadly to include any dwelling-secured open-end credit transaction, whether for consumer or commercial purposes, and regardless of who was extending or receiving the credit. In general, then, the proposed definition of open-end line of credit included all transactions covered by the existing definition of home-equity line of credit in Sec. 1003.2. For the reasons discussed below, the final rule removes the term ``home-equity line of credit'' from the regulation, reserves Sec. 1003.2(h), and retains the term ``open-end line of credit.'' As discussed in the section-by-section analysis of Sec. 1003.2(o), the Bureau received a large number of comments about its proposal to require reporting of all dwelling-secured open-end lines of credit, and those comments are addressed in that section. One commenter specifically addressed the Bureau's proposal to define both ``home- equity line of credit'' and ``open-end line of credit.'' The commenter supported adding a definition for ``open-end line of credit'' but believed that distinguishing between open-end lines of credit and home- equity lines of credit was confusing. The commenter suggested that the Bureau streamline the types of covered transactions into dwelling- secured closed-end mortgage loans, dwelling-secured open-end lines of credit, and reverse mortgages (whether closed- or open-end). As discussed in the section-by-section analysis of Sec. 1003.2(o), the final rule adopts the proposed definition of open-end line of credit largely as proposed. For simplicity, the final rule removes the defined term ``home-equity line of credit'' and retains the defined term ``open-end line of credit'' to refer to all open-end credit transactions covered by the regulation. The final rule requires financial institutions to report whether a transaction is an open-end line of credit (Sec. 1003.4(a)(37)), a commercial- or business-purpose transaction (Sec. 1003.4(a)(38)), or a reverse mortgage (Sec. 1003.4(a)(36)). Using this information, it will be possible to determine whether a given open-end line of credit primarily is for consumer purposes (i.e., a home-equity line of credit) or primarily is for commercial or business purposes, and also whether it is a reverse mortgage. The Bureau thus believes that it is unnecessary to retain the defined term ``home-equity line of credit.'' 2(i) Home Improvement Loan Proposed Sec. 1003.2(i) provided that a home improvement loan was any covered loan made for the purpose, in whole or in part, of repairing, rehabilitating, remodeling, or improving a dwelling, or the real property on which the dwelling is located. Pursuant to the Bureau's authority under HMDA section 305(a), the proposal revised Sec. 1003.2(i) and its accompanying commentary to conform to the proposal to remove non-dwelling-secured home improvement loans from coverage, and to clarify when to report dwelling-secured home improvement loans. For the reasons discussed below, the Bureau is finalizing Sec. 1003.2(i) largely as proposed, with certain technical revisions to the regulation text,\179\ and with revisions to the commentary to streamline it and to add examples or details requested by commenters. --------------------------------------------------------------------------- \179\ For example, the final rule replaces the term ``covered loan'' in Sec. 1003.2(i) with the terms ``closed-end mortgage loan'' and ``open-end line of credit.'' This change reflects the fact that, under final Sec. Sec. 1003.2(e) and 1003.3(c)(10), business- or commercial-purpose transactions are covered loans only if they are for the purpose of home purchase, home improvement, or refinancing. Retaining the term ``covered loan'' in the definition of home improvement loan would cause circularity in the definition of commercial-purpose transactions. --------------------------------------------------------------------------- The Bureau received numerous comments from consumer advocacy groups, financial institutions, trade associations, and other industry participants concerning proposed Sec. 1003.2(i). Most of the comments focused on the proposal to exclude non-dwelling-secured home improvement loans from reporting, with nearly all industry participants supporting the proposal and consumer advocacy groups generally opposing it. A few commenters requested that the Bureau clarify certain aspects of the commentary to the home improvement loan definition. Non-Dwelling-Secured Home Improvement Lending Consumer advocacy groups uniformly stated that the Bureau should maintain reporting of home improvement lending, because such lending has been particularly important to low- and [[Page 66154]] moderate-income borrowers and borrowers of color as a way to finance home repairs. Most of these commenters did not specifically distinguish between dwelling-secured and non-dwelling-secured home improvement lending or specify how they use non-dwelling-secured home improvement lending data, in particular, to achieve HMDA's purposes. One financial institution urged the Bureau to retain reporting of non-dwelling-secured home improvement lending, at least on an optional basis. This commenter stated that non-dwelling-secured home improvement lending can be critical in revitalizing low-to-moderate income communities, including in rural areas, and for financing manufactured home improvements. The commenter expressed concern that financial institutions might stop offering non-dwelling-secured home improvement loans if they were no longer HMDA-reportable. The commenter believed that borrowers would be steered toward home-equity lines of credit, which might be unavailable to low- and moderate-income borrowers with inadequate home equity. The commenter argued that optional reporting would relieve burden for institutions that choose not to report, while allowing institutions that do report to receive credit for serving the housing needs of their communities. All other industry commenters that addressed proposed Sec. 1003.2(i) supported excluding non-dwelling-secured home improvement loans from coverage.\180\ Many of these commenters stated that reporting such loans is burdensome and costly because it is difficult to determine whether the loan will be used for a housing-related purpose, because reporting errors occur frequently, and because examiners have not treated non-dwelling-secured home improvement lending consistently. Other commenters noted that the value of non- dwelling-secured home improvement loan data is limited. Interest rates and terms can vary dramatically depending on the loan and non-dwelling collateral used, and consumers now often use home-equity lines of credit. One commenter stated that the burdens of reporting can outweigh the benefits of making the loans, because non-dwelling-secured home improvement loan amounts tend to be small. --------------------------------------------------------------------------- \180\ Various commenters recommended eliminating the home improvement purpose category for all loans. Those comments are addressed in the section-by-section analysis of Sec. 1003.4(a)(3). --------------------------------------------------------------------------- The Bureau is finalizing Sec. 1003.2(i) as proposed, without a requirement to report non-dwelling-secured home improvement loans. At this time, the Bureau does not believe that the benefits of requiring reporting of such loans justify the burdens. For example, many consumer advocacy group commenters urged the Bureau to retain reporting of all home improvement loans because such loans are important to low-to- moderate income communities. These commenters, however, did not state that they or others have used HMDA data about non-dwelling-secured home improvement loans to further HMDA's purposes. Moreover, as discussed in the proposal, non-dwelling-secured home improvement loans may have been common when HMDA was enacted. However, such loans now comprise only a small fraction of transactions reported,\181\ and borrowers have other non-dwelling-secured credit options, such as credit cards, to fund home improvement projects.\182\ Data about credit card usage for home improvement purposes, however, is not reported under HMDA. Without such data, it is not clear that HMDA users can evaluate fairly the non- dwelling-secured home improvement loan data that is reported. --------------------------------------------------------------------------- \181\ See 79 FR 51731, 51755 (Aug. 29, 2014) (noting that non- dwelling-secured home improvement loans comprised only approximately 1.8 percent of all HMDA records in 2012). \182\ See id. at 51755, 51765-66 (Aug. 29, 2014) (citing Arthur Kennickell & Martha Starr-McCluer, Bd. of Governors of the Fed. Reserve Sys., 80 Fed. Reserve Bulletin 861, Changes in Family Finances from 1989 to 1992: Evidence from the Survey of Consumer Finances, at 874-75 (Oct. 1994), available at http://www.federalreserve.gov/econresdata/scf/files/1992_bull1094.pdf; Arthur Kennickell & Janice Shack-Marquez, Bd. of Governors of the Fed. Reserve Sys., 78 Fed. Reserve Bulletin 1, Changes in Family Finances from 1983 to 1989: Evidence from the Survey of Consumer Finances, at 13 (Jan. 1992), available at http://www.federalreserve.gov/econresdata/scf/files/bull0192.pdf; and David Evans & Richard Schnakebsee, Paying With Plastic, Massachusetts Institute of Technology Press 98-100 (1991)). --------------------------------------------------------------------------- On the other hand, the burdens of reporting such transactions appear to be significant. As discussed in the proposal, these loans are processed, underwritten, and originated through different loan origination systems than are used for dwelling-secured lending.\183\ As noted above, many industry commenters confirmed and elaborated on the burdens of reporting non-dwelling-secured home improvement loans discussed in the Bureau's proposal. --------------------------------------------------------------------------- \183\ See id. at 51755 (Aug. 29, 2014) (citing Chicago Hearing, supra note 46 and Atlanta Hearing, supra note 40). --------------------------------------------------------------------------- On balance, the Bureau concluded that the compliance burden that will be eliminated by streamlining the regulation to require reporting only of dwelling-secured loans justifies the relatively small data loss that will accompany the change. The Bureau considered, as one commenter suggested, maintaining optional reporting of non-dwelling-secured home improvement loans. However, one of the proposal's goals was to simplify Regulation C's transactional coverage. Maintaining optional reporting of non-dwelling-secured home improvement loans would inhibit the Bureau's ability to reduce regulatory complexity by focusing on dwelling-secured lending for an apparently small benefit.\184\ Thus, the final rule requires financial institutions to report only dwelling- secured loans. Unsecured home improvement loans and home improvement loans secured by collateral other than a dwelling (e.g., a vehicle or savings account), are not reportable. --------------------------------------------------------------------------- \184\ The Bureau acknowledges that removing non-dwelling-secured home improvement lending will affect some institutions' reported transaction volumes, which in turn will affect CRA reporting. The Bureau will work with other regulators during the Regulation C implementation period to address these issues. --------------------------------------------------------------------------- One commenter objected that the Bureau's proposal to eliminate reporting of non-dwelling-secured home improvement loans did not address the fact that the HMDA statute still requires reporting of home improvement loans. The Bureau believes, however, that requiring financial institutions to report dwelling-secured home improvement loans satisfies the statutory requirement. As the proposal noted, HMDA does not expressly define ``home improvement loan.'' Although non- dwelling-secured home improvement loans traditionally have been reported, the Bureau believes that it is reasonable to interpret HMDA section 303(2) to include only loans that are secured by liens on dwellings, as that interpretation aligns with common definitions of the term mortgage loan, and such loans will include home improvement loans.\185\ --------------------------------------------------------------------------- \185\ See 79 FR 51731, 51755-56 (Aug. 29, 2014). --------------------------------------------------------------------------- The Bureau also is eliminating reporting of non-dwelling-secured home improvement loans pursuant to its authority under section 305(a) of HMDA, as the Bureau believes that this adjustment and exception is necessary and proper to effectuate HMDA's purposes and to facilitate compliance. Specifically, the Bureau believes that non-dwelling-secured home improvement loan data may distort the overall quality of the HMDA dataset for the reasons described above. The Bureau believes that eliminating reporting of non-dwelling-secured home improvement loans will improve the [[Page 66155]] quality of HMDA data, which will provide citizens and public officials of the United States with sufficient information to enable them to determine whether financial institutions are meeting the housing needs of their communities and to assist public officials in determining how to distribute public sector investments in a manner designed to improve the private investment environment. The Bureau further believes that eliminating these loans will facilitate compliance by removing a significant reporting burden. Home Improvement Loan Definition A few commenters asked the Bureau to clarify certain aspects of home improvement loan reporting as addressed in the commentary. The final rule adopts the commentary generally as proposed, but with several revisions and additions to address commenters' questions, as well as certain other modifications for clarity, as discussed below. Proposed comment 2(i)-1, which provided general guidance about home improvement loans, is adopted as proposed, but with several non- substantive revisions for clarity and with an additional example of a transaction that meets the home improvement loan definition. Consistent with the final rule's requirement under Sec. 1003.2(d) to report loans completed pursuant to a New York CEMA, final comment 2(i)-1 explains that, where all or a portion of the funds from a CEMA transaction will be used for home improvement purposes, the loan is a home improvement loan under Sec. 1003.2(i). One commenter asked whether loans that are not ``classified'' by an institution as home improvement loans nonetheless should be reported as home improvement loans if the supporting documents show that they were for home improvement purposes. The classification test in existing Regulation C applies only to non- dwelling-secured home improvement loans. As discussed, the final rule eliminates such loans from coverage. Under the final rule, there no longer is any requirement that a loan be ``classified'' by a financial institution as a home improvement loan to be a home improvement loan under Sec. 1003.2(i). The Bureau did not propose to revise existing comment Home improvement loan-3. The final rule adopts this comment as comment 2(i)- 3 with minor revisions to reflect the fact that the final rule requires reporting of both closed- and open-end transactions. Proposed comment 2(i)-4 concerning mixed-use properties is adopted largely as proposed. The comment is revised for clarity and to eliminate the statement that a home improvement loan for a mixed-use property is reported as such only if the property itself is primarily residential in nature. Under Sec. 1003.2(e) and (f), a transaction is a covered loan and subject to Regulation C only if it is secured by a dwelling, which by definition is property that is primarily residential in nature. Thus, financial institutions need not separately consider whether a dwelling primarily is residential in nature when determining whether a loan is a home improvement loan under Sec. 1003.2(i). The proposal would have removed existing comment Home improvement loan-5, which discusses how to report a home improvement loan that also is a home purchase loan or a refinancing, because the proposal consolidated all such reporting instructions in Sec. 1003.4. The final rule retains existing comment Home improvement loan-5 and adopts it as comment 2(i)-5 to explain that a transaction with multiple purposes may meet multiple loan-type definitions. The comment provides an example to illustrate that a transaction that meets the definition of a home improvement loan under Sec. 1003.2(i) may also meet the definition of a refinancing under Sec. 1003.2(p). Comment 2(i)-5 also specifies that instructions for reporting a multiple-purpose covered loan are in the commentary to Sec. 1003.4(a)(3). A few commenters asked the Bureau to clarify further how a financial institution determines whether a loan is a home improvement loan. For example, one commenter asked whether a cash-out refinance also is a home improvement loan if the borrower states that some of the cash may be used for home improvement. Another asked whether a loan is a home improvement loan when a consumer states that a loan is for home improvement purposes but it is in fact for purchasing a household item. This commenter also requested that ``small-dollar'' home improvement loans be exempt from reporting. In response to these comments, the final rule includes comment 2(i)-6, which provides that a financial institution relies on the borrower's stated purpose for the loan when the application is received or the credit decision is made, and need not confirm that the borrower actually uses any of the funds for home improvement purposes. If the borrower does not state that any of the funds will be used for home improvement purposes, or does not state any purpose for the funds, the loan is not a home improvement loan. Section 1003.4(a)(3) and related commentary provide instructions about how to report such loans. See the section-by-section analysis of Sec. 1003.4(a)(3). The final rule does not specifically exempt small-dollar home improvement loans, because the Bureau believes that information about such loans is valuable, but the final rule retains in Sec. 1003.3(c)(7) the current exclusion from coverage for transactions for less than $500. 2(j) Home Purchase Loan Regulation C currently provides that a home purchase loan is a loan secured by and made for the purpose of purchasing a dwelling. Proposed Sec. 1003.2(j) revised the definition to provide that a home purchase loan is a ``covered loan'' extended for the purpose of purchasing a dwelling. The proposal also revised the commentary to proposed Sec. 1003.2(j) in several ways, primarily to conform the commentary to the proposal's overall shift to covering only dwelling-secured transactions. Only a handful of commenters addressed proposed Sec. 1003.2(j) or its accompanying commentary, and none of them specifically commented on the proposed regulation text. The Bureau is finalizing Sec. 1003.2(j) largely as proposed, with technical revisions for clarity.\186\ The Bureau is finalizing the commentary to Sec. 1003.2(j) with revisions to address questions that commenters raised regarding assumptions, to clarify how Regulation C applies to multiple- purpose transactions, and to remove certain comments as unnecessary. --------------------------------------------------------------------------- \186\ Specifically, the final rule replaces the term ``covered loan'' in Sec. 1003.2(j) with the terms ``closed-end mortgage loan'' and ``open-end line of credit.'' This change reflects the fact that, under final Sec. Sec. 1003.2(e) and 1003.3(c)(10), business- or commercial-purpose transactions are covered loans only if they are for the purpose of home purchase, home improvement, or refinancing. Retaining the term ``covered loan'' in the definition of home purchase loan would cause circularity in the definition of commercial-purpose transactions. --------------------------------------------------------------------------- First, the Bureau is not adopting proposed comment 2(j)-1 in the final rule. Proposed comment 2(j)-1 provided general guidance about the definition of home purchase loan, including an illustrative example stating that a home purchase loan includes a closed-end mortgage loan but does not include a home purchase completed through an installment contract. No commenters addressed proposed comment 2(j)-1. The final rule incorporates the terms ``closed-end mortgage loan'' and ``open-end credit plan'' in Sec. 1003.2(j). Thus, there is no need to restate in commentary that a closed-end mortgage loan used to purchase a dwelling is a home purchase loan. The Bureau is finalizing the illustrative example discussing installment contracts in commentary to Sec. 1003.2(d), which [[Page 66156]] provides guidance about the term closed-end mortgage loan. See the section-by-section analysis of Sec. 1003.2(d). The proposal renumbered as proposed comment 2(j)-2 existing comment Home purchase loan-1, which provides that a home purchase loan includes a loan secured by one dwelling and used to purchase another dwelling. Two industry commenters stated that ``home purchase loan'' should not include these loan types and recommended that they be defined instead as ``home-equity loans.'' The commenters stated that, under Regulation Z, a loan is not a home purchase loan (i.e., a ``residential mortgage transaction'' under Regulation Z Sec. 1026.2(a)(24)) unless its funds are used to purchase the property securing the dwelling. The commenters stated that industry stakeholders generally view loans secured by one dwelling but used to purchase a different dwelling as home-equity loans, not as purchase loans. Revising Sec. 1003.2(j) in the way that commenters suggested would better align Regulations C and Z. In general, regulatory consistency is desirable; however, HMDA's purposes are different from Regulation Z's purposes. To understand how financial institutions are meeting the housing needs of their communities, it is important to understand the total volume of loans made to purchase dwellings, even if those loans are secured by dwellings other than the ones being purchased. The suggested revision also would require adding a new defined term, home- equity loan, to Regulation C. This term necessarily would lump together loans secured by one dwelling, but used to purchase, improve, or refinance loans on other dwellings; reporting the loans in this way would obscure the valuable information described above. Thus, the Bureau is finalizing proposed comment 2(j)-2 largely as proposed, with certain non-substantive revisions for clarity, and renumbered as comment 2(j)-1. The Bureau received no comments addressing proposed comment 2(j)-3, which made only minor revisions to existing comment Home purchase loan- 2 addressing whether a transaction to purchase a mixed-use property is a home purchase loan. However, the final rule eliminates this comment as unnecessary. As proposed, the comment stated that a transaction to purchase a mixed-use property is a home purchase loan if the property primarily is used for residential purposes, and it provided guidance about how to determine the primary use of the property. Under the final rule, a transaction is not covered by Regulation C unless it is secured by a dwelling, which is defined under Sec. 1003.2(f) to include only mixed-use properties that primarily are used for residential purposes. Because financial institutions will have determined under Sec. 1003.2(f) whether a mixed-used property is a dwelling, there is no need to reevaluate that decision when determining whether a transaction is a home purchase loan. Consistent with the proposal's consolidation of excluded transactions into Sec. 1003.3(c), the proposal moved existing comment Home purchase loan-3, which discusses agricultural-purpose loans, to proposed comment 3(c)(9)-1. No commenters addressed this reorganization, and the Bureau is finalizing proposed comment 3(c)(9)- 1, with revisions, as discussed in the section-by-section analysis of Sec. 1003.3(c)(9). The proposal did not propose to revise existing comments Home purchase loan-4, -5, or -6, and the Bureau received no comments addressing them. These comments are adopted in the final rule as comments 2(j)-2 through -4, respectively, with minor revisions for clarity. As discussed in the section-by-section analysis of Sec. 1003.1(c) regarding Regulation C's scope, the proposal reorganized the commentary to Sec. 1003.1(c). Consistent with that reorganization, the proposal incorporated a revised version of existing comment 1(c)-9, which discusses coverage of assumptions, as comment 2(j)-7 to the definition of home purchase loan. One industry commenter addressed this comment. The commenter argued that proposed comment 2(j)-7 should specify, consistent with Regulation Z, that a successor-in-interest transaction is not an assumption. The final rule adopts proposed comment 2(j)-7 as comment 2(j)-5, with revisions to address the comment received, and with other clarifying revisions, as follows. First, comment 2(j)-5 states that an assumption is a home purchase loan only if the transaction is to finance the new borrower's acquisition of the property (and not, e.g., if the borrower has succeeded in interest to ownership).\187\ Also, consistent with Sec. 1003.2(d) and comment 2(d)-2.ii, which provide that transactions documented pursuant to New York consolidation, extension and modification agreements are extensions of credit, comment 2(j)-5 clarifies that a transaction in which borrower B finances the purchase of borrower A's dwelling by assuming borrower A's existing debt obligation is a home purchase loan even if the transaction is documented pursuant to a New York consolidation, extension, and modification agreement. --------------------------------------------------------------------------- \187\ However, as discussion in the section-by-section analysis of Sec. 1003.2(d), the final rule provides that successor-in- interest transactions are assumptions for purposes of Regulation C. --------------------------------------------------------------------------- The Bureau proposed to remove existing comment Home purchase loan- 7, which described how to report multiple-purpose home-purchase loans, because the proposal consolidated all such reporting instructions in Sec. 1003.4. The final rule retains as comment 2(j)-6 a variation of existing comment Home purchase loan-7 to explain that a transaction with multiple purposes may meet multiple loan-type definitions. The comment provides an illustrative example and specifies that instructions for reporting a multiple-purpose loan are in the commentary to Sec. 1003.4(a)(3). Two commenters requested additional guidance about the definition of home purchase loan. One commenter stated that additional guidance is necessary because there are several ways to transfer property ownership to third parties, not all of which are called a ``purchase.'' The commenter did not specify the other methods it was referencing. As discussed, comment 2(j)-5 provides guidance about two additional methods of title transfer. The Bureau can address additional scenarios in the future, if necessary. Another commenter requested guidance about whether a loan to one sibling to purchase half of another sibling's home, which the other sibling owns outright, is a reportable home purchase loan or a refinancing when the loan is secured by the portion of the home being purchased. Based on the details provided, such a transaction is reportable, because it is a dwelling-secured loan and is not excluded under Sec. 1003.3(c). Because it is for the purpose of purchasing a dwelling, and it does not satisfy and replace an existing, dwelling-secured debt obligation, it is a home purchase loan but it is not a refinancing. See the section-by-section analysis of Sec. 1003.2(p). 2(k) Loan/Application Register Regulation C requires financial institutions to collect and record reportable data in a format prescribed by the regulation. The Bureau proposed to refer to this format as the ``loan application register'' to improve the readability of the regulation and proposed to define it as a register in the format prescribed in appendix A. The Bureau did not receive any comments on this proposed definition. As [[Page 66157]] explained in part I.B above, in order to streamline the regulation, the final rule removes appendix A. Therefore, the Bureau is revising proposed Sec. 1003.2(k) to remove references to appendix A and defining loan/application register to mean both the record of information required to be collected pursuant to Sec. 1003.4 and the record submitted annually or quarterly, as applicable, pursuant to Sec. 1003.5(a). In addition, the Bureau is adding ``/'' to maintain consistency with the term as currently used and to clarify that the data recorded represents applications as well as loan originations. Accordingly, the Bureau is adopting Sec. 1003.2(k) with revisions. 2(l) Manufactured Home The Bureau proposed to make technical corrections and minor wording changes to the definition of manufactured home. Commenters generally supported aligning the definition of manufactured home with the HUD standards and clarifying that other factory-built homes and recreational vehicles are excluded. Other comments related to coverage and reporting of manufactured homes and similar residential structures are discussed in the section-by-section analysis of Sec. 1003.2(f) and Sec. 1003.4(a)(5). The Bureau is finalizing Sec. 1003.2(l) generally as proposed, with minor revisions. The definition is revised to clarify that, for purposes of the construction method reporting requirement under Sec. 1003.4(a)(5), a manufactured home community should be reported as manufactured home. The Bureau received no specific feedback on proposed comments 2(l)-1 and -2, which are adopted as proposed. 2(m) Metropolitan Statistical Area (MSA) and Metropolitan Division (MD) Section 1003.2 of Regulation C sets forth a definition for the terms ``metropolitan statistical area or MSA'' and ``Metropolitan Division or MD.'' The Bureau is adopting a technical amendment to this definition and its commentary. No substantive change is intended. 2(n) Multifamily Dwelling The Bureau proposed to add a new definition of multifamily dwelling as Sec. 1003.2(n). Commenters supported the Bureau's proposal to define a multifamily dwelling as one that includes five or more individual dwelling units. A few commenters also supported the inclusion of manufactured home parks, as discussed in the proposal.\188\ Some commenters requested clarification on the reporting requirements for multifamily dwellings. Other comments related to multifamily residential structures are addressed in the section-by- section analysis of Sec. 1003.2(f). The Bureau is finalizing Sec. 1003.2(n) as proposed. In response to the requests for clarification, the Bureau is also adding two comments related to the definition of multifamily dwelling. Comment 2(n)-1 clarifies how the definition interacts with the definition of dwelling and its reference to multifamily residential structures. Comment 2(n)-2 clarifies the special reporting requirements applicable to multifamily dwellings. --------------------------------------------------------------------------- \188\ 79 FR 51731, 51749 (Aug. 29, 2014); Fed. Fin. Insts. Examination Council, CRA/HMDA Reporter, Changes Coming to HMDA Edit Reports in 2010 (Dec. 2010), available at http://www.ffiec.gov/hmda/pdf/10news.pdf. --------------------------------------------------------------------------- 2(o) Open-End Line of Credit HMDA section 303(2) defines ``mortgage loan'' as a residential real property-secured loan or a home improvement loan but does not specifically address coverage of open-end lines of credit secured by dwellings. Regulation C also currently does not define the term ``open- end line of credit.'' However, as discussed in the section-by-section analysis of Sec. 1003.2(h), Regulation C currently defines the term ``home-equity line of credit'' as an open-end credit plan secured by a dwelling as defined in Regulation Z. Under existing Regulation C Sec. 1003.4(c)(3), financial institutions may, but are not required to, report home-equity lines of credit made in whole or in part for home purchase or home improvement purposes.\189\ Commercial-purpose lines of credit secured by a dwelling fall outside of Regulation Z's definition of open-end credit plan and thus are not optionally reported as home- equity lines of credit under existing Regulation C. --------------------------------------------------------------------------- \189\ Under existing Regulation C Sec. 1003.4(a)(7) and comment 4(a)(7)-3, if a financial institution opts to report home-equity lines of credit, it reports only the portion of the line intended for home improvement or home purchase. --------------------------------------------------------------------------- In 2000, in response to the increasing importance of open-end lending in the housing market, the Board proposed to revise Regulation C to require mandatory reporting of all home-equity lines of credit.\190\ The Board's proposal was based on research showing that about 70 percent of all home-equity lines of credit were being used at least in part for home improvement purposes. The Board believed that requiring reporting of all home-equity lines of credit would provide more complete information about the home improvement market, one of HMDA's original purposes.\191\ The Board's 2002 final rule concluded that, while collecting data on home-equity lines of credit would give a more complete picture of the home mortgage market, the benefits of mandatory reporting relative to other proposed changes (such as collecting information about higher-priced loans) did not justify the increased burden.\192\ The Board thus decided to retain optional reporting. --------------------------------------------------------------------------- \190\ Home-equity lines of credit were rare in the 1970s and early 1980s when Regulation C was first implemented. Regulation C first addressed home-equity lines of credit in 1988, when it permitted financial institutions to report home-equity lines of credit that were home improvement loans. See 53 FR 31683, 31685 (Aug. 19, 1988). \191\ See 65 FR 78656, 78659-60 (Dec. 15, 2000). \192\ See 67 FR 7222, 7225 (Feb. 15, 2002). --------------------------------------------------------------------------- Open-end mortgage lending continued to increase in the years following the Board's 2002 final rule, and the Board continued to receive feedback urging such lending to be reported in HMDA.\193\ The Bureau received similar feedback after it assumed rulemaking authority for HMDA from the Board in 2011.\194\ The feedback suggested that home- equity lines of credit have become increasingly important to the housing market and that requiring such lending to be reported under Regulation C would help to understand how financial institutions are meeting the housing needs of communities. The Bureau thus proposed to require financial institutions to report all home-equity lines of credit, as well as all commercial-purpose lines of credit secured by a dwelling. --------------------------------------------------------------------------- \193\ See, e.g., Donghoon Lee et al., Fed. Reserve Bank of New York, Staff Report No. 569, A New Look at Second Liens, at 11 (Aug. 2012) (approximately $20 billion in home-equity lines of credit were originated in the fourth quarter of 1999; by the fourth quarter of 2005, approximately $125 billion in HELOCs were originated). See generally, e.g., Atlanta Hearing, supra note 40; San Francisco Hearing, supra note 42; Chicago Hearing, supra note 46; Washington Hearing, supra note 39. \194\ See, e.g., National Community Reinvestment Coalition et al., Creating Comprehensive HMDA and Loan Performance Databases: White Paper Submitted to the Consumer Financial Protection Bureau at 15 (Feb. 15, 2013), available at http://www.empirejustice.org/assets/pdf/policy-advocacy/consumer-organizations-urge.pdf. --------------------------------------------------------------------------- Specifically, the Bureau proposed new Sec. 1003.2(o) to define the term ``open-end line of credit,'' which included any dwelling-secured open-end credit plan, as described under Regulation Z Sec. 1026.2(a)(20), even if the credit was issued by someone other than a creditor (as defined in Regulation Z Sec. 1026.2(a)(17)), to someone other than a consumer (as defined in Regulation Z Sec. 1026.2(a)(11)) and for business rather than consumer purposes (as defined in [[Page 66158]] Regulation Z Sec. 1026.2(a)(12)). Together with proposed Sec. 1003.2(e), which provided that all open-end lines of credit were ``covered loans,'' proposed Sec. 1003.2(o) provided that financial institutions must report: (1) all consumer-purpose home-equity lines of credit, which currently are optionally reported, and (2) all dwelling- secured commercial-purpose lines of credit, which currently are not reported. In short, the proposal provided for reporting of all dwelling-secured open-end lines of credit.\195\ --------------------------------------------------------------------------- \195\ 79 FR 51731, 51757-59 (Aug. 29, 2014). --------------------------------------------------------------------------- As discussed below and in the section-by-section analyses of Sec. 1003.2(e) and (g) and of Sec. 1003.3(c)(10) and (12), the Bureau is finalizing mandatory reporting of open-end lines of credit, but with certain modifications from the proposal to: (1) Limit the number of institutions that will report; (2) limit the number of transactions that will be reported; (3) clarify certain reporting requirements for open-end lines of credit; and (4) clarify the definition of ``open-end line of credit.'' As discussed below, the Bureau believes that finalizing mandatory reporting of open-end lines of credit will provide information critical to HMDA's purposes. The Bureau understands that, notwithstanding the modifications described above, financial institutions may incur significant costs as a result of open-end line of credit reporting. However, the Bureau believes that the benefits of reporting justify the burdens. The Bureau received a large number of comments about proposed Sec. 1003.2(o). The vast majority of the comments discussed whether reporting of dwelling-secured open-end lines of credit should be mandatory and, if so, the scope of transactions that should be reported. A few commenters raised specific questions about the proposed definition of open-end line of credit. Consumer advocacy group commenters and researchers favored mandatory reporting, while the majority of industry commenters strongly opposed it. Among industry commenters that addressed mandatory reporting, most objected to reporting any open-end lines of credit. Some, however, specifically objected to mandatory reporting of commercial-purpose lines of credit. For organizational purposes, the Bureau addresses in this section-by- section analysis comments about: (1) Open-end line of credit coverage generally; (2) consumer-purpose line of credit coverage specifically; and (3) the definition of ``open-end line of credit'' in proposed Sec. 1003.2(o).\196\ Comments specific to commercial-purpose lines of credit are addressed in the section-by-section analysis of Sec. 1003.3(c)(10) concerning commercial-purpose transactions. --------------------------------------------------------------------------- \196\ Many commenters used the common phrase ``home-equity lines of credit'' or ``HELOC'' to discuss all open-end mortgage lending. For simplicity and to align with the final rule's deletion of the defined term ``home-equity line of credit'' from Regulation C (see the section-by-section analysis of Sec. 1003.2(h)), the Bureau hereinafter refers to covered (i.e., dwelling-secured) open-end transactions simply as consumer- or commercial-purpose ``open-end lines of credit.'' --------------------------------------------------------------------------- Consumer advocacy group commenters and researchers favored mandatory reporting of all consumer-purpose open-end lines of credit. A large number of these commenters stated that data about open-end lines of credit would be valuable in assessing whether neighborhoods are receiving the full range of credit that they need on nondiscriminatory terms. The commenters stated that open-end lines of credit are much more widely used today than when HMDA was enacted, that problematic practices were associated with these products during the 2000s, that defaults on open-end credit lines contributed significantly to the foreclosure crises in many neighborhoods, and that open-end credit lines are important sources of home improvement financing, particularly in minority and immigrant communities. The commenters stated that fully understanding the mortgage market, including problems relating to overextension of credit in minority and immigrant neighborhoods, requires more detailed information about such transactions. They stated that information about home-equity products, for example, is important for understanding the total amount of debt and, in turn, default risk on a property. A few consumer advocacy group commenters noted that open-end lines of credit, especially when fully drawn at account opening, can be interchangeable with closed-end products such as closed-end, subordinate-lien loans and cash-out refinancings. All such products provide borrowers with cash to do something, borrowers face the same risks of discrimination, and borrowers put their homes on the line in exchange for the funds. Commenters argued that requiring reporting of all dwelling-secured closed-end mortgage loans while continuing optional reporting of open-end lines of credit only would encourage more open-end lending, which in turn would decrease visibility into home-secured lending. Finally, one commenter noted that there is a lack of other publicly available information about dwelling-secured open-end lines of credit. A minority of industry commenters either supported (or stated that they did not oppose) reporting consumer-purpose open-end lines of credit.\197\ A few of these commenters argued that eliminating optional open-end line of credit reporting for consumer-purpose credit lines would reduce confusion and compliance costs by streamlining reporting obligations, or would improve data for HMDA users. Some industry commenters believed that data about consumer-purpose open-end lines of credit would serve HMDA's purposes. For example, one industry commenter acknowledged that, even though reporting open-end lines of credit would be burdensome, the data reported would provide additional information for fair lending use. --------------------------------------------------------------------------- \197\ Industry commenters unanimously opposed reporting dwelling-secured commercial-purpose open-end lines of credit. The Bureau addresses those comments in the section-by-section analysis of Sec. 1003.3(c)(10). --------------------------------------------------------------------------- A large number of industry commenters objected to mandatory reporting of consumer-purpose open-end lines of credit; a few of these commenters suggested that only credit lines for home purchase, home improvement or refinancing should be reported. Commenters generally asserted that mandatory reporting would impose significant burdens for little benefit. Some argued that the burdens are what have kept most financial institutions from voluntarily reporting home-equity line of credit data under current Regulation C. Financial institutions of various types and sizes objected to mandatory reporting, but smaller- or medium-sized banks and their industry associations, and credit unions and their industry associations, generally expressed the greatest concerns, with some stating that open-end coverage was their primary concern with the proposal. A primary concern among many financial institutions and industry associations, and particularly among many credit unions and credit union associations, was the operational costs and burdens of collecting and reporting data about open-end lines of credit. The most commonly cited operational difficulty was that financial institutions treat open-end lines of credit more like consumer loans than mortgage loans. Thus, financial institutions frequently originate and maintain data about open-end lines of credit on different computer systems than traditional mortgages, or use different software vendors. Commenters asserted that upgrading, replacing, or programming their systems [[Page 66159]] to enable open-end reporting would be difficult, expensive, and time- consuming. For example, financial institutions could use their mortgage loan origination systems for open-end reporting, but those systems are more expensive than the consumer systems typically used for credit lines. Commenters noted that, if financial institutions decided to keep separate systems for open- and closed-end credit, they would incur costs from programming and adding data fields in multiple systems, as well as from compiling and aggregating the data. For some (smaller) institutions, aggregating the data would mean manually entering data from two different systems. Some commenters similarly observed that financial institutions use different departments, staff, and processes to originate open-end lines of credit and traditional mortgages. Commenters argued that open-end reporting would require financial institutions to incur costs to change their operations. For example, consumer lending staff either would need to be trained on HMDA reporting, or credit line originations would need to be moved from the consumer- to the mortgage-lending divisions of financial institutions. Some commenters also argued that reporting open-end lines of credit would require institutions to spend even more time and money on quality control and pre-submission auditing and would increase the risk of errors. A number of commenters perceived other types of operational burdens. For example, a few commenters emphasized that the reporting burden would be particularly great because it would be entirely new even for most current HMDA reporters, so infrastructure would need to be built from the ground up. A few commenters similarly worried that some institutions that focus on open-end lending would become HMDA reporters for the first time and would incur significant start-up expenses to begin reporting. Finally, some commenters noted that aligning open-end lending with the MISMO data standards would be burdensome. Many industry commenters argued that reporting all open-end line of credit applications and originations would increase institutions' ongoing HMDA reporting costs because their volume of reportable transactions would increase significantly. Some commenters asserted that the proposal underestimated the increase. Only a handful of commenters specifically estimated how many additional applications and originations they would be required to report. Estimated increases ranged from 20 percent to 200 percent per institution, or from hundreds to thousands of transactions, depending on the institution's size and volume of open-end mortgage lending. Many commenters, particularly smaller institutions, stated that they would need to hire additional staff, or that they would need to allocate more money to technology and staff, to handle the volume increase. A few commenters estimated that collecting data about all dwelling-secured open-end lines of credit would double or triple their ongoing compliance costs. Several commenters also argued that reporting open-end lines of credit would be burdensome because gathering and accurately reporting information about credit lines would be difficult. For example, several industry associations stated that fewer data are gathered from consumers for small-dollar, open-end credit lines than for traditional mortgages, so lenders would need to create systems and procedures to collect the data. One commenter further noted that lines of credit are consumer loan products with different offerings by different institutions and are less standardized than traditional mortgages. Another commenter pointed out that open-end lines of credit are exempt from other regulations because they are different than closed-end loans. Some commenters stated that it would be burdensome to determine whether, and if so how, data points apply to open-end lines of credit. These commenters asserted that reporting open-end lines of credit thus could increase reporting errors. A few of these commenters were particularly concerned about the Bureau's proposal to require information about the first draw on a home-equity line of credit. Many commenters argued that, in addition to being burdensome, reporting open-end lines of credit would have few benefits. First, many commenters asserted that mandatory reporting would exceed HMDA's mission and that the data reported would not serve HMDA's purposes. They argued that the data would not show whether financial institutions were meeting the housing needs of their communities because open-end lines of credit often are used for personal, non-housing-related, purposes (e.g., vacations, education, and bill consolidation). Some commenters argued that data about credit lines used for non-housing- related purposes would produce misleading information about mortgage markets and that reporting should be limited, at most, to credit lines for home purchase, home improvement, or refinancing purposes. Others asserted that, even if a consumer intended to use a line of credit for a housing-related purpose, such as home improvement, financial institutions could not know at account opening whether the borrower ever actually drew on the account or, if so, whether the funds were used for housing or other purposes. The commenters thus asserted that the data reported would not be useful. Some commenters argued that data about open-end lines of credit would not serve HMDA's fair lending purpose, because borrowers taking out open-end credit lines borrow against the equity in their homes and are not fully assessed as new borrowers. A few commenters asserted that it was inappropriate for the Bureau to require open-end reporting for market monitoring and research purposes or to address safety and soundness concerns. One commenter argued that open-end lines of credit are less risky for consumers than closed-end loans, because they often are smaller, with smaller payments that are easier to make. Another argued that the change was not required by the Dodd-Frank Act. Many commenters also argued that mandating reporting of open-end lines of credit would be of little benefit, because certain current and proposed data points (e.g., results from automated underwriting systems, some pricing data, and whether a transaction has non- amortizing features) would not apply to open-end credit. In addition, many commenters stated that mixing data about open-end, ``consumer- purpose'' transactions with traditional, closed-end mortgage loans will skew HMDA data and impair its integrity for HMDA users. Finally, a few commenters noted that the Board previously had considered and rejected mandatory reporting of open-end lines of credit; these commenters asserted that the Board had found that open-end reporting would not serve HMDA's purposes. A few smaller financial institutions, credit unions, and credit union leagues predicted that they or other small institutions could be forced to stop offering open-end lines of credit. Others argued that adding open-end line of credit reporting would strain the limited resources of smaller banks and credit unions already struggling with burdensome compliance requirements, would inhibit such institutions from serving their customers, would increase costs to consumers and credit union members, or could force such institutions to exit the market for home-equity lines of credit, thereby reducing consumers' low-cost credit options. Commenters suggested various alternatives to mandatory reporting of open-end lines of credit. Some urged the [[Page 66160]] Bureau to maintain optional reporting, while others asserted that open- end lines of credit should be excluded from reporting altogether. Some argued that smaller- or medium-sized banks and credit unions should be exempt from reporting, because small institutions did not cause the financial crisis and reporting would burden them unfairly. As noted, a few commenters urged the Bureau to require reporting only of open-end lines of credit for home purchase, home improvement, or refinancing purposes. The Bureau has considered the comments concerning mandatory reporting of open-end lines of credit, and the Bureau is finalizing Sec. 1003.2(o) largely as proposed, but without covering certain commercial-purpose lines of credit.\198\ The Bureau is finalizing separate open-end line of credit coverage thresholds under Sec. 1003.2(g) and Sec. 1003.3(c)(12) to ensure that only financial institutions with a minimum level of open-end line of credit originations will be required to report.\199\ The Bureau acknowledges that, even with these modifications, many financial institutions may incur significant costs to report their open-end lines of credit, and that one-time costs may be particularly large. However, the Bureau believes that the benefits of mandatory reporting justify those costs. --------------------------------------------------------------------------- \198\ See the section-by-section analysis of Sec. 1003.3(c)(10). \199\ See the section-by-section analyses of Sec. 1003.2(g)(1)(v)(B) and (2)(ii)(B) and of Sec. 1003.3(c)(12). --------------------------------------------------------------------------- As discussed in the proposal, the Bureau believes that including dwelling-secured lines of credit within the scope of Regulation C is a reasonable interpretation of HMDA section 303(2), which defines ``mortgage loan'' as a loan secured by residential real property or a home improvement loan. The Bureau interprets ``mortgage loan'' to include dwelling-secured lines of credit, as they are secured by residential real property and they may be used for home improvement purposes.\200\ Moreover, pursuant to section 305(a) of HMDA, the Bureau believes that requiring reporting of all dwelling-secured, consumer- purpose open-end lines of credit is necessary and proper to effectuate the purposes of HMDA and to prevent circumvention of evasion thereof.\201\ HMDA and Regulation C are designed to provide citizens and public officials sufficient information about mortgage lending to ensure that financial institutions are serving the housing needs of their communities, to assist public officials in distributing public sector investments, and to identify possible discriminatory lending patterns. The Bureau believes that collecting information about all dwelling-secured, consumer-purpose open-end lines of credit serves these purposes.\202\ --------------------------------------------------------------------------- \200\ See 79 FR 51731, 51758-59 (Aug. 29, 2014). \201\ See id. at 51759. \202\ Contrary to some commenters' assertions, the Board did not find that open-end reporting would not serve HMDA's purposes; rather, the Board in 2002 determined that the burdens of open-end reporting did not justify the benefits at that time. --------------------------------------------------------------------------- First, financial institutions will know, and the data will show, when an open-end line of credit is being taken out for the purpose of purchasing a home. This data alone will serve HMDA's purposes by providing information about how often, on what terms, and to which borrowers' institutions are originating open-end lines of credit to finance home purchases. Although many commenters argued that dwelling- secured lines of credit are used for purposes unrelated to housing, in the years leading up to the financial crisis, they often were made and fully drawn more or less simultaneously with first-lien home-purchase loans (i.e., as piggybacks), essentially creating high loan-to-value ratio home-purchase transactions that were not visible in the HMDA dataset.\203\ Some evidence suggests that piggyback lending may be on the rise again now that the market has begun to recover from the crisis.\204\ --------------------------------------------------------------------------- \203\ See, e.g., Donghoon Lee et al., Fed. Reserve Bank of New York, Staff Report No. 569, A New Look at Second Liens, at 11 (Aug. 2012) (estimating that, prior to the crisis, as many as 45 percent of purchasers in coastal and bubble areas used a piggyback loan to subsidize the down payment on a first mortgage, hoping to eliminate the need for mortgage insurance). \204\ See, e.g., Joe Light and AnnaMaria Andriotis, Borrowers Tap Their Homes at a Hot Clip, Wall St. J., May 29, 2014), available at http://www.wsj.com/articles/borrowers-tap-their-homes-at-a-hot-clip-1401407763 (discussing the recent increase in home-equity line of credit lending and noting that some lenders have begun to bring back piggyback loans, which ``nearly vanished'' during the mortgage crisis). --------------------------------------------------------------------------- Second, the data will help to understand how often, on what terms, and to which borrowers institutions are originating open-end lines of credit for home improvement purposes. It is true, as many commenters argued, that funds from lines of credit may be used for many purposes, and that lenders cannot track how funds ultimately are used. However, the same is true of funds obtained through cash-out refinancings, which currently are reported under Regulation C, and through closed-end home- equity loans and reverse mortgages, some of which are reportable today and all of which will be reportable under the final rule (unless an exception applies).\205\ Funds from all of these products may be used for personal purposes, but they may also be used for home improvement (and home purchase) purposes. Citizens and public officials long have analyzed data about such products to understand how financial institutions are satisfying borrowers' needs for home improvement lending.\206\ --------------------------------------------------------------------------- \205\ As discussed in the section-by-section analysis of Sec. 1003.2(q), commenters raised some of the same concerns about reverse mortgages. The final rule requires reporting of all reverse mortgages for the reasons discussed in the section-by-section analysis of Sec. 1003.2(q). \206\ For example, financial institutions currently report closed-end home-equity loans when borrowers indicate that some or all of the funds will be used for home improvement purposes. Financial institutions, however, do not track what portion (if any) of the funds ultimately are used for that purpose. No data reporting regime can provide perfect information; the information that is reported nevertheless assists in serving HMDA's purposes. --------------------------------------------------------------------------- The Bureau believes that financial institutions serve the housing needs of their communities not only by providing fair and adequate financing to purchase and improve homes, but also by ensuring that neither individual borrowers nor particular communities are excessively overleveraged through open-end home-equity borrowing. The Bureau thus declines to limit reporting of open-end mortgage lending to transactions for home purchase, home improvement, or refinancing purposes, as some commenters suggested. Open-end home-equity lending, regardless of how the funds are used, liquefies equity that borrowers have built up in their homes, which often are their most important assets. Borrowers who take out dwelling-secured credit lines increase their risk of losing their homes to foreclosure when property values decline. Indeed, as discussed in the proposal, open-end line of credit originations expanded significantly during the mid-2000s, particularly in areas with high home-price appreciation, and research indicates that speculative real estate investors used open-end lines of credit to purchase non-owner-occupied investment properties, which correlated with higher first mortgage defaults and home-price depreciation.\207\ In short, overleverage due to open-end mortgage lending and defaults on dwelling-secured open-end lines of credit contributed to the foreclosure crises that many communities experienced in the late 2000s. Communities' housing needs would have been better served if these crises could have been avoided (or remedied more quickly).\208\ Had open- [[Page 66161]] end line of credit data been reported in HMDA, the public and public officials could have had a much earlier warning of potential risks. The Bureau believes that obtaining data about open-end mortgage lending remains critical, with open-end lending on the rise once again as home prices have begun to recover from the financial crisis.\209\ --------------------------------------------------------------------------- \207\ See 79 FR 51731, 51757 (Aug. 29, 2014). \208\ As noted in the proposal, many public and private mortgage relief programs encountered unique difficulties assisting distressed borrowers who had obtained subordinate-lien loans, including dwelling-secured open-end lines of credit. See 79 FR 51731, 51757 (Aug. 29, 2014). \209\ See, e.g., Press Release, Equifax, First Quarter Mortgage Originations Soar (June 29, 2015), http://investor.equifax.com/releasedetail.cfm?ReleaseID=919892 (stating that more than 285,700 new accounts were originated during the first quarter of 2015, a year-over-year increase of 21.2 percent and the highest level since 2008); CBA, Icon Market Analysis Finds Growing Consumer Demand for Home Equity Lines of Credit (Mar. 23, 2015) (home-equity line of credit originations have increased in each of the past 13 quarters, with annual growth of nearly 22 percent in both 2013 and 2014 and an increase of 36 percent for the first quarter of 2015 versus the first quarter of 2014); Joe Light and AnnaMaria Andriotis, Borrowers Tap Their Homes at a Hot Clip, Wall St. J., May 29, 2014), available at http://www.wsj.com/articles/borrowers-tap-their-homes-at-a-hot-clip-1401407763 (quoting the chief economist of Equifax Inc. that lenders had begun marketing more aggressively in areas where home prices had recovered and that originations had picked up as consumers had returned to home improvement projects postponed during the crisis). --------------------------------------------------------------------------- Finally, mandatory reporting of open-end lines of credit will help to understand whether all dwelling-secured credit is extended on equitable terms. It may be true, as some commenters asserted, that borrowers are not necessarily evaluated for open-end credit in the same manner as for traditional mortgage loans and that adequate home equity is the key consideration. Lending practices during the financial crisis demonstrated, however, that during prolonged periods of home-price appreciation, lenders became increasingly comfortable originating home- equity products to borrowers with less and less equity to spare. The more leveraged the borrower, the more at risk the borrower is of losing his or her home. Obtaining data about open-end mortgage lending could show, during future housing booms, whether such risky lending practices are concentrated among certain borrowers or communities and permit the public and public officials to respond appropriately. In this and other ways, data about open-end lines of credit will help to assist in identifying possible discriminatory lending patterns. Certain commenters pointed out that several data points will not apply to open-end lines of credit. However, the Bureau believes that the public and public officials will receive valuable information from all of the data points that do apply. With applicable data points, HMDA users will have, for the first time, good information about which financial institutions are originating open-end lines of credit, how frequently, on what terms, and to which borrowers. HMDA users will be able to evaluate whether, and how, financial institutions are using open-end lines of credit to serve the housing needs of their communities. Moreover, as discussed below, the final rule adopts several measures to minimize the burdens to financial institutions of determining whether and how data points apply to open-end lines of credit.\210\ The final rule also requires financial institutions to flag whether a transaction is for closed- or open-end credit. See Sec. 1003.4(a)(37). This flag addresses commenters' concerns about commingling information about closed-end mortgage loans and open-end lines of credit.\211\ --------------------------------------------------------------------------- \210\ Some commenters were concerned that financial institutions would be required to report the portion of the open-end line of credit that would be used for home purchase, home improvement, or refinancing purposes. However, the final rule, like the proposal, requires financial institutions to report the total amount of the line at account opening. See the section-by-section analysis of Sec. 1003.4(a)(7). \211\ Indeed, commingling of information is more a of problem under existing Regulation C than it will be under the final rule, because there currently is no way for users of HMDA data to distinguish information about optionally reported open-end lines of credit from the rest of the HMDA dataset. --------------------------------------------------------------------------- Not only will data about open-end lines of credit help to serve HMDA's purposes, but the Bureau believes that expanding the scope of Regulation C to include dwelling-secured, consumer-purpose lines of credit is necessary to prevent evasion of HMDA. As discussed in the proposal, consumer-purpose open-end lines of credit may be interchangeable with consumer-purpose closed-end home-equity products, many of which currently are reported, and all of which will be reported, under final Sec. 1003.2(d) and (e). The Bureau believes that, if open- and closed-end consumer-purpose home-equity products are treated differently under the final rule, there is a heightened risk that financial institutions could steer borrowers to open-end products to avoid HMDA reporting.\212\ The Bureau believes that steering could be particularly attractive (and risky for borrowers) given that open- end lines of credit are not subject to the Bureau's 2013 ATR Final Rule and currently are subject to less complete disclosure requirements than closed-end products under Regulation Z. The Bureau believes that some financial institutions likely would attempt to evade Regulation C if mandatory reporting were not adopted for open-end lines of credit. The Bureau thus has determined that, in addition to being a reasonable interpretation of the statute, requiring reporting of dwelling-secured, consumer-purpose open-end lines of credit also is authorized as an adjustment that is necessary and proper to prevent evasion of HMDA. --------------------------------------------------------------------------- \212\ See 79 FR 51731, 51758 (Aug. 29, 2014). The Bureau believes the risk of steering is highlighted by lending practices described during the Board's 2010 Hearings; for example, one individual described how a loan officer persuaded her to open a home-equity line of credit simultaneously with her primary mortgage, even though she had not inquired about or been interested in opening a line of credit. See id. --------------------------------------------------------------------------- The Bureau acknowledges that reporting open-end lines of credit will impose one-time and ongoing operational costs on reporting institutions. The proposal estimated that the one-time costs of modifying processes and systems and training staff to begin open-end line of credit reporting likely would impose significant costs on some institutions, and that institutions' ongoing reporting costs would increase as a function of their open-end lending volume.\213\ As discussed above, many commenters emphasized both these one-time and ongoing costs.\214\ The Bureau acknowledges these costs and understands that many institutions' reportable transaction volume many increase significantly. --------------------------------------------------------------------------- \213\ See id. at 51825-26, 51836-37 (estimating the one-time and ongoing costs, respectively, to low-, medium-, and high-complexity institutions of reporting open-end lines of credit, all dwelling- secured home-equity loans, and reverse mortgages). \214\ Certain commenters argued that the proposal underestimated the costs of reporting open-end lines of credit. Those comments are addressed in part VII, along with the methodology the Bureau has used to estimate the costs of open-end reporting, and the challenges the Bureau has faced in developing its estimates. --------------------------------------------------------------------------- As discussed in the proposal, in the section-by-section analysis of Sec. 1003.2(g), and in part VII below, the Bureau has faced challenges developing accurate estimates of the likely impact on institutional and transactional coverage of mandatory reporting of open-end lines of credit due to the lack of available data concerning open-end lending. These challenges affect the Bureau's ability to develop reliable one- time and ongoing cost estimates, as well, because such costs are a function of both the number of institutions reporting open-end data and the number of transactions each of those institutions reports.\215\ After careful analysis, the [[Page 66162]] Bureau has developed estimates of open-end line of credit origination volumes by institutions and, as discussed in part VII, has used those estimates to estimate both the overall one-time and overall ongoing costs to institutions of open-end reporting.\216\ The Bureau expects that both one-time and ongoing costs will be larger for more complex financial institutions that have higher open-end lending volume and that will need to integrate separate business lines, data platforms, and systems, to begin reporting open-end lending. Precisely because no good source of publicly available data exists concerning dwelling- secured open-end lines of credit, it is difficult to predict the accuracy of the Bureau's cost estimates, but the Bureau believes that they are reasonably reliable and acknowledges that, for many lenders, the costs of open-end reporting may be significant. As discussed further below, the final rule revises the proposal in several ways to reduce open-end reporting costs for certain financial institutions.\217\ --------------------------------------------------------------------------- \215\ The Bureau solicited information that would assist it in making these estimates and in determining whether the estimates provided in the proposal were accurate, but commenters generally did not provide responsive data. See 79 FR 51731, 51754 (Aug. 29, 2014). Some commenters argued, based on their particular institution's lending volume, that the Bureau underestimated the number of open- end lines of credit that institutions would be required to report. As discussed in part VII, the proposal's and the final rule's estimates of transaction volumes are averages. Thus, they may be low for some financial institutions and high for others. Moreover, some industry commenters did not distinguish between consumer- and commercial-purpose credit lines. As discussed in the section-by- section analysis of Sec. 1003.3(c)(10), the final rule requires financial institutions to report only a subset of commercial-purpose lines of credit. Thus, it is possible that some commenters overestimated the number of open-end transactions that they would report under the final rule. Based on available information, including the feedback provided by commenters, the Bureau cannot definitively conclude whether the proposal significantly underestimated reportable open-end line of credit volume as a general matter. \216\ As noted in part VII, with currently available sources, the Bureau can reliably estimate: (1) Total open-end line of credit originations in the market and (2) subordinate-lien open-end line of credit originations by credit union. Both of these estimates are under- and over-inclusive of the open-end transactions that are reportable under the final rule. Neither includes applications that do not result in originations, which will be reported, and both include commercial-purpose lines of credit, many of which will be excluded under final Sec. 1003.3(c)(10). For banks and thrifts, the Bureau's estimates of open-end line of credit originations have been extrapolated from several data sources using simplified assumptions and may not accurately reflect open-end lending by such institutions. \217\ The Bureau does not believe that open-end reporters will incur burden from aligning with MISMO. As discussed in part VII, the Bureau did not propose to require, and the final rule does not require, any financial institution to use or become familiar with the MISMO data standards. Rather, the rule merely recognizes that many financial institutions are already using the MISMO data standards for collecting and transmitting mortgage data and uses similar definitions for certain data points to reduce burden for those institutions. --------------------------------------------------------------------------- A few commenters argued that reporting open-end lines of credit will be difficult because financial institutions collect less information from consumers when originating open-end products than when originating traditional, closed-end mortgage loans. In part, this may be because open-end lines of credit are not subject to the Bureau's 2013 ATR Final Rule. However, the Bureau believes that this lack of substantive regulation only strengthens the need for open-end line of credit reporting in HMDA so that the public and policymakers have sufficient data about the dwelling-secured open-end credit market to understand whether lenders offering open-end products are serving the housing needs of their communities. Methods To Reduce the Burden of Open-End Line of Credit Reporting The Bureau is finalizing mandatory reporting of dwelling-secured consumer-purpose open-end lines of credit because of the many benefits discussed above. The Bureau is adopting several measures to address commenters' concerns about the burdens of implementing open-end reporting and their concerns about ongoing open-end reporting costs. Institutional coverage threshold. As discussed in the section-by- section analysis of Sec. 1003.2(g), the Bureau is finalizing a separate, open-end institutional coverage threshold to determine whether an institution is a HMDA reporter. As discussed in that section, the Bureau concluded that its proposed institutional coverage test achieved appropriate market coverage of closed-end mortgage lending. However, in light of the costs associated with open-end reporting, the Bureau was concerned that finalizing the proposed institutional coverage test would have required institutions with sufficient closed-end--but very little open-end--mortgage lending to incur costs to begin open-end reporting. The Bureau thus is adopting an institutional coverage test that covers a financial institution only if (in addition to meeting the other criteria under Sec. 1003.2(g)) it originated either (1) 25 or more closed-end mortgage loans or (2) 100 or more open-end lines of credit in each of the two preceding calendar years. As discussed in the section-by-section analysis of Sec. 1003.2(g), the Bureau believes that the 25 closed-end and 100 open-end loan-volume origination tests appropriately balance the benefits and burdens of covering institutions based on their closed- and open-end mortgage lending, respectively. Specifically, as discussed further in the section-by-section analysis of Sec. 1003.2(g) and in part VII, the Bureau estimates that adopting a 100-open-end line of credit threshold will avoid imposing the burden of establishing open-end reporting on approximately 3,000 predominantly smaller-sized institutions with low open-end lending compared to the proposal, while still requiring reporting of a significant majority of dwelling-secured, open-end line of credit originations. As discussed in those sections, the Bureau also believes that all institutions that will be required to report open-end line of credit data are current HMDA reporters. Transactional coverage threshold. The final rule also adds in Sec. 1003.3(c)(12) a transactional coverage threshold for open-end mortgage lending. The transactional coverage threshold is designed to work in tandem with the open-end institutional coverage threshold in Sec. 1003.2(g). Specifically, Sec. 1003.3(c)(12) provides that a financial institution that originated fewer than 100 open-end lines of credit in each of the preceding two calendar years is not required to report data about its open-end lines of credit, even if the financial institution otherwise is a financial institution under Sec. 1003.2(g) because of its closed-end lending (i.e., even if the institution will be reporting data about closed-end mortgage loans).\218\ --------------------------------------------------------------------------- \218\ For balance, the Bureau is adopting a parallel transactional coverage threshold for closed-end mortgage loans in Sec. 1003.3(c)(11). Under Sec. 1003.3(c)(11), a financial institution that originated fewer than 25 closed-end mortgage loans in each of the preceding two calendar years is not required to report data about its closed-end mortgage loans, even if the financial institution otherwise is a financial institution under Sec. 1003.2(g) because of its open-end mortgage lending (i.e., even if the institution will be reporting data about open-end lines of credit). --------------------------------------------------------------------------- Effective date. The Bureau is mindful that most financial institutions have never reported open-end mortgage lending data, that collecting and reporting such data for the first time will be time- consuming and complex, and that implementation costs may be sensitive to the time permitted to complete the required changes. The Bureau thus is providing financial institutions approximately two years to complete the changes necessary to begin collecting the data required under the final rule, including data about open-end lines of credit. As noted in part VI, financial institutions will report the data required under the final rule for actions taken on covered loans on or after January 1, 2018. Other efforts to mitigate burden. Some of the anticipated burdens of reporting open-end lines of credit also likely will be mitigated by the operational [[Page 66163]] enhancements and modifications that the Bureau is exploring for HMDA reporting generally. For example, as discussed elsewhere in the final rule, the Bureau is improving the edit and submission process, which should reduce reporting burden for all covered loans. While these improvements will not reduce the costs that financial institutions will incur to adapt their systems and processes to report open-end lines of credit, they should reduce ongoing costs to institutions by reducing the amount of time financial institutions may spend submitting and editing this data. Clarifying which data points apply to open-end lines of credit, and how they apply, also will alleviate compliance burden. For example, commenters expressed concern about reporting information about initial draws under open-end lines of credit. As discussed in the section-by- section analysis of Sec. 1003.4(a)(39), the Bureau is not finalizing that data point, in part in response to commenters' concerns. The final rule also provides that several other data points do not apply to open- end lines of credit.\219\ Finally, the final rule provides guidance about how several data points apply to open-end lines of credit.\220\ --------------------------------------------------------------------------- \219\ See Sec. 1003.4(a)(4) (preapproval request); Sec. 1003.4(a)(18) (origination charges); Sec. 1003.4(a)(19) (discount points); and Sec. 1003.4(a)(20) (lender credits). \220\ See Sec. 1003.4(a)(7)(ii) and comment 4(a)(7)-6 (loan amount); comments 4(a)(12)-3 and -4 (rate spread); Sec. 1003.4(a)(17) (total points and fees); comment 4(a)(25)-4 (amortization term); and comment 4(a)(26)-1 (introductory rate). --------------------------------------------------------------------------- Open-End Line of Credit Definition The Bureau is adopting a few technical revisions to streamline Sec. 1003.2(o) and to align it with revisions made elsewhere in the final rule. Proposed Sec. 1003.2(o) provided that an open-end line of credit was a dwelling-secured transaction that was neither a closed-end mortgage loan under proposed Sec. 1003.2(d) nor a reverse mortgage under proposed Sec. 1003.2(q). To align with lending practices, to streamline the definitions of closed-end mortgage loan and open-end line of credit, and to streamline Sec. 1003.4(a)(36) (which requires financial institutions to identify reverse mortgages), the final rule eliminates the mutual exclusivity between open-end lines of credit and reverse mortgages. Final Sec. 1003.2(o) thus provides that an open-end line of credit is an extension of credit that (1) is secured by a lien on a dwelling; and (2) is an open-end credit plan as defined in Regulation Z, 12 CFR 1026.2(a)(20), but without regard to whether the credit is consumer credit, as defined in Sec. 1026.2(a)(12), is extended by a creditor, as defined in Sec. 1026.2(a)(17), or is extended to a consumer, as defined in Sec. 1026.2(a)(11). Consistent with Sec. 1003.2(d), final Sec. 1003.2(o) provides that an open-end line of credit is a dwelling-secured ``extension of credit.'' New comment 2(o)-2 clarifies the meaning of the term ``extension of credit'' for open-end transactions for purposes of Sec. 1003.2(o). It states that financial institutions may cross-reference the guidance concerning ``extension of credit'' under Sec. 1003.2(d) and comment 2(d)-2, and it provides an example of an open-end transaction that is not an extension of credit and thus not covered under the final rule. It further clarifies that, for purposes of Sec. 1003.2(o), each draw on an open-end line of credit is not an extension of credit. Thus, financial institutions report covered open-end lines of credit only once, at account opening. 2(p) Refinancing Prior to the proposal, the Bureau received feedback that Regulation C's definition of refinancing was confusing. To address those concerns, the Bureau proposed Sec. 1003.2(p) and related commentary. Proposed Sec. 1003.2(p) streamlined the existing definition of refinancing by moving the portion of the definition that addresses institutional coverage to proposed Sec. 1003.2(g), the definition of ``financial institution.'' For the reasons discussed below, the Bureau is adopting Sec. 1003.2(p) largely as proposed, and is adopting revised commentary to Sec. 1003.2(p) to provide additional guidance about the types of transactions that are refinancings under Regulation C.\221\ --------------------------------------------------------------------------- \221\ Prior to the proposal and in public comments on the proposal, the Bureau received feedback that agricultural-purpose refinancings should be excluded from Regulation C's coverage. The final rule clarifies that all agricultural-purpose transactions, whether for home purchase, home improvement, refinancing, or some other purpose, are excluded transactions. See the section-by-section analysis of Sec. 1003.3(c)(9). --------------------------------------------------------------------------- The Bureau received a number of comments on proposed Sec. 1003.2(p) and its accompanying commentary from financial institutions, industry trade associations, and other industry participants. The comments generally supported the Bureau's proposed revisions, but several commenters suggested different definitions or additional clarifications. The Bureau received only a few comments addressing proposed Sec. 1003.2(p)'s regulation text, all from industry participants. One commenter specifically supported the Bureau's proposal to move the ``coverage prong'' of Sec. 1003.2(p) to the definition of financial institution in Sec. 1003.2(g) and stated that the move would reduce confusion. Another commenter suggested that the Bureau could reduce compliance costs by aligning the definition of refinancing in proposed Sec. 1003.2(p) with Regulation Z Sec. 1026.37(a)(9), so that a refinancing is any transaction that is not a home purchase loan and that satisfies and replaces an existing obligation secured by the same property. For the reasons set forth in the section-by-section analysis of Sec. 1003.4(a)(3), the final rule does not include this modification. The Bureau is finalizing comment 2(p)-1 generally as proposed, but with several non-substantive revisions for clarity. In addition, final comment 2(p)-1 is modified to provide that a refinancing occurs only when the original debt obligation has been satisfied and replaced by a new debt obligation, based on the parties' contract and applicable law. This is consistent with the definition of refinancing in Regulation Z Sec. 1026.20(a) and comment 20(a)-1. The comment further specifies that satisfaction of the original lien, as distinct from the debt obligation, is irrelevant in determining whether a refinancing has occurred. A few commenters requested that the Bureau provide additional guidance concerning loan modifications and renewals, stating that examiners provide inconsistent guidance about whether to report renewal transactions when there is no new note. Accordingly, final comment 2(p)-1 specifies that a new debt obligation that renews or modifies the terms of, but does not satisfy and replace, an existing debt obligation is not a refinancing under Sec. 1003.2(p).\222\ --------------------------------------------------------------------------- \222\ To further address uncertainty about the types of transactions that are reportable under Regulation C, the final rule also clarifies in the commentary to Sec. 1003.2(d) (definition of closed-end mortgage loan) and (o) (definition of open-end line of credit) that loan modifications and renewals are not ``extensions of credit'' under Regulation C and thus are not reportable transactions under the final rule. See the section-by-section analysis of Sec. 1003.2(d) and (o). --------------------------------------------------------------------------- As discussed in the section-by-section analysis of Sec. 1003.2(d), the final rule considers a transaction completed pursuant to a New York State consolidation, extension, and modification agreement and classified as a supplemental mortgage under N.Y. Tax Law Sec. 255 such that the borrower owes reduced or no mortgage recording taxes to be an ``extension of credit'' and therefore reportable. The final rule adds new comment 2(p)-2 to provide that a transaction is considered a refinancing under Sec. 1003.2(p) where: (1) The [[Page 66164]] transaction is completed pursuant to a New York State consolidation, extension, and modification agreement and is classified as a supplemental mortgage under N.Y. Tax Law Sec. 255 such that the borrower owes reduced or no mortgage recording taxes, and (2) but for the agreement the transaction would have met the definition of a refinancing under Sec. 1003.2(p). The Bureau received one comment addressing proposed comment 2(p)-2. The comment requested that the Bureau eliminate from the definition of refinancing the requirement that both the existing and the new debt obligations be dwelling-secured, because it is burdensome to confirm whether the new transaction pays off an existing mortgage. This requirement, however, is consistent with Regulation Z's definition of refinancing. The Bureau notes that, under the final rule, whether a consumer-purpose transaction meets this test (or, for that matter, whether such a transaction otherwise is a refinancing) no longer determines whether the transaction is a covered loan.\223\ Thus, for consumer-purpose transactions, when a financial institution originates a dwelling-secured debt obligation that satisfies and replaces an existing debt obligation, the financial institution no longer needs to determine whether the existing debt obligation was dwelling-secured to know that the transaction is HMDA-reportable. The financial institution will, however, need to determine whether the existing debt obligation was dwelling-secured to determine whether to report the transaction as a refinancing or an ``other purpose'' transaction. See Sec. 1003.4(a)(3). --------------------------------------------------------------------------- \223\ As discussed in the section-by-section analysis of Sec. 1003.3(c)(10), Regulation C's existing purpose-based coverage test applies to business- or commercial-purpose transactions under the final rule. --------------------------------------------------------------------------- The Bureau is finalizing proposed comment 2(p)-3 generally as proposed, with minor modifications for clarity, and renumbered as comment 2(p)-4. The Bureau received a few comments addressing proposed comment 2(p)-3. One financial institution specifically supported the proposed commentary, but another asked for additional guidance for situations, such as a divorce, where only one of the original borrowers is obligated on the new loan. As proposed, comment 2(p)-3 addressed this scenario. It specified that, if one debt obligation to two borrowers was satisfied and replaced by a new debt obligation to either one of the original borrowers, then the new obligation was a refinancing, assuming the other requirements of proposed Sec. 1003.2(p) were met. Proposed comment 2(p)-3 also specified that, if two spouses were divorcing, and a debt obligation of only one spouse was satisfied and replaced by a new debt obligation of only the other spouse, then the transaction was not a refinancing under proposed Sec. 1002.3(p). Final comment 2(p)-4 retains these examples but revises and expands them for clarity. Several commenters asked whether two or more new loans that are originated to satisfy and replace one existing loan are refinancings. The final rule adopts new comment 2(p)-5 to clarify that each of the two new obligations is a refinancing if, taken together, they satisfy and replace the existing obligation. Comment 2(p)-5 also specifies that the same rule applies when one new loan satisfies and replaces two or more existing debt obligations. The final rule adds new comment 2(p)-6 to clarify that a transaction that meets the definition of a refinancing may also be used for other purposes. The comment provides an illustrative example and specifies that instructions for reporting a multiple-purpose covered loan are in the commentary to Sec. 1003.4(a)(3). 2(q) Reverse Mortgage Proposed Sec. 1003.2(q) added a ``reverse mortgage'' definition to Regulation C. Regulation C currently requires financial institutions to report a reverse mortgage if it otherwise is reportable as a home purchase loan, a home improvement loan, or a refinancing. The current regulation, however, does not define ``reverse mortgage'' or require financial institutions to identify which applications or loans are for reverse mortgages. The proposed definition generally provided that a reverse mortgage is a reverse mortgage transaction as defined under Regulation Z Sec. 1026.33(a). Taken together with proposed Sec. 1003.2(e) (definition of ``covered loan''), proposed Sec. 1003.2(q) effectively provided that all reverse mortgage transactions, regardless of their purpose, were covered loans and HMDA-reportable. The Bureau received a number of comments about proposed Sec. 1003.2(q) and coverage of reverse mortgages. While consumer advocacy group commenters generally supported the proposal, industry participants that discussed proposed Sec. 1003.2(q) generally opposed expanding coverage of reverse mortgages. For the reasons discussed below, the Bureau is finalizing Sec. 1003.2(q) substantially as proposed, with minor technical revisions. A number of consumer advocacy groups supported the Bureau's proposed reverse mortgage definition. They stated that having data about all reverse mortgages would be valuable in assessing whether the neighborhoods that they serve are receiving the full range of credit that the neighborhoods need and would be appropriate to ensure an adequate understanding of the mortgage market. These commenters stated that publicly available data about all reverse mortgages will be essential in the coming years as the country's population ages and older consumers, many of whom are cash-poor but own their homes outright, may increasingly use home equity for living expenses and other purposes. The commenters noted that reverse mortgages often are not reported under current Regulation C because they often are not for the purpose of home purchase, home improvement, or refinancing. The commenters further noted that Regulation C's reverse mortgage data lack information about open-end, reverse mortgage transactions. Having data about ``other purpose'' reverse mortgages, as well as open- end reverse mortgages, will help to determine how the housing needs of seniors are being met. This is particularly true because poorly structured or higher-priced reverse mortgages can result in financial hardship to seniors. The commenters also noted the general importance of having data about housing-related transactions to older consumers, who may be particularly vulnerable to predatory or discriminatory lending practices. Several of these commenters urged the Bureau to adopt a flag to identify reverse mortgages. One industry commenter generally supported proposed Sec. 1003.2(q). The commenter agreed that the proposed definition of reverse mortgage was appropriate because it aligned with Regulation Z. A number of industry commenters, including trade associations, several financial institutions, and a compliance professional, disagreed with the Bureau's proposal to require reporting of all reverse mortgages. Some of these commenters asserted that Regulation C should not apply to reverse mortgages at all, or that reverse mortgages are outside the scope of HMDA. Others argued that the Bureau should maintain current coverage of reverse mortgages and require them to be reported only if they are for home purchase, home improvement, or refinancing. The commenters generally argued that reporting all reverse mortgages would create new costs for financial [[Page 66165]] institutions and that the burdens did not justify the benefits. Regarding burden, commenters stated that reverse mortgage lenders already are exiting the market because of regulatory demands and uncertainties with reverse mortgages, and that requiring reporting of all reverse mortgages under HMDA would continue that trend. A few commenters argued that data for reverse mortgages is kept on separate systems from traditional mortgage loans and that it would be costly and time-consuming to upgrade systems for reporting. Some commenters stated that the burden would be particularly great for reverse mortgage lenders that make fewer than 100 mortgages in a year. These commenters argued that the benefits of reporting all reverse mortgages would be small. They stated that financial institutions already report the necessary data about reverse mortgages (i.e., data about closed-end reverse mortgages for home purchase, home improvement, or refinancing). They stated that HMDA does not require data about other types of reverse mortgages, which are used for purposes unrelated to housing finance. They also stated that many of HMDA's data points (e.g., points and fees and debt-to-income ratio) do not apply, or apply differently, to reverse mortgages than to traditional mortgages. The commenters asserted that the data reported thus would have large gaps and would not clarify whether financial institutions are meeting the housing needs of their communities. Some commenters noted that the reverse mortgage market currently is small and that many financial institutions do not offer reverse mortgages, so the value of the data reported would be low. Some commenters stated that comparing reverse mortgage data with data for traditional mortgage loans or lines of credit would lead only to inaccurate conclusions about reverse mortgage originations because, for example, reverse mortgages are underwritten and priced differently than other mortgages and are for different purposes. Other commenters noted that the Bureau has exempted reverse mortgages from other rulemakings, such as the 2013 ATR Final Rule and the Bureau's Integrated Mortgage Disclosures rule (2013 TILA-RESPA Final Rule),\224\ given their differences from traditional mortgages. Finally, one commenter noted that there would be no harm in the Bureau delaying reverse mortgage reporting until after the Bureau has reviewed and considered other reverse mortgage rulemakings. --------------------------------------------------------------------------- \224\ 78 FR 79730 (Dec. 31, 2013). --------------------------------------------------------------------------- The Bureau is finalizing Sec. 1003.2(q) generally as proposed, with minor technical revisions. The Bureau acknowledges that requiring reporting of data on additional transactions will impose burden on financial institutions, but the Bureau believes that the benefits of reporting justify the burdens. As discussed in the proposal and in comments from consumer advocacy groups, the reverse mortgage market currently may be small, but it may become increasingly important as the country's population ages.\225\ While reverse mortgages may provide important benefits to homeowners, they also pose several risks to borrowers, including that they may be confusing, may have high costs and fees, and may result in elderly borrowers or their heirs or non- borrowing spouses losing their homes to foreclosure.\226\ As discussed in the proposal, communities have faced risks due to reverse mortgage lending, particularly communities with sizable populations of borrowers eligible for reverse mortgages programs,\227\ and many State officials have focused on harmful practices associated with reverse mortgage lending.\228\ --------------------------------------------------------------------------- \225\ See 79 FR 51731, 51759 (Aug. 29, 2014) (citing Lisa Prevost, Retiring on the House: Reverse Mortgages for Baby Boomers, N.Y. Times, Feb. 13, 2014, at RE5, available at http://www.nytimes.com/2014/02/16/realestate/reverse-mortgages-for-baby-boomers.html?_r=0). See also Nora Caley, Aging In Place, With A Loan: The State of the Reverse Mortgage Industry, Mortgage Orb, Vol. 2, Issue 17 (May 8, 2013), http://www.mortgageorb.com/e107_plugins/content/content.php?content.13765. \226\ See 79 FR 51731, 51759 (Aug. 29, 2014) (citing Consumer Fin. Prot. Bureau, Report to Congress on Reverse Mortgages 110-145 (June 28, 2012)), http://files.consumerfinance.gov/a/assets/documents/201206_cfpb_Reverse_Mortgage_Report.pdf). \227\ See id. (citing Susan Taylor Martin, Complexities of Reverse Mortgages Snag Homeowners, Tampa Bay Times, May 30, 2014; Kevin Burbach & Sharon Schmickle, As State Ages, Minnesota Braces for Problems With Risky Reverse-Mortgages, MinnPost (April 5, 2013), http://www.minnpost.com/business/2013/04/state-ages-minnesota-braces-problems-risky-reverse-mortgages; and HUD Presentation, Nat'l Reverse Mortgage Lenders Ass'n Eastern Regional Meeting (Mar. 26, 2012) (noting that 8.1 and 9.4 percent of active Home Equity Conversion Mortgage loans were in default in July 2011 and February 2012, respectively). \228\ See id. at 51759-60 (citing Press Release, Illinois Attorney General, Madigan Sues Two Reverse Mortgage Brokers For Using Deceptive Marketing to Target Seniors (Feb. 8, 2010), http://www.illinoisattorneygeneral.gov/pressroom/2010_02/20100208.html; Press Release, Washington State Office of the Attorney General, Ferguson Files Complaint Against Bellevue Insurance Agent and His Company for Targeting Elderly Widows (July 29, 2013), http://www.atg.wa.gov/news/news-releases/ferguson-files-complaint-against-bellevue-insurance-agent-and-his-company. --------------------------------------------------------------------------- Information on all reverse mortgages, regardless of purpose, would help communities understand the risks posed to local housing markets, thereby providing the citizens and public officials of the United States with sufficient information to enable them to determine whether financial institutions are filling their obligations to serve the housing needs of the communities and neighborhoods in which they are located. Furthermore, private institutions and nonprofit organizations, as well as local, State, and Federal programs, traditionally have facilitated or engaged in reverse mortgage lending. However, the proprietary market for reverse mortgages has substantially declined in recent years. Thus, requiring improved information regarding all reverse mortgages would assist public officials in their determination of the distribution of public sector investments in a manner designed to improve the private investment environment. Indeed, it is particularly important to obtain better information about the reverse mortgage market because it serves older consumers, a traditionally vulnerable population. State officials provided feedback during the Board's 2010 Hearings that expanding the transactional coverage of Regulation C to include all reverse mortgages would assist in the identification of discriminatory and other potentially harmful practices against this protected class.\229\ In this regard, the Bureau notes that requiring reporting of all reverse mortgages dovetails with the Dodd-Frank Act's requirement to report age for all covered loans. The Bureau believes that the currently small size of the market, and the fact that the Bureau may address reverse mortgages in future, substantive rulemakings, further support the decision to require reverse mortgage reporting as soon as possible. The flow of information to the public and policymakers will better position them to identify housing needs and market developments as they occur. --------------------------------------------------------------------------- \229\ See id. at 51760 (citing New York State Banking Department comment letter, Board of Governors of the Fed. Reserve System docket no. OP-1388, p. 5, submitted Aug. 6, 2010; San Francisco Hearing, Remarks of Preston DuFauchard, Commissioner of the California Department of Corporations). --------------------------------------------------------------------------- The Bureau acknowledges that, as commenters observed, reverse mortgages are underwritten and priced differently than other mortgages, some data points apply differently to reverse mortgages, and some do not apply at all. However, this is just as true for the reverse mortgages that currently are reported (and that most commenters agree should be reported) as for the reverse mortgages that will be added under the final rule. Where possible, the [[Page 66166]] Bureau has provided additional guidance to instruct financial institutions how particular data points apply to reverse mortgages. Finally, the Bureau is adopting a flag to ensure that data reported for reverse mortgages will not be commingled unknowingly with data reported for other covered loans. See the section-by-section analysis of Sec. 1003.4(a)(36). The final rule modifies proposed Sec. 1003.2(q) to specify that a reverse mortgage is a reverse mortgage transaction as defined in Regulation Z, 12 CFR 1026.33(a), but without regard to whether the security interest is created in a principal dwelling. Thus, under Regulation C, a transaction that otherwise meets the definition of a reverse mortgage must be reported even if the security interest is taken in, for example, the borrower's second residence. Section 1003.2(q) also contains one revision to align the definition with other changes being adopted in the final rule. As discussed in the section-by-section analysis of Sec. 1003.2(d) and (o), the proposal provided that closed-end mortgage loans and open-end lines of credit were mutually exclusive of reverse mortgages, and thus a covered loan under proposed Sec. 1003.2(e) was a closed-end mortgage loan, an open-end line of credit, or a reverse mortgage that was not otherwise excluded under proposed Sec. 1003.3(c). The final rule eliminates the mutual exclusivity between: (1) Closed-end mortgage loans and open-end lines of credit and (2) reverse mortgages. Thus, the final rule both eliminates reverse mortgages as a category of covered loans under Sec. 1003.2(e) and eliminates the cross-reference to Sec. 1003.2(e) from the reverse mortgage definition. Final Sec. 1003.2(q) is adopted pursuant to the Bureau's authority under section 305(a) of HMDA. For the reasons given above, the Bureau believes that including reverse mortgages within the scope of the regulation is a reasonable interpretation of HMDA section 303(2), which defines ``mortgage loan'' to mean a loan which is secured by residential real property or a home improvement loan. The Bureau interprets that term to include reverse mortgages, as those transactions are secured by residential real property, and they may be used for home improvement. In addition, pursuant to its authority under section 305(a) of HMDA, the Bureau believes that this proposed adjustment is necessary and proper to effectuate the purposes of HMDA, to prevent circumvention or evasion thereof, and to facilitate compliance therewith. For the reasons given above, by requiring all financial institutions to report information regarding reverse mortgages, this proposed modification would ensure that the citizens and public officials of the United States are provided with sufficient information to enable them to determine whether depository institutions are filling their obligations to serve the housing needs of the communities and neighborhoods in which they are located. Furthermore, as reverse mortgages are a common method of obtaining credit, this proposed modification would assist in identifying possible discriminatory lending patterns and enforcing antidiscrimination statutes. Section 1003.3 Exempt Institutions and Excluded Transactions 3(c) Excluded Transactions Regulation C currently excludes several categories of transactions from coverage, but the exclusions are scattered throughout the regulation text, appendix A, and commentary. To streamline the regulation, the Bureau proposed to consolidate all existing exclusions in new Sec. 1003.3(c). The Bureau also proposed guidance concerning two categories of excluded transactions: Loans secured by liens on unimproved land and temporary financing. The Bureau received no comments opposing, and one comment supporting, the consolidation of excluded transactions into Sec. 1003.3(c) and is finalizing the reorganization as proposed. The Bureau received a number of comments addressing specific categories of excluded transactions and suggesting additional categories of transactions that should be excluded. For the reasons discussed below, the Bureau is finalizing Sec. 1003.3 to clarify that certain categories of transactions, including all agricultural-purpose transactions and commercial-purpose transactions not for home purchase, home improvement, or refinancing purposes, are excluded from reporting. The final rule also revises Sec. 1003.3 and its accompanying commentary for clarity and to address questions raised by commenters. Suggested Exclusions Not Adopted A few commenters suggested specifically excluding loans made by financial institutions to their employees. The commenters stated that it is and will continue to be difficult to report such loans and that, because such loans typically are offered on better terms than loans to non-employees, their inclusion in HMDA data will skew the dataset and will serve no purpose for fair lending testing. The final rule does not specifically exclude loans made to financial institutions' employees. It is not clear why such loans are more difficult to report than other loans, and commenters did not provide any details to explain the difficulty. Loans to employees may be made on more favorable terms than other loans, but the Bureau doubts that employee loans are originated in sufficient quantities to skew the overall HMDA data. Finally, as always, HMDA data are used only as the first step in conducting a fair lending analysis. Examiners conducting fair lending examinations will be able to identify by looking at loan files when differences in loan pricing, for example, are attributable to an applicant's or borrower's status as a financial institution's employee. Commenters suggested excluding a number of other types of transactions from coverage. The section-by-section analysis of Sec. 1003.2(f) (definition of dwelling) discusses coverage of transactions secured by other than a single-family, primary residence; the section- by-section analysis of Sec. 1003.3(c)(10) discusses coverage of loans made to trusts; and the section-by-section analysis of Sec. 1003.4(a) (reporting of purchases) discusses coverage of repurchased loans. 3(c)(1) Proposed Sec. 1003.3(c)(1) and comment 3(c)(1)-1 retained Regulation C's existing exclusion for loans originated or purchased by a financial institution acting in a fiduciary capacity, which currently is located in Sec. 1003.4(d)(1). The Bureau received no comments concerning proposed Sec. 1003.3(c)(1) or comment 3(c)(1)-1 and finalizes them as proposed, with several technical revisions for clarity. 3(c)(2) Proposed Sec. 1003.3(c)(2) retained Regulation C's existing exclusion for loans secured by liens on unimproved land, which currently is located in Sec. 1003.4(d)(2). The Bureau proposed new comment 3(c)(2)-1 to clarify that the exclusion: (1) Aligns with the exclusion from RESPA coverage of loans secured by vacant land under Regulation X Sec. 1024.5(b)(4), and (2) does not apply if the financial institution ``knows or reasonably believes'' that within two years after the loan closes, a dwelling will be constructed or placed on the land using the loan proceeds. For the reasons discussed below, the Bureau is finalizing Sec. 1003.3(c)(2) as proposed but is finalizing comment 3(c)(2)-1 with certain changes in response to comments received. [[Page 66167]] The Bureau received a number of comments from financial institutions, trade associations, and other industry participants about proposed comment 3(c)(2)-1. Commenters agreed that loans secured by unimproved land should be excluded, but they stated that the proposed comment was inappropriate and that the Bureau either should remove it entirely or should clarify it. A few commenters stated that aligning with Regulation X was unnecessary and advocated a simple rule that would exclude all loans secured only by land when made. Other commenters stated that, if retained, the exemption should be based on the financial institution's actual knowledge, rather than on a ``knows or reasonably believes'' standard that would require lenders to speculate about whether a dwelling would be constructed. Commenters argued that examiners later could second-guess such speculative decisions. Some commenters stated that, as written, the proposed comment would make almost all consumer lot loans reportable, because they generally are built on within two years. The Bureau believes that providing guidance about the types of transactions covered by the exclusion for loans secured by liens on unimproved land is preferable to eliminating the proposed comment, and that aligning with Regulation X helps to achieve regulatory consistency. Moreover, where a loan's funds will be used to construct a dwelling in the immediate future, having information about that loan serves HMDA's purposes of understanding how financial institutions are meeting the housing needs of their communities. On the other hand, the Bureau acknowledges that the Regulation X standard does not provide sufficient specificity for purposes of HMDA reporting, because it does not state how and when a financial institution must know that a dwelling will be constructed on the land. The final rule adopts comment 3(c)(2)-1 without the cross-reference to Regulation X but with a statement, consistent with the spirit of Regulation X, that a loan is secured by a lien on unimproved land if the loan is secured by vacant or unimproved property at the time that is originated, unless the financial institution knows, based on information that it receives from the applicant or borrower at the time the application is received or the credit decision is made, that the loan's proceeds will be used within two years after closing or account opening to construct a dwelling on the land or to purchase a dwelling to be placed on the land. If the applicant or borrower does not provide the financial institution this information at the time the application is received or the credit decision is made, then the exclusion applies. Financial institutions should note that, even if a loan is not exempt under Sec. 1003.3(c)(2), it may be exempt under another Sec. 1003.3(c) exclusion, such as the temporary financing exclusion under Sec. 1003.3(c)(3). 3(c)(3) Proposed Sec. 1003.3(c)(3) retained Regulation C's existing exclusion for temporary financing, which currently is located in Sec. 1003.4(d)(3). Comments 3(c)(3)-1 and -2 were proposed to clarify the scope of the exclusion. For the reasons discussed below, the Bureau is adopting Sec. 1003.3(c)(3) as proposed but is finalizing the commentary to Sec. 1003.3(c)(3) with revisions to address questions and concerns that commenters raised. Consumer advocacy group commenters generally argued that construction loans should not be excluded as temporary financing. Financial institutions, trade associations, and other industry participants generally argued that temporary financing should be excluded from coverage. Several of these commenters argued that all construction loans should be excluded as temporary financing. Most such commenters agreed that guidance about the scope of the temporary financing exclusion would be helpful, but many found the guidance in proposed comments 3(c)(3)-1 and -2 confusing or objected that it relied on a subjective standard. Commenters suggested several methods to clarify the proposed guidance. Regarding proposed comment 3(c)(3)-1, which provided general guidance about the temporary financing exclusion, a few commenters objected to the cross-reference to Regulation X. They stated that the Regulation X standard is unclear and ambiguous and that cross- referencing it would create confusion about which construction loans qualify for Regulation C's exclusion. Some construction loans would be reported (e.g., construction loans involving title transfer) and others would not (e.g., construction-only loans). Similarly, one commenter suggested that long-term construction loans should be excluded regardless of whether they were made to ``bona fide builders.'' Another commenter argued that all construction loans should be exempt, except for construction loans with one-time closings, where the construction loan automatically rolls into permanent financing after a predetermined time. On the other hand, at least one commenter stated that aligning with Regulation X was helpful. Still others suggested that Regulation C should align with Regulation Z and that the Bureau either should adopt a bright-line test (similar to Regulation Z's) to define any loan with a term shorter than a prescribed period of time (e.g., one or two years) as temporary financing, or should adopt a bright-line test to exclude all short-term construction loans. One commenter requested that the Bureau specifically define the term ``bridge loan,'' which is listed as an example of temporary financing in both existing Sec. 1003.4(d)(3) and proposed comment 3(c)(3)-1. Several commenters also argued that proposed comment 3(c)(3)-2 was confusing. Comment 3(c)(3)-2 explained that loans designed to convert to (i.e., rather than designed to be replaced by) permanent financing were not temporary financing and thus were reportable. Consistent with Regulation X, the comment provided that loans issued with a commitment for permanent financing, with or without conditions, were considered loans that would ``convert'' to permanent financing and thus were not excluded transactions. Some commenters urged the Bureau to remove this statement or to clarify further the difference between a loan ``replaced by'' permanent financing and a loan ``converted'' to permanent financing. One commenter observed that a loan issued with a commitment for permanent financing could encompass a situation covered under proposed comment 3(c)(3)-1's first sentence (i.e., a loan designed to be replaced by permanent financing at a later time). The commenter argued that such transactions would be excluded as temporary financing under proposed comment 3(c)(3)-1 but would lose the exemption under proposed comment 3(c)(3)-2. Other commenters questioned the meaning of the term ``designed'' and asked the Bureau to clarify whether construction-only loans that eventually are refinanced into longer-term financing must be reported. Some commenters stated that proposed comment 3(c)(3)-1's first sentence provided clear and sufficient guidance and that proposed comment 3(c)(3)-2 should be removed altogether. The Bureau is finalizing the commentary to Sec. 1003.3(c)(3) with revisions to address the foregoing concerns. Final comment 3(c)(3)-1 provides that temporary financing is excluded from coverage and provides that a loan or line of credit is temporary financing if it is designed to be replaced by permanent financing at a later time. The comment provides several [[Page 66168]] illustrative examples designed to clarify whether a loan or line of credit is designed to be replaced by permanent financing. The final rule does not provide for reporting of all construction loans, as some consumer advocacy group commenters recommended. The Bureau believes that the benefits of requiring all construction loans to be reported do not justify the burdens given that the permanent financing that replaces such loans will be reported. The Bureau believes that comment 3(c)(3)-1 achieves HMDA's purposes while providing better guidance to financial institutions than existing Regulation C. Specifically, the comments should help to ensure that transactions involving temporary financing are not reported more than once; instead, such transactions will be captured by the separate reporting of the longer-term financing, if it otherwise is covered by Regulation C. At the same time, the comments will help to ensure reporting of short-term transactions that function as permanent financing (e.g., a loan with a nine-month term to enable an investor to purchase a home, renovate, and re-sell it before the term expires).\230\ --------------------------------------------------------------------------- \230\ The final rule thus is consistent with the existing FFIEC FAQ concerning temporary financing, which acknowledges that temporary financing is exempt and states that ``financing is temporary if it is designed to be replaced by permanent financing of a much longer term. A loan is not temporary financing merely because its term is short. For example, a lender may make a loan with a 1- year term to enable an investor to purchase a home, renovate it, and re-sell it before the term expires. Such a loan must be reported as a home purchase loan.'' See Fed. Fin. Insts. Examination Council, Regulatory & Interpretive FAQ's, Temporary Financing, http://www.ffiec.gov/hmda/faqreg.htm#TemporaryFinancing. --------------------------------------------------------------------------- After considering the comments received, the Bureau believes that neither aligning with Regulation X or Z, nor creating a new, bright- line rule centered around a loan's term, would serve HMDA's purposes as well as the guidance provided in final comment 3(c)(3)-1. Regulation Z generally excludes loans with terms of less than one year from, for example, the regulation's ability-to-repay rules. Conducting a full ability-to-repay analysis may not be critical for such short-term financing. However, it is important for HMDA purposes to know how often and under what circumstances such financing is granted, for example, to investors to purchase property and then to sell it for occupancy before the term expires. Similarly, the Bureau believes that it is important for HMDA purposes to ensure that construction loans are not double- counted when they are replaced by permanent financing. Thus, the Bureau has not aligned with Regulation X's guidance concerning construction loans, which would have required, for example, some longer-term construction loans to be reported. Two commenters requested that the Bureau clarify whether a loan's purpose is ``construction'' or ``home improvement'' when improvements to an existing dwelling are so extensive that they fundamentally change the nature of the dwelling. The commenters suggested that, if a loan's purpose was ``construction,'' then the loan would be excluded from coverage, whereas if its purpose was ``home improvement,'' it would be included. Under the final rule, the temporary financing exclusion depends on whether the loan is or is not designed to be replaced by longer-term financing at a later time. Thus, for example, if a financial institution originates a short-term loan to a borrower to add a second floor to a dwelling or to complete extensive renovations, the loan is temporary financing if it is designed to be replaced by longer- term financing at a later time (e.g., financing completed through a separate closing that will pay off the short-term loan). If the loan is, for example, a traditional home-equity loan that is not designed to be replaced by longer-term financing, or if it is a construction-to- permanent loan that automatically will convert to permanent financing without a separate closing, then it is not temporary financing and is not excluded under Sec. 1003.3(c). 3(c)(4) Proposed Sec. 1003.3(c)(4) and comment 3(c)(4)-1 retained Regulation C's existing exclusion for the purchase of an interest in a pool of loans, which currently is located in Sec. 1003.4(d)(4). The Bureau received no comments concerning proposed Sec. 1003.3(c)(4) or comment 3(c)(4)-1 and finalizes them as proposed, with technical revisions for clarity. 3(c)(5) Proposed Sec. 1003.3(c)(5) retained Regulation C's existing exclusion for the purchase solely of the right to service loans, which currently is located in Sec. 1003.4(d)(5). The Bureau received no comments concerning proposed Sec. 1003.3(c)(5) and finalizes it as proposed, with technical revisions for clarity. 3(c)(6) Proposed Sec. 1003.3(c)(6) and comment 3(c)(6)-1 retained Regulation C's existing exclusion for loans acquired as part of a merger or acquisition, or as part of the acquisition of all of the assets and liabilities of a branch office, which currently is located in Sec. 1003.4(d)(6) and comment 4(d)-1. The Bureau received no comments concerning proposed Sec. 1003.3(c)(6) or comment 3(c)(6)-1 and finalizes them generally as proposed, with technical revisions for clarity. 3(c)(7) Proposed Sec. 1003.3(c)(7) retained Regulation C's existing exclusion for loans and applications for less than $500, which currently is located in paragraph I.A.7 of appendix A. The Bureau received no comments concerning proposed Sec. 1003.3(c)(7) and finalizes it as proposed, with technical revisions for clarity. 3(c)(8) Proposed Sec. 1003.3(c)(8) retained Regulation C's existing exclusion for the purchase of a partial interest in a loan, which currently is located in comment 1(c)-8. The Bureau received no comments concerning proposed Sec. 1003.3(c)(8) and finalizes it generally as proposed, with technical revisions for clarity. 3(c)(9) As proposed, Sec. 1003.3(c)(9) stated that a loan used primarily for agricultural purposes was an excluded transaction. Proposed comment 3(c)(9)-1, in turn, retained the existing exclusion of home purchase loans secured by property primarily for agricultural purposes, which currently is located in comment Home purchase loan-3. For the reasons discussed below, the Bureau is adopting Sec. 1003.3(c)(9) with technical revisions for clarity and is adopting comment 3(c)(9)-1 with revisions to clarify that all agricultural-purpose loans are excluded transactions. The Bureau received a number of comments from financial institutions, industry associations, and other industry participants about proposed Sec. 1003.3(c)(9) and comment 3(c)(9)-1. Some commenters stated that the proposed regulation text appeared to exclude all agricultural loans, while the commentary appeared to exclude only home-purchase agricultural loans. These commenters stated that all agricultural loans should be excluded, because they are not comparable to other loans reported under HMDA, and reporting them does not serve HMDA's purposes. Other commenters noted that proposed comment 3(c)(9)-1 retained a cross-reference to Regulation X Sec. 1024.5(b)(1), which had exempted loans on property of 25 acres or more from coverage, even though that provision since had been removed from Regulation X. A few of these commenters argued that the [[Page 66169]] Bureau should retain an independent 25-acre test in Regulation C, while others stated that the 25-acre test should be removed altogether because smaller properties can be primarily agricultural and thus should be excluded from coverage, while larger properties can be primarily consumer-purpose and thus should be included in coverage. The Bureau is finalizing Sec. 1003.3(c)(9) and comment 3(c)(9)-1 with revisions to address commenters' concerns. First, final comment 3(c)(9)-1 clarifies that all primarily agricultural-purpose transactions are excluded transactions, whether they are for home purchase, home improvement, refinancing, or another purpose. The comment also clarifies that an agricultural-purpose transaction is a transaction that is secured by a dwelling located on real property used primarily for agricultural purposes or that is secured by a dwelling and whose funds will be used primarily for agricultural purposes. The final rule eliminates from the comment both the proposed cross- reference to Regulation X and the 25-acre test. The comment instead provides that financial institutions may consult Regulation Z comment 3(a)-8 for guidance about what is an agricultural purpose. Comment 3(c)(9)-1 provides that a financial institution may use any reasonable standard to determine whether a transaction primarily is for an agricultural purpose and that a financial institution may change the standard used on a case-by-case basis. This flexible standard should provide sufficient latitude for a financial institution to justify its determination that a property was, or that a loan's funds were, intended to be used primarily for agricultural purposes. 3(c)(10) Unlike certain other consumer protection statutes such as TILA and RESPA, HMDA does not exempt business- or commercial-purpose transactions from coverage. Thus, Regulation C currently covers closed- end, commercial-purpose loans made to purchase, refinance, or improve a dwelling. Examples of commercial-purpose loans that currently are reported are: (1) A loan to an entity to purchase or improve an apartment building (or to refinance a loan secured thereby); and (2) a loan to an individual to purchase or improve a single-family home to be used either as a professional office or as a rental property (or to refinance a loan secured thereby). Dwelling-secured, commercial-purpose lines of credit currently are not required to be reported. Regulation C currently does not provide a mechanism, such as a commercial-purpose flag, to distinguish commercial-purpose loans from other loans in the HMDA dataset, but it appears that commercial-purpose loans currently represent a small percentage of HMDA-reportable loans.\231\ --------------------------------------------------------------------------- \231\ For example, applications and originations for multifamily housing represented about 0.4 percent of all applications and originations reported for 2013. See Neil Bhutta & Daniel R. Ringo, Bd. of Governors of the Fed. Reserve Sys., 100 Fed. Reserve Bulletin 6, The 2013 Home Mortgage Disclosure Act Data, at 4 (Nov. 2014). --------------------------------------------------------------------------- As discussed in the section-by-section analysis of Sec. 1003.2(d), (e) and (o), the proposal provided for dwelling-secured transactional coverage and for mandatory reporting of open-end lines of credit. Under the proposal, financial institutions would have reported applications for, and originations of, all dwelling-secured, commercial-purpose closed-end mortgage loans and open-end lines of credit. For example, a financial institution would have reported all closed-end mortgage loans or open-end lines of credit to a business or sole proprietor secured by a lien on the business owner's dwelling, even if only out of an abundance of caution (i.e., in addition to other collateral such as a storefront, inventory, or equipment) and regardless of how the funds would be used (e.g., to purchase the storefront, inventory, or equipment). A financial institution also would have been required to report any transaction secured by a multifamily dwelling, such as an apartment building, even if the loan or line of credit was for non- housing-related business expansion. The proposal thus would have expanded Regulation C's coverage of commercial-purpose transactions. For the reasons discussed below, the Bureau is maintaining Regulation C's existing purpose-based transactional coverage scheme for commercial-purpose transactions. A large number of comments addressed the proposal's coverage of dwelling-secured commercial-purpose transactions. Consumer advocacy groups favored covering all such transactions, while a significant number of industry commenters, a government agency commenter, and a group of State regulators, urged the Bureau to exclude some or all of these transactions. Numerous consumer advocacy groups generally asserted that having information about dwelling-secured commercial transactions would help them to understand whether neighborhoods are receiving the full range of credit they need. Some consumer advocacy groups specifically urged the Bureau to collect data about all transactions secured by multifamily properties, to understand whether financial institutions are supporting the development of affordable rental housing. Others argued that dwelling-secured commercial-purpose reporting would help to understand the full range of liens against single-family properties. Some of these commenters asserted that, during the mortgage crisis, dwelling-secured commercial lending contributed to overleveraging and foreclosures in many communities, and that HMDA data about such loans could have warned policymakers and advocates of potential concerns. Some consumer advocacy group commenters specified that dwelling- secured commercial lending is an important source of small business financing, particularly in minority and immigrant communities, and that having information about the availability and pricing of such transactions would help to understand those communities' economies, including the total amount of debt and default risk on properties and potential problems related to overextension of credit. A few consumer advocacy commenters noted that information about all dwelling-secured commercial lending also would provide insight into the demand for, and use of, credit for expansion of small businesses.\232\ --------------------------------------------------------------------------- \232\ Some of these commenters also asserted that the Bureau should include in the final rule a flag to distinguish commercial- and consumer-purpose transactions. The Bureau is finalizing such a flag in Sec. 1003.4(a)(38). --------------------------------------------------------------------------- A significant number of industry commenters addressed the proposal's expanded coverage of commercial-purpose transactions, and they all opposed the change. Indeed, many commenters who objected to dwelling-secured transactional coverage cited expanded reporting of commercial-purpose transactions as their main concern. Industry commenters argued that implementing reporting of all dwelling-secured, commercial-purpose transactions would be burdensome, that the data reported would be of little value, and that requiring such reporting would exceed the Bureau's authority under HMDA.\233\ --------------------------------------------------------------------------- \233\ A subset of industry commenters specifically objected to reporting commercial-purpose open-end lines of credit. Indeed, even the small group of industry commenters that did not object to reporting consumer-purpose lines of credit argued that commercial- purpose lines should not be covered. Commenters' concerns about the burdens and benefits of reporting commercial-purpose lines of credit were similar to those raised about commercial-purpose transactions generally. --------------------------------------------------------------------------- Regarding burden, industry commenters stated that removing the purpose test for commercial-purpose [[Page 66170]] applications and originations would increase significantly financial institutions' reportable transactions. A subset of commenters specifically estimated the increase, which varied widely (i.e., from 10 percent to over 900 percent) depending on institution type and the extent of an institution's engagement in dwelling-secured, small- business lending. Some institutions argued that many community banks focus on small-business lending, so expanded commercial coverage particularly could burden smaller institutions. A number of commenters worried about ongoing costs from collecting, quality checking, and reporting information for such a large number of transactions, and some worried about incurring penalties for errors that likely would occur in the commercial data.\234\ --------------------------------------------------------------------------- \234\ Some commenters argued that the Bureau's proposal to expand HMDA-reportable data points only compounded their concerns about increased volume. Others argued that any reporting burden that might be mitigated by aligning Regulation C's data reporting with MISMO standards would not apply to commercial-purpose transactions, because MISMO has not been widely adopted in commercial and multifamily financing. --------------------------------------------------------------------------- Industry commenters also argued that reporting all dwelling-secured commercial transactions would be difficult operationally. Different staff and systems typically handle commercial and residential mortgage loans, and lenders may have relied on manual processes for reporting and assembling data for the limited set of commercial-purpose transactions traditionally reported. Commenters argued that expanded coverage, particularly when combined with new data points, would require updating systems or software, implementing new policies and procedures, and training or hiring new staff. These would be expensive and time-consuming processes, with costs passed to consumers. Industry commenters asserted that the benefits of reporting all commercial-purpose transactions would not justify the burdens. A significant number of commenters argued that reporting data about all commercial-purpose transactions would not serve HMDA's purposes. Some industry commenters asserted that commercial-purpose transactions often are provided to non-natural persons. In such cases, no race, ethnicity, and sex data would be collected and no fair lending analysis could be done (except of the demographics of the dwelling's census tract). Commenters argued that reporting data about such transactions would not help to uncover discriminatory lending practices. Many commenters focused on what they referred to as ``abundance of caution'' transactions and asserted that such transactions would not help to determine whether financial institutions are serving community housing needs. Commenters argued that, in abundance of caution transactions, the home is added to an already adequately secured transaction (to over-collateralize the loan), is secondary to business collateral, and is an insignificant piece of the overall loan structure.\235\ In contrast, commenters argued, consumer-purpose loans typically are fully collateralized by the home. Commenters also argued that there is only a tangential relationship between the loan and housing because the loan's funds are used for business, not housing, purposes.\236\ --------------------------------------------------------------------------- \235\ Commenters explained that, when lenders originate small business loans, they routinely rely on a business owner's dwelling as supplemental collateral out of an abundance of caution, even if other (business) collateral fully collateralizes the loan. Several commenters emphasized that abundance of caution transactions occur frequently, noting that the SBA as a matter of course requires a lien on the borrower's residence when guaranteeing loans. One commenter elaborated that the likelihood that a dwelling would be part of the workout of a distressed commercial loan is ``slim-to- none.'' The commenter asserted that lenders take dwellings as collateral as a matter of safety and soundness, merely to ensure that the borrower has ``skin in the game.'' \236\ A few commenters expressed similar concerns about loans subject to cross-collateralization agreements, which commonly occur in commercial lending and in which all of the collateral for multiple loans secures all of the loans. Commenters worried that non-dwelling-secured commercial transactions would be HMDA- reportable merely because they were cross-collateralized by dwelling-secured loans. --------------------------------------------------------------------------- Regarding data collection, some commenters argued that the application, documentation, and underwriting processes are different for commercial- and consumer-purpose transactions, so data for many of the Bureau's proposed data points are not gathered in a systematic way for commercial-purpose transactions. Some commenters similarly asserted that reporting data for all dwelling-secured commercial transactions would be challenging because Regulation C's existing and the Bureau's newly proposed data points focus on consumer lending. Commenters argued that many data points would not apply to, or would be difficult to define for, commercial transactions.\237\ --------------------------------------------------------------------------- \237\ Commonly cited examples included: application and application date; applicant's income; credit score; pricing data such as points and fees; debt-to-income ratio; combined loan-to- value ratio; property value; and ethnicity, race, sex, and age data. --------------------------------------------------------------------------- Other commenters worried that even correctly reported data would be of little value in understanding commercial-purpose transactions. For example, some commenters observed that numerous data points would be reported ``not applicable'' for commercial-purpose transactions and argued that the limited number of reportable data points would not further HMDA's purposes or assist policymakers in preventing or responding to future mortgage crises. Others observed that much information that would be relevant to understanding the economics of commercial-purpose loans, such as the debt service coverage ratio, leasing requirements and expirations, zoning restrictions, environmental regulations, and cash flow, would not be reported. Some commenters also asserted that there would be little value in comparing all dwelling-secured commercial- and consumer-purpose transactions, because they are underwritten and priced differently (e.g., based on cash flow rather than income), and they have different loan terms and features (e.g., rate and fee structures, balloon, interest-only and prepayment penalty terms). Finally, some industry commenters worried that mixing data about all dwelling-secured, commercial-purpose transactions with traditional mortgage loans would distort or skew the HMDA dataset and impair its integrity for HMDA users. Numerous industry commenters argued that HMDA does not authorize the Bureau to require reporting of all dwelling-secured commercial- purpose transactions. They argued that HMDA itself focuses on home mortgage lending and that Congress understood, but opted not to revise, Regulation C's current coverage when it passed the Dodd-Frank Act.\238\ Some commenters similarly argued that, when Congress intended to grant the Bureau authority to collect business lending data, it did so explicitly.\239\ Other commenters argued [[Page 66171]] that HMDA reporting of all commercial-purpose transactions would duplicate CRA reporting or would negatively affect CRA performance. --------------------------------------------------------------------------- \238\ A group of State regulators similarly argued that the expansion into commercial lending was outside of HMDA's scope and would burden financial institutions for little benefit. They argued that Federal and State regulators should determine whether financial institutions are structuring transactions to evade reporting or other disclosure requirements, and that regulators could assess evasion efforts through risk-scoping and examinations. \239\ For example, section 1071 of the Dodd-Frank Act amended ECOA to authorize the Bureau to obtain data about loans and lines of credit to women-owned, minority-owned, and small businesses. Some commenters argued that reporting commercial transactions in HMDA was unnecessary because data about small-business lending would be reported when the Bureau implements section 1071. --------------------------------------------------------------------------- Finally, some commenters expressed concerns that reporting all dwelling-secured commercial-purpose transactions could be particularly burdensome for smaller institutions, because small-business loans may represent a large portion of their lending activity. A few commenters asserted that some small institutions exited consumer mortgage lending to focus on small-business lending specifically to avoid the costs of complying with Dodd-Frank Act regulations and that the proposal unfairly would burden such institutions with HMDA reporting. Others expressed concern that financial institutions would stop taking small- business borrowers' homes as collateral to avoid reporting, or would increase borrowers' fees to cover reporting costs, in turn decreasing small businesses' access to credit and harming local and national economies. Industry commenters provided a number of alternatives for coverage of commercial-purpose transactions. A significant number of commenters urged the Bureau specifically to exclude all dwelling-secured commercial-purpose transactions. These commenters cited the benefits and burdens already discussed, asserted that such an exclusion would reduce burden significantly, and argued that it would align coverage across Regulations C, X, and Z. A number of commenters urged the Bureau specifically to exclude transactions for multifamily housing (or alternatively to non-natural persons), emphasizing the differences in underwriting between multifamily and other lending, and asserting that multifamily loan data is particularly ill-suited to serving HMDA's purposes because multifamily loans typically are made to corporate borrowers rather than to consumers.\240\ A few commenters expressed concern about the privacy of multifamily borrowers, fearing that multifamily loans easily could be identified in the dataset because relatively few are made each year and they have unique characteristics. --------------------------------------------------------------------------- \240\ Commenters cited other differences, such as the lack of standardized underwriting criteria in multifamily lending, and heavy reliance on a property's income-producing capacity, on the borrower's cash flow, and on an evaluation of the strength of the overall market. --------------------------------------------------------------------------- Other commenters variously urged that reporting of commercial applications and originations should be required only for: (1) Multifamily transactions; (2) closed-end mortgage loans; (3) first-lien transactions; or (4) transactions for home purchase, home improvement, or refinancing.\241\ Commenters who recommended retaining Regulation C's home purchase, home improvement, and refinancing test for commercial-purpose transactions argued that: (1) The purpose test reasonably limits the scope of reportable commercial transactions and better serves HMDA's purposes; and (2) financial institutions easily can identify their dwelling-secured commercial- and consumer-purpose transactions, because they are accustomed to making a similar determination for coverage under Regulations X and Z. --------------------------------------------------------------------------- \241\ Several commenters discussed commercial- and agricultural- purpose loans together and urged the Bureau to exclude both categories of loans entirely from Regulation C. For the reasons discussed in the section-by-section analysis of Sec. 1003.3(c)(9), the final rule excludes agricultural-purpose transactions from reporting. --------------------------------------------------------------------------- As discussed in the proposal, the Bureau believes that HMDA's scope is broad enough to cover all dwelling-secured commercial-purpose transactions and that collecting information about all such transactions would serve HMDA's purposes. HMDA section 303(2) defines ``mortgage loan'' as a loan secured by residential real property or a home improvement loan. While the Board historically interpreted HMDA section 303(2) to refer to loans for home purchase, home improvement, or refinancing purposes, the Bureau believes that the definition is broad enough to include all dwelling-secured mortgage loans and lines of credit, even if their funds are used in whole or in part for commercial (or for other, non-housing-related) purposes.\242\ --------------------------------------------------------------------------- \242\ As noted in the Bureau's proposal, when the Board first proposed to implement HMDA, it proposed to require reporting of all loans secured by residential real property. See 41 FR 13619, 13620 (Mar. 31, 1976). The Board subsequently decided to adopt a narrower scope based on loan purpose, because the Board believed that focusing on loan purpose would provide more useful data. See 41 FR 23931, 23932 (June 14, 1976). --------------------------------------------------------------------------- Moreover, the Bureau believes that collecting data about all such transactions would serve HMDA's purposes by showing not only the availability and condition of multifamily housing units, but also the full extent of leverage on single-family homes, particularly in communities that may rely heavily on dwelling-secured loans to finance small-business expenditures. The Bureau believes that financial institutions serve the housing needs of their communities not only by providing fair and adequate financing to purchase and improve homes, but also by ensuring that neither individual borrowers nor particular communities are excessively overleveraged through business-related home-equity borrowing, and that all such credit is extended on equitable terms.\243\ --------------------------------------------------------------------------- \243\ See also 79 FR 51731, 51747-48 (Aug. 29, 2014). --------------------------------------------------------------------------- The Bureau nevertheless has determined at this time to require reporting only of applications for, and originations of, dwelling- secured commercial-purpose loans and lines of credit for home purchase, home improvement, or refinancing purposes. After considering the comments, the Bureau concluded that it is unclear whether the benefits of reporting all dwelling-secured commercial-purpose transactions justify the burdens, particularly in light of the many other changes required under the final rule. While the Bureau has no data with which to estimate specifically how many additional transactions would have been reported under the proposal, it seems clear that some financial institutions' HMDA reports would have expanded dramatically. The Bureau is concerned that the impact could be greatest for smaller institutions that specialize in small-business lending. The Bureau considered other burdens, as well, including the unique burdens of collecting and reporting information about commercial-purpose transactions (relative to consumer-purpose transactions) and the burdens of addressing loans subject to cross-collateralization agreements. Against these burdens, the Bureau weighed commenters' arguments that abundance of caution transactions likely would pose less risk to borrowers' homes than consumer-purpose equity lending and that data reporting for commercial- purpose lending could be addressed in a future Bureau rulemaking to implement section 1071 of the Dodd-Frank Act. The Bureau concluded that, at this time, maintaining purpose-based reporting of dwelling-secured commercial-purpose transactions appropriately balances reporting benefits and burdens. The final rule thus adds to Regulation C new Sec. 1003.3(c)(10), which provides that loans and lines of credit made primarily for a commercial or business purpose are excluded transactions unless they are for the purpose of home purchase under Sec. 1003.2(j), home improvement under Sec. 1003.2(i), or refinancing under Sec. 1003.2(p). New comment 3(c)(10)-1 explains the general rule and clarifies that Sec. 1003.3(c)(10) does not exclude all dwelling-secured business- or commercial-purpose loans or credit lines from coverage. New comment 3(c)(10)-2 explains how financial [[Page 66172]] institutions should determine whether a transaction primarily is for a commercial or business purpose. Specifically, comment 3(c)(10)-2 provides that a loan or line of credit that is business, commercial, or organizational credit under Regulation Z Sec. 1026.3(a) and related commentary also is business or commercial credit under Regulation C and subject to special reporting under Sec. 1003.3(c)(10).\244\ Comments 3(c)(10)-3 and -4 provide illustrative examples of business- or commercial-purpose loans and credit lines that are covered loans under the final rule, or that are excluded transactions under Sec. 1003.3(c)(10). --------------------------------------------------------------------------- \244\ The commentary to Regulation Z Sec. 1026.3(a) discusses some transactions (such as credit card transactions) that are not subject to Regulation C at all, and others (such as agricultural- purpose loans) that are excluded from Regulation C under final Sec. 1003.3(c)(9) regardless of whether they are for home purchase, home improvement, or refinancing purposes. The Bureau believes that the burden relief achieved through regulatory alignment supports relying on Regulation Z's commentary to the extent applicable. --------------------------------------------------------------------------- The Bureau intends Sec. 1003.3(c)(10) to maintain coverage of commercial-purpose transactions generally at its existing level. Section 1003.3(c)(10) does expand coverage of dwelling-secured commercial-purpose lines of credit, which are not currently required to be reported, by requiring them to be reported if they primarily are for home purchase, home improvement, or refinancing purposes.\245\ For the reasons discussed in the section-by-section analysis of Sec. 1003.2(o), the final rule equalizes reporting of closed-end loans and open-end credit lines. Section 1003.3(c)(10) thus treats all dwelling- secured, commercial-purpose transactions the same, whether closed- or open-end. The Bureau believes that relatively few dwelling-secured, commercial-purpose open-end lines of credit are used for home purchase, home improvement, or refinancing purposes.\246\ The Bureau thus expects that reporting them will impose a relatively small burden on financial institutions. And, for the reasons given, the Bureau concludes that coverage of dwelling-secured, commercial-purpose credit lines for home improvement, home purchase, or refinancing purposes, as finalized in this rule, is necessary to further HMDA's purposes, especially because this is a segment of the mortgage market for which the public and public officials lack significant data. --------------------------------------------------------------------------- \245\ A few commenters specifically requested that the Bureau exclude from coverage dwelling-secured, agricultural-purpose lines of credit. The final rule excludes such transactions under Sec. 1003.3(c)(9). See the section-by-section analysis of Sec. 1003.3(c)(9). \246\ As discussed in part VII below, the Bureau has faced challenges estimating institutions' open-end lending volume given limitations in publicly available data sources. For example, it is difficult to estimate commercial-purpose open-end lending volume because available data sources do not distinguish between consumer- and commercial-purpose lines of credit. --------------------------------------------------------------------------- Section 1003.3(c)(10) also expands coverage of applications by, or originations to, certain trusts. For simplicity and regulatory consistency, final comment 3(c)(10)-2 aligns the definition of business or commercial credit under Regulation C with that definition under Regulation Z Sec. 1026.3(a). In the 2013 TILA-RESPA Final Rule, the Bureau revised comments 3(a)-9 and -10 to Sec. 1026.3(a) to provide that certain trusts made primarily for personal, family, or household purposes are transactions to natural persons in substance if not in form. Thus, transactions involving trusts as described in Regulation Z comment 3(a)-10 are subject to general dwelling-secured reporting under Regulation C.\247\ The Bureau believes that the benefits of aligning the Sec. 1003.3(c)(10) test with Regulation Z justify the burdens of reporting these transactions.\248\ --------------------------------------------------------------------------- \247\ Section 1003.3(c)(10) sets forth rules only concerning coverage. When determining whether and how to report particular data points for covered trust transactions, financial institutions should rely on the guidance set forth in Sec. 1003.4 and accompanying commentary and instructions. \248\ In aligning with Regulation Z's interpretation of trusts for coverage purposes, the Bureau is declining to exclude trusts from reporting as some commenters urged. As discussed in the 2013 TILA-RESPA Final Rule, the Bureau believes that many dwelling- secured loans made to trusts are consumer-focused transactions in substance and that data about such transactions will fulfill HMDA's purposes of understanding how financial institutions are serving the housing needs of their communities, even if particular data points like age or credit score may not apply to all trust transactions. The final rule includes specific guidance about whether and how to report age (comment 4(a)(10)(ii)-4) or ethnicity, race, and sex (appendix B, instruction 7) for transactions involving trusts. --------------------------------------------------------------------------- Maintaining commercial reporting roughly at its existing level will burden financial institutions more than eliminating reporting of all commercial-purpose transactions, as many commenters suggested. Financial institutions will continue to report transactions for home purchase, home improvement or refinancing purposes, and they will incur some burden distinguishing commercial-purpose transactions subject to Sec. 1003.3(c)(10) from non-commercial-purpose transactions subject to the general dwelling-secured coverage test. Like the commercial-purpose test under Regulation Z Sec. 1026.3(a), the Sec. 1003.3(c)(10) test requires financial institutions to determine the primary purpose of the transaction by looking at a variety of factors (and not, for example, by applying a bright-line rule). In some cases, for transactions that have multiple purposes, this approach will require financial institutions to exercise their judgment about the transaction's primary purpose. The Bureau believes that the benefits of maintaining purpose-based reporting of commercial transactions, however, justify these burdens. As noted at the beginning of this section-by-section analysis, HMDA, unlike TILA and RESPA, does not exempt business- or commercial-purpose transactions from coverage. Rather, HMDA, like ECOA, as implemented by the Bureau's Regulation B, and the CRA, provides authority to cover commercial-purpose transactions. HMDA's scope reflects that HMDA has a somewhat broader-based, community-level focus than certain other consumer financial laws. Specifically, while HMDA endeavors to ensure that applicants and borrowers are not discriminated against in particular transactions, it also seeks to ensure that financial institutions are meeting the housing needs of their communities and that public-sector funds are distributed to improve private investments in areas where they are needed. HMDA's broader purposes are served by gathering data both about individual transactions to applicants or borrowers and, for example, about the available stock of multifamily rental housing in particular communities.\249\ The final rule achieves these goals without requiring institutions to report all dwelling-secured commercial-purpose transactions. The final rule also addresses commenters' concerns about commingling consumer- and commercial-purpose data by adding a commercial-purpose flag in Sec. 1003.4(a)(38).\250\ Finally, the final rule clarifies whether and how certain data points apply to commercial- purpose transactions.\251\ --------------------------------------------------------------------------- \249\ It is also for this reason that the final rule does not exclude particular categories of commercial-purpose lending, such as multifamily or subordinate-lien commercial lending, from coverage. \250\ See the section-by-section analysis of Sec. 1003.4(a)(38). \251\ See, e.g., comments 4(a)(10)(iii)-7 and 4(a)(23)-5, specifying that a financial institution reports ``not applicable'' for income relied on and debt-to-income ratio when the applicant or co-applicant is not a natural person or when the covered loan is secured by a multifamily dwelling. See also Sec. 1003.2(n) and comment 2(n)-2, which list special reporting requirements for multifamily dwellings. --------------------------------------------------------------------------- [[Page 66173]] 3(c)(11) As discussed in the section-by-section analysis of Sec. 1003.2(g), the final rule provides that a financial institution is covered under Regulation C and must report data about covered loans if, among other things, the financial institution originated more than 100 open-end lines of credit in the preceding two years. The Bureau recognizes that some financial institutions may be covered financial institutions because they meet the open-end line of credit threshold in Sec. 1003.2(g)(1)(v)(B) or (2)(ii)(B), but that these institutions may have closed-end mortgage lending volume that falls below the 25-loan coverage threshold in Sec. 1003.2(g)(1)(v)(A) or (2)(ii)(A). Section 1003.3(c)(11) provides that such institutions' closed-end mortgage loans are excluded transactions. The Bureau does not believe that it is useful to burden such institutions with reporting closed-end mortgage data merely because their open-end lending exceeded the separate, open- end loan-volume threshold in Sec. 1003.2(g). Comment 3(c)(11)-1 provides an illustrative example of the rule. 3(c)(12) As discussed in the section-by-section analysis of Sec. 1003.2(g), the final rule provides that a financial institution is covered under Regulation C and must report data about covered loans if, among other things, the financial institution originated more than 25 closed-end mortgage loans in the preceding two years. The Bureau recognizes that some financial institutions may be covered financial institutions because they meet the closed-end mortgage loan threshold in Sec. 1003.2(g)(1)(v)(A) or (2)(ii)(A), but that these institutions may have open-end line of credit volume that falls below the 100-line of credit coverage threshold in Sec. 1003.2(g)(1)(v)(B) or (2)(ii)(B). Section 1003.3(c)(12) provides that such institutions' open-end lines of credit are excluded transactions. The Bureau does not believe that it is useful to burden such institutions with reporting data about open-end lines of credit merely because their closed-end lending exceeded the separate, closed-end loan-volume threshold in Sec. 1003.2(g). Comment 3(c)(12)-1 provides an illustrative example of the rule. Section 1003.4 Compilation of Reportable Data 4(a) Data Format and Itemization Section 1003.4(a) requires financial institutions to collect and record specific information about covered loans, applications for covered loans, and purchases of covered loans. As discussed in detail below, the Bureau proposed several changes to Sec. 1003.4(a) to implement the Dodd-Frank Act amendments to HMDA and to exercise its discretionary authority under the Dodd-Frank Act to require collection of certain additional information. In addition, the Bureau proposed modifications to Regulation C to reduce redundancy, provide greater clarity, and make the data more useful. The Bureau proposed modifications to Sec. 1003.4(a) and comments 4(a)-1 and 4(a)-4 through -6. These revisions addressed reporting transactions involving more than one institution, reporting repurchased loans, and other technical modifications. In addition, the proposal solicited feedback on the number and type of data proposed to be collected. These issues are discussed below separately. Reporting Transactions Involving More Than One Institution Currently, commentary to Sec. 1003.1(c) describes the ``broker rule,'' which explains a financial institution's reporting responsibilities when a single transaction involves more than one institution. Proposed comments 4(a)-4 and -5 modified and consolidated current comments 1(c)-2 through -7 and 4(a)-1.iii and.iv. Proposed comment 4(a)-4 described which financial institution reports a covered loan or application when more than one institution is involved in reviewing a single application and provided illustrative examples. Proposed comment 4(a)-5 discussed reporting responsibilities when a financial institution makes a credit decision through the actions of an agent. The Bureau is adopting comment 4(a)-4, renumbered as comments 4(a)-2 and -3, with changes to address certain industry comments, discussed below. The Bureau received no comments on proposed comment 4(a)-5 and is adopting it as proposed, renumbered as comment 4(a)-4. Two industry commenters stated that they supported proposed comment 4(a)-4. Other industry commenters expressed concerns with proposed comment 4(a)-4. One industry commenter pointed out that loans originated as part of a State housing finance agency (HFA) program may not be reported under the proposed commentary because under those programs the State HFA, which the commenter asserted may not be required to report HMDA data, usually makes the credit decision. Another industry commenter urged the Bureau to allow more than one institution to report the same origination. The Bureau recognizes that some applications and loans will not be reported under proposed comment 4(a)-4, finalized as comments 4(a)-2 and -3, if the institution making the credit decision is not a financial institution required to report HMDA data. However, the Bureau believes that it is appropriate to limit reporting responsibilities to the financial institution that makes the credit decision. Requiring that only one institution report the origination of a covered loan eliminates duplicate data. For example, if more than one financial institution reported the same origination, the total origination volume for a particular census tract would appear higher than the actual number of loans originated in that tract. On balance, the Bureau concludes that only the financial institution that makes the credit decision should report an origination. Other industry commenters asked for examples of how to report a loan or application involving more than two institutions. The Bureau has added an example to proposed comment 4(a)-4, finalized as comment 4(a)-3, to illustrate financial institutions' reporting responsibilities when multiple institutions are involved. The example demonstrates that more than one financial institution will report the action taken on the same application if the same application is forwarded to multiple institutions. However, only one financial institution will report the loan as an origination. An industry commenter sought clarification about what is meant by application for the purposes of the proposed comment. Section 1003.2(b) defines application for purposes of Regulation C and, accordingly, for purposes of Sec. 1003.4(a) and its commentary. The Bureau is modifying proposed comment 4(a)-4, finalized as comments 4(a)-2 and -3, to clarify that Sec. 1003.4(a) requires a financial institution to report data on applications that it receives even if the financial institution received an application from another financial institution rather than directly from an applicant. In addition, a trade association asked the Bureau to clarify the reporting responsibilities when a credit union contracts a credit union service organization (CUSO) to perform loan origination services. The commentary to the final rule addresses these situations. Comment 4(a)-2 explains that the institution that makes the credit decision prior to closing or account opening reports that decision. [[Page 66174]] Accordingly, if a credit union makes a credit decision prior to closing or account opening, then the credit union reports that decision. In addition, comment 4(a)-3.v addresses situations when a financial institution (in this case the CUSO) makes a credit decision using the underwriting criteria of a third party (in this case the credit union). In that case, if the CUSO makes a credit decision without the credit union's review before closing, the CUSO reports the credit decision. However, if the CUSO approves the application acting as the credit union's agent under State law, comment 4(a)-4 clarifies that the credit union is required to report the actions taken through its agent. Purchased Loans In 2010, the FFIEC issued a publication in which it noted that repurchases qualify as purchases for Regulation C, and provided guidance on how and when to report such purchases.\252\ The Bureau proposed to incorporate this guidance into Regulation C by adding new comment 4(a)-5 to clarify that covered loans that had been originated by a financial institution, sold to another entity, and subsequently repurchased by the originating institution should be reported under Regulation C unless the sale, purchase, and repurchase occurred within the same calendar year. When the FFIEC publication was issued, data users could not reliably identify repurchased loans within HMDA data because each loan was reported with a unique application or loan number, even if it was a loan being repurchased. Thus loans repurchased and reported multiple times within the same calendar year would distort the annual HMDA data, because the characteristics of those loans would be represented multiple times within the data. For the reasons discussed below, the Bureau is not adopting comment 4(a)-5 as proposed and, instead, is revising it to require the reporting of most repurchases as purchased loans regardless of when the repurchase occurs. --------------------------------------------------------------------------- \252\ Fed. Fin. Insts. Examination Council, CRA/HMDA Reporter, Changes Coming to HMDA Edit Reports in 2010, at 5 (Dec. 2010), available at http://www.ffiec.gov/hmda/pdf/10news.pdf. --------------------------------------------------------------------------- Most commenters opposed the Bureau's proposal. Some industry commenters argued that repurchases should never be reported, even outside of the calendar year in which the loan was originated. Some industry commenters argued that the calendar year exception would negatively affect CRA ratings for some financial institutions that temporarily purchase CRA-eligible loans under certain lending arrangements. Other industry commenters argued that any reporting of repurchases would inflate CRA ratings by allowing the loans to appear in a financial institution's HMDA data more than once. However, a few commenters supported the Bureau's proposal and argued that repurchases should be considered purchases for purposes of HMDA except for when the repurchase occurs within the same calendar year as the loans were originated. The Bureau recognizes that the one-calendar-year reporting exception in the FFIEC guidance has led to inconsistent reporting of repurchased loans, because loans originated late in a calendar year and repurchased early in the succeeding calendar year are reported as loan purchases, while loans originated early in a calendar year and repurchased within the same calendar year are not reported. The Bureau also understands that there have been questions concerning the scope of the guidance and whether various scenarios constitute a repurchase or are addressed by the guidance. The Bureau has determined that repurchases of covered loans should be reported as loan purchases, with only a narrow exception discussed below. The Bureau believes that the one-calendar-year reporting exception, which was based on guidance originally published by the FFIEC, will no longer be needed in light of other elements of the final rule.\253\ The universal loan identifier (ULI), as adopted in Sec. 1003.4(a)(1)(i), will enable a loan to be identified in the HMDA dataset through multiple HMDA reporting events and the repurchase reporting event could be identified and not included in an analysis or compilation of HMDA data focused on originated loans or annual market volume. Repurchases after the origination and sale of a covered loan to a secondary market investor still effect a transfer of legal title to the covered loan, which then could be held in portfolio by the originating institution or sold to another secondary market investor later. Information about these transfers should be reflected in HMDA as purchases, just as the original purchase is, so that the information may be included in the HMDA dataset to further the purposes of HMDA, and so that the ULI may be used effectively to monitor covered loans through their lifecycle. --------------------------------------------------------------------------- \253\ Id. --------------------------------------------------------------------------- In addition, if repurchase data are not included, there could be gaps in the history of a covered loan. The Dodd-Frank Act also requires the U.S. Securities and Exchange Commission to prescribe regulations that require securitizers to disclose fulfilled and unfulfilled repurchase requests across all trusts aggregated by the securitizer.\254\ The Bureau believes that the usefulness of the HMDA data would be enhanced by having repurchases included so that information could be available through multiple HMDA reporting events. --------------------------------------------------------------------------- \254\ Dodd-Frank Act section 943; see also 17 CFR 240.15Ga-1. --------------------------------------------------------------------------- For the reasons discussed above, the Bureau is adopting comment 4(a)-5 with modifications. However, the Bureau is creating an exception for certain assignments of legal ownership of covered loans through interim funding arrangements that operate as the functional equivalent of warehouse lines of credit because they may not truly reflect sales and purchases of covered loans. These interim funding agreements are used as functional equivalents of warehouse lines of credit where legal title to the covered loan is acquired by the party providing interim funding, subject to an obligation of the originating institution to repurchase at a future date, rather than taking a security interest in the covered loan as under the terms of a more conventional warehouse line of credit. The Bureau does not believe that these arrangements should require reporting under Regulation C given the temporary nature of the transfer and the intent of the arrangement. Therefore, pursuant to HMDA section 305(a) the Bureau is incorporating an exception into comment 4(a)-5 for such agreements so that such activity will not be reported under Regulation C. This exception is necessary and proper to effectuate HMDA's purposes, because reporting of these transfers in addition to reporting of the underlying originations, subsequent purchases, and any repurchase at a later date may distort HMDA data without providing meaningful information that furthers HMDA's purposes. This exception will also facilitate compliance for financial institutions. Other Technical Modifications The Bureau also proposed technical modifications to 4(a) and proposed comment 4(a)-1. The Bureau received no comments on the proposed changes to 4(a) and proposed comment 4(a)-1 and is adopting them as proposed, with minor modifications. The Bureau is also moving comments 4(a)-1.iv, -2, and -3 to the commentary to Sec. 1003.4(f) to clarify a financial institution's [[Page 66175]] obligation to record data on a quarterly basis. Number of Data Points As detailed in the section 1022 discussion below, currently Regulation C requires reporting of approximately 35 separate pieces of information, and allows for optional reporting of three denial reasons. The Dodd-Frank Act amended HMDA by enhancing two existing data points (rate spread and application ID) and identifying 11 new data points, which the Bureau proposed to implement with 22 data fields. The Bureau also proposed to require financial institutions to report 13 additional data points not identified in the Dodd-Frank Act, implemented with 28 data fields, and to modify and expand some of the existing Regulation C data fields. Also detailed in the section 1022 discussion below, while the Bureau estimates that the incremental cost of each additional data point and associated data fields is small, the Bureau acknowledges that there are variable costs, one-time costs, and ongoing costs associated with the additional data points when considered collectively. The Bureau considered this in developing the proposal and proposed only those additional data points that the Bureau believes have sufficient value to justify the costs. As discussed below, the Bureau is not dramatically changing the number of the proposed data points, either by not adopting a substantial number of those that were proposed or by adopting substantially more than the number that were proposed. The number of data fields implementing some of the data points has increased based on changes the Bureau has adopted for the final rule. Some industry commenters stated that the Bureau should only require data points that were specifically defined in the Dodd-Frank Act. Some industry commenters also suggested removing data points currently required under Regulation C. Some industry commenters stated that the Bureau should only require certain financial institutions to report data points not specifically defined in the Dodd-Frank Act, such as institutions that had been found to be in violation of fair lending laws, HMDA, or the CRA, or institutions that exceed certain asset-size or loan-volume thresholds. Some industry commenters stated that the Bureau should conduct additional analysis on the value of the proposed data points before deciding whether to adopt them. Many consumer advocate commenters argued that the Bureau's proposal did not require enough information to be reported, and that additional information would be required to fulfill HMDA's purposes. Some industry and other commenters also suggested additional data points. Collectively these commenters suggested more than 45 additional data points. Some industry commenters and consumer advocate commenters stated that the Bureau's proposal was reasonable and measured in terms of the number of data points and made sense given the current mortgage market. The Bureau has analyzed the proposed data points carefully in light of the comments received and other considerations and believes that the data points adopted in this final rule each significantly advance the purposes of HMDA and are warranted in light of collection burdens. Each such data point is discussed below in the section-by-section analysis. The Bureau has authority to expand the data points collected to include such other information as it may require under HMDA section 304(b)(5)(D) and (b)(6)(J). As discussed below throughout the section- by-section analysis, the Bureau is adopting many of the data points proposed, modifying certain data points based on feedback received from commenters, and not finalizing certain proposed data points. Regarding the comments suggesting criteria or thresholds for reporting additional data points, the Bureau does not believe that it would be appropriate to condition the reporting of such data points on such criteria. The Bureau believes that the data points proposed to be reported fulfill HMDA's purposes and that limiting reporting of them to only some financial institutions would limit the usefulness of the data. Limiting reporting of certain information to financial institutions that had a history of violating certain laws would compromise the usefulness of the HMDA data because that information would not be available from other financial institutions, precluding the generating of a representative (presumptively non-violative) sample of the market for statistical comparison. Limiting reporting of certain information by asset size or loan volume would also undermine the utility of the HMDA data, because financial institutions that would fall under any threshold may have different characteristics and lending practices that would then not be visible through HMDA data. Financial institutions have different business models and underwriting practices which can, in part, be based on their asset size or loan volume. Excluding certain financial institutions would potentially exclude information about covered loans with different characteristics or information related to differences in underwriting practices and would create data that is not uniform. This would not only undermine HMDA's purposes, but limit information available to policymakers in considering how legal requirements should apply to different business models and underwriting practices. The Bureau also considered the additional data points suggested by commenters. As discussed below throughout the section-by-section analysis, certain data points have been modified to take into account some of these suggestions. The Bureau is not adopting many of these data points because it does not believe it has sufficient information at this time to determine whether adding them would serve HMDA's purposes and be warranted in light of collection burdens. Others the Bureau believes would be duplicative of, or would provide information only marginally different than, data points adopted in the final rule. Because many of these comments proposed data points similar to ones proposed by the Bureau, the responses to many of these comments are discussed below in the section-by-section analysis for the data point being finalized most relevant to those suggestions. 4(a)(1) 4(a)(1)(i) HMDA section 304(b)(6)(G), as amended by Dodd-Frank Act section 1094(3)(A)(iv), authorizes the Bureau to require a universal loan identifier, as it may determine to be appropriate.\255\ Existing Sec. 1003.4(a)(1) requires financial institutions to report an identifying number for each loan or loan application reported. The current commentary to Sec. 1003.4(a)(1) strongly discourages institutions from using the applicant's or borrower's name or Social Security number in the application or loan number. The current commentary also requires the number to be unique within the institution, but does not provide guidance on how institutions should select ``unique'' identifiers. The Bureau proposed to implement HMDA section 304(b)(6)(G) by replacing the current HMDA loan identifier with a new self-assigned loan or application identifier that would be unique throughout the industry rather than just within the reporting financial institution, would be used by all financial institutions that report the loan or application for HMDA purposes, [[Page 66176]] and could not be used to directly identify the applicant or borrower. The Bureau believes a reasonable interpretation of ``universal loan identifier'' in HMDA section 304(b)(6)(G) is that the identifier would be unique within the industry. For the reasons discussed below, the Bureau is adopting Sec. 1003.4(a)(1)(i) generally as proposed requiring entities to provide a universal loan identifier (ULI) for each covered loan or application. The Bureau is adding separate paragraphs to address purchased covered loans and applications that are reconsidered or reinstated during the same calendar year. In addition, as discussed below, the Bureau is adding a paragraph requiring a check digit as part of the ULI. --------------------------------------------------------------------------- \255\ 12 U.S.C. 2803(b)(6)(G). --------------------------------------------------------------------------- The Bureau solicited comment on whether the proposed changes to the loan or application identifiers used for HMDA reporting are appropriate. Most industry commenters expressed concern that the proposed ULI would introduce unnecessary complexities in the HMDA reporting process. Several industry commenters stated that requiring institutions to reinvent current loan numbering procedures would result in significant implementation costs because it would require a programming change to current operation systems, such as an institution's loan origination software. Industry commenters pointed out that most institutions assign loan numbers based on a certain order, such as the order the application was received, and furthermore that creditors may include information within the loan number that is pertinent to the institution's operations. For example, an industry commenter stated that its loan origination software assigns numbers randomly but uses a unique identifier for originations and a unique identifier for all other loans not originated. The Bureau acknowledges that the proposed ULI may pose operational challenges for financial institutions. However, the Bureau believes that the benefits that can be gained from the use of a ULI, including the potential ability to track an application or loan over its life and to help in accurately identifying lending patterns across various markets justify the burden associated with implementing a ULI. Additionally, the Bureau understands that financial institutions need flexibility for organizational purposes, such as the flexibility to assign loan numbers that include numbers that would represent product type. With this in mind, the Bureau proposed that the ULI would consist of up to an additional 25 characters that follow the Legal Entity Identifier (LEI) to identify the covered loan or application. The Bureau believes that this approach provides financial institutions with the flexibility to accommodate organizational purposes when assigning loan numbers, except that the additional 25 characters must not include any information that could be used to directly identify the applicant or borrower. Currently, institutions assign alphanumeric identifiers, with up to 25 characters, to identify a covered loan or application. The Bureau proposed a maximum 45-character ULI. The first 20 characters would be comprised of the LEI followed by up to 25 characters, which would represent the unique sequence of characters to identify the covered loan or application, and may be letters, numerals, symbols, or a combination of letters, numerals, and symbols. A trade association recommended that the ULI be lengthened to 65 characters, as opposed to the proposed 45. An industry commenter stated that an institution could run out of identifiers quickly with the proposed maximum. The Bureau believes that lengthening the proposed ULI may benefit some institutions with large loan volumes that may use certain characters in the ULI to represent business lines or branches, but, at the same time, a ULI longer than 45 characters may be burdensome for other financial institutions. The Bureau believes the right balance between flexibility and usability is a maximum of 45 characters in the ULI, with the first 20 characters representing the LEI. A few commenters expressed concerns regarding the potential errors that could arise in a loan identifier as long as 45 characters. One commenter stated that manual input of a 45-digit loan identifier will likely result in typos while another commenter suggested that manual input would need to take place to ensure accurate information because there is potential room for error with a 45-character loan identifier. To address the potential errors that could arise, an industry commenter recommended that the Bureau consider adding a check-digit requirement to the ULI. A check digit is used to validate or verify that a sequence of numbers or characters, or numbers and characters, are correct. A mathematical function is applied to the sequence of numbers or characters, or numbers and characters, to generate the check digit. This mathematical methodology could then be performed at a point in the HMDA process to ensure that the check digit resulting from performing the mathematical methodology on the sequence of letters or numerals, or letters and numerals, matches the check digit in the ULI. Implementation of a check digit can help ensure that the sequence of characters assigned to identify the covered loan or application are persistent throughout the HMDA process. For example, at the application stage, a financial institution assigns the ULI, which consists of the financial institution's LEI, a 23-character unique sequence of letters and numerals that identify the application, and a 2-character check digit. Once the application is complete, the file is transferred to another division of the financial institution where it will be handled by other staff. To ensure that the ULI was transferred correctly, the mathematical function could be performed to obtain the check digit and ensure that it matches the check digit in the ULI. This would ensure that the ULI does not contain an error due to typos or transposition of characters as a result of manual entry or file transfer errors. If the check digit resulting from the performed mathematical function does not match the check digit in the ULI, then it would be an indication to staff that an error in the ULI exists. Adding a check digit requirement in the ULI also benefits the file transfer process between financial institutions. For example, a file transfer process could be initiated because the loans are sold to another financial institution. The financial institution that originated the loans electronically transmits to the financial institution that purchases the loans the applicable information, including the ULI, related to the loans. Although an electronic transmission reduces the incidence of errors, it is not guaranteed because of the likelihood that the institutions use different systems to capture the data and therefore, the financial institution that purchased the loans may need to implement specific software to intake the data. In addition, unlike other information related to the loan that can undergo a quality control process through the implementation of business logic and statistical analyses, the ULI does not contain information that would make it possible to ensure that the ULI transferred is valid through the application of business logic or statistical analyses. Therefore, implementation of a check digit can help ensure that the ULI was transferred correctly. The check-digit requirement would enable financial institutions to quickly identify and correct errors in the ULI, which would ensure a valid ULI, and therefore enhance data quality. Check digits are currently implemented in [[Page 66177]] certain identifiers, such as vehicle identification numbers, which function as a check against transcription errors.\256\ The national unique health plan identifier implemented by the U.S. Department of Health and Human Services also incorporates a check digit.\257\ The Bureau believes that the benefits of a check digit in the ULI justifies the additional burden associated with implementing a check digit. --------------------------------------------------------------------------- \256\ See 73 FR 23367, 23369 (Apr. 30, 2008). \257\ See 77 FR 54664, 54675 (Sept. 5, 2012). --------------------------------------------------------------------------- The Bureau is publishing in this final rule new appendix C that includes the methodology for generating a check digit and instructions on how to validate a ULI using the check digit. The methodology is adapted from Mod 97-10 \258\ in the international standard ISO/IEC 7064, which is published by the International Organization for Standardization (ISO).\259\ ISO/IEC 7064 specifies check character systems that can detect errors in a string of characters that are the result of data entry or copy errors.\260\ Specifically, ISO/IEC 7064 check character systems can detect errors caused by substitution or transposition of characters. For example, the check digit can detect a transposition error such as when two adjacent numbers are transposed or when a single character is substituted for another. The Bureau believes that the identification of these types of errors will enhance data quality and reduce burden in the long run for institutions because the errors can be identified early in the process. To reduce burden, the Bureau plans to develop a tool that financial institutions may use, at their option, to assist with check digit generation. --------------------------------------------------------------------------- \258\ Mod 97-10 applies the mathematical function modulus, which is defined by ISO as an integer used as a divisor of an integer dividend in order to obtain an integer remainder. \259\ ISO is the world's largest developer of international standards and has published over 19,500 standards that cover aspects of business and technology. ISO is comprised of national standards bodies from 162 member countries. More information about ISO and the standards is available at http://www.iso.org/iso/home.html. \260\ Int'l. Org. for Standards, ISO/IEC 7064:2003, Information technology-Security techniques-Check character systems (Feb. 15, 2003), http://www.iso.org/iso/home.html. --------------------------------------------------------------------------- For the reasons stated above, the Bureau adopts as final the requirement to include a check digit to the ULI. In order to maintain the maximum 45-character ULI, the Bureau is also modifying the maximum number of additional characters to identify the covered loan or application and reducing it from the proposed 25 to 23. Several industry commenters suggested that the Bureau should consider using the MERS Mortgage Identification Number (MIN) as the core of the ULI.\261\ The MIN is an 18-digit number registered on the MERS System. The first seven digits of the 18-digit MIN number would be the financial institution's identification number assigned by MERS. The next 10 digits would be assigned by the financial institution and the last digit serves as a check digit. One commenter stated that uniqueness is important in a loan number and that the MIN could guarantee uniqueness because it is registered with the MERS System. The MIN is usually issued at origination but may be issued at application. For the reasons discussed below, the Bureau is not adopting a ULI that uses the MIN as the core. --------------------------------------------------------------------------- \261\ The MERS System is owned and managed by MERSCORP Holdings, Inc., an industry-owned and privately held corporation. According to MERSCORP, the MERS System is a national electronic database that tracks changes in mortgage servicing and beneficial ownership interests in residential mortgage loans on behalf of its members. --------------------------------------------------------------------------- First, a rule that prescribes the MIN as the core would require all financial institutions reporting HMDA data to register with MERSCORP and obtain an organization number assigned by MERSCORP. This organization number would not be able to serve the same function as the LEI described in the section-by-section analysis of Sec. 1003.5(a)(3) below because there would not be a way to link HMDA-reporting institutions with their corporate families using the MERS identification number. Second, the 10-digit number assigned by the institution that would serve as the identification number that can be used to identify and retrieve the loan application would not provide the same flexibility as the maximum 23-character that the ULI provides. Some financial institutions may need more than 10 digits to identify and retrieve a loan application because certain characters in the loan number may represent branches or business lines. For these reasons, the Bureau is not adopting a ULI that uses the MERS MIN as the core. Some industry commenters suggested that the ULI should be identical to the loan identification number prescribed by the 2013 TILA-RESPA Final Rule. That rule provides that the loan identification number is a number that may be used by the creditor, consumer, and other parties to identify the transaction.\262\ See Regulation Z Sec. 1026.37(a)(12). Although the burden on industry would be mitigated if the Bureau required that financial institutions use the same loan identification number for HMDA reporting as the loan identification number in the TILA-RESPA disclosures, the Bureau believes that an application number that may meet the TILA-RESPA standards may not be appropriate for HMDA reporting. Section 1026.37(a)(12) does not limit the number of characters in the loan application number. The lack of limitation enables creditors to assign as many characters in the loan application number as they want, which could result in compliance challenges for users of the ULI. For example, if an institution purchases a loan with a 60-character application number assigned by the institution that originated the loan pursuant to Sec. 1026.37(a)(12), the institution that purchased the loan would need to make updates to their system to accommodate a 60-character ULI in order to report the purchased loan under HMDA if the purchasing institution's system was programmed to handle ULIs with a maximum number of 45 characters pursuant to Regulation C. For these reasons, the Bureau is not adopting a rule that would enable institutions to use the TILA-RESPA loan application number for the ULI. The Bureau notes, however, that the loan application number requirements in the TILA-RESPA rule are not necessarily incompatible with the ULI. Therefore, a financial institution may generate a ULI for both HMDA and TILA-RESPA. --------------------------------------------------------------------------- \262\ See 78 FR 79730 (Dec. 31, 2013). The rule is effective on October 3, 2015 and applies to transactions for which the creditor or mortgage broker receives an application on or after that date. --------------------------------------------------------------------------- The Bureau also proposed that the ULI may consist of letters, numbers, symbols, or a combination of letters, numbers, and symbols. While the Bureau did not receive any comments regarding the use of letters or numbers, the Bureau received a comment from industry stating that symbols may contain embedded special characters that could potentially result in interference with applications or programs that use the ULI. In addition, certain symbols may not be recognized by certain programs that use HMDA data. The commenter suggested that the Bureau should provide a list of symbols that are permissible in the ULI or provide a list of symbols that are not permissible in the ULI. After considering the comment, the Bureau concluded that symbols in the ULI can potentially present challenges for financial institutions and data when reporting or analyzing HMDA data. Therefore, the final rule does not permit the use of symbols, as in proposed Sec. 1003.4(a)(1)(i)(B)(1). The Bureau is adopting a final rule that provides that [[Page 66178]] the maximum number of characters in the ULI must be 45, with the first 20 characters representing the LEI followed by up to 23 additional characters that may be letters, numerals, or a combination of both, and a 2-character check digit. The Bureau explained in the proposal that the current identifier requirement makes it difficult to track an application or loan over its life. Commenters, including industry, consumer advocates, and trade associations, supported the proposed ULI because it would require a financial institution that reports HMDA data and that reports a purchased loan to report the same ULI that was previously reported under HMDA by the financial institution that originated the loan. One commenter stated that the ULI will enable a much better understanding of how the market works and how loans perform. Another commenter pointed out that the ULI is the single most useful addition for regulators to assess what happens after a loan is originated, from servicer changes to secondary mortgage market activity. Another commenter supporting the proposed ULI argued that a ULI that follows a loan through various permutations may help shed light into which racial and ethnic minority homeowners may be disproportionately subjected to predatory lending, foreclosure, fraud, and underwater mortgages. A commenter that supported the ULI stated that issues regarding the ULI could arise in a transaction that involves a purchased covered loan. Specifically, the commenter noted that the proposal did not specify which entity assigns the ULI at the initial reporting of the covered loan, particularly if a quarterly reporter purchased the loan and reports it prior to the annual reporter that originated the loan. The Bureau recognizes that the proposal may have created confusion regarding the ULI on purchased covered loans. To eliminate the confusion, the Bureau is adding Sec. 1003.4(a)(1)(i)(D) to address purchased covered loans. Section 1003.4(a)(1)(i)(D) provides that a financial institution that reports a purchased covered loan must use the ULI that was assigned or previously reported for the covered loan. For example, if a quarterly reporter pursuant to Sec. 1003.5(a)(1)(ii) purchases a covered loan from a financial institution that is an annual reporter and that submits data annually pursuant to Sec. 1003.5(a)(1)(i), the quarterly reporter that purchased the covered loan must use the ULI that the financial institution that is an annual reporter assigned to the covered loan. Additionally, the Bureau is adding Sec. 1003.4(a)(1)(i)(E) to address the option for using the same ULI for an original and reinstated or reconsidered application that occur during the same calendar year. For example, assume a quarterly reporter pursuant to Sec. 1003.5(a)(1)(ii) takes final action on an application in the first quarter and submits it with its first quarter information. If in the second quarter during the same calendar year, the financial institution reconsiders the application and takes final action in the second quarter that is different from that in the first quarter, the financial institution may use the same ULI that was reported in its first quarter data. The Bureau believes that providing this option for financial institutions will reduce burden associated with assigning a new ULI for a later transaction that a financial institution considers as a continuation of an earlier transaction. The Bureau proposed Sec. 1003.5(a)(3) to require a financial institution to provide an LEI when the financial institution reports its data. Section 1003.5(a)(3) also describes the issuance of the LEI. The Bureau is adopting the requirement in Sec. 1003.5(a)(3) to require a financial institution to provide its LEI when reporting its data, as discussed in detail below in the section-by-section analysis of Sec. 1003.5(a)(3). However, the Bureau is making a technical change and moving the description of the issuance of the LEI to Sec. 1003.4(a)(1)(i)(A) for ease of reference. See the section-by-section analysis of Sec. 1003.5(a)(3) below for more information. For these reasons and those above, the Bureau is adopting Sec. 1003.4(a)(1)(i) generally as proposed, with modifications related to symbols and the number of characters, the issuance of the LEI, additional clarification related to purchased covered loans and previously reported applications, and the addition of the check digit requirement. The Bureau solicited feedback regarding hashing as an encryption method for the ULI. The Bureau also solicited feedback on salting in addition to hashing to enhance the encryption. One industry commenter recommended that the Bureau finalize hashing and salting while most other industry commenters opposed such a requirement arguing that it would not provide any benefit but would entail an additional cost, including expertise and resources. After considering the comments, the Bureau has concluded that the benefits of hashing and salting would not be sufficient to justify the costs of such requirements. Accordingly, the Bureau is not adopting a requirement that the ULI must be encrypted using a hash algorithm. Proposed comment 4(a)(1)(i)-1 clarified the uniqueness requirement of the ULI. The Bureau did not receive any comments on proposed comment 4(a)(1)(i)-1, which is adopted generally as proposed, but with technical modifications. The Bureau did not receive feedback on comment 4(a)(1)(i)-2, which provided guidance on the ULI's privacy requirements, and is adopted as proposed. The Bureau is also adopting new comments 4(a)(1)(i)-3 through -5 to provide guidance and illustrative examples for the ULI on purchased covered loans and reinstated or reconsidered applications, and guidance on the check digit. 4(a)(1)(ii) The Bureau proposed Sec. 1003.4(a)(1)(ii) to provide for reporting of the date the application was received or the date shown on the application form. For the reasons discussed below, the Bureau is finalizing Sec. 1003.4(a)(1)(ii) as proposed with minor revisions to the associated commentary. Some commenters requested additional guidance on reporting application date. Many of these comments stated that application date is difficult to report for commercial loans, because the application process is much more fluid than in consumer lending and an application form may not be formally completed until the end of the application process for some commercial loans. These concerns will be reduced by the Bureau's decision to generally maintain reporting of dwelling- secured, commercial-purpose transactions at its current level as discussed above in the section-by-section analysis of Sec. 1003.3(c)(10). For those commercial loans that will be required to be reported, the definition of application, combined with the ability to rely on the date shown on the application form, permits sufficient flexibility for financial institutions to report application date for commercial loans. A commenter suggested that instead of reporting application date financial institutions should report only the month of application to ease compliance. The Bureau believes such a change would reduce the data's utility. Because interest rates can change more rapidly than monthly, and policies or criteria that affect the action taken on applications can change during a calendar month, it is important to have a more complete application date reporting requirement so that loans can be grouped appropriately for analysis. [[Page 66179]] Therefore, the Bureau is finalizing Sec. 1003.4(a)(1)(ii) as proposed, and finalizing comment 4(a)(1)(ii)-1 as proposed with minor revisions to provide additional guidance on reporting application date when multiple application forms are processed. The Bureau received no specific feedback on comment 4(a)(1)(ii)-2 and is finalizing it as proposed. The Bureau is adding additional language to comment 4(a)(1)(ii)-3 for clarity. The Bureau is deleting comment 4(a)(ii)-4, because it is duplicative of comment 4(a)(8)(i)-14. 4(a)(2) HMDA section 304(b)(1) requires financial institutions to report the number and dollar amount of mortgage loans which are insured under Title II of the National Housing Act or under Title V of the Housing Act of 1949 or which are guaranteed under chapter 37 of Title 38. The Bureau proposed to retain the current reporting requirement, but incorporate the text of the statutory provision, with conforming modifications, directly into Regulation C. For the reasons discussed below, the Bureau is finalizing Sec. 1003.4(a)(2) with modifications to maintain consistency with the current reporting requirement. Commenters suggested various changes to the requirement, including aligning it with similar categories in other regulations, including new categories, or exempting certain types of covered loans from the requirement. A few commenters suggested adding an additional enumeration for State housing agency loans. Because many loans that State housing agencies are involved with are also insured or guaranteed by FHA or another government entity, the Bureau does not believe that adding an additional enumeration would accurately capture State housing agency loans without requiring financial institutions to select multiple categories, which would add additional burden and complexity. Other commenters suggested aligning to the Regulation Z Sec. 1026.37(a)(10)(iv) loan type categories, which would remove the category for USDA Rural Housing Service and Farm Service Agency loans and combine it with State housing agency loans under an ``other'' category. The Bureau believes that the less burdensome approach is to maintain the current category for USDA Rural Housing Service and Farm Service Agency loans and not adopt a new category incorporating multiple types of covered loans. Some commenters also argued that commercial loans should be exempted from this requirement, or that a Small Business Administration enumeration should be added. The Bureau is adopting a reporting requirement to identify covered loans primarily for a business or commercial purpose as discussed in the section-by-section analysis of Sec. 1003.4(a)(38) below and therefore believes it would be largely duplicative to add a reporting requirement specifically for Small Business Administration loans, especially considering that such loans are not specifically identified by HMDA section 304(b)(1). After considering the comments and conducting additional analysis, the Bureau is finalizing Sec. 1003.4(a)(2) with modifications. The Bureau is specifying the name of the government insurer or guarantor instead of the chapter or title of the United States Code or statute under which the loan is insured or guaranteed as specified in the statutory text to maintain consistency with current reporting requirements provided in appendix A to Regulation C. Federal Housing Administration Title I loans would be reported as FHA loans in addition to Title II loans. Because Title I loans include many manufactured housing loans, the Bureau is concerned that if the proposal were finalized as proposed, Title I manufactured housing loans would have been reported as conventional loans which would not clearly distinguish them from home-only manufactured home loans not insured by FHA. 4(a)(3) Current Sec. 1003.4(a)(3) requires financial institutions to report the purpose of a loan or application using the categories home purchase, home improvement, or refinancing. The Bureau proposed only technical modifications to Sec. 1003.4(a)(3) to conform to proposed changes in transactional coverage and to add an ``other'' category, but sought comment regarding whether the loan purpose reporting requirement should be modified with respect to home improvement loans and cash-out refinancings. For the reasons discussed below, the Bureau is adopting Sec. 1003.4(a)(3) with modifications to include a cash-out refinancing category and to make changes to the commentary to implement this additional category and provide instructions for reporting covered loans with multiple purposes. Some commenters addressed the home improvement loan purpose reporting requirement. One commenter suggested that the loan purpose be simplified to track only whether a loan was for purchase of a dwelling or not, as discerning a borrower's intent can be difficult. Other commenters also stated that determining home improvement purpose can be difficult for cash-out refinancings and other loans, and various commenters recommended eliminating the home improvement purpose category. However, some commenters supported requiring financial institutions to identify loans and applications with a home improvement purpose. The Bureau believes that the home improvement purpose continues to be an important indicator of home financing available for home improvements, and therefore is preserving that loan purpose category in this final rule. The Bureau solicited comment on the utility and feasibility of requiring a cash-out refinancing purpose, as distinct from refinancings generally. Many commenters stated that cash-out refinancings do not have a standardized definition in the industry and can vary by loan program or financial institution. Some commenters argued that definitional problems would make any reporting requirement difficult. A few commenters argued that the most the Bureau should require would be to report whether the financial institution considered the loan or application to be a cash-out refinancing rather than trying to establish a specific definition for HMDA purposes alone. Other commenters stated that reporting of cash-out refinancings would enhance the HMDA data by shedding light on borrowers taking equity out of their homes and differentiate these refinancings from rate-and-term refinancings in the data. Some commenters also noted that there is often a pricing difference between cash-out refinancings and other refinancings and that differentiating them in the data would be helpful. One commenter stated that the Bureau should adopt an additional data point for Regulation C indicating the amount of cash received by the consumer at closing. The Bureau does not believe it would be appropriate to adopt a specific additional data point for cash received by the consumer at closing at this time. The amount of cash received might not be a true indicator of whether the loan was considered or priced as a cash-out refinancing, because some financial institutions and loan programs allow for a limited amount of cash to be received in rate-and-term refinancings. However, the Bureau believes that differentiating cash-out refinancings in HMDA data will be valuable because there are often significant differences in rates or fees between cash-out refinancings and rate- [[Page 66180]] and-term refinancings.\263\ These differences might not otherwise be distinguishable in the HMDA data and could appear to be a result of discrimination in a fair lending analysis if the distinction could not be controlled for. --------------------------------------------------------------------------- \263\ See, for example, Fannie Mae, Loan-Level Price Adjustment Matrix (July 1, 2015), available at https://www.fanniemae.com/content/pricing/llpa-matrix.pdf; Freddie Mac, Bulletin 2015-6 Ex. 19 Postsettlement Delivery Fees (Apr. 17, 2015), available at http://www.freddiemac.com/singlefamily/pdf/ex19.pdf. --------------------------------------------------------------------------- Therefore, pursuant to HMDA sections 305(a) and 304(b)(6), the Bureau is finalizing Sec. 1003.4(a)(3) with the addition of a cash-out refinancing loan purpose. The Bureau believes this addition will carry out HMDA's purposes, by, for example, assisting in enforcing antidiscrimination statutes. The Bureau is adopting new comment 4(a)(3)-2 to provide guidance on reporting cash-out refinancings. This comment provides that a financial institution reports a covered loan or an application as a cash-out refinancing if it is a refinancing as defined by Sec. 1003.2(p) and the institution considered it to be a cash-out refinancing in processing the application or setting the terms under its guidelines or an investor's guidelines. This comment also provides illustrative examples. Some commenters stated that the Regulation C loan purpose categories should be aligned with the loan purpose categories in Regulation Z Sec. 1026.37(a)(9). HMDA section 304(b) requires the disclosure of home improvement loans, which is not a loan purpose under Regulation Z Sec. 1026.37(a)(9). Further, the Bureau is adopting a cash-out refinancing loan purpose category for Regulation C as discussed above, whereas Regulation Z Sec. 1026.37(a)(9) contains only a refinancing purpose. Because these differences are important for the purposes of Regulation C, the Bureau does not believe that aligning Sec. 1003.4(a)(3) with Regulation Z Sec. 1026.37(a)(9) would be appropriate. After considering the comments and conducting additional analysis, the Bureau is finalizing Sec. 1003.4(a)(3) with modifications to include cash-out refinancings. Comment 4(a)(3)-1, which is part of current Regulation C but was not included in the proposal, is adopted with changes to provide additional guidance for reporting the ``other'' category. Comment 4(a)(3)-2 is generally adopted as proposed, with conforming changes related to the addition of the cash-out refinancing purpose and renumbered as 4(a)(3)-3. Comment 4(a)(3)-3 provides guidance on reporting covered loans that would qualify under multiple categories under the Sec. 1003.4(a)(3) reporting requirement. The revised comment would provide that a covered loan that is both a cash- out refinancing or a refinancing and a home improvement loan should be reported as a cash-out refinancing or refinancing. The Bureau believes that this will make the cash-out refinancing and refinancing reporting categories more valuable by clearly identifying loans that are considered cash-out refinancings or refinancings whether or not they are for home improvement. Proposed comment 4(a)(3)-3 is adopted with modifications related to the addition of the cash-out refinancing purpose and is renumbered as 4(a)(3)-4. The Bureau is adopting new comment 4(a)(3)-5 to provide guidance on reporting loan purpose under Regulation C for loans with a business or commercial purpose when such loans are not excluded from coverage. 4(a)(4) Current Sec. 1003.4(a)(4) requires financial institutions to identify whether the application is a request for a covered preapproval. The Bureau proposed to continue this requirement and proposed minor technical revisions to the instructions in appendix A. Comments related to preapprovals are discussed in the section-by- section analysis of Sec. 1003.2(b)(2) and Sec. 1003.4(a). The Bureau is finalizing Sec. 1003.4(a)(4) with modifications to clarify the requirement. Based on additional analysis, the Bureau is also finalizing new comment 4(a)(4)-1 to provide guidance on the requirement and to simplify the current reporting requirement. Currently appendix A provides three codes for reporting this requirement: Preapproval requested, preapproval not requested, and not applicable. The instructions provide that preapproval not requested should be used when an institution has a preapproval program but the applicant did not request a preapproval through that program and that not applicable should be used when the institution does not have a preapproval program and for other types of loans and applications that are not part of the definition of a preapproval program under Regulation C. The Bureau has found that it is a common error for financial institutions to incorrectly report not applicable instead of preapproval not requested. The information provided by distinguishing these situations is of limited value, and the Bureau believes that it will reduce compliance burden to no longer have separate reporting options based on this distinction. Comment 4(a)(4)-1 provides that an institution complies with the reporting requirement by reporting that a preapproval was not requested regardless of whether the institution has such a program and the applicant did not apply through that program or if the institution does not have a preapproval program as defined by Regulation C. The Bureau is also finalizing new comment 4(a)(4)-2 to provide guidance on the scope of the reporting requirement. 4(a)(5) Regulation C currently requires reporting of the property type to which the loan or application relates as one- to four-family dwelling (other than manufactured housing), manufactured housing, or multifamily dwelling. The Bureau proposed to replace the requirement to report property type under Sec. 1003.4(a)(5) with the requirement to report the construction method for the dwelling related to the property identified in Sec. 1003.4(a)(9). For the reasons discussed below, the Bureau is adopting Sec. 1003.4(a)(5) with modifications to remove the ``other'' reporting category and finalizing a new comment providing guidance on reporting construction method for manufactured home communities. Some commenters supported the proposed changes and the treatment of modular housing. Other commenters argued that the current property type reporting requirement should be retained. A few commenters argued that the construction method and property type reporting requirement should be removed entirely. The Bureau does not agree that combining construction method and number of units as the current Sec. 1003.4(a)(5) property requirement does is appropriate, and believes separating these concepts into two distinct requirements will provide data that better reflects how financial institutions are serving the housing needs of their communities. The Bureau is therefore, pursuant to HMDA sections 305(a) and 304(b)(6)(J), finalizing Sec. 1003.4(a)(5) generally as proposed, but with modifications. The Bureau believes that the modifications will carry out HMDA's purposes and facilitate compliance therewith by providing more detail regarding whether institutions are serving the housing needs of their communities and by better aligning reporting to industry standards. The Bureau is removing the ``other'' option for reporting of construction method, because, as discussed in the section- by-section analysis of Sec. 1003.2(f), the Bureau is [[Page 66181]] finalizing the exclusion for many types of structures (such as recreational vehicles, houseboats, and pre-1976 mobile homes) that do not meet the definition of a manufactured home under Sec. 1003.2(l). In light of this change, the Bureau believes that an ``other'' category is unnecessary. Proposed comment 4(a)(5)-1 is being adopted generally as proposed, with minor revisions for clarity. Proposed comment 4(a)(5)-2 is being adopted as proposed, renumbered as comment 4(a)(5)- 3. The Bureau is also adopting new comment 4(a)(5)-2 to provide guidance on reporting the construction method for manufactured home communities. As discussed in the supplementary information to the proposed rule, the FFIEC had previously provided guidance to report the property type for manufactured home communities as manufactured housing.\264\ Based on a review of recent HMDA data, the Bureau believes that, while some financial institutions are following this prior guidance, some financial institutions may not be. The Bureau therefore believes it will facilitate compliance to include a comment specifically on the topic of reporting construction method for covered loans secured by manufactured home communities. --------------------------------------------------------------------------- \264\ 79 FR 51731, 51768 (Aug. 29, 2014); Fed. Fin. Insts. Examination Council, CRA/HMDA Reporter, Changes Coming to HMDA Edit Reports in 2010 (Dec. 2010), available at http://www.ffiec.gov/hmda/pdf/10news.pdf. --------------------------------------------------------------------------- A few commenters argued that additional information related to the construction of the dwelling should be reported. One trade association argued that the age of the dwelling should be reported in order to provide public data about housing finance as the housing stock ages, which would be helpful for understanding housing demand. Another commenter argued that individual condominium or cooperative units should be identified as such in HMDA data, which would facilitate housing research in large metropolitan areas. While both suggested modifications would improve the data, the Bureau does not believe that the benefits of these data would justify the burden at this time. However, the Bureau believes that with the requirement to report property address under Sec. 1003.4(a)(9), it may be possible to derive a proxy for condominium and cooperative units from the fact that unit numbers generally are included as part of the property address for such units. The Bureau may explore whether it would be possible to include such data in the release of HMDA data. 4(a)(6) HMDA section 304(b)(2) requires the disclosure of the number and dollar amount of mortgage loans made to mortgagors who did not, at the time of execution of the mortgage, intend to reside in the property securing the mortgage loan. Current Sec. 1003.4(a)(6) requires reporting the owner occupancy status of the property as owner-occupied as a principal dwelling, not owner-occupied as a principal dwelling, or not applicable. The Bureau proposed to require financial institutions to report whether a property will be used as a principal residence, as a second residence, as an investment property with rental income, or as an investment property without rental income. The Bureau proposed changes to appendix A to require distinguishing between investment properties with rental income and investment properties without rental income. For the reasons discussed below, the Bureau is finalizing Sec. 1003.4(a)(6) with modifications to require reporting of whether the property is a principal residence, second residence, or investment property. Some commenters generally supported reporting based on borrower occupancy rather than owner occupancy. Some commenters supported the additional category for second residences. Many commenters addressed the proposed investment property reporting requirement. Some commenters argued that the distinction between rental income and other investment properties would be burdensome and unnecessary. Some commenters also believed the example provided in comment 4(a)(6)-4 was inconsistent with the general exclusion for transitory residences in proposed comment 2(f)-2 (final comment 2(f)-3). Other commenters believed that the distinction would be helpful for research. Some commenters stated that investment properties with rental income would not be sufficient, that in addition it would be important for research to identify multi- unit dwellings where the borrower occupies one unit and rents the remaining units. The Bureau believes that multi-unit owner-occupied rental properties would be identifiable under the proposed reporting requirement as principal residences with more than one unit reported under the requirements of Sec. 1003.4(a)(31). The Bureau recognizes that the proposal's investment property distinction may pose compliance challenges and is inconsistent with some industry standards for categorizing occupancy. The Bureau is therefore finalizing Sec. 1003.4(a)(6) with modifications. The Bureau is combining investment properties into a single category. The Bureau is also finalizing comment 4(a)(6)-4 with modifications to clarify that the example refers to a long-term residential property and to replace the proposed term ``owner'' with ``borrower or applicant'' for consistency with Sec. 1003.4(a)(6) and comments 4(a)(6)-2 and -3. The Bureau is finalizing proposed comment 4(a)(6)-5 regarding multiple properties as final comment 4(a)(6)-1. Current comment 4(a)(6)-1 also deals with multiple properties and the Bureau believes that the comments should be consolidated into final comment 4(a)(6)-1. For the reasons stated in the preamble to the proposed rule, the Bureau believes that the finalized reporting requirement will provide valuable information about owner-occupancy for determining how financial institutions are serving the housing needs of their communities and the requirement as adopted will further understanding of how second homes and investment properties affect housing affordability and affect local communities.\265\ The Bureau is therefore finalizing Sec. 1003.4(a)(6) with modifications as discussed above to implement section 304(b)(2) of HMDA and pursuant to its authority under sections 305(a) and 304(b)(6)(J) of HMDA. The Bureau believes requiring this level of detail about residency status is a reasonable interpretation of HMDA section 304(b)(2). Furthermore, for the reasons [[Page 66182]] given above and in the preamble to the proposed rule, the Bureau believes this change is necessary and proper to effectuate HMDA's purposes, because this information will help determine whether financial institutions are serving the housing needs of their communities and will assist in decisions regarding the distribution of public sector investments. --------------------------------------------------------------------------- \265\ The Bureau adopts its discussion of the benefits of this change provided in the preamble to the proposed rule. See 79 FR 51731 at 51768-69; see also Deborah Halliday, You Can't Eat the View: The Loss of Housing Affordability in the West, The Rural Collaborative at 9-10 (2003); Linda Venturoni, Northwest Council of Governments, The Economic and Social Effects of Second Homes-- Executive Summary at 4-5 (June 2004) (stating that as the number of second homes in a community increases, the more the local economy will shift towards serving the needs of the second homes); Andrew Haughwout et al., Fed. Reserve Bank of New York, Staff Report No. 514, Real Estate Investors, the Leverage Cycle, and the Housing Market Crisis, at 21 (Sept. 2011); see also, e.g., Allan Mallach, Urban Institute, Investors and Housing Markets in Las Vegas: A Case Study, at 32-34 (2013) (discussing that foreign real estate investors in Las Vegas are crowding out potential domestic purchasers); Robert D. Cruz and Ebony Johnson, Miami-Dade Cnty. Regulatory and Economic Resources Dept., Research Notes on Economic Issues: Impact of Real Estate Investors on Local Buyers, (2013) (analyzing how domestic first-time home purchasers are at a competitive disadvantage compared to foreign real estate investors); Kathleen M. Howley, Bloomberg, Families Blocked by Investors from Buying U.S. Homes (2013) (discussing that the rise of all-cash purchases, among other things, has prevented many potential homeowners from purchasing homes). --------------------------------------------------------------------------- 4(a)(7) Section 304(a) and (b) of HMDA requires the disclosure of the dollar amount of covered loans and applications.\266\ Section 1003.4(a)(7) of Regulation C requires financial institutions to report the amount of the loan or the amount applied for. Paragraph I.A.7 in appendix A instructs financial institutions to report loan amount to the nearest thousand, among other things. The Bureau proposed Sec. 1003.4(a)(7), which provided that financial institutions shall report the amount of the covered loan or the amount applied for and clarified how to determine and report loan amount with respect to various types of transactions. In addition, the Bureau proposed to delete the requirement to round the loan amount to the nearest thousand, and also proposed several technical, conforming, and clarifying modifications to Sec. 1003.4(a)(7) and its corresponding comments. --------------------------------------------------------------------------- \266\ 12 U.S.C. 2803(a), (b). --------------------------------------------------------------------------- Proposed Sec. 1003.4(a)(7)(i) provided that for a closed-end mortgage loan, other than a purchased loan or an assumption, a financial institution shall report the amount to be repaid as disclosed on the legal obligation. The Bureau received a few comments regarding reporting the exact dollar amount, rather than the loan amount rounded to the nearest thousand. Some industry commenters suggested that the Bureau maintain the current rounding requirement, explaining that the change to reporting the exact loan amount in dollars will have limited value and will present an increased opportunity for clerical errors. Other industry commenters recommended that loan amount be reported in ranges rather than an exact loan amount in order to eliminate potential reporting errors and to better protect the privacy of applicants. On the other hand, a few commenters supported the proposal to report the exact loan amount, agreeing with the Bureau's proposed rationale that this would allow for a more precise calculation of loan- to-value ratio. One industry commenter indicated that reporting loan amount in dollars would also eliminate the potential for errors associated with incorrect rounding. Another industry commenter stated that while rounding has been the standard for reporting loan amount, it has been known to cause problems with data integrity. The Bureau has considered this feedback and determined that requiring reporting of the exact dollar amount is the more appropriate approach. Reporting of the exact dollar amount will facilitate HMDA compliance because such information is evident on the face of the loan documents and financial institutions will no longer need to make an additional calculation required for rounding. In addition, when coupled with Sec. 1003.4(a)(28), which requires a financial institution to report the value of the property relied on in making the credit decision, a requirement to report the exact dollar amount under Sec. 1003.4(a)(7) will allow for the calculation of loan-to-value ratio, an important underwriting variable. A rounded loan amount would render these calculations less precise, undermining their utility for data analysis. Proposed Sec. 1003.4(a)(7)(i) further provides that, for a purchased closed-end mortgage loan or an assumption of a closed-end mortgage loan, the financial institution shall report the unpaid principal balance at the time of purchase or assumption. An industry commenter indicated that reporting the unpaid principal balance at the time of purchase for a purchased closed-end mortgage loan would present operational difficulties since payments may sometimes be in process and reconciliation may be required and such reconciliation would be complicated with quarterly reporting. The Bureau does not believe that requiring a financial institution to report the unpaid principal balance of a purchased closed-end mortgage loan at the time of purchase would result in significant difficulties. Moreover, the Bureau simply moved this existing reporting requirement into the text of proposed Sec. 1003.4(a)(7)(i), which prior to the proposal, was found in an instruction and comment. With respect to quarterly reporting, those requirements are described further below in the section-by-section analysis of Sec. 1003.5(a)(1). The Bureau received no other feedback regarding this proposed requirement. Consequently, the Bureau is adopting Sec. 1003.4(a)(7)(i) generally as proposed, with technical and clarifying modifications. In addition, the Bureau is adopting new comment 4(a)(7)-5, which clarifies the loan amount that a financial institution reports for a closed-end mortgage loan as set forth in Sec. 1003.4(a)(7)(i). Proposed Sec. 1003.4(a)(7)(ii) provides that for an open-end line of credit, including a purchased open-end line of credit or an assumption of an open-end line of credit, a financial institution shall report the amount of credit available to the borrower under the terms of the plan. With respect to open-end lines of credit, the Bureau proposed to collect the full line, rather than only the portion intended for home purchase or improvement, as is currently required. One commenter supported this modification, indicating that it would reduce burdens on financial institutions associated with determining the purposes of open-end lines of credit. Another industry commenter asked the Bureau to expressly clarify that the requirement to report loan amount for a home-equity line of credit is the amount of the line of credit, regardless of any amounts drawn. No clarification is necessary because the commentary provides that the loan amount that must be reported for an open-end line of credit is the entire amount of credit available to the borrower under the terms of the plan. The Bureau is adopting Sec. 1003.4(a)(7)(ii) generally as proposed, with one modification to clarify that reverse mortgage open-end lines of credit are subject to Sec. 1003.4(a)(7)(iii), discussed below. The Bureau is also adopting new comment 4(a)(7)-6, which clarifies that for a purchased open-end line of credit and an assumption of an open-end line of credit, a financial institution reports the entire amount of credit available to the borrower under the terms of the plan. Regulation C is currently silent as to how loan amount should be determined for a reverse mortgage. Proposed Sec. 1003.4(a)(7)(iii) provides that, for a reverse mortgage, the amount of the covered loan is the initial principal limit, as determined pursuant to section 255 of the National Housing Act (12 U.S.C. 1715z-20) and implementing regulations and mortgagee letters prescribed by HUD. The Bureau specifically solicited feedback on how to determine loan amount for non-federally insured reverse mortgages but received no comments. One industry commenter requested that the Bureau clarify upon which basis financial institutions should report non-federally insured reverse mortgages. The Bureau believes that industry is familiar with HUD's Home Equity Conversion Mortgage Insurance Program and its implementing regulations and mortgagee letters. Applying this well-known calculation to both federally insured and non-federally insured [[Page 66183]] reverse mortgages will produce more consistent and reliable data on reverse mortgages. Consequently, the Bureau is adopting Sec. 1003.4(a)(7)(iii) generally as proposed, but with technical modifications for clarity. In addition, the Bureau is adopting new comment 4(a)(7)-9, which clarifies that a financial institution reports the initial principal limit of a non-federally insured reverse mortgage as set forth in Sec. 1003.4(a)(7)(iii). The Bureau also proposed comments 4(a)(7)-2, -5, and -6. The Bureau received no specific feedback regarding these comments. Accordingly, the Bureau is adopting these comments generally as proposed, with several technical amendments for clarity and renumbered as 4(a)(7)-3, - 7, and -8. The Bureau is adopting proposed comment 4(a)(7)-3 generally as proposed and renumbered as 4(a)(7)-4, but clarifies that for a multiple-purpose loan, a financial institution reports the entire amount of the covered loan, even if only a part of the proceeds is intended for home purchase, home improvement, or refinancing. In addition, the Bureau is adopting new comment 4(a)(7)-2, which clarifies the loan amount that a financial institution reports for an application or preapproval request approved but not accepted under Sec. 1003.4(a)(7). 4(a)(8) 4(a)(8)(i) Current Sec. 1003.4(a)(8) requires reporting of the action taken on the covered loan or application and the date of action taken. The Bureau proposed to revise the commentary under Sec. 1003.4(a)(8) with respect to rescinded loans, conditional approvals, and applications received by third parties. The Bureau proposed to require that rescinded loans be reported as loans approved but not accepted. In addition, the Bureau proposed guidance on reporting action taken for loans involving conditional approvals and on reporting action taken for applications received by third parties. Comments regarding reporting for applications involving multiple parties are discussed in the section-by-section analysis of Sec. 1003.4(a). For the reasons discussed below, the Bureau is adopting Sec. 1003.4(a)(8) with modifications by providing separate paragraphs for the requirements to report action taken and date of action taken and to incorporate material from current appendix A into Sec. 1003.4(a)(8)(i) and the associated commentary. The Bureau did not propose changes to Sec. 1003.4(a)(8). To clarify and streamline the regulation, and to provide separate paragraph citations for the action taken reporting requirement and the action taken date reporting requirement, the Bureau is incorporating material from current appendix A into new Sec. 1003.4(a)(8)(i) and new Sec. 1003.4(a)(8)(ii). The Bureau is also adopting several comments which incorporate material previously contained in appendix A into the commentary in order to facilitate compliance. These comments 4(a)(8)(i)-1 through -8 primarily incorporate existing appendix A material, but contain some modifications to align with other changes and new comments discussed below. Because the material was previously contained in appendix A, no substantive change is made. Few commenters addressed the proposal regarding rescinded loans. One commenter supported the proposal because it provided a consistent reporting rule. Another commenter stated that the proposal would provide consistency, but argued that the number of rescinded loans is so small that the change would not be worth the regulatory compliance cost. The Bureau believes that approved but not accepted most accurately reflects the outcome of a rescinded transaction, and that a consistent reporting rule for rescinded loans is appropriate and justifies any compliance burden. Therefore, it is finalizing comment 4(a)(8)-2 generally as proposed, but with minor technical revisions, renumbered as comment 4(a)(8)(i)-10. Some commenters addressed the proposal to clarify conditional approvals in comment 4(a)(8)-5. The proposal amended the commentary to clarify the types of conditions that are considered credit conditions and those that are customary commitment or closing conditions, and to clarify which action taken categories should be reported in certain circumstances involving conditional approvals. One industry commenter stated that the revised commentary was helpful. A few commenters stated that the conditional approval rules were generally confusing and did not reflect a financial institution's true credit decision in all circumstances. The Bureau believes that the general framework established by the conditional approvals commentary serves HMDA's purposes and provides a reasonable way for reflecting financial institutions' actions on covered loans and applications. While some financial institutions may view any type of approval, even one with many outstanding conditions, as an approved loan and wish to report it as such under Regulation C, the Bureau believes this would be an inappropriate result for applications that ultimately did not result in originations and were conditioned on underwriting or creditworthiness conditions. The Bureau is finalizing comment 4(a)(8)-5 as proposed, renumbered as comment 4(a)(8)(i)-13. One commenter argued that financial institutions should not report purchased loans under Regulation C and cited legislative history the commenter believed demonstrated that Congress intended to exclude loans purchased. HMDA section 304(a)(1)(B) has included a requirement to compile and make available information about loans ``purchased by that institution'' since HMDA was enacted in 1975.\267\ The legislative history referred to by the commenter does not address whether purchased loans should be reported, but rather, whether secondary market entities that only purchase loans but do not also originate loans should be required to report under HMDA; Congress ultimately enacted a requirement for financial institutions to report the class of purchaser of loans.\268\ The Bureau believes that HMDA section 304(a)(1)(B) clearly authorizes reporting of loans purchased by financial institutions covered by HMDA. The Bureau is finalizing comment 4(a)(8)- 3 related to purchased loan as proposed, renumbered as comment 4(a)(8)(i)-11. --------------------------------------------------------------------------- \267\ Home Mortgage Disclosure Act of 1975, Public Law 94-200, section 304(a)(1)(B). 12 U.S.C. 2803(a). \268\ H. Rept. 101-222 (1989), at 460. 12 U.S.C. 2803(h)(1)(C). --------------------------------------------------------------------------- The Bureau is finalizing comment 4(a)(8)-1 with modifications for clarity, renumbered as comment 4(a)(8)(i)-9. The Bureau is finalizing comment 4(a)(8)-4 as proposed, renumbered as comment 4(a)(8)(i)-12. The Bureau is also adopting new comments 4(a)(8)(i)-1 through 4(a)(8)(i)-8 which incorporate material in existing appendix A with some modifications for clarity. The Bureau is also adding new comment 4(a)(8)(i)-15 to provide guidance on reporting action taken when a financial institution has provided a notice of incompleteness followed by an adverse action notice on the basis of incompleteness under Regulation B.\269\ The comment provides that an institution may report the action taken as either file closed for incompleteness or application denied in such a circumstance. --------------------------------------------------------------------------- \269\ 12 CFR 1002.9(c)(1)(i) and (ii). --------------------------------------------------------------------------- 4(a)(8)(ii) The Bureau proposed only technical changes and modifications to the [[Page 66184]] current Regulation C requirement to report the date of action taken by a financial institution on a covered loan or application. The Bureau did not receive many comments related to the requirement to report action taken date. Comments related generally to the definition of application or reporting of applications are discussed in the section- by-section analysis of Sec. 1003.2(b). The Bureau is finalizing the requirement to report the date of action taken as new Sec. 1003.4(a)(8)(ii) to provide a separate paragraph for the requirement. The Bureau is adopting comments 4(a)(8)-7, -8, and -9 as proposed, renumbered as comments 4(a)(8)(ii)-4, -5, and -6. The Bureau is also adopting new comments 4(a)(8)(ii)-1, -2, and -3, which incorporate existing requirements in appendix A related to reporting of action taken date. 4(a)(9) The Bureau proposed to require financial institutions to report the address of the property securing the covered loan, discussed below in the section-by-section analysis of Sec. 1003.4(a)(9)(i), and to continue to require financial institutions to report the State, MSA or MD, county, and census tract of most reported covered loans, discussed below in the section-by-section analysis of Sec. 1003.4(a)(9)(ii). The Bureau is adopting proposed Sec. 1003.4(a)(9) with the modifications discussed below. Covered Loans Related to Multiple Properties The Bureau proposed to revise existing comments 4(a)(9)-1 and -2 to provide a single framework clarifying how to report a covered loan related to multiple properties. Proposed comment 4(a)(9)-1 discussed reporting when a covered loan relates to more than one property but only one property secures or would secure the loan. Proposed comment 4(a)(9)-2 provided that if more than one property secures or would secure the covered loan, a financial institution may report one of the properties using one entry on its loan/application register or the financial institution may report all of the properties using multiple entries on its loan/application register. Proposed comment 4(a)(9)-3 discussed reporting multifamily properties with more than one address. A few commenters provided feedback on proposed comment 4(a)(9)-2. One consumer advocate suggested that the Bureau should require financial institutions to report information concerning all of the properties securing the loan. A few industry commenters took the opposite position and urged the Bureau to require financial institutions to report information about only one of the properties. After considering the comments, the Bureau concludes that optional reporting is not advisable because HMDA data would provide inconsistent information about these types of transactions. At the same time, requiring financial institutions to report information about all of the properties securing the loan is also problematic because it would present additional burden for financial institutions. In addition, defining what constitutes multiple properties may present challenges for some multifamily complexes, which may sit on one parcel but have multiple addresses. For those reasons, the final rule requires financial institutions to report information about only one of the properties securing the loan. Accordingly, the Bureau is finalizing proposed comments 4(a)(9)-1 through -3 with modifications to require reporting of one property when a covered loan is secured by more than one property. The Bureau also proposed technical modifications to existing comments 4(a)(9)-4 and -5. The Bureau received no comments on comments 4(a)(9)-4 and -5 and is finalizing them as proposed. 4(a)(9)(i) The Dodd-Frank Act amended HMDA to authorize the Bureau to collect ``as [it] may determine to be appropriate, the parcel number that corresponds to the real property pledged or proposed to be pledged as collateral.'' \270\ The Bureau proposed to implement this authorization with proposed Sec. 1003.4(a)(9)(i), which provided that financial institutions were required to report the postal address of the physical location of the property securing the covered loan or, in the case of an application, proposed to secure the covered loan. The proposal indicated that the Bureau anticipated that postal address information would not be publicly released if proposed Sec. 1003.4(a)(9)(i) were finalized. The Bureau solicited feedback on whether collecting postal address was an effective way to implement the Dodd-Frank amendment. --------------------------------------------------------------------------- \270\ HMDA section 304(b)(6)(H) authorizes the Bureau to include in the HMDA data collection ``the parcel number that corresponds to the real property pledged or proposed to be pledged as collateral.'' 12 U.S.C. 2803(b)(6)(H). --------------------------------------------------------------------------- For the reasons discussed below, the Bureau is adopting Sec. 1003.4(a)(9)(i) as proposed with the technical modifications discussed below. The Bureau is also adopting new comments 4(a)(9)(i)-1 through -3 to clarify the reporting requirements. The Bureau received several comments on proposed Sec. 1003.4(a)(9)(i). Several consumer advocate commenters supported reporting postal address.\271\ These commenters highlighted that postal addresses would improve the ability to detect localized discrimination, noting that discrimination can occur in areas smaller than census tracts or other geographic boundaries. In addition, some explained that relying on census tracts for geographic analysis creates challenges for longitudinal analysis of the data because census tracts change over time. They also noted that collecting address in HMDA would enable tracking of multiple liens on the same property and thereby identifying risks for borrowers who may be over-leveraged. --------------------------------------------------------------------------- \271\ During the Board's hearings, a consumer advocate urged the Board to add information that uniquely identifies the property related to the loan to the HMDA data. See, e.g., Washington Hearing, supra note 39 (remarks of Lisa Rice, Vice President, National Fair Housing Alliance). --------------------------------------------------------------------------- Several industry commenters raised objections to reporting postal address. Some of these commenters suggested that postal address would not provide any valuable information because census tract information provides sufficient information to conduct fair lending or other statistical analysis of the property location. Other commenters asserted that reporting postal address would not support HMDA's purposes. Some industry commenters also expressed concerns about the burden of reporting postal address. In addition, many industry commenters raised concerns about the privacy implications of including postal address in the HMDA data set. Commenters expressed concerns both about collecting the information and about disclosing the information. Commenters explained that address can be used to link the financially sensitive information included in the HMDA data with an individual borrower. Commenters suggested that the Bureau's data security systems would not adequately protect the information from accidental disclosure during the transmission of the information to the Bureau and while the information is stored on the Bureau's systems. Some industry commenters noted that information on census tract was preferable to postal address because it protects privacy. Most commenters urged the Bureau not to release the reported postal address information if [[Page 66185]] collected. A consumer advocate also urged the Bureau to consider protections for specific populations, such as victims of domestic violence, when considering whether to release address information. A few consumer advocate commenters, on the other hand, urged the Bureau to release address, or point-specific information, to trusted researchers. The Bureau is finalizing the proposal to collect the postal address, changed to property address for the reasons discussed below, of the property securing or proposed to secure a covered loan. Collecting property address will enrich the HMDA data and will support achieving HMDA's purposes. With these data, Federal officials will be able to track multiple liens on the same property. In addition, property address will help officials better understand access to credit and risks to borrowers in particular communities and better target programs to reach vulnerable borrowers and communities. Using these data, Federal officials may be able to detect patterns of geographic discrimination not evident from census tract data, which will assist in identifying violations of fair lending laws. In addition, as census tracts change over time, collecting property address will facilitate better longitudinal analysis of geographic lending trends. However, the Bureau recognizes that collecting property address presents some challenges. As noted in the proposal, including property address in the HMDA data raises privacy concerns because property address can easily be used to identify a borrower. The Bureau is sensitive to the privacy implications of including property address in the HMDA data and has considered these implications carefully. Although the Bureau's privacy analysis is ongoing, as discussed in part II.B above, the Bureau anticipates that property address will not be included in the publicly released HMDA data. Due to the significant benefits of collecting this information, the Bureau believes it is appropriate to collect property address in spite of the privacy concerns and other concerns raised by commenters about collecting this information. Parcel Number Many commenters discussed whether postal address was an appropriate way to implement the Dodd-Frank authorization to collect a parcel number. Most of these commenters, including both industry and consumer advocate commenters, expressed support for using postal address to implement the authorization to collect a parcel number. Commenters noted that collecting postal address, while imperfect, is the best available option, because it is less burdensome to report than reporting a local parcel number and uniquely identifies most properties. A few commenters specifically stated that other alternatives discussed in the proposal, such as geospatial coordinates or local parcel number, present greater reporting burdens than postal address. Commenters also noted the current absence of a national universal parcel numbering system. One commenter stated that local parcel numbers are not used by lenders and are used solely by professionals that manage property records. Another commenter described the burden associated with reporting a local parcel number, stating that address, unlike a local parcel number, is stored in the same system as the other HMDA data. Other commenters stated that postal address would provide more complete information than a local parcel number for loans related to manufactured housing because manufactured homes located in mobile home parks may be placed on the same parcel but have unique property addresses. Some consumer advocate commenters stated that postal address was currently an appropriate way to collect a parcel number, but asked the Bureau to consider replacing postal address with a universal parcel identifier if one is developed in the future. In addition, one commenter urged the collection of local parcel numbers because of their value for analysis at the local level. A few commenters that represented geospatial vendors recommended collecting both postal address and local parcel information. They explained that this would allow the Bureau, using both the reported address and local parcel information, to establish a national parcel database with mapping capabilities. Some of these commenters noted that collecting this information would also facilitate the creation of a national parcel numbering system. The Bureau concludes that collecting property address is an appropriate way to implement the Dodd-Frank authorization to collect a parcel number. As noted by commenters, address is the least burdensome way to collect information that will uniquely identify a property. Financial institutions currently collect property address during the mortgage origination and application process if the address is available, and store that information with the other application and loan data that is reported in HDMA. In addition, most properties, including manufactured homes, have property addresses. In a small number of cases, a property address may not be available at the time of origination for some properties. Nonetheless, property address is an efficient and effective way to implement the authorization to collect a parcel number. Currently, no universal standard exists for identifying a property so that it can be linked to related mortgage data. Parcel data are collected and maintained by individual local governments with limited State or Federal involvement. Local jurisdictions do not use a standard way to identify properties. In addition, local parcel data are not easily linked to the location of the property, which, as discussed above, substantially amplifies the usefulness of a parcel identifier. Local parcel information would provide some value for local analysis, but property address also provides valuable information at the local level. Therefore, compared with collecting property address, collecting a local parcel number would substantially increase the burden associated with reporting a parcel identifier and would substantially decrease the utility of the data. The Bureau is not at this time pursuing commenters' suggestions for using Regulation C to develop a national parcel database. The Bureau may consider in the future whether and how it could work with other regulators and public officials to explore a national parcel identification system or other similar systems. The final rule does not require financial institutions to collect a local parcel number in addition to property address. The Bureau concludes that collecting property address strikes the appropriate balance between improving the data's utility and minimizing undue burden on data reporters. For the reasons discussed above, the Bureau is implementing the Dodd-Frank authorization to collect the ``parcel number that corresponds to the real property pledged or proposed to be pledged as collateral'' by requiring financial institutions to report the property address of the property securing the covered loan or, in the case of an application, proposed to secure the covered loan.\272\ As discussed above, there is no universal parcel number system; therefore, the Bureau believes it is reasonable to interpret the Dodd-Frank Act amendment to refer to information that uniquely identifies a dwelling pledged or proposed to be [[Page 66186]] pledged as collateral. The Bureau is also adopting Sec. 1003.4(a)(9)(i) pursuant to the Bureau's HMDA section 305(a) authority to provide for adjustments because, for the reasons given above, the Bureau believes the provision is necessary and proper to effectuate HMDA's purposes and facilitate compliance therewith. --------------------------------------------------------------------------- \272\ HMDA section 304(b)(6)(H) authorizes the Bureau to include in the HMDA data collection ``the parcel number that corresponds to the real property pledged or proposed to be pledged as collateral.'' 12 U.S.C. 2803(b)(6)(H). --------------------------------------------------------------------------- Reporting Issues Some industry commenters discussed situations when reporting a postal address is not possible or should not be required. A few of these commenters asked what to report if the property does not have an address. Others urged the Bureau not to require reporting of postal address information for purchases or for applications withdrawn or denied. The Bureau recognizes that in some cases address information will not be known. Consequently, address information will not be reported for all HMDA entries, as indicated in new comment 4(a)(9)-3. As discussed above, however, because property address greatly enriches the utility of HMDA data, financial institutions must report property address if the information is available. Therefore, the Bureau is not adopting commenters' suggestions to exclude certain types of entries from the requirement to report property address. Some commenters suggested that Regulation C require reporting of the physical location of the property, instead of the mailing address, which may be different from the physical location of the property in some cases. Proposed Sec. 1003.4(a)(9) and proposed instruction 4(a)(9)-1 directed financial institutions to report the postal address that corresponds to the physical location of the property, not the mailing address. To eliminate the confusion about whether to report the mailing address or the physical location of the property, the Bureau is modifying Sec. 1003.4(a)(9)(i) to replace the term postal address, which may have been misunderstood to mean mailing address, with the term property address, which is understood to refer to the physical location of the property. In addition, the Bureau is adopting new comment 4(a)(9)(i)-1 to clarify that the financial institution reports the property address of the physical location of the property. One commenter urged revising the requirement to include primary street address points, sub-address points, and geographic coordinates. The commenter also urged the Bureau to partner with States as they build addresses to meet the requirements of Next Generation 9-1-1 systems. The Bureau recognizes that in some cases, addresses may not convey full information about a property's location. These enhanced addressing standards would enrich the quality of the geographic information reported in HMDA data in those cases where address does not precisely identify a property's location, such as for dwellings located on rural routes. However, importing these standards for HMDA reporting seems likely to result in new burden for financial institutions that currently collect address during the application process but may not be collecting the information required by these standards. At the same time, any benefit from using these standards in HMDA would be limited only to a subset of HMDA reportable transactions. The Bureau's judgment is that reporting property address is less burdensome for institutions than enhanced standards, and will provide benefits sufficient to justify any burden that might be imposed on financial institutions. Some industry commenters noted the challenges of reporting postal address in a standard format. To resolve those challenges, one commenter suggested requiring reporting the information in the same format as the closing disclosure. Another commenter noted that reporting postal address would have risks of input errors and suggested that the Bureau allow good faith errors for the address information. Other commenters sought clarification about how to report and whether abbreviations were allowed. In response to these comments, the final rule clarifies institutions' reporting obligations to help minimize the risk of inadvertent reporting errors. Accordingly, new comment 4(a)(9)(i)-2 provides guidance on how to report the property address. In addition, Sec. 1003.6, discussed below, addresses bona fide errors. Final Rule Having considered the comments received and for the reasons discussed above, the Bureau is finalizing Sec. 1003.4(a)(9)(i) as proposed with the modifications discussed above. In addition, for the reasons discussed above, the Bureau is adopting new comments 4(a)(9)(i)-1 through -3 to provide illustrative examples and to incorporate information included in proposed instruction 4(a)(9). 4(a)(9)(ii) Under HMDA and current Regulation C, a financial institution is required to report the location of the property to which the covered loan or application relates by MSA or MD, State, county, and census tract if the loan is related to a property located in an MSA or MD in which the financial institution has a home or branch office and a county with a population of more than 30,000.\273\ In addition, Sec. 1003.4(e) requires banks and savings associations that are required to report data on small business, small farm, and community development lending under regulations that implement the CRA to collect the location of property located outside MSAs and MDs in which the institution has a home or branch office or outside of any MSA. The Bureau proposed to renumber existing Sec. 1003.4(a)(9) as Sec. 1003.4(a)(9)(ii) and to make certain nonsubstantive technical modifications for clarification. The Bureau did not propose any changes to Sec. 1003.4(e). --------------------------------------------------------------------------- \273\ See Sec. 1003.4(a)(9); HMDA section 304(a)(2). Part I.C.3 of appendix A directs financial institutions to enter ``not applicable'' for census tract if the property is located in a county with a population of 30,000 or less. A for-profit mortgage-lending institution is deemed to have a branch office in an MSA or MD if in the preceding calendar year it received applications for, originated, or purchased five or more home purchase loans, home improvement loans, or refinancings related to property located in that MSA or MD, respectively. See Sec. 1003.2 (definition of branch office). --------------------------------------------------------------------------- The Bureau explained in the proposal that it was exploring ways to reduce the burden associated with reporting the State, county, MSA, and census tract of a property, such as operational changes that may enable the Bureau to perform geocoding (i.e., identifying the State, county, MSA, and census tract of a property) for financial institutions. The Bureau suggested that it might create a system where a financial institution reports only the address and the Bureau provides the financial institution with the census tract, county, MSA or MD, and State. The Bureau solicited feedback on the potential operational improvements. For the reasons discussed below, the Bureau is adopting Sec. 1003.4(a)(9)(ii), which requires financial institutions to report the State, county, and census tract of the property securing or proposed to secure a covered loan if the property is located in an MSA in which the institution has a home or branch office or if Sec. 1003.4(e) applies. The final rule eliminates the requirement to report the MSA or MD of the property securing or proposed to secure a covered loan. The Bureau is also adopting new comments 4(a)(9)(ii)(B)-1 and 4(a)(9)(ii)(C)-1 to provide guidance on how to report county and census tract information, respectively. Many commenters provided feedback on whether the Bureau should assume geocoding responsibilities for reporters. Some commenters, including a few industry commenters and many consumer advocate commenters, expressed support for the Bureau [[Page 66187]] assuming geocoding responsibilities. Many of those commenters noted that such a change would improve the accuracy of geocoding information. Most industry commenters, however, raised concerns with the Bureau assuming geocoding responsibilities for reporters. Some asserted that such an operational change would not reduce their burden because financial institutions already have geocoding systems in place and would continue to use those systems even if the Bureau assumed geocoding responsibilities. Some of these commenters explained that financial institutions would not want to wait until they submit their HMDA data to obtain the geocoding information because they need on demand geocoding for business purposes such as evaluating their lending penetration. In addition, some commenters raised some practical issues with the Bureau assuming geocoding, such as developing a system for the Bureau and financial institutions to communicate back-and-forth about geocoding results. Commenters also stated that geocoding would be more accurate if performed by the financial institution because the institution is probably more familiar with the particular geographic area and likely could identify errors in geocoding more readily than the Bureau could. In addition, industry commenters raised concerns about whether financial institutions would be held responsible for the accuracy of the Bureau's geocoding and about whether the Bureau would assume responsibility for identifying the census tracts of properties that return an error in the Bureau's geocoding database. A few industry commenters asked the Bureau to allow them to report their geocoded information even if the Bureau decides to take the geocoding on itself. A few other industry commenters suggested that instead of geocoding for financial institutions, that the Bureau develop a free geocoding database or tool for financial institutions. The Bureau has concluded that it should not geocode for financial institutions and instead should focus on the best way to achieve accuracy in the property location information reported in HMDA. Property location data is more likely to be accurate if the financial institution reporting the covered loan or application also geocodes the property. In addition, based on comments from financial institutions, it appears that assuming geocoding responsibilities for financial institutions might not achieve the burden reduction that the Bureau hoped to achieve when it issued the proposal. Therefore, the Bureau does not plan to pursue assuming geocoding responsibilities in the manner discussed in the proposal. Instead, the Bureau is exploring other ways that it can assist reporters with geocoding, such as developing an improved geocoding tool for financial institutions. Consumer advocate commenters also discussed the value of the currently reported property location information and urged the Bureau to continue to require reporting of information by census tract and to continue to make that information available in the publicly disclosed data. The Bureau is generally retaining reporting of the currently required property location information because it provides valuable information. The Bureau believes that it can reduce the burden of reporting by eliminating the requirement to report the MSA or MD in which the property is located. If a financial institution reports the county, the regulators can identify the MSA or MD because MSAs and MDs are defined at the county level. The MSA or MD can be inserted into the publicly available data so that the data's utility is preserved. Finally, it appears that financial institutions do not report MSA or MD information when they have incomplete property location information. In the past five years, no financial institutions have reported the MSA or MD of a property without other property location information.\274\ Therefore, retaining this field only for cases when the financial institution does not know the county in which the property securing, or proposed to secure, the covered loan is located would also not provide valuable information. Therefore, the final rule eliminates the burden of reporting this information to facilitate compliance. --------------------------------------------------------------------------- \274\ It is not clear why a financial institution does not report property location information for a particular entry. It could be because the information is not required, because, for example, the property is not located in an MSA or MD in which the institution has a home or branch office. See Sec. 1003.4(a)(9). In the past five years, some financial institutions reported the State in which the property is located without other property location information, which may suggest that the financial information had incomplete information about the location of the property. --------------------------------------------------------------------------- For the reasons discussed above, the Bureau is finalizing proposed Sec. 1003.4(a)(9)(ii), with modifications to eliminate the requirement included in proposed Sec. 1003.4(a)(9)(ii)(C) as discussed above. 4(a)(10) 4(a)(10)(i) HMDA section 304(b)(4) requires the reporting of racial characteristics and gender for borrowers and applicants.\275\ Section 1003.4(a)(10) of Regulation C requires a financial institution to collect the ethnicity, race, and sex of the applicant or borrower for applications and loan originations for each calendar year. The Bureau proposed to renumber this requirement as Sec. 1003.4(a)(10)(i), and also proposed several technical and clarifying amendments to the instructions in appendix A and the associated commentary. --------------------------------------------------------------------------- \275\ 12 U.S.C. 2803(b)(4); see also 79 FR 51731, 51775 (Aug. 29, 2014), n. 340. --------------------------------------------------------------------------- The Bureau's proposal solicited feedback regarding the challenges faced by both applicants and financial institutions by the data collection instructions prescribed in appendix B and specifically solicited comment on ways to improve the data collection of the ethnicity, race, and sex of applicants and borrowers. The Bureau also conducted a voluntary, small-scale survey to solicit suggestions from financial institutions on ways to improve the process of collecting the ethnicity, race, and sex of applicants that may potentially relieve burden and help increase the response rates by applicants, in particular, for applications received by mail, internet, or telephone. The Bureau selected nine financial institutions for participation in the survey which, according to recent HMDA data, generally exhibited relatively high incidences of applicants providing ethnicity, race, and sex in applications made by mail, internet, or telephone. The Bureau was interested to learn what factors may have contributed to these higher response rates and also to identify potential improvements to appendix B. Five financial institutions chose to participate in the survey and the Bureau considered their responses as part of the HMDA rulemaking. In response to the proposal's solicitation for feedback, a few industry commenters recommended that the Bureau remove the proposed requirement, which currently exists under the rule, that financial institutions collect an applicant's ethnicity, race, and sex on the basis of visual observation and surname when an application is taken in person and the applicant does not provide the information. In general, these industry commenters did not support this collection requirement for the following reasons. First, commenters expressed the belief that loan originators should [[Page 66188]] not have to guess, on the basis of visual observation or surname, as to what is an applicant's ethnicity, race, and sex. Second, commenters expressed the belief that such guessing results in inaccurate and unreliable data. Lastly, commenters expressed the belief that an applicant's decision not to provide his or her demographic information should be respected and that a loan originator should not override that decision by being required to collect the information on the basis of visual observation or surname. On the other hand, several consumer advocate commenters provided feedback emphasizing that data on an applicant's ethnicity, race, and sex is vital to HMDA's utility. A few of these commenters also emphasized the need for HMDA data to reflect whether such demographic information was self-reported by applicants or the result of a loan originator collecting the information on the basis of visual observation or surname. For example, one commenter stated that information on ethnicity and race is crucial for discovering potential patterns of discrimination and recommended that the loan/application register include a flag indicating whether ethnicity and race information was provided by the applicant, allowing independent researchers and community advocates to undertake important fair lending analyses. Another commenter stated that in order for the Bureau to better understand whether the visual observation or surname requirement is producing useful information, it urged the Bureau to require financial institutions to report whether the borrowers have furnished the race, ethnicity, and sex data. Lastly, another commenter stated that information regarding how often borrowers refuse to voluntarily report demographic data or how often lenders report such information on the basis of visual observation or surname is not easily found and therefore, at the very least, the Bureau should flag applicant or borrower versus financial institution reporting of demographic information. The Bureau has considered this feedback and determined that the appropriate approach to further HMDA's purposes is to continue to require that financial institutions collect the ethnicity, race, and sex of applicants on the basis of visual observation and surname when an application is taken in person and the applicant does not provide the information. The Bureau agrees with both industry and consumer advocate commenters that recognized the importance of data on an applicant's or borrower's ethnicity, race, and sex to the purposes of HMDA. The Bureau has determined that removing the visual observation or surname requirement from the final rule would diminish the utility of the HMDA data to further HMDA's purposes. The Bureau has also determined that requiring financial institutions to report whether the applicant's ethnicity, race, and sex was collected on the basis of visual observation or surname improves the utility of HMDA data. Accordingly, the Bureau is maintaining the current requirement in appendix B that when an applicant does not provide the requested information for an application taken in person, a financial institution is required to collect the demographic information on the basis of visual observation or surname. In addition, the Bureau is adopting a new requirement in Sec. 1003.4(a)(10)(i) of the final rule that requires financial institutions to report whether the applicant's ethnicity, race, or sex was collected on the basis of visual observation or surname. The Bureau is adopting new instructions and modifications to the sample data collection form in appendix B to capture this new reporting requirement. In response to the proposal's solicitation for feedback on ways to improve the data collection of an applicant's ethnicity, race, and sex, and in response to the Bureau's survey which sought, among other things, suggestions on ways to help increase the response rates by applicants, the Bureau received feedback urging the Bureau to disaggregate the ethnicity category as well as two race categories--the Asian category and the Native Hawaiian and Other Pacific Islander category. Before discussing this feedback, it is important to first describe the data standards on ethnicity and race issued by the Office of Management and Budget (OMB). The OMB has issued the standards for the classification of Federal data on ethnicity and race.\276\ OMB's current government-wide standards provide ``a minimum standard for maintaining, collecting, and presenting data on race and ethnicity for all Federal reporting purposes. . . . The standards have been developed to provide a common language for uniformity and comparability in the collection and use of data on race and ethnicity by Federal agencies.'' \277\ The OMB standards provide the following minimum categories for data on ethnicity and race: Two minimum ethnicity categories (Hispanic or Latino; Not Hispanic or Latino) and five minimum race categories (American Indian or Alaska Native; Asian; Black or African American; Native Hawaiian or Other Pacific Islander; and White). The categories for ethnicity and race in existing Regulation C conform to the OMB standards. --------------------------------------------------------------------------- \276\ Office of Mgmt. and Budget, Revisions to the Standards for the Classification of Federal Data on Race and Ethnicity, 62 FR 58782-90 (Oct. 30, 1997) [hereinafter OMB Federal Data Standards on Race and Ethnicity]. \277\ See id. --------------------------------------------------------------------------- In addition to the minimum data categories for ethnicity and race, the OMB Federal Data Standards on Race and Ethnicity provide additional key principles. First, self-identification is the preferred means of obtaining information about an individual's ethnicity and race, except in instances where observer identification is more practical.\278\ Second, the collection of greater detail is encouraged as long as any collection that uses more detail is organized in such a way that the additional detail can be aggregated into the minimum categories for data on ethnicity and race. More detailed reporting, which can be aggregated to the minimum categories, may be used at the agencies' discretion. Lastly, Federal agencies must produce as much detailed information on ethnicity and race as possible; however, Federal agencies shall not present data on detailed categories if doing so would compromise data quality or confidentiality standards.\279\ --------------------------------------------------------------------------- \278\ See id. \279\ See id. --------------------------------------------------------------------------- In addition to the OMB standards, it is also important to describe the data standards used in the 2000 and 2010 Decennial Census. The U.S. Census Bureau (Census Bureau) collects Hispanic origin and race information following the OMB standards and guidance discussed above.\280\ Responses to the Hispanic origin question and race question in the 2000 and 2010 Decennial Census were based on self- identification.\281\ --------------------------------------------------------------------------- \280\ U.S. Census Bureau, C2010BR-02, Overview of Race and Hispanic Origin: 2010, at 2 (2011) [hereinafter Census Bureau Overview], available at http://www.census.gov/prod/cen2010/briefs/c2010br-02.pdf. \281\ See id. --------------------------------------------------------------------------- The OMB definition of Hispanic or Latino origin used in the 2010 Census refers to a person of Cuban, Mexican, Puerto Rican, South or Central American, or other Spanish culture or origin regardless of race.\282\ Hispanic or Latino origin can be viewed as the heritage, nationality group, lineage, or country of birth of the person or the person's parents or ancestors before their arrival in the United States.\283\ The [[Page 66189]] 2010 Census disaggregated ethnicity into four categories (Mexican, Puerto Rican, Cuban, Other Hispanic or Latino) and included one area where respondents could write-in a specific Hispanic or Latino origin group.\284\ As required by the OMB, the response categories and the write-in answers for the Census Bureau's ethnicity question can be combined to create the two minimum OMB categories for ethnicity, discussed above. --------------------------------------------------------------------------- \282\ See OMB Federal Data Standards on Race and Ethnicity; Census Bureau Overview at 2. \283\ See Census Bureau Overview at 2. \284\ See id. --------------------------------------------------------------------------- The OMB definitions of the race categories used in the 2010 Census, plus the Census Bureau's definition of Some Other Race, are discussed in footnote 285 below.\285\ For respondents who are unable to identify with any of the five minimum OMB race categories, OMB approved the Census Bureau's inclusion of a sixth race category--Some Other Race--on the 2000 and 2010 Census questionnaires. The 2010 Census disaggregated the Asian race into seven categories (Asian Indian, Chinese, Filipino, Japanese, Korean, Vietnamese, Other Asian), the Native Hawaiian and Other Pacific Islander race into four categories (Native Hawaiian, Guamanian or Chamorro, Samoan, Other Pacific Islander) and included three areas where respondents could write-in a specific Asian race, a specific Pacific Islander race, and the name of his or her enrolled or principal tribe in the American Indian or Alaska Native category.\286\ As required, the response categories and the write-in answers for the Census Bureau's race question can be combined to create the five minimum OMB categories for race, discussed above, plus Some Other Race. --------------------------------------------------------------------------- \285\ ``White'' refers to a person having origins in any of the original peoples of Europe, the Middle East, or North Africa. It includes people who indicated their race(s) as ``White'' or reported entries such as Irish, German, Italian, Lebanese, Arab, Moroccan, or Caucasian. ``Black or African American'' refers to a person having origins in any of the Black racial groups of Africa. It includes people who indicated their race(s) as ``Black, African Am., or Negro'' or reported entries such as African American, Kenyan, Nigerian, or Haitian. ``American Indian or Alaska Native'' refers to a person having origins in any of the original peoples of North or South America (including Central America) and who maintains tribal affiliation or community attachment. The category includes people who indicated their race(s) as ``American Indian or Alaska Native'' or reported their enrolled or principal tribe, such as Navajo, Blackfeet, Inupiat, Yup'ik, or Central American Indian groups or South American Indian groups. ``Asian'' refers to a person having origins in any of the original peoples of the Far East, Southeast Asia, or the Indian subcontinent, including, for example, Cambodia, China, India, Japan, Korea, Malaysia, Pakistan, the Philippine Islands, Thailand, and Vietnam. It includes people who indicated their race(s) as ``Asian'' or reported entries such as ``Asian Indian,'' ``Chinese,'' ``Filipino,'' ``Korean,'' ``Japanese,'' ``Vietnamese,'' and ``Other Asian'' or provided other detailed Asian responses. ``Native Hawaiian or Other Pacific Islander'' refers to a person having origins in any of the original peoples of Hawaii, Guam, Samoa, or other Pacific Islands. It includes people who indicated their race(s) as ``Pacific Islander'' or reported entries such as ``Native Hawaiian,'' ``Guamanian or Chamorro,'' ``Samoan,'' and ``Other Pacific Islander'' or provided other detailed Pacific Islander responses. ``Some Other Race'' includes all other responses not included in the White, Black or African American, American Indian or Alaska Native, Asian, and Native Hawaiian or Other Pacific Islander race categories described above. Respondents reporting entries such as multiracial, mixed, interracial, or a Hispanic or Latino group (for example, Mexican, Puerto Rican, Cuban, or Spanish) in response to the race question are included in this category. See Census Bureau Overview at 2-3. \286\ See Census Bureau Overview at 1-2. --------------------------------------------------------------------------- Another Federal agency has already begun to require more detailed data collection on ethnicity and race as is encouraged by the OMB and as has been used by the Census Bureau for 15 years. On October 31, 2011, the U.S. Department of Health and Human Services (HHS) published data standards for ethnicity and race that it now uses in its national population health surveys undertaken pursuant to the Affordable Care Act. These data standards are based on the disaggregation of the OMB standard and the 2000 and 2010 Decennial Census discussed above. Many of the commenters that provided feedback on the Bureau's proposal, discussed below, urged the Bureau to follow the data collection standards being used by the HHS and require financial institutions to collect and report more detailed ethnicity and race information. In addition, the American Housing Survey, which is a comprehensive national housing survey sponsored by HUD and conducted biennially by the Census Bureau, will similarly provide more detailed country of origin information for the first time ever in 2015.\287\ According to HUD's ``Priority Program Goals for the Asian American and Pacific Islander Community,'' one of the agency's five program goals is to improve the data collected on Asian American and Pacific Islander (AAPI) communities and it is working to disaggregate data for all major programs, including homeownership, tenant based rental assistance, and public housing. HUD's goal to disaggregate data extends not only to the AAPI community, but also to the Hispanic or Latino community.\288\ --------------------------------------------------------------------------- \287\ See https://www.whitehouse.gov/the-press-office/2015/05/12/fact-sheet-white-house-summit-asian-americans-and-pacific-islanders. \288\ See U.S. Dep't. of Housing and Urban Dev., Priority Program Goals for the Asian American and Pacific Islander Community, available at http://www.hud.gov/offices/hsg/mfh/trx/meet/raceethnicdatacollexecorder.pdf. --------------------------------------------------------------------------- The Bureau received many comments in response to its solicitation regarding the challenges faced by both applicants and financial institutions by the HMDA data collection instructions regarding an applicant's ethnicity, race, and sex, and on ways to improve that data collection. The comment letters of many consumer advocacy groups-- reinforced in subsequent communications and outreach--recommended disaggregation of the Asian and Native Hawaiian or Other Pacific Islander categories. A handful of these organizations also recommended disaggregation of data on the ethnicity category. These recommendations generally align with the 2000 and 2010 Decennial Census, the approach that HHS has been using since 2011 in its national population health surveys, and the approach HUD will be taking in all of its major programs. In general, these commenters urged the Bureau to disaggregate the ethnicity and race categories under HMDA for the following reasons. First, commenters stated that disaggregated data will more accurately reflect the borrowing experiences of various AAPI and Hispanic or Latino communities across the country. For example, some commenters stated that newer immigrants are likely to have different experiences in the mortgage market than earlier immigrants. In addition, since many subpopulation groups include limited-English proficient communities, commenters supported disaggregated data as a vehicle to better understanding of lending to these vulnerable groups and perhaps improved access to homeownership. Second, commenters expressed the belief that the aggregate OMB categories for ethnicity and race may mask discriminatory practices that are occurring against subpopulation groups that fall within these aggregate categories. For example, one consumer advocate commenter described the efforts made by one of its member organizations to manually disaggregate the HMDA data using borrowers' last name, census tract information in Queens, New York, and public court records to determine that more than 50 percent of defaults were among South Asians in many neighborhoods. In response, the organization assessed the needs of this particular Asian subpopulation group and prioritized building a foreclosure prevention program, which helped stabilize these minority neighborhoods. Overall, many commenters stated that expanding the [[Page 66190]] aggregate ethnicity and race categories to include specific subpopulations will assist regulators and the public in determining whether discrimination against certain subpopulations is occurring in minority communities. Lastly, commenters stated that the importance of ethnicity and race data to HMDA's purposes is critical and as such, the Bureau should do what it can to encourage applicants to provide their demographic information. These commenters expressed the belief that the aggregate OMB categories for ethnicity and race are often too broad and do not provide applicants within subpopulation groups with the opportunity of self-identification. One industry participant in the Bureau's survey expressed a similar perspective after speaking to several of its originators indicating that applicants opt to skip the ethnicity and race questions altogether when the options do not accurately describe their ethnic or racial identity. As discussed above, the OMB encourages the collection of greater detail beyond the two minimum categories for ethnicity and the five minimum categories for race, and as such, agencies may use more detailed reporting at their discretion so long as any collection that uses more detail is organized in such a way that the additional detail can be aggregated into the minimum categories for data on ethnicity and race. The Bureau has considered the feedback it received in response to its solicitation on ways to improve the data collection of an applicant's ethnicity, race, and sex under appendix B and determined, as discussed below, that the appropriate approach to further HMDA's purposes is to build upon the OMB standards by adding the type of granularity for subpopulations that was used in the 2000 and 2010 Decennial Census, with the exception that the Bureau is not adopting the sixth race category used by Census--Some Other Race--which cannot be aggregated to the five minimum OMB categories for race. First, the Bureau believes that disaggregated data on applicants' ethnicity and race will provide meaningful data, which will further HMDA's purposes--in determining whether financial institutions within a particular market are serving the housing needs of specific communities; in distributing public-sector investments so as to attract private investment to areas or communities where it is needed; and in identifying possible discriminatory lending patterns. Consumer advocates have been urging the Bureau for years to gather disaggregated information, which will enable them to determine whether institutions are filling their obligations to serve the housing needs of the communities and neighborhoods in which they are located. Data on subpopulation groups in the residential mortgage market will substantially advance the ability to better understand the market for particular subgroups and monitor access to credit. The Bureau recognizes that disaggregated data may not be useful in analyzing potential discrimination where financial institutions do not have a sufficient number of applicants or borrowers within particular subgroups to permit reliable assessments of whether unlawful discrimination may have occurred. However, in situations in which the numbers are sufficient to permit such fair lending assessments, disaggregated data on ethnicity and race will help identify potentially discriminatory lending patterns. Improved data will not only assist in identifying potentially discriminatory practices, but will also contribute to a better understanding of the experiences that members within subpopulations may share in the mortgage market. Second, as a 21st century, data-driven agency, the Bureau believes that its rules should recognize the nation's changing ethnic and racial diversity. By aligning the ethnicity and race categories in HMDA with the questions on Hispanic origin and race used by the Census Bureau during the last 15 years, the Bureau is taking a step forward in updating its data collection requirements. Lastly, as pointed out by commenters, disaggregation will also encourage self-reporting by applicants by offering, as the Census does, categories which promote self-identification. The Bureau recognizes that financial institutions may have concerns about this change to the collection and reporting of ethnicity and race under HMDA. This change may increase the burden of collection and reporting HMDA data. Disaggregation, as described here, may also result in financial institutions having to expand their data systems, update their application forms and processes, and provide additional training to loan originators to ensure compliance with the new requirements. There may also be questions as to what the Bureau expects of financial institutions with respect to their compliance management systems and challenges they may face in conducting fair lending analyses with the new data on ethnicity and race. The Bureau has considered these potential concerns, among others, and nonetheless believes that the utility of disaggregated HMDA data on applicants' ethnicity and race justifies the potential burdens and costs. Accordingly, the Bureau is adopting new data standards for the collection and reporting of ethnicity and race by modifying the instructions in appendix B and the sample data collection form. As such, the final rule requires financial institutions to use the following data standards for the collection and reporting of an applicant's ethnicity and race. [[Page 66191]] [GRAPHIC] [TIFF OMITTED] TR28OC15.000 As discussed above, with regard to the current requirement in appendix B that a financial institution collect an applicant's ethnicity, race, and sex on the basis of visual observation or surname when the applicant does not provide the requested information for an application taken in person, the Bureau has determined that it will maintain this requirement as is. However, the concerns with the visual observation and surname requirement expressed by commenters discussed above, would arguably be magnified due to the difficulties loan originators would potentially encounter in determining an applicant's ethnicity and race with the expanded categories the Bureau is finalizing. Thus, to reduce the potential burden of this change on financial institutions, the Bureau has determined that, at this point in time, the appropriate approach is to only permit self-identification of the disaggregated categories. That is, only an applicant may use the disaggregated categories to identify his or her ethnicity or race. When an application is taken in person and the applicant does not provide the information, the final rule will continue to require loan originators to collect, on the basis of visual observation or surname, the minimum OMB categories of ethnicity and race. The Bureau believes that this approach balances the value of disaggregated data on ethnicity and race to further HMDA's purposes with the potential burdens on financial institutions. Accordingly, the Bureau is modifying appendix B by adding a new instruction to require a financial institution to collect an applicant's ethnicity, race, and sex on the basis of visual observation or surname when the applicant does not provide the requested information for an application taken in person, by selecting from the following OMB minimum categories: Ethnicity (Hispanic or Latino; not Hispanic or Latino); race (American Indian or Alaska Native; Asian; Black or African American; Native Hawaiian or Other Pacific Islander; White). The Bureau is also modifying appendix B by adding a new instruction to provide that only an applicant may self-identify as being of a particular Hispanic or Latino subcategory (Mexican, Puerto Rican, Cuban, Other Hispanic or Latino) or of a particular Asian subcategory (Asian Indian, Chinese, Filipino, Japanese, Korean, Vietnamese, Other Asian) or of a particular Native Hawaiian or Other Pacific Islander subcategory (Native Hawaiian, Guamanian or Chamorro, Samoan, Other Pacific Islander) or of a particular American Indian or Alaska Native enrolled or principal tribe. The [[Page 66192]] Bureau recognizes the change to the collection and reporting of ethnicity and race under HMDA may raise concerns regarding applicant and borrower privacy. See part II.B above for a discussion of the Bureau's approach to protecting applicant and borrower privacy with respect to the public disclosure of HMDA data. Similar to the Census questionnaire and as outlined above in the new data standards the Bureau is adopting for the collection and reporting of an applicant's ethnicity and race, the Bureau is modifying the sample data collection form in appendix B to allow an applicant to provide a particular Hispanic or Latino origin when ``Other Hispanic or Latino'' is selected by the applicant, a particular Asian race when ``Other Asian'' is selected by the applicant, a particular Other Pacific Islander race when ``Other Pacific Islander'' is selected by the applicant, and lastly, the name of the enrolled or principal tribe when the applicant selects American Indian or Alaska Native race. The Bureau believes that this may encourage self-reporting by applicants by offering, as the Census does, an option for applicants to provide a specific Hispanic/Latino origin and race, which promotes self- identification and will improve the HMDA data's usefulness. In addition, in order to facilitate compliance, the Bureau has determined that it will limit the number of particular racial designations of applicants that are required to be reported by financial institutions. The Bureau reviewed recent Census data to consider the occurrence of respondents that self-identify as being of more than one particular race. For example, the 2010 Census data shows that of the Asian population where only Asian was reported as the respondents' race, only 0.11 percent of those self-identified as being of three particular Asian races, while only 0.02 percent self- identified as being of seven particular Asian races. Regulation C currently requires financial institutions to report up to five racial designations of an applicant. The Bureau believes that the likelihood of applicants self-identifying as being of more than five particular racial designations is low. Accordingly, the Bureau is adopting a new instruction 9 in appendix B, which provides that a financial institution must offer the applicant the option of selecting more than one particular ethnicity or race. The new instruction provides that if an applicant selects more than one particular ethnicity or race, a financial institution must report each selected designation, subject to the limits described in the instruction. With respect to ethnicity, the instruction requires a financial institution to report each aggregate ethnicity category and each ethnicity subcategory selected by the applicant. In addition, the instruction explains that if an applicant selects the Other Hispanic or Latino ethnicity subcategory, the applicant may also provide a particular Hispanic or Latino ethnicity not listed in the standard subcategories. In such a case, the instruction requires a financial institution to report both the selection of Other Hispanic or Latino and the additional information provided by the applicant. With respect to race, the instruction requires a financial institution to report every aggregate race category selected by the applicant. If the applicant also selects one or more race subcategories, the instruction requires the financial institution to report each race subcategory selected by the applicant, except that the financial institution must not report more than a total of five aggregate race categories and race subcategories combined. The instruction provides illustrative examples to facilitate HMDA compliance. In addition, the instruction explains that if an applicant selects the Other Asian race subcategory or the Other Pacific Islander race subcategory, the applicant may also provide a particular Other Asian or Other Pacific Islander race not listed in the standard subcategories. In either such case, the instruction requires a financial institution to report both the selection of Other Asian or Other Pacific Islander, as applicable, and the additional information provided by the applicant, subject to the five-race maximum. In all such cases where the applicant has selected an Other race subcategory and also provided additional information, for purposes of the five-race maximum, the Other race subcategory and additional information provided by the applicant together constitute only one selection. The instruction provides an illustrative example to facilitate compliance. The Bureau is also modifying the introductory paragraph in the sample data collection form in appendix B in an effort to improve the explanation provided to applicants by financial institutions as to why their demographic information is being collected. In response to the Bureau's solicitation for feedback on ways to improve the data collection on ethnicity, race, and sex, a few commenters stated that applicants may be reluctant to provide their demographic information because they do not understand why it is being collected or for what purposes. For example, an industry commenter suggested that the language explaining to the applicant why the information is being requested should be in plain language and contain less legalese in order for an applicant to feel more comfortable in responding to the request. Another industry commenter suggested that applicants who choose not to provide their demographic information may be concerned that by doing so, such information may negatively influence the credit decision made by a financial institution. The Bureau believes that the explanation provided to applicants by financial institutions should clearly state why their demographic information is being collected and for what purposes such information is requested by the Federal government. Accordingly, the Bureau is modifying the introductory paragraph in the sample data collection form in appendix B to include the following sentences: ``The purpose of collecting this information is to help ensure that all applicants are treated fairly and that the housing needs of communities and neighborhoods are being fulfilled. For residential mortgage lending, Federal law requires that we ask applicants for their demographic information (ethnicity, race, and sex) in order to monitor our compliance with equal credit opportunity, fair housing, and home mortgage disclosure laws.'' The Bureau is adopting other changes to the introductory paragraph in the sample data collection form to align with the new data standards on collection and reporting of ethnicity and race. In order to align with the modified introductory paragraph in the sample data collection form, the Bureau is also adopting new instruction 2, which clarifies that a financial institution must inform applicants that Federal law requires collection of their demographic information in order to protect consumers and to monitor compliance with Federal statutes that prohibit discrimination against applicants on the basis of ethnicity, race, and sex. The Bureau is also modifying the title of the sample data collection form. A few commenters stated that ``Information for Government Monitoring Purposes'' may discourage applicants from providing their demographic information. For example, by using the words ``government monitoring,'' a few industry commenters suggested that applicants may view the collection of this information as intrusive or intimidating, as opposed to ensuring that they are protected from discrimination. Another industry [[Page 66193]] commenter stated that some applicants are not aware that Federal statutes and regulations protect them from discrimination and that ``government monitoring information'' promotes a sense among applicants that the financial institution's credit decision is based, at least in part, on their demographic information. The Bureau has considered this feedback and determined that the title of the sample data collection form should be modified in order to address the concern that the current title may discourage applicants from providing their demographic information. Accordingly, the Bureau is modifying the title of the sample data collection form to ``Demographic Information of Applicant and Co-Applicant.'' The Bureau has determined that modifying the introductory paragraph in the sample data collection form and its title, as well as adopting new instruction 2 in appendix B, will assist financial institutions in explaining to applicants the purposes of collecting their demographic information and how the information is used. The Bureau believes that these changes may improve the HMDA data's usefulness by encouraging applicants to provide their demographic information. The Bureau is also modifying instruction 1 in appendix B, which currently provides that for applications taken by telephone, the information in the collection form must be stated orally by the lender, except for that information which pertains uniquely to applications taken in writing. The Bureau has received questions regarding the meaning of the phrase ``except for that information which pertains uniquely to applications taken in writing.'' The Bureau has modified this instruction in the final rule and provides an illustrative example, which will address confusion regarding this phrase. The Bureau is also modifying the sample data collection form by allowing applicants to select ``I do not wish to provide this information'' separately for ethnicity, race, and sex. Previously, the sample data collection form provided a ``I do not wish to furnish this information'' box at the top of the form, which applied to ethnicity, race, and sex as a group. The Bureau believes that modifying the selection to include a ``I do not wish to provide this information'' box following the request for the applicant's ethnicity, race, and sex will allow an applicant to more clearly articulate a decision to decline to provide certain information but not other information. Additional guidance on this topic had been published in the FFIEC FAQs.\289\ The Bureau believes it is appropriate to modify the sample data collection form in appendix B, adapted from the FFIEC FAQs, to improve the collection of this information and assist financial institutions with HMDA compliance. --------------------------------------------------------------------------- \289\ See http://www.ffiec.gov/hmda/faqreg.htm#threeboxes. --------------------------------------------------------------------------- The Bureau is also proposing to add four new instructions to appendix B to provide additional guidance regarding the reporting requirement under Sec. 1003.4(a)(10)(i). First, the Bureau received feedback requesting that it clarify whether a financial institution must report the demographic information of a guarantor. To help facilitate HMDA compliance, the Bureau is adopting new instruction 4 in appendix B, which clarifies that for purposes of Sec. 1003.4(a)(10)(i), if a covered loan or application includes a guarantor, a financial institution does not report the guarantor's ethnicity, race, and sex. While the terms ``applicant'' and ``borrower'' may include guarantors in other regulations,\290\ the Bureau believes the inclusion of information regarding the ethnicity, race, and sex of guarantors in the HMDA data would be unnecessarily burdensome and potentially lead to inconsistencies in the data. --------------------------------------------------------------------------- \290\ For example, Regulation B defines the term ``applicant'' to include guarantors, sureties, endorsers, and similar parties for some purposes. See 12 CFR 1002.2(e). --------------------------------------------------------------------------- Second, an industry commenter pointed out that the Bureau's proposed instruction 4(a)(10)-2.a provides ``You need not collect or report this information for covered loans purchased. If you choose not to report this information for covered loans that you purchase, use the Codes for not applicable.'' However, the Bureau's proposed instructions 4(a)(10)(i)-2.c, 4(a)(10)(i)-3.b, 4(a)(10)(i)-4.a, and 4(a)(10)(ii)-1.d instructed financial institutions to report the corresponding code for ``not applicable'' for ethnicity, race, sex, age, and income ``when the applicant or co-applicant information is unavailable because the covered loan has been purchased by your institution.'' The Bureau agrees that these instructions do not align and has determined that a clarification will facilitate HMDA compliance. Consequently, the Bureau is adopting new instruction 6 in appendix B, which requires that when a financial institution purchases a covered loan and chooses not to report the applicant's or co-applicant's ethnicity, race, and sex, the financial institution reports that the requirement is not applicable. Third, prior to the Bureau's proposal, financial institutions had expressed uncertainty as to whether a trust is a non-natural person for purposes of HMDA. In response, the Bureau proposed to add ``trust'' to the list of examples in the technical instructions in appendix A, which direct financial institutions to report the code for ``not applicable'' if the borrower or applicant is not a natural person. A few commenters supported the proposed clarification. The Bureau has determined that the proposed clarification will facilitate HMDA compliance. Consequently, the Bureau is adopting new instruction 7, which provides, in part, a financial institution reports that the requirement to report the applicant's or co-applicant's ethnicity, race, and sex is not applicable when the applicant or co-applicant is not a natural person (for example, a corporation, partnership, or trust). The new instruction clarifies that for a transaction involving a trust, a financial institution reports that the requirement is not applicable if the trust is the applicant. On the other hand, if the applicant is a natural person, and is the beneficiary of a trust, a financial institution reports the applicant's ethnicity, race, and sex. Lastly, the Bureau is adopting new instruction 13 in appendix B, which clarifies how a financial institution should report partial demographic information provided by an applicant. Additional guidance on this topic had been published in the FFIEC FAQs.\291\ The Bureau believes it is appropriate to include an instruction in appendix B, adapted from the FFIEC FAQs, to assist financial institutions with HMDA compliance. --------------------------------------------------------------------------- \291\ See http://www.ffiec.gov/hmda/faqreg.htm#collectioninfo. --------------------------------------------------------------------------- For the reasons discussed above, the Bureau is adopting proposed Sec. 1003.4(a)(10)(i), with the following substantive change. The Bureau is requiring financial institutions to report whether the applicant's or co-applicant's ethnicity, race, and sex was collected on the basis of visual observation or surname. Consequently, Sec. 1003.4(a)(10)(i) and appendix B of the final rule require a financial institution to collect and report the applicant's or co-applicant's ethnicity, race, and sex, and whether this information was collected on the basis of visual observation or surname. In addition, for the reasons discussed above, the Bureau is adding new instructions, as well as modifying a few of the current instructions, in appendix B and the sample data collection form [[Page 66194]] in order to facilitate compliance with the new collection and reporting requirements relating to an applicant's ethnicity, race, and sex. The Bureau is adopting proposed comments 4(a)(10)(i)-1, -2, -3, -4, and -5 as new instructions 8, 10, 12, 5, and 3, respectively, in appendix B, modified to conform to the changes the Bureau is finalizing in Sec. 1003.4(a)(10)(i) and to provide additional clarity as to the data collection requirements. In addition, as discussed above, the Bureau is adopting new instructions 4, 6, 7, 9, 11, and 13 in appendix B. The Bureau has modified proposed comment 4(a)(10)(i)-1, which directs financial institutions to refer to appendix B for instructions on collection of an applicant's ethnicity, race, and sex. By placing all of the data collection instructions with respect to an applicant's ethnicity, race, and sex in one location--appendix B--the Bureau has streamlined the regulatory requirements in an effort to reduce compliance burden. The Bureau has determined that these data collection instructions in appendix B and the revised sample data collection form, discussed above, will help facilitate HMDA compliance by providing additional guidance regarding the reporting requirements under Sec. 1003.4(a)(10)(i). Lastly, in order to facilitate compliance with the new collection and reporting requirements in Sec. 1003.4(a)(10)(i) and appendix B relating to an applicant's ethnicity, race, and sex, the Bureau added new comment 4(a)(10)(i)-2 in the final rule and provides an illustrative example. Comment 4(a)(10)(i)-2 provides that if a financial institution receives an application prior to January 1, 2018, but final action is taken on or after January 1, 2018, the financial institution complies with Sec. 1003.4(a)(10)(i) and (b) if it collects the information in accordance with the requirements in effect at the time the information was collected. For example, if a financial institution receives an application on November 15, 2017, collects the applicant's ethnicity, race, and sex in accordance with the instructions in effect on that date, and takes final action on the application on January 5, 2018, the financial institution has complied with the requirements of Sec. 1003.4(a)(10)(i) and (b), even though those instructions changed after the information was collected but before the date of final action. However, if, in this example, the financial institution collected the applicant's ethnicity, race, and sex on or after January 1, 2018, Sec. 1003.4(a)(10)(i) and (b) requires the financial institution to collect the information in accordance with the amended instructions. 4(a)(10)(ii) Section 1094(3)(A)(i) of the Dodd-Frank Act amended HMDA section 304(b)(4) to require financial institutions to report an applicant's or borrower's age.\292\ The Bureau proposed to implement the requirement to collect and report age by adding this characteristic to the information listed in proposed Sec. 1003.4(a)(10)(i). In light of potential applicant and borrower privacy concerns related to reporting date of birth, the Bureau proposed that financial institutions enter the age of the applicant or borrower, as of the date of application, in number of years as derived from the date of birth as shown on the application form. --------------------------------------------------------------------------- \292\ 12 U.S.C. 2803(b)(4). --------------------------------------------------------------------------- The Bureau solicited feedback regarding whether this was an appropriate manner of collecting the age of applicants. Many commenters expressed concern about potential privacy implications if the Bureau requires financial institutions to report an applicant's age or if the Bureau were to release such data to the public. As with other proposed data points like credit score, commenters were concerned that if information regarding an applicant's or borrower's age is made available to the public, such information could be coupled with other publicly available information, such as the security instrument and other local records, in a way that compromises an applicant's or borrower's privacy. A national trade association commented that by increasing the scope of HMDA reporting, the Bureau would increase potential privacy risks of consumers. The commenter argued that expansive new data elements, like age, result in an unjustifiable privacy intrusion by providing information that allows someone to identify applicants and borrowers along with a detailed picture of their financial state. Similarly, an industry commenter suggested that in addition to the potential for criminal misuse of a borrower's financial information, the availability of the expanded data released under HMDA will very likely permit marketers to access the information which will result in aggressive marketing that is ``personalized'' to unsophisticated and vulnerable consumers for potentially harmful financial products and services. Another State trade association recommended that the Bureau strengthen its data protection as it relates to the selective disclosure of HMDA data to third parties and specifically recommended that the Bureau convert actual values to ranges or normalize values before sharing the data with a third party. The Bureau has considered this feedback. See part II.B above for a discussion of the Bureau's approach to protecting applicant and borrower privacy with respect to the public disclosure of HMDA data. In contrast, many consumer advocate commenters stated that requiring financial institutions to report an applicant's age is vital information that allows the public to evaluate age biases in lending, especially in conjunction with reverse mortgages. These commenters stated that the public needs to know the extent of reverse mortgage lending for various categories of older adults to ensure that various age cohorts are being served and are not being abused. Another commenter stated that an applicant's age is an important element for understanding patterns of mortgage lending and noted that mortgage underwriting standards may contribute to disparate outcomes in homeownership among different age cohorts. Another commenter stated that requiring financial institutions to report a borrower's age is important to ensure that borrowers in any particular age category are not experiencing undue barriers to mortgage credit. Many commenters also provided feedback regarding the Bureau's request as to whether there was a less burdensome way for financial institutions to collect such information for purposes of HMDA. For example, many industry commenters recommended that the Bureau require financial institutions to report age as a ``range of values'' rather than an applicant's or borrower's actual age. The commenters suggested that reporting an applicant's age as a range of values will eliminate a substantial number of potential errors on financial institutions' loan/ application registers, would better protect the privacy of applicants, and would not compromise the integrity of the HMDA data. Another industry commenter generally agreed that applicants' age information would be useful to users of the HMDA data when analyzing housing trends and a financial institution's fair lending performance, but recommended that the Bureau require reporting of an applicant's date of birth and not the actual age of the applicant. Another industry commenter explained that it only requires date of birth on its applications and not age specifically. If the Bureau implements the requirement to report the applicant's age in years, the commenter stated that the consequence would be that [[Page 66195]] customized loan application forms would need to be amended to include this additional information or institutions would need to manually calculate an applicant's age, which will significantly increase both the burden of this reporting requirement and errors. A few industry commenters stated that the costs of the proposed requirement would not be justified. Other industry commenters stated that calculating an applicant's actual age will be an unnecessary burden and an area of potentially high error rate, and as such, the Bureau should require reporting of the applicant's year of birth. The Bureau has considered this feedback and determined that requiring financial institutions to report the applicant's actual age-- and not the applicant's date of birth, year of birth, or a range within which an applicant's age falls--is the appropriate method of implementing HMDA section 304(b)(4) and carrying out HMDA's purposes. In light of potential applicant and borrower privacy concerns related to reporting date of birth or year of birth, the Bureau has determined that requiring financial institutions to report the applicant's actual age is the proper approach. The Bureau has also determined that requiring financial institutions to report age as a range of values would diminish the utility of the data to further HMDA's purposes. By requiring financial institutions to report the applicant's actual age, this information will assist in identifying whether financial institutions are serving the housing needs of their communities, identifying possible discriminatory lending patterns, and enforcing antidiscrimination statutes. The Bureau recognizes that a requirement to collect and report the applicant's age may impose some burden on financial institutions and that requiring financial institutions to calculate the age of an applicant in number of years by referring to the date of birth as shown on the application form may result in potential calculation errors. However, the Bureau has determined that the benefits of this reporting requirement justify any burdens and financial institutions will have to manage the risk of an error in calculating an applicant's age to ensure HMDA compliance. The final rule renumbers proposed Sec. 1003.4(a)(10)(i) and moves the requirement to collect the age of the applicant or borrower to Sec. 1003.4(a)(10)(ii). The new numbering is intended only for ease of reference and is not a substantive change. In addition, in order to help facilitate HMDA compliance, the Bureau is moving the proposed commentary regarding the reporting requirements for an applicant's and borrower's age into new comments. The Bureau is adopting new comments 4(a)(10)(ii)-1, -2, -3, -4, and -5. The Bureau is adopting new comment 4(a)(10)(ii)-1, which explains that a financial institution complies with Sec. 1003.4(a)(10)(ii) by reporting the applicant's age, as of the application date under Sec. 1003.4(a)(1)(ii), as the number of whole years derived from the date of birth as shown on the application form, and provides an illustrative example. This requirement aligns with the definition of age under Regulation B.\293\ --------------------------------------------------------------------------- \293\ The Bureau's Regulation B requires, as part of the application for credit, a creditor to request the age of an applicant for credit primarily for the purchase or refinancing of a dwelling occupied or to be occupied by the applicant as a principal dwelling, where the credit will be secured by the dwelling. Regulation B Sec. 1002.13(a)(1)(iv). Age has been a protected category under ECOA and Regulation B since 1976, and a creditor may not discriminate against an applicant on the basis of age regarding any aspect of a credit transaction, including home mortgage lending. See Regulation B Sec. Sec. 1002.1(b), 1002.4(a)(b), 15 U.S.C. 1691(a)(1). Under Regulation B, ``age'' refers ``only to the age of natural persons and means the number of fully elapsed years from the date of an applicant's birth.'' Regulation B Sec. 1002.2(d). --------------------------------------------------------------------------- Similar to the requirement applicable to an applicant's ethnicity, race, and sex, the Bureau is adopting new comment 4(a)(10)(ii)-2, which clarifies that if there are no co-applicants, a financial institution reports that there is no co-applicant. On the other hand, if there is more than one co-applicant, the financial institution reports the age only for the first co-applicant listed on the application form. The comment also explains that a co-applicant may provide the absent co- applicant's age on behalf of the absent co-applicant. The Bureau is adopting new comment 4(a)(10)(ii)-3, which clarifies when a financial institution reports that the requirement is not applicable. Similar to the requirement applicable to an applicant's ethnicity, race, and sex, comment 4(a)(10)(ii)-3 explains that for a covered loan that the financial institution purchases and for which the institution chooses not to report the applicant's or co-applicant's age, the financial institution reports that the requirement is not applicable. In addition, comment 4(a)(10)(ii)-4 explains that a financial institution reports that the requirement to report the applicant's or co-applicant's age is not applicable when the applicant or co-applicant is not a natural person (for example, a corporation, partnership, or trust), and provides an illustrative example. Lastly, the Bureau received feedback requesting that it clarify whether a financial institution must report the demographic information of a guarantor. Similar to the requirement applicable to an applicant's ethnicity, race, and sex, the Bureau is adopting new comment 4(a)(10)(ii)-5, which clarifies that for purposes of Sec. 1003.4(a)(10)(ii), if a covered loan or application includes a guarantor, a financial institution does not report the guarantor's age. These five new comments will help facilitate HMDA compliance by providing guidance on the reporting requirements regarding an applicant's or borrower's age. 4(a)(10)(iii) HMDA section 304(b)(4) requires the reporting of income level for borrowers and applicants. Section 1003.4(a)(10) of Regulation C implements this requirement by requiring collection and reporting of the applicant's gross annual income relied on in processing the application. Proposed Sec. 1003.4(a)(10)(ii) revised the current rule to require the reporting of gross annual income relied on in making the credit decision requiring consideration of income or, if a credit decision requiring consideration of income was not made, the gross annual income collected as part of the application process. The Bureau also proposed amendments to the commentary and two new illustrative comments. The Bureau is adopting Sec. 1003.4(a)(10)(iii), renumbered from proposed Sec. 1003.4(a)(10)(ii), and comments 4(a)(10)(iii)-1 through -10. The Bureau received feedback on proposed Sec. 1003.4(a)(10)(ii) and its commentary from a small number of commenters. A handful of commenters, including consumer advocates and industry commenters, expressed support for proposed Sec. 1003.4(a)(10)(ii). As information about an applicant's or borrower's income provides information about underwriting decisions and access to credit, the Bureau believes that collecting it is important for achieving HMDA's purposes: to identify possible fair lending violations, to understand whether financial institutions are meeting the housing needs of their communities, and to help policymakers allocate public investments so as to attract private capital. Therefore, it is appropriate to continue to require financial institutions to report information about an applicant's or borrower's gross annual income. A few industry commenters addressed challenges associated with reporting the gross annual income relied on in making the credit decision. One commenter suggested requiring [[Page 66196]] reporting of the income obtained from a readily verifiable source instead of the gross annual income relied on in making the credit decision. Others asked for clarification about what is meant by gross annual income, including whether gross annual income requires reporting of the income that the financial institution has verified. It is not necessary to modify proposed Sec. 1003.4(a)(10)(ii) to allow financial institutions that rely on the verified gross annual income to report the verified gross annual income. Proposed Sec. 1003.4(a)(10)(ii) provided flexibility for the financial institution to report the gross annual income that the financial institution relied on in making the credit decision for the loan or application that the institution is reporting. Under the proposal, if a financial institution relied on the verified gross annual income, then the institution would report the verified gross annual income. In addition, in circumstances when a financial institution did not rely on the verified gross annual income, the financial institution would report the gross annual income that it relied on in making the credit decision. The Bureau believes that it is important to maintain this flexibility in the final rule and accordingly is not adopting commenters' suggestions to change the requirement. However, in response to the comments, the Bureau is modifying proposed comment 4(a)(10)(ii)-1, renumbered as comment 4(a)(10)(iii)-1, to clarify that a financial institution reports the verified gross annual income when the financial institution relied on the verified gross annual income in making the credit decision. Some industry commenters also raised concerns about public disclosure of this information. See part II.B above for a discussion of the Bureau's approach to protecting applicant and borrower privacy with respect to the public disclosure of the data. Other industry commenters urged the Bureau to consider excluding certain types of loans, such as multifamily loans, business purpose loans, and purchased loans, from the requirement to report income in proposed Sec. 1003.4(a)(10)(ii). The final rule effectively excludes these loans from income reporting. New comment 4(a)(10)(iii)-7 excludes loans to non-natural persons and new comment 4(a)(10)(iii)-8 excludes those related to multifamily dwellings from the requirement to report income information. New comment 4(a)(10)(iii)-9 provides that reporting income information is optional for purchased loans. However, as discussed in comments 4(a)-3 and -4, a financial institution that reviews an application for a covered loan, makes a credit decision on that application prior to closing, and purchases the covered loan after closing will report the covered loan that it purchases as an origination, not a purchase. Accordingly, in those circumstances, the final rule requires the financial institution to report the gross annual income that it relied on in making the credit decision. Other industry commenters expressed concerns about the proposed requirement to report the gross annual income collected as part of the application process. One commenter urged the Bureau to only require reporting of income information if it is relied on in making a credit decision. Another commenter urged the Bureau to require reporting of the most recent verified income, instead of the income stated by the borrower, because institutions update income throughout the application process to take into account new information. Another commenter suggested that collecting income information that is not verified is inconsistent with the Bureau's 2013 ATR Final Rule, which the commenter stated requires income to be verified. Information concerning income on applications when no credit decision was made provides valuable data to understand access to credit and underwriting decisions. The Bureau recognizes, however, as suggested by commenters, that the proposal's description of the requirement to report income in those circumstances created confusion about what income information to report. To respond to the concerns raised by the commenters, the Bureau is not adopting the language in proposed Sec. 1003.4(a)(10)(ii) that describes reporting income on applications when no credit decision was made. Instead, the Bureau is retaining the language currently used in Sec. 1003.4(a)(10) to describe what to report in that circumstance. The final rule provides that if a credit decision is not made, a financial institution reports the gross annual income relied on in processing the application for a covered loan that requires consideration of income. In that case, the financial institution should report whatever income information it was relying on when the application was withdrawn or closed for incompleteness, which could include the income information provided by the applicant initially, any additional income information provided by the applicant during the application process, and any adjustments to that information during the application process due to the institution's policies and procedures. These adjustments may include, for example, reducing the income amount to reflect verified income or to eliminate types of income not considered by the financial institution. In addition, proposed comment 4(a)(10)(ii)-5, finalized as comment 4(a)(10)(iii)-5, is revised to clarify that a financial institution is not necessarily required to report the income information initially provided on the application. Rather, the financial institution may update the income information initially provided by the applicant with additional information collected from the applicant if it relies on that additional information in processing the application. Another industry commenter expressed concerns about proposed comment 4(a)(10)(ii)-4, which explained that an institution should not include as income, amounts considered in making a credit decision based on factors that an institution relies on in addition to income. For example, the proposal directed financial institutions not to include as income any amounts derived from annuitization or depletion of an applicant's remaining assets. The commenter noted that proposed comment 4(a)(10)(ii)-4 would be difficult to implement because lenders would have to create new data fields to identify and exclude annuitized income. In addition, the commenter stated that adopting the proposed comment would create a distorted picture of an applicant's cash flow. The Bureau is finalizing proposed comment 4(a)(10)(ii)-4, renumbered as comment 4(a)(10)(iii)-4, to focus on applicant income as distinct from an applicant's assets or other resources. Although financial institutions may rely on assets or other resources in underwriting a loan, including amounts other than income, such as assets, would result in data that is less useful and less accurate. Therefore, it would not be appropriate to report that information as income. For the reasons discussed above, the Bureau is finalizing proposed Sec. 1003.4(a)(10)(ii), renumbered as Sec. 1003.4(a)(10)(iii), with technical modifications for clarification. The Bureau is also finalizing proposed comments 4(a)(10)(ii)-1 through -6, renumbered as comments 4(a)(10)(iii)-1 through -6, with clarifying modifications to provide illustrative examples. The Bureau is also moving proposed instruction 4(a)(10)-2.a into new comment 4(a)(10)(iii)-9 and proposed instruction 4(a)(10)(ii)-1 into new comments 4(a)(10)(iii)-7, -8, and - 10. [[Page 66197]] 4(a)(11) Current Sec. 1003.4(a)(11) requires financial institutions to report the type of entity purchasing a loan that the financial institution originates or purchases and then sells within the same calendar year, and provides that this information need not be included in quarterly updates.\294\ In conjunction with the Bureau's proposal to require quarterly data reporting by certain financial institutions as described further below in the section-by-section analysis of Sec. 1003.5(a)(1)(ii), the Bureau proposed to modify Sec. 1003.4(a)(11) by deleting the statement that the information about the type of purchaser need not be included in quarterly updates. In addition, the Bureau proposed technical modifications to current comments 4(a)(11)-1 and -2 and also proposed to add six new comments to provide additional guidance regarding the type of purchaser reporting requirement. --------------------------------------------------------------------------- \294\ 12 CFR 1002.4(a)(11); see also 12 U.S.C. 2803(h)(1)(C) (authorizing regulations that ``require disclosure of the class of the purchaser of such loans''). --------------------------------------------------------------------------- The Bureau solicited feedback regarding whether the proposed comments were appropriate and specifically solicited feedback regarding whether additional clarifications would assist financial institutions in complying with proposed Sec. 1003.4(a)(11). The Bureau received a few comments. With respect to the Bureau's proposal that the type of purchaser data be included in quarterly reporting by certain financial institutions, one industry commenter stated that the proposal did not specify how a quarterly reporter would report a loan it originated in one quarter and sold in another quarter during the same year. The Bureau proposed an instruction, which it is adopting as new comment 4(a)(11)-9 with the following clarifications: A financial institution records that the requirement is not applicable if the institution originated or purchased a covered loan and did not sell it during the calendar quarter for which the institution is recording the data; if the financial institution sells the covered loan in a subsequent quarter of the same calendar year, the financial institution records the type of purchaser on its loan/application register for the quarter in which the covered loan was sold; if the financial institution sells the covered loan in a succeeding year, the institution should not record the sale. For clarity, the Bureau also adopts new comment 4(a)(11)-10, which provides that a financial institution reports that the requirement is not applicable for applications that were denied, withdrawn, closed for incompleteness or approved but not accepted by the applicant; and for preapproval requests that were denied or approved but not accepted by the applicant. The new comment also provides that a financial institution reports that the requirement is not applicable if the institution originated or purchased a covered loan and did not sell it during that same calendar year. The Bureau proposed comment 4(a)(11)-3, which clarifies when a financial institution shall report the code for ``affiliate institution'' by providing a definition of the term ``affiliate'' and clarifying that for purposes of proposed Sec. 1003.4(a)(11), the term ``affiliate'' means any company that controls, is controlled by, or is under common control with, another company, as set forth in the Bank Holding Company Act of 1956 (12 U.S.C. 1841 et seq.). One industry commenter stated that it is difficult for a financial institution to determine the correct code to report for the type of purchaser, especially when mergers, acquisitions, and affiliates are involved in the transaction, and recommended that financial institutions simply report ``sold'' or ``kept in portfolio'' for this requirement. Another industry commenter stated that the proposed definition of ``affiliate'' remains unclear and urged the Bureau to align the definition with existing regulations, including the Secure and Fair Enforcement of Mortgage Licensing Act of 2008 (SAFE Act). The Bureau considered the recommendation to require reporting of whether a particular loan has been ``sold'' within the same calendar year or ``kept in portfolio,'' but has determined that requiring reporting of the type of purchaser is the more appropriate approach. The type of purchaser information reported under HMDA provides valuable information, for example, by helping data users understand the secondary mortgage market. A requirement to simply report whether a particular loan was ``sold'' or ``kept in portfolio'' would greatly diminish the utility of this HMDA data. In addition, the Bureau has determined that the proposed definition of ``affiliate'' is appropriate and provides clarity as to when a financial institution should report that the type of purchaser is an affiliate institution. The Bureau considered other definitions of ``affiliate'' across various laws and regulations and has concluded that for purposes of reporting the type of purchaser under HMDA, the definition of ``affiliate'' established in the Bank Holding Company Act is appropriate. Appendix A to Sec. 1003.4(a)(11) groups ``life insurance company, credit union, mortgage bank, or finance company'' into one category when reporting type of purchaser. The Bureau did not propose to change this grouping. However, one commenter recommended that ``insurance companies'' be separated from ``life insurance company, credit union, mortgage bank, or finance company.'' The commenter argued that separating insurance companies from other types of purchasers would result in improved data with respect to both information about the ultimate source of financing in the multifamily market and information about secondary-market financing provided by credit unions, mortgage banks, and finance companies. In response, the Bureau is adopting a new modification that will permit reporting that the purchaser type is a life insurance company separately from other purchaser types. The Bureau is also modifying proposed comment 4(a)(11)-5 by replacing ``mortgage bank'' with ``mortgage company'' and clarifying that for purposes of Sec. 1003.4(a)(11), a mortgage company means a nondepository institution that purchases mortgage loans and typically originates such loans. Additional guidance on this topic had been published in the FFIEC FAQs.\295\ The Bureau believes this clarification, adapted from the FFIEC FAQs, will facilitate compliance with the type of purchaser reporting requirement. --------------------------------------------------------------------------- \295\ See http://www.ffiec.gov/hmda/faqreg.htm#mrtgbanks. --------------------------------------------------------------------------- The Bureau is adopting Sec. 1003.4(a)(11) as proposed. The Bureau is also adopting comments 4(a)(11)-1 through -8, with several technical and clarifying modifications, and new comments 4(a)(11)-9 and -10 to help facilitate HMDA compliance by providing additional guidance regarding the type of purchaser reporting requirement. 4(a)(12) HMDA section 304(b)(5)(B) requires financial institutions to report mortgage loan information, grouped according to measurements of ``the difference between the annual percentage rate associated with the loan and a benchmark rate or rates for all loans.'' \296\ Currently, Regulation C requires financial institutions to report the difference between a loan's annual [[Page 66198]] percentage rate (APR) and the average prime offer rate (APOR) for a comparable transaction, as of the date the interest rate is set, if the difference equals or exceeds 1.5 percentage points for first-lien loans, or 3.5 percentage points for subordinate-lien loans. The Bureau proposed to implement HMDA section 304(b)(5)(B) in Sec. 1003.4(a)(12), by requiring financial institutions to report, for covered loans subject to Regulation Z, 12 CFR part 1026, other than purchased loans and reverse mortgage transactions, the difference between the covered loan's annual percentage rate and the average prime offer rate for a comparable transaction as of the date the interest rate is set. For the reasons discussed below, the Bureau is adopting Sec. 1003.4(a)(12) generally as proposed, but with a modification to exclude assumptions. --------------------------------------------------------------------------- \296\ Section 1094(3)(A)(iv) of the Dodd-Frank Act amended HMDA by adding section 304(b)(5)(B), which expanded the rate spread reporting requirement beyond higher-priced mortgage loans. --------------------------------------------------------------------------- The Bureau solicited comment on the general utility of the revised rate spread data and on the costs associated with collecting and reporting. Several industry commenters and a few trade associations opposed the Bureau's proposal requiring rate spread information. One commenter stated that certain financial institutions should be exempted from the rate spread reporting requirement on covered loans and applications. Industry commenters were generally concerned about the burden associated with reporting rate spread data for more transactions than what is currently collected and reported. In particular, commenters pointed to the expense or additional work required to calculate the rate spread, such as the need to update software. One industry commenter stated that current systems determine rate spread and provide a numerical difference if the difference exceeds a predetermined trigger. The Bureau's proposal that the rate spread should be reported for all loans and not just the ones whose rate spread exceeds a certain threshold will require systems updates or a manual updates, according to the commenter. One commenter stated that rate spread information would not provide any meaningful data regarding access to credit on fair terms and another commenter stated that the additional regulatory burden would not be beneficial to consumers or for the purposes of antidiscriminatory monitoring. As noted in the proposal, Congress found that improved pricing information would bring greater transparency to the market and facilitate enforcement of fair lending laws.\297\ Feedback from the Board's 2010 Hearings suggested that requiring rate spread information for all loans, not just certain loans considered higher-priced, would provide a more complete understanding of the mortgage market and also improve loan analyses across various markets and communities.\298\ Furthermore, the proposal noted that recent enforcement actions pursued by the U.S. Department of Justice indicated that price discrimination can occur even at levels that fall below the current higher-priced thresholds. Based on the findings of Congress, feedback from the Board's 2010 Hearings, and enforcement actions, the Bureau concluded that requiring the rate spread for most loans or applications by all financial institutions will enhance the HMDA data by providing the information that could improve loan analyses and therefore enable a better understanding of the mortgage market. The Bureau believes that such benefits will justify any additional burden imposed by the final rule. --------------------------------------------------------------------------- \297\ H.R. Rep. No. 111-702, at 191 (2011). \298\ See Atlanta Hearing, supra note 40; Chicago Hearing, supra note 46; see also Neil Bhutta & Glenn B. Canner, Bd. of Governors of the Fed. Reserve Sys., 99 Fed. Reserve Bulletin 4, Mortgage Market Conditions and Borrower Outcomes: Evidence from the 2012 HMDA Data and Matched HMDA-Credit Record Data, at 31-32 (Nov. 2013) (noting that gaps in the rate spread data limit its current usefulness for assessing fair lending compliance). --------------------------------------------------------------------------- Several industry commenters asked for clarification on whether the rate spread field will be required to be completed on loans subject to Regulation Z but exempted from the higher-priced loan category in Regulation Z Sec. 1026.35, such as a home-equity lines of credit. The Bureau believes that the rate spread data on most transactions, including open-end lines of credit, would be beneficial by providing data to contribute to a more complete understanding of the mortgage market. One industry commenter questioned whether reporting a covered loan's or application's APR would be a better alternative than reporting rate spread data. This commenter pointed out that reporting APR is much less burdensome than calculating the rate spread and therefore less prone to errors, such as the use of the wrong date on which to compare APR to the APOR. In addition to the risk of errors, the commenter stated that requiring the financial institution to report the rate spread information will increase the cost of preparing the report. A trade association questioned why it would not be sufficient for the APR to be reported, which would then allow the data user to select a benchmark of their choice for comparison. Although reporting the APR on the covered loan or application would reduce the burden on financial institutions reporting the rate spread data, based on the language in the Dodd-Frank Act, the Bureau believes a reasonable interpretation of HMDA section 304(b)(5)(B) is that financial institutions should report the difference between the APR and APOR. In addition, the rate spread provides a more accurate picture of a loan's price relative to the rate environment at the time of the lender's pricing decision because the date the loan's interest rate was set is not publicly available. A few commenters warned that rate spread data could be misleading if viewed out of context. For example, a trade association commented that some loans may have higher rate spreads but offer special features, such as lower down payment requirements or waiver of an institution's private mortgage insurance requirement. Another commenter suggested that users need to be aware of the issues regarding rate spread data and pointed out that lender credits do not impact the APR and therefore the rate spread will look higher in comparison to similar loans without lender credits. Although there may be issues regarding rate spread data, the Bureau believes that it would be less burdensome on financial institutions to calculate the difference between APR, which is already a calculation performed by the financial institutions for TILA-RESPA purposes, and APOR. The Bureau does not believe that the additional burden of requiring financial institutions to take into account other factors, such as lender credits, when calculating the APR for the purposes of the rate spread would outweigh any benefit provided by this adjusted method of calculation. In addition, the Bureau believes that a reasonable interpretation of HMDA section 304(b)(5)(B) is that financial institutions should report the difference between the APR on the loan and the APOR for a comparable transaction. The Bureau also solicited comment on the scope of the rate spread reporting requirement, including whether the requirement should be expanded to cover purchased loans. One trade association agreed with the Bureau's proposal that reverse mortgages should be exempted from rate spread reporting. A few trade associations agreed with the Bureau and commented that the rate spread reporting requirement should not be expanded to include purchased loans. One trade association reasoned this this would require a manual retroactive process to determine the APOR for the financial institution reporting the purchased loan. The Bureau recognizes the burden that [[Page 66199]] would be imposed on the financial institution reporting the purchased loan to also report the rate spread and therefore is excluding purchased covered loans from the rate spread reporting requirement as proposed. One industry commenter asked the Bureau to clarify whether rate spread should be reported on commercial loans that do not have an APR. The Bureau did not propose to, and the final rule does not, require a financial institution to report the rate spread for commercial loans because these loans are not covered by Regulation Z, and therefore creditors are not required to calculate and disclose an APR to borrowers. Many commenters noted that the Bureau's proposal contained inconsistent rounding methodologies across various data points, including the rate spread, and recommended that the Bureau provide a consistent rounding method. The technical instructions in current appendix A provides that the rate spread should be reported to two decimal places. If the rate spread figure is more than two decimal places, the figure should be rounded or truncated to two decimal places. The Bureau proposed that the rate spread should be rounded to three decimal places. One commenter questioned the Bureau's proposal to report the rate spread to three decimal places and stated that APR is typically disclosed to two decimal places. The Bureau acknowledges that the proposed instruction may pose some challenges for financial institutions. After considering the feedback, the Bureau has determined that the proposed instruction may be unduly burdensome on financial institutions. Consequently, the Bureau is not adopting the proposed instruction in the final rule. The Bureau proposed comment 4(a)(12)-4.iii to provide guidance on the rounding method for calculating the rate spread for a covered loan with a term to maturity that is not in whole years. The proposed comment specifically provided that when the actual loan term is exactly halfway between two whole years, the shorter loan term should be used. This proposed comment was based on guidance published in an FFIEC FAQ.\299\ One commenter pointed out that this rounding method does not follow the typical method of rounding up when a number is exactly halfway in between two others. This commenter suggested that unnecessary errors can occur as a result of this rounding method. The Bureau considered this feedback and believes that the benefit of adopting a rounding method inconsistent with the guidance published in the FFIEC FAQ for this specific calculation does not outweigh the burden because it would require a change in a financial institution's systems or processes for calculating the rate spread for the specific scenario that the proposed comment addresses. For example, financial institutions may have already instituted processes for rounding down when a loan term is exactly halfway between two years based on current FFIEC guidance. Accordingly, the Bureau is adopting comment 4(a)(12)- 4.iii as proposed. --------------------------------------------------------------------------- \299\ See http://www.ffiec.gov/hmda/faqreg.htm#rate. --------------------------------------------------------------------------- The Bureau proposed comment 4(a)(12)-5.i to illustrate the relevant date to use to determine the APOR if the interest rate in the transaction is set pursuant to a ``lock-in'' agreement between the financial institution and the borrower. The proposed comment also explained that the relevant date to use if no lock-in agreement is executed. Several industry commenters asked the Bureau to clarify the rate spread lock-in date where the transaction did not include an option to lock the loan's rate. The guidance provided in comment 4(a)(12)-5.i clarifies that, in a transaction where no lock-in agreement is executed, the relevant date to use to determine the applicable APOR is the date on which the financial institution sets the rate for the final time before closing. Except for technical amendments to comments 4(a)(12)-3, -4.i and .ii, and -5.iii, the Bureau is adopting the commentary to Sec. 1003.4(a)(12) substantially as proposed. In addition, the Bureau is adopting two comments that incorporate material contained in proposed appendix A into the commentary to Sec. 1003.4(a)(12). Comments 4(a)(12)-7 and -8 primarily incorporate proposed appendix A instructions and do not contain any substantive changes. The Bureau is making a technical change and incorporating the exclusion of assumptions from rate spread reporting in Sec. 1003.4(a)(12), which was included in proposed appendix A and was based on FFIEC guidance. The Bureau believes that the utility that the rate spread would provide on assumptions does not justify the burden in collecting the information. Therefore, the Bureau is adopting Sec. 1003.4(a)(12) generally to require financial institutions to report the difference between a loan's APR and APOR for a comparable transaction as of the date the interest rate is set, except for purchased loans, reverse mortgages, and loans that are not subject to Regulation Z, 12 CFR part 1026, with a modification that excludes assumptions from the scope of the rate spread reporting requirement. The Bureau believes that rate spread information on loans that are both below and above the threshold for higher-priced mortgage loans will reveal greater detail about the extent of the availability of prime lending in all communities. Pursuant to HMDA section 305(a), the Bureau is excluding purchased loans, reverse mortgages, assumptions, and loans that are not subject to Regulation Z, 12 CFR part 1026 from rate spread reporting to facilitate compliance and because information about the rate spread for such transactions could be potentially misleading. 4(a)(13) Regulation C Sec. 1003.4(a)(13) currently requires financial institutions to report whether a loan is subject to HOEPA, as implemented by Regulation Z Sec. 1026.32. Prior to the proposal, the Bureau received feedback suggesting that information regarding the reason for a loan's HOEPA status might improve the usefulness of the HMDA data. Pursuant to HMDA sections 305(a) and 304(b)(5)(D), the Bureau proposed to require financial institutions to report for covered loans subject to HOEPA, whether the covered loan is a high-cost mortgage under Regulation Z Sec. 1026.32(a), and the reason that the covered loan qualifies as a high-cost mortgage, if applicable. For the reasons discussed below, the Bureau is adopting Sec. 1003.4(a)(13) with modifications to remove the requirement to report information concerning the reasons for a loan's HOEPA status. The Bureau solicited feedback on the general utility of the modified data and on the costs associated with reporting the data. A few commenters stated that the expanded HOEPA flag would create an unnecessary burden. Several industry commenters suggested removing the HOEPA status field from HMDA reporting. They argued that the Bureau's 2013 ATR Final Rule eliminated the origination of HOEPA loans. One financial institution stated that the proposed HOEPA flag is either not applicable to it or would offer little benefit. Another commenter stated that the HOEPA status field is unnecessary because a user should be able to determine using the rate spread whether the loan's APR meets the HOEPA trigger. Another industry commenter stated that the proposal would require financial institutions to report points and fees, final rate, and origination charges as well as the rate spread. Data users could use these data points to determine whether a loan is higher-cost. [[Page 66200]] A few commenters supported the HOEPA flag but suggested that the Bureau should not collect the additional information regarding the reason(s) for whether the loan is subject to HOEPA. They pointed to the burden associated with reporting the information and the Bureau's proposal to collect other information about loan pricing, such as points and fees. An expanded HOEPA reporting requirement would have the potential to provide greater insight into which specific triggers are most prevalent among high-cost mortgages. However, the Bureau acknowledges the compliance burden associated with reporting information concerning the reasons for a loan's HOEPA status. As commenters pointed out, pricing information is available in other data fields, such as the rate spread, total points and fees, and interest rate. The benefits that would be provided by an expanded HOEPA reporting requirement does not justify the burden associated with reporting the information, particularly because other HMDA data fields capture pricing information that could be used to determine the reason for a loan's HOEPA status. In response to concerns raised by commenters regarding burden, the Bureau will only require financial institutions to report whether a loan is subject to HOEPA, as implemented by Regulation Z Sec. 1026.32. The Bureau believes that requiring financial institutions to report whether a loan is subject to HOEPA is necessary to carry out the purposes of HMDA because an indication of a loan's HOEPA status will help determine whether financial institutions are serving the housing needs of their communities. Accordingly, pursuant to HMDA sections 305(a) and 304(b)(5)(D), the Bureau is adopting Sec. 1003.4(a)(13) with modifications to remove the requirement to report information concerning the reasons for a loan's HOEPA status. In addition, the Bureau is adopting new comment 4(a)(13)-1 to clarify when a financial institution reports that the HOEPA status reporting requirement is not applicable. Comment 4(a)(13)-1 explains that a financial institution reports that the requirement to report the HOEPA status is not applicable if the covered loan is not subject to the Home Ownership and Equity Protection Act of 1994, as implemented in Regulation Z, 12 CFR 1026.32. Comment 4(a)(13)-1 also explains that, if an application did not result in an origination, a financial institution complies with Sec. 1003.4(a)(13) by reporting that the requirement is not applicable. 4(a)(14) Current Sec. 1003.4(a)(14) requires financial institutions to report the lien status of the loan or application (first lien, subordinate lien, or not secured by a lien on a dwelling). The technical instructions in current appendix A provide that, for loans that a financial institution originates and for applications that do not result in an origination, a financial institution shall report the lien status as one of the following: Secured by a first lien, secured by a subordinate lien, not secured by a lien, or not applicable (purchased loan). The Bureau proposed to modify Sec. 1003.4(a)(14) to require reporting of the priority of the lien against the subject property that secures or would secure the loan in order to conform to the MISMO industry data standard, which provides the following enumerations: First lien, second lien, third lien, fourth lien, or other. The proposal also removed the current exclusion of reporting lien status on purchased loans. The Bureau proposed technical modifications to the instruction in appendix A regarding how to enter lien status on the loan/application register. In addition, in order to provide clarity on proposed Sec. 1003.4(a)(14), the Bureau proposed technical modifications to comment 4(a)(14)-1 and proposed new comment 4(a)(14)-2. The Bureau solicited feedback regarding whether the Bureau should maintain the current reporting requirement (secured by a first lien or subordinate lien) modified to conform to the proposed removal of unsecured home improvement loans, or whether financial institutions prefer to report the actual priority of the lien against the property (secured by a first lien, second lien, third lien, fourth lien, or other). In response, a consumer advocate commenter supported the proposal to require reporting of the priority of the lien against the subject property and a few industry commenters stated that alignment with the MISMO industry data standard would help ensure consistency. However, most of the commenters that responded to this solicitation of feedback opposed the proposal to require reporting of the priority of the lien against the subject property and recommended that the Bureau continue to require reporting the lien status of the loan or application as either first lien or subordinate lien. In general, industry commenters stated that very few loans are secured by liens beyond a second lien and that as a result, the additional burden of reporting the actual lien priority would outweigh the potential utility of the data. For example, an industry commenter argued that a lien status beyond a second lien is rare and that reporting the actual lien status will not add much value to the HMDA data. A State trade association suggested that requiring financial institutions to specify the exact lien priority of the mortgage would result in little useful data and yet the burden would be excessive and unnecessary. In addition, with respect to potential privacy implications, a few commenters were concerned that if information regarding lien priority is made available to the public, such information could be coupled with other publicly available information on property sales and ownership records to compromise a borrower's privacy. The Bureau has considered this feedback. See part II.B above for a discussion of the Bureau's approach to protecting applicant and borrower privacy with respect to the public disclosure of HMDA data. While HMDA compliance and data submission can be made easier by aligning the requirements of Regulation C, to the extent practicable, to existing industry standards for collecting and transmitting mortgage data, the Bureau has determined that requiring reporting of the lien status of the loan or application as either first lien or subordinate lien is the appropriate approach. Based on the comments the Bureau received, it appears that the burdens associated with reporting the various enumerations (first lien, second lien, third lien, fourth lien, and other) may not outweigh the benefits discussed in the Bureau's proposal--namely, enhanced data collected under Regulation C and facilitating compliance by better aligning the data collected with industry practice. Accordingly, the Bureau does not adopt Sec. 1003.4(a)(14) as proposed but instead maintains the current reporting requirement (secured by a first lien or subordinate lien) modified to conform to the removal of non-dwelling-secured home improvement loans, and adopts corresponding modifications to the proposed commentary. The Bureau also solicited feedback on the general utility of lien status data on purchased loans and on the unique costs and burdens associated with collecting and reporting the data that financial institutions may face as a result of the proposal. A few industry commenters did not support the Bureau's proposal to remove the current exclusion of reporting lien status on purchased loans. For example, one [[Page 66201]] industry commenter suggested that such data is not an indicator of discriminatory lending and also that such information is better examined on a loan-by-loan basis by bank examiners. Another industry commenter did not support the proposed reporting requirement because it would be a regulatory burden with no particular benefit. While requiring financial institutions to report the lien status of purchased loans would add some burden on financial institutions, the Bureau has determined that such data will further enhance the utility of HMDA data overall. Given that loan terms, including loan pricing, vary based on lien status, and in light of the Bureau's determination to require reporting of certain pricing data for purchased loans, such as the interest rate, lender credits, total origination charges, and total discount points, the Bureau has determined that requiring financial institutions to report the lien status of purchased loans will improve the HMDA data's usefulness overall. In addition, as described in the Bureau's proposal, the liquidity provided by the secondary market is a critical component of the modern mortgage market, and information about the types of loans being purchased in a particular area, and the pricing terms associated with those purchased loans, is needed to understand whether the housing needs of communities are being fulfilled. Furthermore, local and State housing finance agency programs facilitate the mortgage market for low- to moderate- income borrowers, often by offering programs to purchase or insure loans originated by a private institution. Since the HMDA data reported by financial institutions does not include the lien status of purchased loans, it is difficult to determine the pricing characteristics of the private secondary market. Lien status information on purchased loans may help public entities, such as local and State housing finance agencies, understand how to complement the liquidity provided by the secondary market in certain communities, thereby maximizing the effectiveness of such public programs. Requiring that such data be reported may assist public officials in their determination of the distribution of public sector investments in a manner designed to improve the private investment environment. Additionally, providing lien status information to purchasers is standard industry practice. For these reasons, the Bureau has determined that data on the lien status of purchased loans will further the purposes of HMDA in determining whether financial institutions are serving the housing needs of their communities; in distributing public-sector investments so as to attract private investment to areas or communities where it is needed; and in identifying possible discriminatory lending patterns. Pursuant to the Bureau's authority under sections 305(a) and 304(b)(6)(J) of HMDA, the Bureau is adopting the modification to Sec. 1003.4(a)(14) to require reporting of lien status information--whether the covered loan is a first or subordinate lien--for purchased loans. Lastly, in order to facilitate HMDA compliance, the Bureau is modifying comment 4(a)(14)-1.i to clarify that financial institutions are required to report lien status for covered loans they originate and purchase and applications that do not result in originations, which include preapproval requests that are approved but not accepted, preapproval requests that are denied, applications that are approved but not accepted, denied, withdrawn, or closed for incompleteness. The Bureau is also adopting proposed comment 4(a)(14)-2, which directs financial institutions to comment 4(a)(9)-2 regarding transactions involving multiple properties with more than one property taken as security. 4(a)(15) Neither HMDA nor Regulation C historically has required reporting of information relating to an applicant's or borrower's credit score. Section 1094(3)(A)(iv) of the Dodd-Frank Act amended section 304(b) of HMDA to require financial institutions to report ``the credit score of mortgage applicants and mortgagors, in such form as the Bureau may prescribe.'' \300\ The Bureau proposed Sec. 1003.4(a)(15) to implement this requirement.\301\ Except for purchased covered loans, proposed Sec. 1003.4(a)(15)(i) requires financial institutions to report the credit score or scores relied on in making the credit decision and the name and version of the scoring model used to generate each credit score. In addition, the Dodd-Frank Act amendments to HMDA do not provide a definition of ``credit score.'' Therefore, the Bureau proposed in Sec. 1003.4(a)(15)(ii) to interpret ``credit score'' to have the same meaning as in section 609(f)(2)(A) of the Fair Credit Reporting Act (FCRA), 15 U.S.C. 1681g(f)(2)(A). --------------------------------------------------------------------------- \300\ 12 U.S.C. 2803(b)(6)(I). \301\ The Dodd-Frank amendments to HMDA added new provisions directing the Bureau to develop regulations that ``modify or require modification of itemized information, for the purpose of protecting the privacy interests of the mortgage applicants or mortgagors, that is or will be available to the public,'' and identified credit score as a new data point that may raise privacy concerns. HMDA sections 304(h)(1)(E) and (h)(3)(A)(i). See part II.B above for discussion of the Bureau's approach to protecting applicant and borrower privacy in light of the goals of HMDA. --------------------------------------------------------------------------- The Bureau also proposed instruction 4(a)(15)-1, which directed financial institutions to enter the credit scores relied on in making the credit decision and proposed instruction 4(a)(15)-2, which provided the codes that financial institutions would use for each credit score reported to indicate the name and version of the scoring model used to generate the credit score relied on in making the credit decision. In addition, the Bureau proposed four comments to provide clarification on the reporting requirement under proposed Sec. 1003.4(a)(15). The Bureau proposed comment 4(a)(15)-1, which explained that a financial institution relies on a credit score in making the credit decision if the credit score was a factor in the credit decision even if it was not a dispositive factor, and provided an illustrative example. Proposed comment 4(a)(15)-2 addressed circumstances where a financial institution obtains or creates multiple credit scores for a single applicant or borrower, as well as circumstances in which a financial institution relies on multiple scores for the applicant or borrower in making the credit decision, and provided illustrative examples. Proposed comment 4(a)(15)-3 addressed situations involving credit scores for multiple applicants or borrowers and provided illustrative examples. Finally, proposed comment 4(a)(15)-4 clarified that a financial institution complies with proposed Sec. 1003.4(a)(15) by reporting ``not applicable'' when a credit decision is not made, for example, if a file was closed for incompleteness or the application was withdrawn before a credit decision was made. Proposed comment 4(a)(15)- 4 also clarified that a financial institution complies with proposed Sec. 1003.4(a)(15) by reporting ``not applicable'' if it makes a credit decision without relying on a credit score for the applicant or borrower. In order to facilitate HMDA compliance and address concerns that it could be burdensome to identify credit score information for purchased covered loans, the Bureau excluded purchased covered loans from the requirements of proposed Sec. 1003.4(a)(15)(i). The Bureau solicited feedback on whether this exclusion was appropriate and received a few comments. A national trade association supported the Bureau's proposal to exclude purchased covered [[Page 66202]] loans from the proposed reporting requirement under Sec. 1003.4(a)(15)(i) without providing further explanation. One consumer advocate commenter did not oppose the proposal so long as the ULI is included in the final rule, because it can be used to link origination data to purchased loans. Similarly, another consumer advocate commenter recommended that until the ULI is successfully implemented, purchased loans should not be excluded from the credit score data reporting requirement. Finally, two other consumer advocate commenters argued that credit score should be reported for purchased loans. One of these commenters stated that the Bureau's proposed exclusion of purchased loans from Sec. 1003.4(a)(15)(i) will have the negative effect of not requiring financial institutions to report credit score information even when the applicant or borrower's credit score is in its possession or the institution could easily obtain it. The commenter suggested that any exception for purchased loans under proposed Sec. 1003.4(a)(15)(i) should be limited only to instances where the financial institution does not have and cannot reasonably obtain the credit score. The other commenter recommended that purchasers of covered loans should use the ULI to look up credit score information from the HMDA data associated with the loan's origination, or should request the information from the originator if the loan was not made by a financial institution required to report under HMDA. The Bureau has considered this feedback and has determined that it would be unduly burdensome for financial institutions that purchase loans to identify the credit score or scores relied on in making the underlying credit decision and the name and version of the scoring model used to generate each credit score. Consequently, the Bureau is adopting the exclusion of purchased covered loans proposed under Sec. 1003.4(a)(15)(i). The Bureau is also adopting new comment 4(a)(15)&6 which explains that a financial institution complies with Sec. 1003.4(a)(15) by reporting that the requirement is not applicable when the covered loan is a purchased covered loan. The Bureau solicited feedback on whether the Bureau should require any other related information to assist in interpreting credit score data, such as the date on which the credit score was created. In response, a few consumer advocate commenters specifically recommended that the Bureau require disclosure of the date on which the credit score was created. One commenter pointed out that this additional information will provide for richer data for purposes of statistical analysis. Other commenters stated that credit scores are essentially analyses of risk at a given point in time and thus the meaning of the score is relative to the date on which it was created, and that the date on which the credit score was created would allow the Bureau to ensure that financial institutions are treating borrowers equally when using credit score information. In contrast, a few industry commenters did not support requiring the date on which the credit score was created arguing that such additional data is not necessary. For example, one industry commenter stated that while credit scores can change, they usually do not significantly change in a short period of time. A national trade association stated that additional data related to credit score, such as the date, should not be required because it is superfluous information and would be burdensome to report for financial institutions. The Bureau has considered the feedback received and has determined that requiring financial institutions to report the date on which the credit score was created would not add sufficient value to the credit score information that will be required to be reported to warrant the additional burden placed on financial institutions. Accordingly, a financial institution will not be required to report the date on which the credit score was created under Sec. 1003.4(a)(15). In response to the Bureau's solicitation for feedback on whether it should require any other related information to assist in interpreting credit score data, a few consumer advocate commenters recommended that the Bureau also require financial institutions to report the name of the credit reporting agency that provided the underlying data to create the credit score (i.e., Equifax, Experian, or TransUnion). One commenter stated that in some cases, the proposed required disclosure of the ``name and version'' of the credit scoring model by a financial institution will indicate which credit reporting agency's data was used. For example, the disclosure will reveal not only that a ``FICO'' score was used, but that a ``Beacon'' score (the FICO 04 score based on Equifax data) was used. However, in other cases, such as VantageScore, the commenter stated that the name or the version of the credit scoring model will not indicate which credit reporting agency's data was used. In order to address the latter scenario, the commenter recommended that the Bureau require financial institutions to report the credit reporting agency whose data was used to generate the credit score that is reported. The Bureau has considered this feedback and has determined that it will not require financial institutions to report the name of the credit reporting agency that provided the underlying credit score data that institutions report under Sec. 1003.4(a)(15). Requiring that this additional information be reported would add burden on financial institutions, which the Bureau has determined is not justified by the value of the data. In response to the Bureau's general solicitation for feedback, several industry commenters recommended that the Bureau require financial institutions to report credit score as a ``range of values'' rather than an applicant's or borrower's actual credit score. The commenters suggested that reporting credit score as a range of values will eliminate a substantial number of potential errors on financial institutions' loan/application registers, would better protect the privacy of applicants, and would not compromise the integrity of the HMDA data. In contrast, one consumer advocate commenter argued that an applicant's or borrower's precise credit score is important because financial institutions may use different cutoff points in their underwriting processes which may not align with the provided ranges. The Bureau has considered this feedback and determined that requiring financial institutions to report credit score as a range of values would diminish the utility of the data to further HMDA's purposes. The Bureau has determined that requiring financial institutions to report the applicant's or borrower's actual credit score or scores relied on in making the credit decision is the appropriate approach and will assist in identifying whether financial institutions are serving the housing needs of their communities, identifying possible discriminatory lending patterns, and enforcing antidiscrimination statutes. The Bureau solicited feedback on whether the proposed codes that financial institutions would use for each credit score reported to indicate the name and version of the scoring model used to generate the credit score relied on in making the credit decision are appropriate for reporting credit score data, including using a free-form text field to indicate the name and version of the scoring model when the code for ``Other credit scoring model'' is reported by financial institutions. The Bureau also invited comment on any alternative [[Page 66203]] approaches that might be used for reporting this information. In response, a few commenters did not support the Bureau's proposed instruction 4(a)(15)-2.b, which instructs financial institutions to provide the name and version of the scoring model used in a free-form text field if the credit scoring model is one that is not listed. One commenter recommended that the Bureau not require a free-form text field for credit score because the data would be impossible to aggregate and would cause significant confusion. As an alternative, the commenter recommended that the Bureau maintain its proposal that financial institutions report the code for ``Other credit scoring model'' when appropriate but not require institutions to indicate the name and version of the scoring model in a free-form text field. Another industry commenter stated that free-form text fields are illogical because they lack the ability of being sorted and reported accurately. This commenter also opined that the additional staff and/or programming that will be needed on a government level to analyze these free text fields is costly and not justified when looking at the minimal impact these fields have on the overall data collection under HMDA. The Bureau has considered the concerns expressed by industry commenters with respect to the proposed requirement that a financial institution enter the name and version of the scoring model in a free- form text field when ``Other credit scoring model'' is reported but has determined that the utility of this data justifies the potential burden that may be imposed by the reporting requirement. As to the commenters' concern that credit scoring model data reported in the free-form text field would be impossible to aggregate due to the variety of potential names and versions of scoring model reported, the Bureau has determined that the data reported in the free-form text field will be useful even if the data cannot be aggregated. Lastly, with respect to the commenters' recommendation that requiring a financial institution to report the corresponding code for ``Other credit scoring model'' is sufficient and that the Bureau should not also require an institution to enter the name and version of the scoring model in a free-form text field in these circumstances, the Bureau has determined that such an approach would hinder the utility of the credit score data for purposes of HMDA. When a financial institution reports ``Other credit scoring model'' in the loan/ application register without further explanation as to what the other credit scoring model is, it would be difficult to perform accurate analyses of such data since different models are associated with different scoring ranges and some models may even have different ranges depending on the version used. Moreover, the free-form text field will provide key information on credit scoring models that are used by financial institutions to underwrite a loan but are not currently listed. For example, the data reported in the free-form text field for ``Other credit scoring model'' can be used to monitor those credit scoring models or to add commonly used, but previously unlisted, credit scoring models to the list. As such, the Bureau has determined that the HMDA data's usefulness will be improved by requiring financial institutions to report in a free-form text field the name and version of the scoring model when the institution reports ``Other credit scoring model'' on its loan/application register. The Bureau invited comment on whether it was appropriate to request the name and version of the scoring model under proposed Sec. 1003.4(a)(15)(i). For a variety of reasons, several industry commenters did not support the Bureau's proposal to include the name and version of the credit scoring model used to generate the credit score relied on in making the credit decision. In general, the commenters stated that while they support requiring financial institutions to report the credit score relied on in making the credit decision, reporting the name and version of the credit scoring model used to generate that score would impose significant regulatory and operational burden on industry. Commenters also stated that the Bureau had failed to provide compelling reasons for how the collection and reporting of this additional credit score data ensures fair access to credit in the residential mortgage market. In addition, commenters did not support the Bureau's proposal requiring financial institutions to report the credit scoring model used to generate the credit score on the grounds that the Dodd-Frank Act mandated that an applicant's or borrower's credit score be reported, but not additional data on the credit scoring model. In contrast, the vast majority of commenters supported the Bureau's proposal to require financial institutions to report not only the credit score or scores relied on in making the credit decision, but also the name and version of the scoring model used to generate each credit score. Several consumer advocate commenters pointed out that the name and version of the scoring model used to generate the credit score relied on in making the credit decision is needed to accurately interpret the credit score field. These commenters stated that requiring financial institutions to report this information is vital because each credit scoring model may generate different credit scores which may confound simple comparisons. Some industry commenters also supported the Bureau's proposal. One industry commenter stated that for purposes of fair lending analysis, credit score information is vital to understanding a financial institution's credit and pricing decision and that without such information, inaccurate conclusions may be reached by users of HMDA data. The Bureau has considered this feedback and determined that its proposal to require financial institutions to report the credit score or scores relied on in making the credit decision is the appropriate approach and is a reasonable interpretation of HMDA section 304(b)(6)(I). The Bureau has also determined that its interpretation of HMDA section 304(b)(6)(I) to require the name and version of the scoring model is reasonable because, as discussed above, this information is necessary to understand any credit scores that will be reported, as different models are associated with different scoring ranges and some models may even have different ranges depending on the version used.\302\ In addition, the Bureau's implementation is authorized by HMDA sections 305(a) and 304(b)(6)(J), and is necessary and proper to effectuate the purposes of HMDA, because, among other reasons, the name and version of the credit scoring model facilitates accurate analyses of whether financial institutions are serving the housing needs of their communities by providing adequate home financing to qualified applicants. Accordingly, the Bureau is adopting Sec. 1003.4(a)(15)(i) as proposed. --------------------------------------------------------------------------- \302\ For example, the range for VantageScore 3.0 scores is 300 to 850, but earlier VantageScore models have a range of 501 to 990. See VantageScore, How the Scores Range, http://your.vantagescore.com/interpret_scores. --------------------------------------------------------------------------- As discussed above, the Bureau proposed Sec. 1003.4(a)(15)(ii), which provides that ``credit score'' has the meaning set forth in 15 U.S.C. 1681g(f)(2)(A). The Bureau's proposal interpreted ``credit score'' to have the same meaning as in section 609(f)(2)(A) of the Fair Credit Reporting Act (FCRA), 15 U.S.C. 1681g(f)(2)(A). However, the Bureau solicited feedback on whether Regulation C should instead use a different definition of ``credit score.'' [[Page 66204]] For example, the Bureau suggested that it could define ``credit score'' based on the Regulation B definitions of ``credit scoring system'' or ``empirically derived, demonstrably and statistically sound, credit scoring system.'' \303\ Another alternative would be to interpret credit score to mean the probability of default, using a concept similar to the probability of default metric that the FDIC uses in determining assessment rates for large and highly complex insured depository institutions.\304\ --------------------------------------------------------------------------- \303\ According to Regulation B, a credit scoring system is ``a system that evaluates an applicant's creditworthiness mechanically, based on key attributes of the applicant and aspects of the transaction, and that determines, alone or in conjunction with an evaluation of additional information about the applicant, whether an applicant is deemed creditworthy.'' Regulation B Sec. 1002.2(p)(1). The four-part definition of an ``empirically derived, demonstrably and statistically sound, credit scoring system'' in Regulation B Sec. 1002.2(p)(1) establishes the criteria that a credit system must meet in order to use age as a predictive factor. Regulation B comment 2(p)-1. \304\ FDIC Assessments, Large Bank Pricing, 77 FR 66000 (Oct. 31, 2012). --------------------------------------------------------------------------- The commenters that provided feedback on the proposed definition of ``credit score'' supported the Bureau's proposal to use the FCRA section 609(f)(2)(A) definition of credit score. For example, one consumer advocate commenter stated that it supports the Bureau's proposal to use the definition of ``credit score'' set forth in the FCRA because the definition is familiar to industry, regulators, and other stakeholders. Similarly, another consumer advocate commenter stated that it supports the definition because it would facilitate compliance. The Bureau has considered this feedback and determined that the FCRA section 609(f)(2)(A) definition of ``credit score'' is the most appropriate because it provides a general purpose definition that is familiar to financial institutions that are already subject to FCRA and Regulation V requirements. Consequently, the Bureau is adopting Sec. 1003.4(a)(15)(ii) generally as proposed, but with technical modifications for clarity. Lastly, many commenters expressed concern about potential privacy implications if the Bureau collects credit score data or if it were to release credit score data to the public. As with other proposed data points like property value, commenters were concerned that if information regarding credit score data is made available to the public, such information could be coupled with other publicly available information, such as property sales and ownership records, in a way that compromises a borrower's privacy. A State trade association commented that public disclosure of credit score data creates the ability for unscrupulous third parties to specifically identify borrowers and directly market to those borrowers. The commenter suggested that these third parties would have access to a sufficient amount of information disclosed through HMDA and coupled with other information, such as public recordation information, to give the appearance through their marketing that they have some connection to the original lender. Similarly, an industry commenter suggested that in addition to the potential for criminal misuse of a borrower's financial information, the availability of the expanded data released under HMDA will very likely permit marketers to access the information which will result in aggressive marketing that is ``personalized'' to unsophisticated and vulnerable consumers for potentially harmful financial products and services. Another State trade association stated that credit score data should not be released to the public because collecting and releasing credit score data could lead to fraudsters, neighbors, marketers, and others learning very private pieces of information about the applicant or borrower. Another State trade association recommended that the Bureau strengthen its data protection as it relates to the selective disclosure of HMDA data to third parties and specifically recommended that the Bureau convert actual values to ranges or normalize values before sharing the data with a third party. The Bureau has considered this feedback. See part II.B above for a discussion of the Bureau's approach to protecting applicant and borrower privacy with respect to the public disclosure of HMDA data. As discussed above, the Bureau is adopting Sec. 1003.4(a)(15)(i) as proposed and Sec. 1003.4(a)(15)(ii) generally as proposed, but with technical modifications for clarity. The Bureau is adopting comments 4(a)(15)-1 and -2, as proposed. The Bureau is also adopting comment 4(a)(15)-3 as proposed with a clarification that in a transaction involving two or more applicants or borrowers for which the financial institution obtains or creates a single credit score, and relies on that credit score in making the credit decision for the transaction, the institution complies with Sec. 1003.4(a)(15) by reporting that credit score for either the applicant or first co-applicant. With regard to a financial institution reporting that the requirement is not applicable, the Bureau is modifying comment 4(a)(15)-4 by maintaining in that comment the guidance with respect to transactions for which no credit decision was made and moves the guidance with respect to transactions for which credit score was not relied on to new comment 4(a)(15)-5. The Bureau clarifies in comment 4(a)(15)-4 that if a file was closed for incompleteness or the application was withdrawn before a credit decision was made, the financial institution complies with Sec. 1003.4(a)(15) by reporting that the requirement is not applicable, even if the financial institution had obtained or created a credit score for the applicant or co-applicant. As discussed above, the Bureau is also adopting new comment 4(a)(15)-6, which clarifies that a financial institution complies with Sec. 1003.4(a)(15) by reporting that the requirement is not applicable when the covered loan is a purchased covered loan. The Bureau is also adopting new comment 4(a)(15)-7, which clarifies that when the applicant and co-applicant, if applicable, are not natural persons, a financial institution complies with Sec. 1003.4(a)(15) by reporting that the requirement is not applicable. 4(a)(16) Section 1003.4(c)(1) currently permits optional reporting of the reasons for denial of a loan application. However, certain financial institutions supervised by the OCC and the FDIC are required by those agencies to report denial reasons on their HMDA loan/application registers.\305\ The Bureau proposed Sec. 1003.4(a)(16), which requires mandatory reporting of denial reasons by all financial institutions. --------------------------------------------------------------------------- \305\ 12 CFR 27.3(a)(1)(i), 128.6, 390.147. --------------------------------------------------------------------------- The Bureau proposed instruction 4(a)(16) in appendix A, which modified the current instruction and provided technical instructions regarding how to enter the denial reason data on the loan/application register. First, proposed instruction 4(a)(16)-1 provided that a financial institution must indicate the principal reason(s) for denial, indicating up to three reasons. Second, the Bureau explained in proposed instruction 4(a)(16)-2 that, when a financial institution denies an application for a principal reason not included on the list of denial reasons in appendix A, the institution should enter the corresponding code for ``Other'' and also enter the principal denial reason(s) in a free-form text field. Third, the Bureau added a code for ``not applicable'' and explained in proposed instruction 4(a)(16)-3 that this code should be used by a financial institution if the action taken on the application was not a denial pursuant to Sec. 1003.4(a)(8), such as if the application [[Page 66205]] was withdrawn before a credit decision was made or the file was closed for incompleteness. Lastly, the Bureau also proposed to renumber current instruction I.F.2 of appendix A as proposed instruction 4(a)(16)-4, which explains how a financial institution that uses the model form for adverse action contained in appendix C to Regulation B (Form C-1, Sample Notice of Action Taken and Statement of Reasons) should report the denial reasons for purposes of HMDA, including entering the principal denial reason(s) in a free-form text field when the financial institution enters the corresponding code for ``Other.'' In addition, the Bureau proposed comment 4(a)(16)-1 to provide clarity as to what the Bureau requires with respect to a financial institution reporting the principal reason(s) for denial. The Bureau also proposed comment 4(a)(16)-2 to align with proposed instructions 4(a)(16)-2 and -4. A few industry commenters did not support the Bureau's proposal and recommended that reporting of denial reasons remain optional under Regulation C. The main reason offered by commenters was that a mandatory requirement to report denial reasons would increase regulatory burden on financial institutions. In contrast, most consumer advocate commenters supported the Bureau's proposed Sec. 1003.4(a)(16). For example, several consumer advocate commenters pointed out that different types of housing counseling and intervention is needed depending on the most frequent reasons for denial. These commenters stated that denial reason data is important to housing counseling agencies because it helps identify the most significant impediments to homeownership and provide more effective counseling. A government commenter noted that denial reasons will be particularly effective for fair lending analyses. Another consumer advocate commenter pointed out that denial reason data will be helpful for understanding why a particular loan application was denied and identifying potential barriers in access to credit. The Bureau has determined that maintaining the current requirement of optional reporting of denial reasons is not the appropriate approach given the value of the data in furthering HMDA's purposes. The reasons an application is denied are critical to understanding a financial institution's credit decision and to screen for potential violations of antidiscrimination laws, such as ECOA and the Fair Housing Act.\306\ Denial reasons are important for a variety of purposes including, for example, assisting examiners in their reviews of denial disparities and underwriting exceptions. The Bureau has determined that requiring the collection of the reasons for denial will facilitate more efficient, and less burdensome, fair lending examinations by the Bureau and other financial regulatory agencies, thereby furthering HMDA's purpose of assisting in identifying possible discriminatory lending patterns and enforcing antidiscrimination statutes. --------------------------------------------------------------------------- \306\ 15 U.S.C. 1691 et seq.; 42 U.S.C. 3601 et seq. ECOA and Regulation B require all financial institutions to provide applicants the reasons for denial, or a notice of their right to receive those denial reasons, and to maintain records of compliance. See Regulation B Sec. Sec. 1002.9 and 1002.12, 15 U.S.C. 1691(d). --------------------------------------------------------------------------- The Bureau acknowledges that mandatory reporting of denial reasons will contribute to certain financial institutions' compliance burden. However, the statistical value of optionally reported data is lessened because of the lack of standardization across all HMDA reporters. Moreover, as discussed above, certain financial institutions supervised by the OCC and the FDIC are already required by those agencies to report denial reasons.\307\ A requirement that all financial institutions report reasons for denial of an application is the proper approach for purposes of HMDA. For these reasons, pursuant to its authority under HMDA sections 305(a) and 304(b)(6)(J), the Bureau is finalizing the requirement that all financial institutions report reasons for denial of an application. This information is necessary to carry out HMDA's purposes, because it will provide more consistent and meaningful data, which will assist in identifying whether financial institutions are serving the housing needs of their communities, as well as assist in identifying possible discriminatory lending patterns and enforcing antidiscrimination statutes. --------------------------------------------------------------------------- \307\ See supra note 306. --------------------------------------------------------------------------- The Bureau solicited feedback on the proposed requirement that a financial institution enter the principal denial reason(s) in a free- form text field when ``Other'' is entered in the loan/application register. Several commenters did not support the proposed requirement for a variety of reasons, including, for example, concerns about having sufficient space to accurately or adequately capture the denial reason with the limited space available for reporting on the loan/application register, concerns that denial reason data reported in the free-form text field would be impossible to aggregate due to the variety of potential denial reasons reported, and concerns that such reporting would cause significant confusion and regulatory burden. A few industry commenters suggested that requiring a financial institution to report the corresponding code for ``Other'' would be sufficient when the institution denies an application for a principal reason not included on the list of denial reasons in appendix A or on the model form for adverse action contained in appendix C to Regulation B. The commenters suggested that the Bureau should not also require an institution to enter the principal denial reason(s) in a free-form text field in these circumstances for the reasons listed above. The Bureau has considered the concerns expressed by industry commenters with respect to the proposed requirement that a financial institution enter the principal denial reason(s) in a free-form text field when a financial institution reports the denial reason as ``Other'' in the loan/application register but has determined that the utility of this data justifies the potential burden that may be imposed by the reporting requirement. In addition, with respect to the concern that financial institutions will not have sufficient space in the loan/ application register to accurately or adequately capture the denial reasons, the Bureau believes that the free-form text field will provide institutions with sufficient space to comply with proposed Sec. 1003.4(a)(16). As explained in proposed comment 4(a)(16)-1, the denial reasons reported by a financial institution must be specific and accurately describe the principal reason or reasons an institution denied the application. The free-form text field will not limit a financial institution's ability to comply with proposed Sec. 1003.4(a)(16). As to the commenters' concern that denial reason data reported in the free-form text field would be impossible to aggregate due to the variety of potential denial reasons reported, the Bureau has determined that the data reported in the free-form text field will be useful even if the data cannot be aggregated. The Bureau also proposed comment 4(a)(16)-2, which provides clarification as to the proposed requirement that a financial institution enter the principal denial reason(s) in a free-form text field when ``Other'' is entered in the loan/application register. The Bureau is finalizing this comment, modified for additional clarity, to address any potential confusion. Lastly, with respect to the commenters' recommendation that it be sufficient to require a financial institution to report ``Other'' as the denial reason and that the Bureau [[Page 66206]] should not also require an institution to enter the principal denial reason(s) in a free-form text field in these circumstances, the Bureau has determined that such an approach would hinder the utility of the denial reason data for purposes of HMDA. Many consumer advocate commenters pointed out that transparency about denial reasons provides the public as well as regulators with the information needed to better understand challenges to access to credit. One commenter specifically pointed out the reporting accuracy of denial reasons will be improved in two ways if financial institutions are required to explain the denial reason in the free-form text field when the institution indicates ``Other'' as a reason for denial. First, the commenter suggested that this reporting requirement will prevent the misuse of the ``Other'' category when financial institutions report the denial reason as ``Other'' when in fact the denial reason may more appropriately fall into one or more of the listed denial reasons. Without further explanation as to what the ``Other'' denial reason actually is, the commenter stated that it has been impossible to tell if the financial institution accurately reported the denial reason. Second, the commenter stated that the free-form text field will provide key information on denial reasons that are not currently listed. For example, the denial reason data can be used to monitor other denial reasons or to add common, but previously unlisted, denial reasons to the list. The Bureau has determined that the HMDA data's usefulness will be improved by requiring financial institutions to report the principal reason(s) it denied the application in a free-form text field when the institution reports the denial reason as ``Other'' in the loan/application register. The Bureau solicited feedback regarding whether additional clarifications would assist financial institutions in complying with the proposed requirement. A few industry commenters pointed out that while the proposal requires a financial institution to report up to three principal reasons for denial, the commenters read Regulation B as providing that a creditor may provide up to four principal reasons for denial and such inconsistency between regulations adds to the compliance burden imposed by the Bureau's new mandatory reporting requirement under proposed Sec. 1003.4(a)(16). The adverse action notification provisions of Regulation B do not mandate that a specific number of reasons be disclosed when a creditor denies an application but instead provides that disclosure of more than four reasons is not likely to be helpful to the applicant.\308\ In light of the feedback on the proposal and in an effort to help facilitate compliance and consistency between regulations, the Bureau is modifying proposed comment 4(a)(16)-1 to provide that a financial institution complies with Sec. 1003.4(a)(16) by reporting the principal reason or reasons it denied the application, indicating up to four reasons. --------------------------------------------------------------------------- \308\ See Regulation B Sec. 1002.9, Supp. I., Sec. 1002.9, comment 9(b)(2)-1. The Bureau noted in its proposal that ECOA and Regulation B require creditors to provide applicants the reasons for denial, or a notice of their right to receive those denial reasons, and to maintain records of compliance. See 79 FR 51731, 51775 (Aug. 29, 2014), note 381. See also 15 U.S.C. 1691(d), Regulation B Sec. Sec. 1002.9 and 1002.12. --------------------------------------------------------------------------- In order to help facilitate compliance with proposed Sec. 1003.4(a)(16), the Bureau also adopts two new comments. The Bureau is adopting new comment 4(a)(16)-2, which clarifies that a request for a preapproval under a preapproval program as defined by Sec. 1003.2(b)(2) is an application and therefore, if a financial institution denies a preapproval request, the financial institution complies with Sec. 1003.4(a)(16) by reporting the reason or reasons it denied the preapproval request. The Bureau also adopts new comment 4(a)(16)-4, which clarifies that a financial institution complies with Sec. 1003.4(a)(16) by reporting that the requirement is not applicable if the action taken on the application, pursuant to Sec. 1003.4(a)(8), is not a denial. For example, a financial institution complies with Sec. 1003.4(a)(16) by reporting that the requirement is not applicable if the loan is originated or purchased by the financial institution, or the application or preapproval request was approved but not accepted, or the application was withdrawn before a credit decision was made, or the file was closed for incompleteness. Several commenters were also concerned that if information regarding denial reasons were made available to the public, such information could be coupled with other publicly available information, which would result in not only compromising a borrower's privacy but also potentially place consumers at greater risk of financial harm through unlawful marketing to consumers by unscrupulous parties, such as identify thieves, other scammers, or criminals. For example, one industry commenter suggested that ``unsophisticated consumers could be vulnerable to aggressive marketing techniques, which may appear even more `personalized' to their situation because of the availability of their specific financial picture through the LAR data.'' The Bureau has considered this feedback. See part II.B above for a discussion of the Bureau's approach to protecting applicant and borrower privacy with respect to the public disclosure of HMDA data. The Bureau is adopting Sec. 1003.4(a)(16) as proposed, with minor technical modifications. The Bureau is adopting proposed comments 4(a)(16)-1 and 4(a)(16)-2, with several technical and clarifying modifications, and renumbers proposed comment 4(a)(16)-2 as 4(a)(16)-3. In addition, as discussed above, the Bureau is adopting new comments 4(a)(16)-2 and -4, which will help facilitate HMDA compliance by providing additional guidance regarding the denial reason reporting requirement. 4(a)(17) Section 304(b)(5)(A) of HMDA \309\ provides for reporting of ``the total points and fees payable at origination in connection with the mortgage as determined by the Bureau, taking into account 15 U.S.C. 1602(aa)(4).'' \310\ The Bureau proposed to implement this provision through proposed Sec. 1003.4(a)(17), which required financial institutions to report the total points and fees charged in connection with certain mortgage loans or applications. Proposed Sec. 1003.4(a)(17) defined total points and fees by reference to TILA, as implemented by Regulation Z Sec. 1026.32(b)(1) or (2). Section 1026.32(b)(1) defines ``points and fees'' for closed-end credit transactions, while Sec. 1026.32(b)(2) defines ``points and fees'' for open-end credit transactions. Proposed Sec. 1003.4(a)(17) would have applied to applications for and originations of certain closed-end mortgage loans and open-end lines of credit, but not to reverse mortgages or commercial-purpose loans or lines of credit. --------------------------------------------------------------------------- \309\ Section 1094(3)(A)(iv) of the Dodd-Frank Act amended section 304(b) of HMDA to provide for the reporting of total points and fees. \310\ 15 U.S.C. 1602(aa)(4) is part of TILA. Prior to amendments made by the Dodd-Frank Act, that section generally defined ``points and fees'' for the purpose of determining whether a transaction was a high-cost mortgage. See 15 U.S.C. 1602(aa)(4). Section 1100A of the Dodd-Frank Act redesignated subsection 1602(aa)(4) as subsection 1602(bb)(4), where it is currently codified. In light of that redesignation, the Bureau interprets HMDA section 304(b)(5)(A) as directing it to take into account 15 U.S.C. 1602(bb)(4) and its implementing regulations, as those provisions address ``points and fees'' and because current subsection 1602(aa)(4) is no longer relevant to a determination regarding points and fees. --------------------------------------------------------------------------- The Bureau also solicited comment on the costs and benefits of the proposed [[Page 66207]] definition of total points and fees and on the specific charges that should be included or excluded. Additionally, in discussing proposed Sec. 1003.4(a)(18), the Bureau sought feedback on the merits of a more inclusive measure of the cost of a loan. For the reasons provided below, the Bureau is requiring financial institutions to report the total loan costs for any covered loan that is both subject to the ability-to-repay section of the Bureau's 2013 ATR Final Rule and for which a Closing Disclosure is required under the Bureau's 2013 TILA-RESPA Final Rule. Total loan costs are disclosed pursuant to Regulation Z Sec. 1026.38(f)(4). For a covered loan that is subject to the ability-to-repay section of the Bureau's 2013 ATR Final Rule but for which a Closing Disclosure is not required under the Bureau's 2013 TILA-RESPA Final Rule, financial institutions must report the total points and fees, unless the covered loan is a purchased covered loan. This reporting requirement does not apply to applications or to covered loans not subject to the ability-to-repay requirements in the 2013 ATR Final Rule, such as open-end lines of credit, reverse mortgages, or loans or lines of credit made primarily for business or commercial purposes. Commenters were divided on whether financial institutions should be required to report points-and-fees data. Most consumer advocates generally supported the proposed pricing data points, including total points and fees. These commenters explained that more detailed pricing information will improve their ability to identify potential price discrimination and to understand the terms on which consumers in their communities are being offered credit. One consumer advocate stated that certain groups, such as women, minorities, and borrowers of manufactured housing loans may be unfairly charged higher amounts of points and fees than other borrowers. This commenter also stated that the total amount of points and fees was important for determining a loan's status under HOEPA and the ability-to-repay and qualified mortgage requirements of Regulation Z, and that data about points and fees would clarify any need for further regulation. Industry commenters, on the other hand, generally opposed collection of points-and-fees data. Many commenters stated that reporting the data would be unduly burdensome because of uncertainty regarding the definition of points and fees or because the total is not required to be calculated by other regulations. Other commenters believed that points-and-fees data would mislead users or duplicate data reported pursuant to other provisions of the proposal. Finally, a few commenters claimed that the data would not be valuable for HMDA purposes. Specifically, several industry commenters stated that variance among the fees and charges included in points and fees may result in unclear data. One commenter noted that the points-and-fees calculation adjusts based on factors unrelated to the total loan cost, such as whether a particular charge was paid to an affiliate of the creditor. Similarly, other industry commenters stated that the total amount of points and fees was subject to factors that would prevent effective comparison among borrowers, such as daily market fluctuations, differences in location, and borrower decisionmaking. The Bureau believes that total points-and-fees data, as defined in proposed Sec. 1003.4(a)(17), would have some value in helping HMDA data users to understand certain fees and charges imposed on borrowers. However, after considering the comments, the Bureau concludes that other measures of loan cost, such as total loan costs, as defined in final Sec. 1003.4(a)(17)(i), will be more valuable and nuanced than points and fees, as defined in proposed Sec. 1003.4(a)(17), and will better capture the type of information that HMDA section 304(b)(5)(A) is intended to cover. Total loan costs are the total upfront costs involved in obtaining a mortgage loan. Specifically, for covered loans subject to the disclosure requirements of Regulation Z Sec. 1026.19(f), total loan costs are the sum of the amounts disclosed as borrower-paid at or before closing found on Line D of the Closing Cost Details page of the current Closing Disclosure, as provided for in Regulation Z Sec. 1026.38(f)(4). Final Sec. 1002.4(a)(17)(i) requires financial institutions to report total loan costs because they are a more comprehensive measure than total points and fees, as defined in proposed Sec. 1003.4(a)(17), and because they better facilitate comparisons among borrowers. Total loan costs include all amounts paid by the consumer to the creditor and loan originator for originating and extending credit, all points paid to reduce the interest rate, all amounts paid for third- party settlement services for which the consumer cannot shop, and all amounts paid for third-party settlement services for which the consumer can shop. However, total loan costs omits other closing costs, such as amounts paid to State and local governments for taxes and government fees, prepaids such as homeowner's insurance premiums, initial escrow payments at closing, and other services that are required or obtained in the real estate closing by the consumer, the seller, or another party. In other words, this total generally represents the costs that the financial institution imposes in connection with the mortgage loan, and omits costs controlled by other entities, such as government jurisdictions. Unlike total points and fees as defined in proposed Sec. 1003.4(a)(17), total loan costs may be more easily compared across borrowers because third-party charges are not included or excluded depending on various factors, such as whether they were paid to an affiliate of the creditor. This consistency enables users to better compare loan costs among borrowers and to understand the total upfront costs that borrowers face when obtaining mortgage loans. The amount of total loan costs may also be analyzed in combination with the other pricing data points more readily than the total points and fees. For example, the difference between the total loan costs and total origination charges provides the total amount the borrower paid for third-party services.\311\ Because of the improved utility of total loan costs, for covered loans subject to final Sec. 1003.4(a)(17) for which total loan costs are available, the final rule requires financial institutions to report total loan costs. --------------------------------------------------------------------------- \311\ Some costs, such as certain upfront mortgage insurance premiums, would not be included. --------------------------------------------------------------------------- The Bureau acknowledges that total loan costs do not include all closing costs. For example, total loan costs omit amounts paid to State and local governments for taxes and government fees, prepaids such as homeowner's insurance premiums, initial escrow payments at closing, and other services that are required or obtained in the real estate closing by the consumer, the seller, or another party. Many excluded closing costs, however, are unrelated to the cost of extending credit by the financial institution. Because HMDA focuses on the lending activity of financial institutions, the Bureau has determined that the exclusion of these costs is proper. Total loan costs, as provided for in the final rule, also exclude upfront charges paid by sellers or other third parties if these parties were legally obligated to pay for such costs.\312\ This omission would understate the total loan costs charged by a financial institution for covered loans with seller-paid or other- paid closing costs in certain situations. However, including such costs would require financial institutions to perform [[Page 66208]] a calculation that they are not otherwise performing for purposes of the Closing Disclosure. The Bureau has determined that avoiding requiring such calculations by relying on the description of total loan costs found in Regulation Z reduces burden and facilitates compliance. --------------------------------------------------------------------------- \312\ See 12 CFR 1026.19(f)(1)(i). --------------------------------------------------------------------------- Total loan costs are not currently required to be calculated for certain loans. The Bureau's 2013 TILA-RESPA Final Rule exempted certain loans from the requirement to provide a Closing Disclosure. For example, manufactured housing loans secured by personal property are exempt from the requirements of the Bureau's 2013 TILA-RESPA Final Rule. But such loans are subject to the ability-to-repay provision of the Bureau's 2013 ATR Final Rule. For these loans, final Sec. 1003.4(a)(17) requires financial institutions to report the total points and fees, calculated pursuant to Regulation Z Sec. 1026.32(b)(1). Although total points and fees as defined in final Sec. 1003.4(a)(17)(ii) are a less comprehensive and less comparable measure of cost than total loan costs, requiring financial institutions to calculate the total loan costs for loans outside of the scope of the 2013 TILA-RESPA Final Rule would be overly burdensome because financial institutions would have no regulatory definition or experience on which to rely. Moreover, the Bureau believes that total points and fees as defined in final Sec. 1003.4(a)(17)(ii) will provide valuable information about the upfront cost of a loan that would otherwise be lacking from the data. Total points and fees as defined in final Sec. 1003.4(a)(17)(ii) include many of the same charges that comprise total loan costs, albeit in a less consistent fashion. Moreover, in some cases loans not subject to the Closing Disclosure requirement may be made to vulnerable consumers. For example, the Bureau's research suggests that manufactured-housing borrowers of chattel loans are more likely to be older, to have lower incomes, and to pay higher prices for their loans.\313\ Without points-and-fees data, users would have no insight into the upfront costs associated with such loans. --------------------------------------------------------------------------- \313\ See Bureau of Consumer Fin. Prot., Manufactured-Housing Consumer Finance in the United States at 5-6 (2014), available at http://files.consumerfinance.gov/f/201409_cfpb_report_manufactured-housing.pdf. --------------------------------------------------------------------------- Regarding the commenters' concerns about misleading data, the final rule includes a number of factors that will help users put the data in their proper context. Regarding total loan costs and total points and fees, many of the factors identified by commenters are reflected in the final rule, such as location and product type.\314\ More importantly, however, the HMDA data need not reflect all conceivable determinants of loan pricing to be beneficial to users. The final rule's pricing data will provide important benefits that would be lost if the Bureau were to eliminate it entirely. For example, regulators are able to use pricing data to efficiently prioritize fair lending examinations. Prioritizing examinations based on insufficient data would result in some financial institutions facing unnecessary examination burden while others whose practices warrant closer review would not receive sufficient scrutiny. Overall, the pricing data included in the final rule represent a marked improvement over the current regulation. --------------------------------------------------------------------------- \314\ See 12 CFR 1003.4(a)(2) (loan type); id. at 1003.4(a)(9) (location). --------------------------------------------------------------------------- One trade association stated that points-and-fees data would lead to reduced price competition. However, the Bureau believes, consistent with standard economic theory, that increased transparency regarding price generally increases competition and ultimately benefits consumers. Therefore, the Bureau is not persuaded that the commenter's price competition concern is a basis for not capturing information regarding total loan costs and points and fees, as defined in Sec. 1003.4(a)(17). A more detailed discussion of the benefits, costs, and impacts can be found in the section 1022 discussion below. Other industry commenters expressed concern over the burden associated with proposed Sec. 1003.4(a)(17). For example, several industry commenters pointed out that although financial institutions face limits on points and fees if they wish to avoid coverage under the 2013 HOEPA Final Rule, and if they wish to make a qualified mortgage under the 2013 ATR Final Rule, neither rule expressly requires financial institutions to calculate that total. One industry commenter explained that the total amount of points and fees was not currently recorded electronically. Many industry commenters cited concerns over the uncertainty or complexity of the definition of points and fees. Similarly, some commenters requested guidance on what charges to include within the total points and fees or called on the Bureau to supply a ``standard'' definition of the term. Some industry commenters believed that the reporting the total points and fees would expose them to citations under Regulation C for small errors. In comparison to the proposed rule, final Sec. 1003.4(a)(17) substantially reduces burden while still ensuring that valuable data are reported. Commenters generally stated that the calculation of total points and fees was not completed for all loans subject to HOEPA or the Bureau's 2013 ATR Final Rule, and that, if the calculation was completed, it involved substantial uncertainty and complexity. For the vast majority of covered loans subject to final Sec. 1003.4(a)(17), financial institutions will report the total loan costs. These institutions would have already calculated the total loan costs in order to disclose the total to borrowers pursuant to the 2013 TILA- RESPA Final Rule. Therefore, the burden of reporting for Sec. 1003.4(a)(17) is generally limited to loans for which financial institutions would already have to calculate the total loan costs. Using the same definition across regulations was supported by several commenters with respect to total points and fees, and final Sec. 1003.4(a)(17) does so by using the existing definition of total loan costs found in Regulation Z. For the narrow class of loans subject to the ability-to-repay provision of the Bureau's 2013 ATR Final Rule but which are exempt from the 2013 TILA-RESPA Final Rule, financial institutions must report the total points and fees as defined in final Sec. 1003.4(a)(17)(ii). These loans are generally manufactured housing loans secured by personal property. Because such loans run a greater risk of crossing the high-cost mortgage thresholds than site-built home loans, the Bureau believes that most financial institutions would calculate the total points and fees for these loans for compliance with HOEPA and other laws.\315\ Additionally, the final rule does not increase burden on these same institutions because it uses the existing definition of ``total points and fees'' found in Regulation Z. --------------------------------------------------------------------------- \315\ See Bureau of Consumer Fin. Prot., Manufactured-Housing Consumer Finance in the United States at 32-37 (2014), available at http://files.consumerfinance.gov/f/201409_cfpb_report_manufactured-housing.pdf (comparing the pricing of manufactured home loans and site-built home loans). --------------------------------------------------------------------------- The final rule also avoids increased burden by limiting Sec. 1003.4(a)(17) to covered loans that are subject to the ability-to-repay provision of the 2013 ATR Final Rule, rather than loans subject to either the 2013 ATR Final Rule or HOEPA. The primary effect of this change from the proposal is to exclude open-end lines of credit from the scope of the reporting requirement. The Bureau believes that such loans typically have lower upfront charges than comparable closed-end loans. Additionally, many open-end lines of [[Page 66209]] credit feature bona fide third-party charges that are waived on the condition that the consumer not terminate the line of credit sooner than 36 months after account opening, which are excluded from the total points and fees.\316\ At the same time, such loans are less likely to trigger high-cost mortgage status, which makes financial institutions less likely to complete the points-and-fees calculation for such loans. Therefore, the Bureau believes that on balance, Sec. 1003.4(a)(17) should be limited to covered loans that are subject to the ability-to- repay provision of the 2013 ATR Final Rule. --------------------------------------------------------------------------- \316\ See 12 CFR 1026.32(b)(6)(ii). --------------------------------------------------------------------------- Final Sec. 1003.4(a)(17) will provide a more consistent measure of upfront loan costs than total points and fees as defined in proposed Sec. 1003.4(a)(17). Total loan costs, combined with total origination charges, discount points, and lender credits, will also enable a more detailed understanding of the upfront costs that borrowers pay for their loans. Accordingly, these data will provide significant utility for fair lending analysis and for understanding the terms of credit being offered. With respect to loans made to lower-income consumers, such as some borrowers in manufactured housing communities, final Sec. 1003.4(a)(17) provides information about upfront loan costs by adopting reporting of points and fees. Finally, by substituting total loan costs for most loans and limiting the reporting of points and fees as described above, final Sec. 1003.4(a)(17) represents a substantial decrease in burden from the proposed rule. Therefore, the Bureau is adopting final Sec. 1003.4(a)(17), which requires financial institutions to report, for covered loans subject to the ability-to- repay provision of the 2013 ATR Final Rule, the total loan costs if the loan is subject to the disclosure requirements in Sec. 1026.19(f), or the total points and fees if the loan is not subject to the disclosure requirements in Sec. 1026.19(f) and is not a purchased covered loan. The Bureau believes that final Sec. 1003.4(a)(17) also addresses many of the specific issues or questions that commenters raised regarding the proposed points-and-fees data point. For example, several commenters asked the Bureau for clarification or modification of the scope of the reporting requirement. Two industry commenters asked the Bureau to exclude commercial loans from the scope of proposed Sec. 1003.4(a)(17), or to confirm that commercial loans are excluded. The final rule limits Sec. 1003.4(a)(17) to covered loans subject to Regulation Z Sec. 1026.43(c), which is inapplicable to commercial loans. Therefore, financial institutions are not required to report the total loan costs or the total points and fees for commercial-purpose transactions. The Bureau is adopting final comment 4(a)(17)(i)-1 to clarify that the total loan costs reporting requirement is not applicable to covered loans not subject to Regulation Z Sec. 1026.19(f), and final comment 4(a)(17)(ii)-1 to clarify that the reporting requirement is not applicable to covered loans not subject to Regulation Z Sec. 1026.43(c). One industry commenter recommended that no points and fees be required to be reported for applications that are not approved. This commenter also recommended that, for applications that have been approved by the financial institution but not accepted by the consumer, the total points and fees should be considered accurate if the amount is no less than the amount on which the financial institution relied. Regarding total loan costs, the Closing Disclosure required by Regulation Z Sec. 1026.19(f) is generally not provided for applications that do not result in a closed loan. Regarding total points and fees, elements of points and fees have the highest degree of uncertainty during the application stage, which limits their utility but increases the reporting burden. Therefore, final Sec. 1003.4(a)(17) excludes applications from the scope of the reporting requirement. Final comments 4(a)(17)(i)-1 and 4(a)(17)(ii)-1 explain that applications are not subject to the requirement to report either total loan costs or total points and fees. A few industry commenters suggested that proposed Sec. 1003.4(a)(17) be limited to HOEPA loans and qualified mortgages because the total points and fees would be most readily available for those loans. However, another industry commenter stated that the total points and fees were more likely to be available for loans that exceeded the qualified-mortgage thresholds. Finally, one industry commenter urged the Bureau to restrict the scope of proposed Sec. 1003.4(a)(17) to loans secured by principal dwellings to better fulfill the purposes of HMDA. These comments are largely addressed by the changes the Bureau has made in the final rule. The vast majority of covered loans subject to the requirement in Sec. 1003.4(a)(17) are governed by the scope of Regulation Z Sec. 1026.19(f). For these loans, final Sec. 1003.4(a)(17) requires no calculations that would not otherwise be performed for purposes of the Closing Disclosure. Accordingly, there is no reason to exclude a particular subset of covered loans for which the total loan costs are reported. For the narrow remainder of manufactured home loans for which total points and fees are reported, the risk to consumers warrants maintaining coverage of these loans, and points and fees are a less burdensome requirement than applying regulatory definitions that would not otherwise apply to these loans. Finally, the final rule does not exclude loans secured by secondary dwellings from Sec. 1003.4(a)(17) because HMDA's coverage is not limited to loans secured by the borrower's primary residence and includes loans secured by second homes as well as non-owner-occupied properties. Pricing data about such dwelling-secured homes will provide information necessary to better understand potentially speculative purchases of housing units similar to those that contributed to the recent financial crisis. One industry commenter recommended that the Bureau exclude community banks from the points-and-fees reporting requirement because the calculation is burdensome and may not be completed in all cases, and because community banks avoided the irresponsible lending practices that contributed to the financial crisis.\317\ Another industry commenter suggested that the Bureau require financial institutions to report either the loan's annual percentage rate or the finance charge instead of the total points and fees. This commenter stated that total points and fees require a manual calculation. As explained above, final Sec. 1003.4(a)(17) generally does not require financial institutions to calculate an amount that would not otherwise be calculated for other regulatory requirements or purposes. The Bureau acknowledges that a financial institution may have to report points and fees for a limited set of loans for which the institution does not otherwise calculate the total points and fees, such as for manufactured housing loans secured by personal property. However, as discussed above, the Bureau believes that the burden of performing such a calculation is justified by the benefit of having some measure of fees charged to borrowers. Moreover, the APR and finance charge combine both interest and fees and do not allow users to identify the amount of fees imposed on a borrower in connection with a transaction. Therefore, the final rule does not adopt [[Page 66210]] the changes recommended by these commenters. --------------------------------------------------------------------------- \317\ The Bureau notes that many community banks will be excluded from HMDA reporting altogether under the revised loan- volume threshold. --------------------------------------------------------------------------- Several industry commenters supported the exclusion for purchased covered loans found in proposed Sec. 1003.4(a)(17). In fact, one industry commenter recommended excluding all data points, including pricing data, from purchased covered loans. This commenter explained that the ULI would enable tracking of purchased covered loans and believed that the exclusion of the government-sponsored enterprises, which purchase most of the covered loans, would distort the data. Conversely, a consumer advocate recommended that the Bureau require reporting of data for purchased covered loans unless the purchasing entity is unable to reasonably obtain the relevant information from the original financial institution. This commenter noted that a blanket exception for purchased covered loans would create gaps in the HMDA data, especially if the original financial institution was not subject to HMDA. The Bureau proposed to exclude purchased loans from Sec. 1003.4(a)(17) because the total points and fees are not readily available from the information obtained from the selling entity. Therefore, purchasing entities would be required to calculate the total points and fees, and might lack the information necessary to do so. If the purchasing financial institution required the selling entity to calculate the total points and fees, and the seller was not a HMDA reporter, then the seller would face a difficult and uncertain calculation without the benefit of having to otherwise report the data under HMDA. For these reasons, the Bureau adopts this exclusion with respect to total points and fees, as required by final Sec. 1003.4(a)(17)(ii). However, the same reasoning does not support providing a similar exclusion from purchased loans with respect to total loan costs, as required by final Sec. 1003.4(a)(17)(i). Unlike total points and fees, the total loan costs are calculated for all covered loans subject to the reporting requirement, and are present on the Closing Disclosure. Therefore, the Bureau is including purchased covered loans in the scope of final Sec. 1003.4(a)(17)(ii). Final comments 4(a)(17)(i)-2 and 4(a)(17)(ii)-1 provide guidance on the scope of the total-loan-costs and total-points-and-fees reporting requirements with respect to purchased covered loans. One consumer advocate asked the Bureau to clarify the scope of proposed Sec. 1003.4(a)(17) with respect to covered loans ``subject to'' HOEPA or the Bureau's 2013 ATR Final Rule. This commenter also urged the Bureau to expand Sec. 1003.4(a)(17) to include home-equity lines of credit and reverse mortgages because both types of loans have been subject to abusive pricing. Proposed Sec. 1003.4(a)(17) would have applied to open-end lines of credit secured by the borrower's principal dwelling, but would have excluded other open-end lines of credit and all reverse mortgages. The Bureau believes that the benefit of points-and-fees data on such loans does not justify the burden of reporting for the reasons discussed above. Reverse mortgages are exempt from the ability-to-repay provisions of the 2013 ATR Final Rule and the 2013 TILA-RESPA Final Rule. Therefore, extending final Sec. 1003.4(a)(17) to reverse mortgages would require a calculation using a regulatory definition that would likely require certain modifications. The Bureau believes that this burden does not justify extending coverage to reverse mortgages or open-end lines of credit. However, the final rule will vastly improve upon the current regulation regarding the pricing information for these loans, by requiring reporting of data points such as rate spread,\318\ interest rate, prepayment penalty, and nonamortizing features. Final comments 4(a)(17)(i)-1 and 4(a)(17)(ii)-1 clarify that open-end lines of credit and reverse mortgages are excluded from the scope of the total-loan-costs and total-points-and- fees reporting requirements. --------------------------------------------------------------------------- \318\ Rate spread applies to open-end lines of credit but not reverse mortgages. See Sec. 1003.4(a)(12). --------------------------------------------------------------------------- Finally, many industry commenters and consumer advocates made comments that were broadly applicable to the proposed pricing data points. For example, both industry and consumer advocate commenters urged the Bureau to adopt alternative or additional pricing data points. Several industry commenters suggested that rate spread be reported instead of the other proposed pricing data points. These commenters noted that financial institutions were currently reporting the rate spread under existing Regulation C and believed that it made the other data points unnecessary. Similarly, one industry commenter proposed replacing the pricing data points with the annual percentage rate. The final rule does not adopt these suggestions because neither the rate spread nor the APR allows users to identify and compare fees imposed on borrowers. Two commenters recommended that ``legitimate discount points'' be distinguished from other disguised charges intended to compensate the lender or mortgage broker. One of these commenters recommended different data points for direct fees, yield-spread premiums, and points that are fees. Similarly, one consumer advocate recommended that the Bureau require reporting of loan originator compensation. This commenter explained that loan originator compensation was a factor in disparate pricing, is related to abusive lending practices, and that compensation data is necessary to monitor the appropriateness of the Bureau's loan originator compensation rules. The Bureau believes that the final pricing data points will enable HMDA data users to distinguish many of the costs about which these commenters were concerned. To the extent that additional data points would be necessary to perfectly address these commenters' concerns, the final rule does not adopt them. The final rule includes numerous data points related to loan pricing that will vastly improve the ability of users to understand and evaluate the costs associated with mortgage loans. More pricing data could increase the utility of the data, but not without imposing substantial burden on financial institutions. For example, many of the data points needed to represent various fees and charges or loan originator compensation would not be aligned with an existing regulation or appear consistently on any disclosure. Another commenter urged the Bureau to substantially expand the pricing data required by the final rule by including upfront costs to the lender or originator, less fees for title and settlement services; discount points; lender credits; interest rate; APR; upfront fees for settlement services; and a flag to indicate whether a lender or real estate agent possess an ownership interest in the title company. This commenter explained that the data described above were necessary to examine numerous issues related to loan pricing and cost, including the existence of high title service fees and the use of discount points. The Bureau agrees that including such data would provide value to users and notes that it has adopted many of the recommended data points in the final rule, such as discount points, lender credits, and interest rate. Further expansion at this time, however, would impose an unjustified burden on financial institutions. For example, the recommendations regarding the financial institution's ownership interest in the title company and the exclusion of title and settlement service costs from the total loan costs are absent from existing regulatory definitions, Federal disclosure forms, and standard industry data formats. [[Page 66211]] One industry commenter noted that certain pricing data points were not applicable to open-end lines of credit, such as total origination charges and total discount points. This commenter believed that this exclusion suggested that such data are not valuable. In fact, the exclusion of open-end lines of credit is a consequence of the Bureau's decision to align the data point to the Closing Disclosure and Regulation Z Sec. 1026.38(f)(1) in order to reduce burden. As explained in greater detail below, these data points provide important price information to users. Therefore, the Bureau believes that the scope of these data points balances the benefit of the data with the burden of reporting. For the reasons provided above, the Bureau is adopting new Sec. 1003.4(a)(17), which requires financial institutions to report, for covered loans subject to Regulation Z Sec. 1026.43(c), one of the following measures of loan cost: (i) If a disclosure is provided for the covered loan pursuant to Regulation Z, 12 CFR 1026.19(f), the amount of total loan costs, as disclosed pursuant to Regulation Z, 12 CFR 1026.38(f)(4), or, (ii) if the covered loan is not subject to the disclosure requirements in Regulation Z, 12 CFR 1026.19(f), and is not a purchased covered loan, the total points and fees charged in connection with the covered loan, expressed in dollars and calculated in accordance with Regulation Z, 12 CFR 1026.32(b)(1). This reporting requirement does not apply to applications or to covered loans not subject to the ability-to-repay requirements in the 2013 ATR Final Rule, such as open-end lines of credit, reverse mortgages, or loans or lines of credit made primarily for business or commercial purposes. The Bureau is also adopting several new comments. Final comments 4(a)(17)(i)-1 and 4(a)(17)(ii)-1 clarify the scope of the reporting requirement. Final comment 4(a)(17)(i)-2 explains that purchased covered loans are not subject to this reporting requirement if the application was received by the selling entity prior to the effective date of Regulation Z Sec. 1026.19(f). Final comment 4(a)(17)(ii)-2 provides guidance in situations where a financial institution has cured a points-and-fees overage. Final comment 4(a)(17)(i)-3 provides guidance in situations where a financial institution has issued a revised Closing Disclosure with a new amount of total loan costs. The Bureau believes that final Sec. 1003.4(a)(17) satisfies Congress's direction to provide for reporting total points and fees ``as determined by the Bureau, taking into account'' the definition of total points and fees provided by TILA and implemented in Regulation Z Sec. 1026.32(b).\319\ In requiring reporting of a covered loan's total points and fees, Congress intended to increase transparency regarding mortgage lending and improve fair lending screening.\320\ As defined in proposed Sec. 1003.4(a)(17), total points and fees would provide information about some of the upfront costs paid by borrowers. Similarly, total loan costs, as defined in final Sec. 1003.4(a)(17), also provide information about upfront costs paid by borrowers. Congress recognized the importance of the Bureau's expertise in deciding how to implement this measure by expressing that it should be defined ``as determined by the Bureau.'' The Bureau's implementation is consistent with that broad delegation of discretion. The Bureau has carefully considered the merits of both total points and fees, as defined in Regulation Z Sec. 1026.32(b), and total loan costs, as defined in Regulation Z Sec. 1026.38(f)(4). In proposing to require reporting of the total points and fees, as defined in Regulation Z Sec. 1026.32(b), the Bureau believed that such information would enable users to gain deeper insight into the terms on which different communities are offered mortgage loans. As explained above, after reviewing public comments, the Bureau has determined that total loan costs provide greater analytical value for comparing borrowers and understanding the cost of loans than total points and fees as defined in the proposal, while reducing the burden of reporting for financial institutions. Therefore, for certain loans, total loan costs are more consistent with Congress's goals in amending HMDA than proposed Sec. 1003.4(a)(17). For the reasons given above, final Sec. 1003.4(a)(17) implements HMDA section 304(b)(5)(A), and is also authorized by the Bureau's authority pursuant to HMDA section 304(b)(5)(D) to require such other information as the Bureau may require, and by the Bureau's authority pursuant to HMDA section 305(a) to provide for adjustments and exceptions. For the reasons given above, final Sec. 1003.4(a)(17) is necessary and proper to effectuate the purposes of and facilitate compliance with HMDA, because it will help identify possible discriminatory lending patterns and help determine whether financial institutions are serving the housing needs of their communities, and because it will significantly reduce burden for reporting financial institutions. Accordingly, where total loan costs are available, final Sec. 1003.4(a)(17) requires financial institutions to report them. However, as explained above, where total loan costs are not available, total points and fees, as defined in Sec. 1003.4(a)(17)(ii), will provide useful information that would not otherwise be available. --------------------------------------------------------------------------- \319\ 12 U.S.C. 2803(b)(5)(A). \320\ H. Rept. 111-702 at 191 (2011) (finding that more specific loan pricing information would ``provide more transparency on underwriting practices and patterns in mortgage lending and help improve the oversight and enforcement of fair lending laws.''). --------------------------------------------------------------------------- 4(a)(18) Section 304(b) of HMDA permits the disclosure of such other information as the Bureau may require.\321\ Pursuant to HMDA sections 305(a) and 304(b)(5)(D), the Bureau proposed to require financial institutions to report, for covered loans subject to the disclosure requirements in Regulation Z Sec. 1026.19(f), the total origination charges associated with the covered loan. Origination charges are those costs designated ``borrower-paid'' on Line A of the Closing Cost Details page of the current Closing Disclosure, as provided for in Regulation Z Sec. 1026.38(f)(1). Proposed Sec. 1003.4(a)(18) would have applied to closed-end covered loans and purchases of such loans, but not to applications, open-end lines of credit, reverse mortgages, or commercial-purpose loans. For the reasons provided below, the Bureau is adopting Sec. 1003.4(a)(18) as proposed, with additional clarifying commentary. --------------------------------------------------------------------------- \321\ Section 1094(3)(A)(iv) of the Dodd-Frank Act amended section 304(b) of HMDA. --------------------------------------------------------------------------- Industry commenters generally opposed the adoption of total origination charges. Several industry commenters believed that the total amount of borrower-paid origination charges provided little value, for various reasons. Two industry commenters asserted that the value of origination charges was minimal because they were influenced by factors outside of the financial institution's control, such as the borrower's decisionmaking. Many industry commenters raised similar objections to the proposed pricing data in general. For example, one industry commenter pointed out that the pricing data were incomplete because it omitted additional information about the borrower's overall relationship with the financial institution, such as the borrower's loan payment history or deposit balances. Therefore, these commenters argued, the pricing data points, including borrower-paid origination charges, would mislead users. [[Page 66212]] Despite the presence of other variables that influence loan pricing, information about origination charges offers analytical value. First, the final rule will capture several factors about which commenters were concerned, such as a borrower's decision to trade a higher interest rate for lower closing costs. To the extent that financial institutions lack the ability to unilaterally determine every item of borrower-paid origination charges, the control they exercise is high relative to many of the other elements of the Closing Disclosure, such as taxes and other government fees, prepaids, or the initial escrow payment at closing. Moreover, as stated above, the Bureau believes that the final rule need not provide an exhaustive representation of every factor that might conceivably affect loan pricing in order to benefit users. The final rule's pricing data represents a marked improvement over the existing regulation, and these benefits would be lost if the Bureau were to eliminate any data point that might be influenced by the complexity of the pricing process. Other industry commenters pointed out that proposed Sec. 1003.4(a)(18) omitted certain charges, such as appraisal fees and items paid by the seller. However, Sec. 1003.4(a)(18) is intended to capture the origination charges paid to the financial institution by the borrower; it is not intended to measure the total cost of the transaction. The Bureau is also providing for reporting of total loan costs in final Sec. 1003.4(a)(17), which will provide some of the information about the upfront cost of credit that commenters believed was missing from Sec. 1003.4(a)(18), such as costs associated with appraisal and settlement services. Regarding origination charges paid by the seller, as with total loan costs, seller-paid origination charges would appear on the Closing Disclosure if the seller were legally obligated to pay for such costs.\322\ However, only the sum of borrower-paid origination charges are disclosed on the current Closing Disclosure. Incorporating seller-paid origination charges would increase burden because financial institutions could no longer simply report the amount calculated under Regulation Z. --------------------------------------------------------------------------- \322\ See 12 CFR 1026.19(f)(1)(i). --------------------------------------------------------------------------- Several industry commenters argued that proposed Sec. 1003.4(a)(18) was duplicative because the Bureau had also proposed to require reporting of the total points and fees in Sec. 1003.4(a)(17). These commenters stated that origination charges were included in total points and fees, and that, in many cases, the origination charges would be identical to the total points and fees. Although final Sec. 1003.4(a)(17) requires reporting of the total loan costs rather than the total points and fees, as defined in proposed Sec. 1003.4(a)(17), the two data points overlap somewhat. However, total loan costs and borrower-paid origination charges differ in important respects. Total loan costs include many additional costs that are excluded from borrower-paid origination charges, such as charges for third-party settlement services. In contrast, total origination charges represent the costs that financial institutions themselves are directly imposing on borrowers. Furthermore, a user could take the difference between total loan costs and total origination charges as an approximate measure of total third-party charges. Therefore, final Sec. 1003.4(a)(17) and final Sec. 1003.4(a)(18) are necessary to enable users to gain a more precise understanding of the costs associated with a mortgage loan. Several other industry commenters argued that the total amount of borrower-paid origination charges was too burdensome to report. As mentioned above, the Bureau has aligned Sec. 1003.4(a)(18) to Regulation Z and to the Closing Disclosure in order to reduce burden. As with all pricing data points aligned to the Closing Disclosure, the calculation of origination charges will be required only for covered loans for which a Closing Disclosure is required pursuant to Regulation Z Sec. 1026.19(f). Loans excluded from Regulation Z Sec. 1026.19(f), such as open-end lines of credit, reverse mortgages, and commercial loans, are not subject to this provision. Therefore, the burden of reporting under Sec. 1003.4(a)(18) is limited to loans for which financial institutions would already have to calculate the total loan costs in order to disclose them to consumers. This alignment was supported by two industry commenters. Because using the definition of origination charges found in Regulation Z reduces burden while preserving the utility of the data, the Bureau is adopting this definition in the final rule. These exclusions are stated in final comment 4(a)(18)-1, which clarifies the scope of the reporting requirement. As stated in the proposal, the total amount of borrower-paid origination charges provides a relatively focused measure of the charges imposed on the borrower by the financial institution for originating and extending credit. Furthermore, separate identification of borrower-paid origination charges in addition to total discount points and lender credits facilitates understanding of loan pricing because charges are often interchangeable and may be spread across different elements of loan pricing. The proposed pricing data points, including total origination charges, will help users of HMDA data determine whether different borrowers are receiving fair pricing and develop a better understanding of the ability of borrowers in certain communities to access credit. Therefore, the Bureau is adopting Sec. 1003.4(a)(18) generally as proposed. In response to the Bureau's solicitation of feedback, one consumer advocate urged the Bureau to require the amount listed as the ``total closing costs'' on Line J of the current Closing Disclosure in addition to or instead of the total origination charges. The commenter stated that origination charges represent a small part of total costs and that financial institutions exert some control over other costs through affiliated business arrangements. In contrast, one industry commenter opposed requiring total closing costs because the commenter believed that the number of factors incorporated into the total closing costs made meaningful comparisons among borrowers impossible. The Bureau acknowledges that total closing costs would provide important information about the costs required for consumers to close on a loan, but is not adopting a new data point for total closing costs. As described above, the Bureau is adopting Sec. 1003.4(a)(17), which requires reporting the total loan costs associated with the covered loan. Final Sec. 1003.4(a)(17) addresses many of the concerns this commenter raised regarding a more inclusive, consistent measure of loan costs, and also includes the upfront cost associated with many third- party settlement services. Furthermore, total closing costs, as disclosed pursuant to Regulation Z Sec. 1026.38(h)(1), include many costs unrelated to the charges imposed by financial institutions for extending credit, such as taxes and other government fees. The Bureau believes that many of these costs can be more accurately estimated by users than the total loan costs, because they will be largely determined by the jurisdiction in which the loan was originated. Total origination charges and total loan costs also bear a closer relationship to the lending practices of financial institutions than total closing costs, and therefore better advance the purposes of HMDA. For the reasons provided above, pursuant to HMDA sections 305(a) and 304(b)(5)(D), the Bureau is adopting [[Page 66213]] Sec. 1003.4(a)(18) as proposed. For the reasons given above, data about total origination charges will assist public officials and members of the public in determining whether financial institutions are serving the housing needs of their communities and in identifying potentially discriminatory lending patterns. Final Sec. 1003.4(a)(18) requires financial institutions to report, for covered loans subject to the disclosure requirements in Regulation Z Sec. 1026.19(f), the total of all itemized amounts that are designated borrower-paid at or before closing, as disclosed pursuant to Sec. 1026.38(f)(1). These charges are the total costs designated ``borrower-paid'' on Line A of the Closing Cost Details page of the current Closing Disclosure. The Bureau is also adopting several new comments. Final comment 4(a)(18)-1 clarifies the scope of the reporting requirement. Final comment 4(a)(18)-2 explains that purchased covered loans are not subject to this reporting requirement if the application was received by the selling entity prior to the effective date of Regulation Z Sec. 1026.19(f). Final comment 4(a)(18)-3 provides guidance in situations where a financial institution has issued a revised Closing Disclosure with a new amount of total origination charges. 4(a)(19) Section 304(b) of HMDA permits the disclosure of such other information as the Bureau may require.\323\ Pursuant to HMDA sections 305(a) and 304(b)(5)(D), the Bureau proposed to require financial institutions to report, for covered loans subject to the disclosure requirements in Regulation Z Sec. 1026.19(f), the total discount points paid by the borrower. Discount points are points paid to the creditor to reduce the interest rate, and are listed on Line A.01 of the Closing Cost Details page of the current Closing Disclosure, as described in Regulation Z Sec. 1026.37(f)(1)(i). Proposed Sec. 1003.4(a)(19) would have applied to closed-end covered loans and purchases of such loans, but not to applications, open-end lines of credit, reverse mortgages, or commercial-purpose loans. For the reasons provided below, the Bureau is adopting Sec. 1003.4(a)(19) generally as proposed, with minor technical modifications and new commentary for increased clarity. --------------------------------------------------------------------------- \323\ Section 1094(3)(A)(iv) of the Dodd-Frank Act amended section 304(b) of HMDA. --------------------------------------------------------------------------- Industry commenters generally opposed the requirement to report discount points. Some industry commenters believed that reporting the total discount points was unnecessary or duplicative. Several of these commenters pointed out that the proposal also required financial institutions to report the total points and fees, while other commenters stated that discount points were only applicable to a limited class of loans sold into the secondary market. One industry commenter believed that rate spread and total points and fees could be used to reveal potential unlawful discrimination. Although discount points are included in both total loan costs and total origination charges, these data points are not substitutes for each other. As explained above, total loan costs and total origination charges represent different elements of loan cost. Discount points are also different than the other loan costs because they represent charges directly related to reductions in the interest rate and are necessary to understand the tradeoffs between rates and points. Other measures of pricing, such as rate spread and total loan costs, can be useful for comparing borrowers, but separate reporting of discount points will improve analysis of the value borrowers are receiving for paying discount points. Finally, even if discount points are not present in every loan, studies of loan costs and public comments received before and after the proposal suggest that discount points are an important element of loan pricing.\324\ --------------------------------------------------------------------------- \324\ See, e.g., 79 FR 51731, 51788-89 (Aug. 29, 2014) (describing feedback received prior to the proposal); Susan E. Woodward, A Study of Closing Costs for FHA Mortgages, at 60-69 (2008) (report prepared for the U.S. Dep't. of Hous. and Urban Dev., Office of Policy Dev. and Research) (discussing problems with discount points on FHA loans); David Nickerson & Marsha Courchane, Discrimination Resulting from Overage Practices, 11 J. of Fin. Servs. Research 133 (1997). --------------------------------------------------------------------------- Other industry commenters opposed reporting discount points because they believed that doing so would distort the data or potentially mislead users. One industry commenter noted that the absence of information about lender credits would make comparisons between loans with and without lender credits misleading. Other industry commenters argued that comparisons between borrowers were difficult or impossible because of market fluctuations, differences in product type, and borrower decisionmaking. In response to these comments, the Bureau is adding a requirement for financial institutions to report lender credits. As explained above, however, even though HMDA data are not exhaustive, the data still provide extremely valuable information for the public and public officials that fulfills HMDA's purposes. Regarding the influence of other variables, the final rule includes several data points that will allow users to control for several of the factors mentioned by commenters, including location and product type. Indeed, not requiring reporting of discount points might also mislead users by limiting their ability to explain the lower rates received by borrowers who paid discount points. Several industry commenters argued that the benefit of proposed Sec. 1003.4(a)(19) was unclear and questioned whether there was any evidence of discrimination against borrowers through discount points. As stated in the proposal, reporting discount points benefits users of HMDA data by enabling them to develop a more detailed understanding of loan pricing. This improved information allows for better analyses regarding the value that borrowers receive in exchange for discount points, and determinations of whether similarly situated borrowers are receiving similar value. Existing studies of loan costs and feedback received prior to the proposal suggested that discount points were a sufficiently important element of loan pricing to justify their inclusion in HMDA.\325\ --------------------------------------------------------------------------- \325\ See, e.g., 79 FR 51731, 51788-89 (Aug. 29, 2014) (describing feedback received prior to the proposal); Susan E. Woodward, A Study of Closing Costs for FHA Mortgages, at 60-69 (2008) (report prepared for the U.S. Dep't. of Hous. and Urban Dev., Office of Policy Dev. and Research) (discussing problems with discount points on FHA loans); David Nickerson & Marsha Courchane, Discrimination Resulting from Overage Practices, 11 J. of Fin. Servs. Research 133 (1997). --------------------------------------------------------------------------- Finally, one industry commenter believed that reporting discount points was too burdensome because the definition was uncertain. To minimize any burden associated with reporting discount points, the Bureau is adopting a definition of discount points that aligns to Regulation Z. Loans excluded from Regulation Z Sec. 1026.19(f), such as open-end lines of credit, reverse mortgages, and commercial loans, are not subject to final Sec. 1003.4(a)(19). Therefore, the burden of reporting is limited to loans for which financial institutions would already have to know the amount of discount points in order to disclose it to consumers. These exclusions are stated in final comment 4(a)(19)- 1, which clarifies the scope of the reporting requirement. This alignment was supported by one industry commenter. The TILA-RESPA integrated disclosure forms, including the Closing Disclosure, are the subject of considerable outreach and guidance from the Bureau during the implementation process. As financial institutions become familiar with these [[Page 66214]] forms, the burden of reporting should decrease. For the reasons provided above, pursuant to HMDA sections 305(a) and 304(b)(5)(D), the Bureau is adopting Sec. 1003.4(a)(19) generally as proposed, with minor technical modifications. These technical modifications clarify that, although discount points are described more clearly in Regulation Z Sec. 1026.37(f)(1)(i), financial institutions should report the amount found on the Closing Disclosure, as disclosed pursuant to Regulation Z Sec. 1026.38(f)(1). For the reasons given above, data about discount points will assist public officials and members of the public in determining whether financial institutions are serving the housing needs of their communities and in identifying potentially discriminatory lending patterns. Final Sec. 1003.4(a)(19) requires financial institutions to report, for covered loans subject to the disclosure requirements in Regulation Z, 12 CFR 1026.19(f), the points paid to the creditor to reduce the interest rate, expressed in dollars, as described in Regulation Z, 12 CFR 1026.37(f)(1)(i), and disclosed pursuant to Regulation Z, 12 CFR 1026.38(f)(1). For covered loans subject to the disclosure requirements in Regulation Z Sec. 1026.19(f), the discount points that financial institutions would report are those listed on Line A.01 of the Closing Cost Details page of the current Closing Disclosure. The Bureau is also adopting several new comments. Final comment 4(a)(19)-1 clarifies the scope of the reporting requirement. Final comment 4(a)(19)-2 explains that purchased covered loans are not subject to this reporting requirement if the application was received by the selling entity prior to the effective date of Regulation Z Sec. 1026.19(f). Final comment 4(a)(19)-3 provides guidance in situations where a financial institution has issued a revised Closing Disclosure with a new amount of discount points. 4(a)(20) Proposed 4(a)(20) Section 304(b) of HMDA authorizes the disclosure of such other information as the Bureau may require.\326\ Pursuant to HMDA sections 305(a) and 304(b)(5)(D), the Bureau proposed to require financial institutions to report, for covered loans subject to the disclosure requirements in Regulation Z Sec. 1026.19(f), other than purchased covered loans, the risk-adjusted, pre-discounted interest rate associated with a covered loan. The risk-adjusted, pre-discounted interest rate (RPIR) is the rate that the borrower would have received in the absence of any discount points or rebates and is the same base rate from which a financial institution would exclude ``bona fide discount points'' from the points-and-fees total used to determine qualified mortgage and high-cost mortgage status under Regulation Z. Proposed Sec. 1003.4(a)(20) would have applied to closed-end covered loans, but not to applications or purchased covered loans, or open-end lines of credit, reverse mortgages, or commercial-purpose loans. For the reasons provided below, the Bureau is not finalizing proposed Sec. 1003.4(a)(20). --------------------------------------------------------------------------- \326\ Section 1094(3)(A)(iv) of the Dodd-Frank Act amended section 304(b) of HMDA. --------------------------------------------------------------------------- Most consumer advocates expressed support for the proposed pricing data points collectively, but few commented specifically on the RPIR. One commenter generally stated that the RPIR would be helpful for fair lending analysis. Another consumer advocate believed that, combined with the other proposed data points, the RPIR would better enable users to understand pricing disparities among groups of consumers. This consumer advocate further urged the Bureau to expand Sec. 1003.4(a)(20) to cover home-equity lines of credit because doing so would improve the ability of users to compare pricing across loan types. The Bureau agrees with commenters that the concept of a risk- adjusted, pre-discounted interest rate would have value for fair lending purposes, provided that such a rate was consistently calculated. However, public comments and additional outreach have revealed that the rate proposed to be reported under Sec. 1003.4(a)(20) is less valuable and more unclear than the Bureau initially believed. Several industry commenters cited definitional issues surrounding proposed Sec. 1003.4(a)(20). For example, one commenter noted that a single loan may have multiple rates available to the consumer that would satisfy the description of the RPIR. Another commenter stated that the concept of an RPIR existed only in the realm of informal guidance provided by the Bureau under Regulation Z. Similar feedback was provided by many of the vendors and financial institutions that participated in additional outreach conducted by the Bureau after the proposal's comment period closed. These participants expressed different understandings of the rate that would be required by proposed Sec. 1003.4(a)(20). For example, two participants noted that multiple rates could potentially satisfy the requirements of the RPIR, and that the discretion of a financial institution was required to select a rate that would actually function as the pre-discounted rate, if applicable, for Regulation Z purposes. Other participants cited lack of definitional clarity as a factor that would add significant burden to the proposed reporting requirement. Additionally, several industry commenters questioned the benefit that the RPIR would provide for fair lending purposes. For example, one commenter doubted that the RPIR would produce any fair lending insights beyond those made possible by the current pricing data. As stated in the proposal, the potential value of the RPIR comes from its explanatory power. Pricing outcomes are determined by many factors, including rate-sheet inputs, loan-level pricing adjustments, other discretionary pricing adjustments, and consumer decisionmaking. The RPIR would reflect many of the pricing adjustments for which users would have to control in order to determine whether pricing disparities were explained by legitimate business considerations. Therefore, analyzing the changes to loan pricing that occur after a financial institution has determined the RPIR may provide strong evidence of potential impermissible discrimination with a reduced need to control for multiple legitimate factors that influence loan pricing. However, the Bureau now believes that the RPIR may not provide sufficient value to justify the burden associated with collecting and reporting it. The rate described in proposed Sec. 1003.4(a)(20) is the base rate to which a financial institution would apply any reduction obtained by the payment of discount points in determining whether those points may be excluded as ``bona fide discount points'' from points and fees pursuant to Regulation Z Sec. 1026.32(b). This rate was originally designed to ensure that discount points excluded from the points-and-fees coverage tests actually produced an appropriate reduction in the borrower's interest rate. The rate was not intended to isolate pricing adjustments necessary to facilitate fair lending analysis. Therefore, the Bureau believes that the rate is less beneficial for fair lending purposes than it initially thought. After considering the function of the rate and the burden associated with reporting it, the Bureau has decided not to finalize proposed Sec. 1003.4(a)(20). As part of the additional outreach, the Bureau also sought information about two other measures of loan pricing that might have greater fair lending benefit than the proposed RPIR. These measures are the ``post-LLPA rate'' and the ``discretionary adjustment.'' The [[Page 66215]] post-LLPA rate is the interest rate that reflects all the transaction- specific, nondiscretionary pricing adjustments dictated by the financial institution's standard loan pricing policy. The discretionary adjustment is any alteration by the financial institution of the interest rate or points made for any reason other than the application of the standard loan pricing policy. However, feedback received through the additional outreach process suggested that these measures would be more burdensome to report. For example, they may be calculated and stored less commonly than the RPIR, and neither currently possesses a definition in either existing regulation or industry custom. Therefore, at this time, the Bureau has not identified a suitable alternative base rate that it could substitute for the RPIR proposed in Sec. 1003.4(a)(20). For the reasons provided above, the Bureau is not finalizing proposed Sec. 1003.4(a)(20). Final 4(a)(20) Section 304(b) of HMDA permits the disclosure of such other information as the Bureau may require.\327\ In using its discretionary authority to propose to require financial institutions to report the total discount points paid by the consumer, the Bureau also invited comment on ``whether to include any lender credits, premiums, or rebates in the measure of discount points.'' \328\ For the reasons provided below, the Bureau is adopting new Sec. 1003.4(a)(20), which requires financial institutions to report, for covered loans subject to the disclosure requirements in Regulation Z Sec. 1026.19(f), the total amount of lender credits, as disclosed pursuant to Regulation Z Sec. 1026.38(h)(3). Lender credits are amounts provided to the borrower to offset closing costs and are disclosed under Line J of the Closing Cost Details page of the current Closing Disclosure. Final Sec. 1003.4(a)(20) applies to closed-end covered loans and purchases of such loans, but not to applications, open-end lines of credit, reverse mortgages, or commercial-purpose loans. --------------------------------------------------------------------------- \327\ Section 1094(3)(A)(iv) of the Dodd-Frank Act amended section 304(b) of HMDA. \328\ See 79 FR 51731, 51789 (Aug. 29, 2014). --------------------------------------------------------------------------- The Bureau received several comments in response to its solicitation for feedback regarding lender credits. Some industry commenters requested clarification regarding whether such credits would be included within any of the proposed data points. For example, two commenters asked how offsetting credits associated with an interest rate would be reported, if at all. One industry commenter believed that information regarding lender credits would provide no value to HMDA users. However, other comments suggested that data on lender credits would be valuable even though the commenters did not advocate for reporting of these data. For example, one commenter explained that without some representation of lender credits, the prices of loans with such offsetting credits would appear artificially high. The Bureau believes that lender credits are a basic element of the cost of the loan that should be represented in the HMDA data. Financial institutions often offer borrowers a credit or rebate to offset some or all of the closing costs associated with a loan in return for accepting a higher interest rate. These credits reflect trade-offs similar to those that borrowers make between discount points and the interest rate, and are generally displayed as negative points on the rate sheet. As commenters have pointed out, without accounting for these credits, users of HMDA data would be unable to determine that loans with credits or rebates were not higher priced than similar loans without such credits. As noted above, the final rule cannot provide for reporting of every factor that might conceivably influence loan pricing. However, the Bureau finds that lender credits should be included because they are sufficiently important to understanding the price of a loan. Although the amount of lender credits disclosed under Regulation Z Sec. 1026.38(h)(3) may also include any refunds provided for amounts that exceed the limitations on increases in closing costs, the Bureau believes that an imperfect measure of lender credits is substantially better than no measure at all.\329\ Furthermore, removing such refunds to obtain a pure measure of lender credits would increase burden by forcing lenders to perform a new calculation that they would not otherwise perform under any existing regulation. --------------------------------------------------------------------------- \329\ The lender credits disclosed pursuant to Regulation Z Sec. 1026.38(h)(3) would also exclude any credits attributable to specific loan costs listed in the Closing Disclosure. See 12 CFR 1026.19(f), comment 38(h)(3)-1. --------------------------------------------------------------------------- Two industry commenters opposed reporting lender credits because they would be burdensome to report. However, the Bureau is adopting a definition of lender credits that aligns to Regulation Z Sec. 1026.38(h)(3) and is applying the final reporting requirement only to covered loans for which a Closing Disclosure is required. Loans excluded from Regulation Z Sec. 1026.19(f), such as open-end lines of credit, reverse mortgages, and commercial loans, are not subject to final Sec. 1003.4(a)(20). Therefore, the burden of reporting is limited to loans for which financial institutions would already have to disclose the total amount of lender credits. These exclusions are stated in final comment 4(a)(20)-1, which clarifies the scope of the reporting requirement. For the reasons provided above, pursuant to HMDA sections 305(a) and 304(b)(5)(D), the Bureau is adopting new Sec. 1003.4(a)(20), which requires financial institutions to report, for covered loans subject to the disclosure requirements in Regulation Z Sec. 1026.19(f), the total amount of lender credits, as disclosed pursuant to Regulation Z Sec. 1026.38(h)(3). The total amount of lender credits appears under Line J of the Closing Cost Details page of the current Closing Disclosure. For the reasons given above, data about lender credits will assist public officials and members of the public in determining whether financial institutions are serving the housing needs of their communities and in identifying potentially discriminatory lending patterns. The Bureau is also adopting several comments. Final comment 4(a)(20)-1 clarifies the scope of the reporting requirement. Final comment 4(a)(20)-2 explains that purchased covered loans are not subject to this reporting requirement if the application was received by the selling entity prior to the effective date of Regulation Z Sec. 1026.19(f). Final comment 4(a)(20)-3 provides guidance in situations where a financial institution has issued a revised Closing Disclosure with a new amount of lender credits. 4(a)(21) Section 304(b) of HMDA permits the disclosure of such other information as the Bureau may require.\330\ Pursuant to HMDA sections 305(a) and 304(b)(6)(J), the Bureau proposed to require financial institutions to report the interest rate that is or would be applicable to the covered loan or application at closing or account opening. Proposed comment 4(a)(21)-1 explained the interest rate that financial institutions should report for covered loans subject to certain disclosure requirements in Regulation Z. For the reasons provided below, the Bureau is generally adopting Sec. 1003.4(a)(21) as proposed, with minor modifications and the addition of commentary clarifying the reporting obligations for applications and for adjustable-rate transactions for which [[Page 66216]] the interest rate is unknown at the time final action is taken. --------------------------------------------------------------------------- \330\ Section 1094(3)(A)(iv) of the Dodd-Frank Act amended section 304(b) of HMDA. --------------------------------------------------------------------------- Consumer groups supported the proposed pricing data points, including the interest rate. These commenters stated that such information would help identify potentially unlawful price discrimination and better understand the type and terms of credit offered to different communities. For example, one commenter noted that the interest rate would be particularly valuable for analyzing the impact of discount points. Another commenter stated that the interest rate was necessary to study the terms of the loan. Finally, other consumer advocate commenters noted that the interest rate, when combined with the other pricing variables, would enable a more precise understanding of the elements of loan pricing. Industry commenters generally opposed requiring financial institutions to report the interest rate. Some industry commenters argued that the interest rate had little value or relevance, and one industry commenter disagreed that facilitating comparisons among borrowers was sufficient to justify the reporting requirement. The value of information regarding the interest rate, however, comes not only from comparing the interest rates received by borrowers but from the ability to better understand the relationship between the interest rate and discount points, origination charges, and lender credits. This more detailed understanding will better facilitate identification of potentially discriminatory lending patterns and provide a more complete picture of the credit available to particular communities. Several other industry commenters argued that the interest rate was an unnecessary data point. Most of these commenters pointed out that the rate spread was already reported and would enable some analysis of loan pricing. One industry commenter suggested that the annual percentage rate be reported instead of the interest rate. However, one commenter believed that the APR was often calculated inaccurately and therefore supported reporting of the interest rate. Although the rate spread and the interest rate are related, they are not equivalent measures of loan pricing. As explained in the proposal, the APR is a measure of the cost of credit, including both interest and certain fees, expressed as a yearly rate, while the interest rate is the cost of the loan expressed as a percentage rate. The interest rate enables users to understand the relationship between the interest rate and discount points, origination charges, and lender credits more directly than the rate spread, because the rate spread does not isolate the interest rate. Second, the rate spread and interest rate data points have substantially different scopes. Unlike rate spread, final Sec. 1003.4(a)(21) applies to both reverse mortgages and commercial loans. Indeed, Sec. 1003.4(a)(21) is one of few pricing data points that applies to such loans. Other industry commenters stated that information about the interest rate would be misleading. One industry commenter noted that the interest rate was influenced by factors outside of a financial institution's control, such as market fluctuations and borrower decisionmaking. Two industry commenters believed that proposed Sec. 1003.4(a)(21) would encourage financial institutions to provide ``teaser rates'' to create the illusion of lower-priced loans in their HMDA data. Although financial institutions set interest rates based in part on market factors that they may not control, interest rate data are still valuable, along with other data elements, to help further HMDA's purposes, including as a screen for potential fair lending concerns. For example, the final rule provides for reporting information about the date, product type, location, and certain consumer decisions, such as the choice to pay discount points for a lower rate or receive lender credits in exchange for a higher rate. Moreover, eliminating the interest rate might also undermine the utility of other data points. Users would experience more difficulty understanding the discount points and lender credits among borrowers or groups of borrowers. Finally, the final rule will also provide for reporting of the introductory rate period, which should discourage the type of rate manipulation about which commenters were concerned. One industry commenter believed that reporting the interest rate might allow competitors to gain insight into confidential business information, such as underwriting criteria. This commenter did not explain how a competitor would derive proprietary information regarding its underwriting criteria from the interest rate, and the Bureau is aware of no reliable means of doing so. Several industry commenters raised concerns over the burden of reporting the interest rate. These commenters pointed out that interest rates fluctuate frequently and may be unavailable for loans that are not originated. Similarly, several commenters requested that the Bureau not require financial institutions to report the interest rate for applications because the rate might be unknown. One commenter asked what rate should be reported for an application for which the rate has not been locked. The Bureau notes that, for many applications, a financial institution may not know the interest rate applicable to the covered loan. However, for applications approved by the financial institution but not accepted by the applicant, the interest rate would typically be available. Accordingly, the Bureau is clarifying that Sec. 1003.4(a)(21) requires a financial institution to report the interest rate only if the application has been approved by the financial institution but not accepted by the borrower, or if the financial institution reports the loan as originated. For all other applications or preapprovals, such as applications that have been denied or withdrawn, or files closed for incompleteness, a financial institution reports that no interest rate was applicable. The Bureau is adopting final comment 4(a)(21)-2 to clarify the reporting obligations in the case of applications. This comment removes the burden of attempting to determine the interest rate where the rate is truly unavailable while preserving data utility regarding applications by providing for reporting of the rate where the rate is available. For applications that have been approved but not accepted for which the rate has not been locked, financial institutions would report the rate applicable at the time the application was approved. The Bureau is also adopting comment 4(a)(21)-3, which states that, for adjustable-rate covered loans or applications, if the interest rate is unknown at the time that the application was approved, or at closing or account opening, a financial institution reports the fully-indexed rate. For purposes of Sec. 1003.4(a)(21), the fully-indexed rate is the index value and margin at the time that the application was approved, or, for covered loans, at closing or account opening. This comment mirrors the approach taken by comment 4(a)(21)-1, which clarifies the interest rate to be reported for loans subject to the Bureau's TILA-RESPA Integrated Disclosure Rule. Several industry commenters also requested that the Bureau exclude commercial loans, including multifamily mortgage loans, from the scope of Sec. 1003.4(a)(21). Commercial loans, these commenters explained, typically have interest rates that are variable and based on different indices than consumer loans. Similarly, one industry commenter noted that the interest rates for multifamily mortgage loans were based on a variety of factors that differed among multifamily loans. [[Page 66217]] Regarding variable interest rates, as explained above, the Bureau is adopting comment 4(a)(21)-3, which provides that, for adjustable-rate covered loans or applications, if the interest rate is unknown at the time that the application was approved, or at closing or account opening, a financial institution reports the fully-indexed rate based on the index applicable to the covered loan or application. Regarding loan comparisons, the adoption of a commercial-purpose flag in the final rule will enable HMDA data users to identify these loans and avoid potentially misleading comparisons. Information about multifamily housing continues to be an important component of the HMDA data. Information about the conditions of financing for multifamily dwellings may help public officials in distributing public-sector investment so as to attract private investment to areas where it is needed. Therefore, the Bureau is not excluding such loans from Sec. 1003.4(a)(21). For the reasons provided above, pursuant to HMDA sections 305(a) and 304(b)(6)(J), the Bureau is adopting Sec. 1003.4(a)(21) generally as proposed, with minor modifications and additional clarifying commentary. For the reasons given above, data about the interest rate will assist public officials and members of the public in determining whether financial institutions are serving the housing needs of their communities and in identifying potentially discriminatory lending patterns. The Bureau is adopting commentary identifying the interest rate that should be reported for covered loans subject to the disclosure requirements of Regulation Z Sec. 1026.19(e) or (f). The commentary also explains that, for applications, final Sec. 1003.4(a)(21) requires a financial institution to report the interest rate only for applications that have been approved by the financial institution but not accepted by the borrower. Finally, the Bureau is adopting commentary clarifying the interest rate to be reported for adjustable-rate covered loans or applications for which the initial interest rate is unknown. Final Sec. 1003.4(a)(21) applies to closed- end covered loans, open-end lines of credit, reverse mortgages, and commercial-purpose loans, as well as to purchases of such loans, and applications that have been approved by the lender but not accepted by the borrower. 4(a)(22) Section 304(b) of HMDA \331\ requires reporting of the term in months of any prepayment penalty or other fee or charge payable upon repayment of some portion of principal or the entire principal in advance of scheduled payments.\332\ The Bureau proposed to implement this provision through proposed Sec. 1003.4(a)(22), which required financial institutions to report the term in months of any prepayment penalty, as defined in Regulation Z Sec. 1026.32(b)(6)(i) or (ii), as applicable. Prepayment penalties are charges imposed on borrowers for paying all or part of the transaction's principal before the date on which the principal is due. Proposed Sec. 1003.4(a)(22) would have applied to applications for, and originations of, closed-end loans, open-end lines of credit, reverse mortgages, and commercial-purpose loans, but not to purchases of such loans. For the reasons provided below, the Bureau is adopting Sec. 1003.4(a)(22) generally as proposed, with clarifying commentary, but is limiting its scope to certain covered loans or applications subject to Regulation Z, 12 CFR part 1026. The revised scope of the reporting requirement excludes purchased covered loans, as well as reverse mortgages and loans or lines of credit made primarily for business or commercial purposes. --------------------------------------------------------------------------- \331\ Section 1094(3)(A)(iv) of the Dodd-Frank Act amended section 304(b) of HMDA. \332\ 12 U.S.C. 2803(b)(5)(C). --------------------------------------------------------------------------- The Bureau received few comments supporting or opposing proposed Sec. 1003.4(a)(22). Two industry commenters asserted that reporting information about prepayment penalties was unnecessary because regulatory scrutiny and the requirements of secondary market programs have diminished their prevalence. On the other hand, several consumer advocates supported the improved pricing data, including reporting of the prepayment penalty. One consumer advocate was particularly supportive of proposed Sec. 1003.4(a)(22) because of the importance of understanding whether certain communities were receiving loans with problematic features. The final rule retains the requirement to report data about prepayment penalties, consistent with the Dodd-Frank Act amendments to HMDA. In the lead-up to the financial crisis, prepayment penalties were frequently cited as a risky feature for consumers with subprime loans. Although prepayment penalties may be less prevalent than they were in the years preceding the financial crisis, their use may increase in the future. Prepayment penalty data will allow for the identification of any potential increase in prepayment penalties when considering how institutions are meeting the housing needs of their communities, and when looking for any potentially discriminatory lending practices. Most industry commenters requested certain clarifications or revisions to the scope of the reporting requirement. One industry commenter requested that the final rule not require reporting of the prepayment penalty for applications that do not result in originations. The Bureau is not adopting this suggestion. Both loans and applications for loans with prepayment penalties will provide valuable data for HMDA's purposes, and commenters have not suggested that the prepayment penalty term is more burdensome to determine for an application than for an originated loan. If the loan for which a consumer applied featured a prepayment penalty, the financial institution would report the term of that prepayment penalty. Similarly, if the loan for which the consumer applied featured no prepayment penalty, the financial institution would report that the reporting requirement was not applicable to the transaction. The Bureau has reflected these requirements in final comment 4(a)(22)-2. Two other industry commenters requested clarification regarding certain conditionally-waived charges. Final Sec. 1003.4(a)(22) defines prepayment penalty with reference to Regulation Z Sec. 1026.32(b)(6)(i) or (ii), as applicable. The commentary to Sec. 1026.32(b)(6) discusses waived, bona fide third- party charges imposed under certain conditions and, as explained in final comment 4(a)(22)-2, may be relied on for purposes of Sec. 1003.4(a)(22). Two industry commenters asked the Bureau to exclude commercial loans, including multifamily loans, from the prepayment penalty reporting requirement. These commenters pointed out that prepayment penalties serve different purposes in commercial lending. One commenter explained that multifamily mortgage loans featured various forms of prepayment protection, such as lock-out features, yield maintenance, or prepayment premiums that were not contemplated in the definition of prepayment penalty found in Regulation Z Sec. 1026.32(b)(6)(i) and (ii). This commenter urged the Bureau to either limit Sec. 1003.4(a)(22) to consumer loans or to adopt a new definition that was relevant to the commercial and multifamily lending context. [[Page 66218]] The Bureau understands that commercial loans, particularly multifamily mortgage loans, include forms of prepayment protection which have no analog in the consumer-purpose mortgage context. For example, these loans may feature defeasance, in which the borrower of a multifamily mortgage loan substitutes a new form of collateral, such as bonds or other securities, designed to generate sufficient cash flow to cover future loan payments. In order to capture these complex arrangements, the final rule would have to include a new definition of prepayment penalty. A new definition that is not part of any other existing regulation would likely impose burden on financial institutions. Moreover, consumer mortgage loans with prepayment penalties were most frequently cited as a concern in the lead up to the financial crisis and the Dodd-Frank Act. The Bureau is not aware of similar concerns about commercial loans covered by HMDA. At this time, the Bureau does not believe that applying Sec. 1003.4(a)(22) to commercial loans would provide sufficient benefits to justify the additional burden on financial institutions. Therefore, the Bureau is limiting the scope of final Sec. 1003.4(a)(22) to covered loans or applications subject to Regulation Z, 12 CFR part 1026. For the reasons provided above, to implement HMDA section 304(b)(5)(C), and pursuant to HMDA section 305(a), the Bureau is adopting Sec. 1003.4(a)(22) generally as proposed, but is modifying the scope of the provision to apply to certain covered loans and applications subject to Regulation Z, 12 CFR part 1026. Final Sec. 1003.4(a)(22) applies to applications for, and originations of, closed- end covered loans and open-end lines of credit, but not reverse mortgages and commercial-purpose loans. To facilitate compliance, the Bureau is excepting covered loans that have been purchased by a financial institution. As the Bureau explained in the proposal, it does not believe that the term of a prepayment penalty would be readily available from the information obtained from the selling entity.\333\ The Bureau is also excepting reverse mortgages and commercial-purpose loans, which, as explained above, will facilitate compliance. --------------------------------------------------------------------------- \333\ 79 FR 51731 at 51791-92. --------------------------------------------------------------------------- Final Sec. 1003.4(a)(22) includes commentary clarifying the reporting obligations of financial institutions in certain situations. Final comment 4(a)(22)-1 clarifies the scope of the reporting requirement. Final comment 4(a)(22)-2 provides guidance for reporting the prepayment penalty for applications and allows financial institutions to rely on the commentary to the relevant sections of Regulation Z. 4(a)(23) Proposed Sec. 1003.4(a)(23) provided that a financial institution must report the ratio of the applicant's or borrower's total monthly debt to the total monthly income relied on in making the credit decision (debt-to-income ratio). Proposed Sec. 1003.4(a)(23) applied to covered loans and applications, except for reverse mortgages. The Bureau also proposed new comments 4(a)(23)-1 through -4. Many commenters addressed including the debt-to-income ratio in the HMDA data. Many community advocate commenters expressed support for its inclusion, while many industry commenters raised concerns about reporting the data. For the reasons discussed below, the Bureau is finalizing Sec. 1003.4(a)(23) and comments 4(a)(23)-1 through -4 as proposed with technical modifications discussed below. In addition, the Bureau is adopting new comments 4(a)(23)-5 through -7. Comments Several consumer advocate commenters expressed strong support for proposed Sec. 1003.4(a)(23). Many noted that the debt-to-income ratio will help identify problematic loans where there may be a need for intervention. One commenter stated that higher ratios correspond with higher default rates and suggested that lenders' acceptance of higher debt-to-income ratios in loans originated in the mid-2000s contributed to the high foreclosure rates after 2005. In addition, commenters stated that the debt-to-income ratio will enable users to identify whether the debt-to-income ratio is a barrier to credit and, if so, which consumers are affected. A consumer advocate commenter expressed support for collecting the debt-to-income ratio, but noted limitations to its utility because it can be easily manipulated. The commenter explained that the debt-to- income ratio may overstate a borrower's repayment ability because a borrower may repay an open-end line of credit to reduce their debt in order to qualify, but then immediately re-draw the line. In addition, the debt-to-income ratio may understate a borrower's ability to repay because a financial institution may only consider the minimum income to qualify. Many industry commenters expressed concerns about proposed Sec. 1003.4(a)(23). Many commenters questioned the value of reporting this information. Some noted that the data would be difficult to analyze because the debt-to-income ratio is calculated and weighted differently depending on the loan product, financial institution, and applicant's circumstances. Others stated that the data would not be valuable for different reasons, including that the debt-to-income ratio is not calculated for all loans and that the debt-to-income ratio only factors into denial, and not into pricing decisions. Commenters also expressed concern that the information may be misunderstood because the debt-to- income ratio is one of many factors in an underwriting decision and conveys complex information. Other commenters objected to including this requirement because it is not expressly required by the Dodd-Frank amendments to HMDA. A few commenters asserted that collecting the debt- to-income ratio would not support HMDA's purposes. Others suggested that collecting the debt-to-income ratio was duplicative of other information included in the proposal, including denial reasons. In addition to general concerns about the proposed requirement, some commenters stated that reporting the debt-to-income ratio would be too burdensome for financial institutions. On the other hand, some industry commenters noted that the burden for reporting proposed Sec. 1003.4(a)(23) would be low because it requires reporting of the debt- to-income ratio relied on by the financial institution in making the credit decision instead of prescribing a specific calculation. A few industry commenters stated that they supported reporting the debt-to-income ratio relied on in making the credit decision, rather than requiring financial institutions to report a calculation prescribed by the Bureau. Other commenters urged the Bureau to require reporting of a specific debt-to-income ratio to increase the utility of the data. The Bureau concludes that including the debt-to-income ratio in the HMDA data will provide many benefits and further HMDA's purposes. The debt-to-income ratio will help identify potential patterns of discrimination. The Bureau understands that the debt-to-income ratio is only one factor in underwriting. Nonetheless, the debt-to-income ratio provides important information about the likelihood of default and about access to credit. Reporting debt-to-income information supplements the denial reason field in which financial institutions may indicate whether an application was denied due to the debt-to-income ratio. In addition to information about whether a loan was [[Page 66219]] denied due to the debt-to-income ratio, reporting the debt-to-income ratio will illuminate potential disparate treatment of similarly situated applicants. This information will help to better identify discriminatory practices, better understand whether lenders are meeting their obligations to serve the needs of the communities in which they operate, and, potentially, better target programs and investments to vulnerable borrowers. Requiring the financial institution to report the debt-to-income ratio relied on in making the credit decision would provide these benefits even though, as noted by industry commenters, the debt-to- income ratio is calculated differently depending on the loan product and lender. A prescribed debt-to-income calculation for HMDA purposes may allow for better comparison of debt-to-income information across the data. However, a prescribed calculation would significantly increase the burden associated with reporting the debt-to-income ratio. Therefore, the final rule, like the proposal, does not require a prescribed debt-to-income ratio calculation for HMDA purposes, and, instead, requires financial institutions to report the debt-to-income ratio relied on in making the credit decision. Some consumer advocate commenters urged the Bureau to collect additional information related to the mortgage payment-to-income ratio (front-end debt-to-income ratio). The front-end debt-to-income ratio differs from the information requested by proposed Sec. 1003.4(a)(23), which is commonly referred to as the back-end debt-to-income ratio, in that it, unlike the back-end debt-to-income ratio, does not include debts other than the mortgage debt in the debt-to-income ratio. As a result, the front-end debt-to-income ratio is a less complete measure of a borrower's ability to repay a loan and, accordingly, is a less important factor in underwriting decisions. In addition, using the reported income, discussed above in the section-by-section analysis of Sec. 1003.4(a)(10)(iii), and loan amount, discussed above in the section-by-section analysis of Sec. 1003.4(a)(7), it will be possible to calculate that ratio, if desired. For these reasons, the final rule does not require financial institutions to report the front-end debt- to-income ratio. Several industry commenters also raised concerns about the privacy implications of collecting and disclosing the applicant or borrower's debt-to-income ratio. See part II.B above for a discussion of the Bureau's approach to protecting applicant and borrower privacy with respect to the public disclosure of the data. Due to the significant benefits of collecting this information, the Bureau believes it is appropriate to collect the debt-to-income ratio despite the concerns raised by commenters about collecting this information. Some industry commenters urged the Bureau to exclude certain types of transactions (e.g., applications) or types of financial institutions (e.g., community banks) from the requirement to report the information required by proposed Sec. 1003.4(a)(23). In addition, some commenters believed that the proposal would require a financial institution to calculate a debt-to-income ratio for HMDA reporting purposes even if the financial institution did not calculate or use debt-to-income information in its credit decisions. Proposed Sec. 1003.4(a)(23) does not require reporting the debt- to-income ratio unless the financial institution has calculated and relied upon a debt-to-income ratio in evaluating an application. As discussed above, the debt-to-income ratio is an important aspect in underwriting and reporting this information will provide an important insight into an institution's credit decision. This information is particularly important when a financial institution denies an application due to the debt-to-income ratio. In addition, as discussed above, a financial institution is not required to report a debt-to- income ratio if it has not calculated the debt-to-income ratio for a particular application. The final rule does not require financial institutions to calculate debt-to-income ratios solely for HMDA reporting purposes. Therefore, the debt-to-income ratio should be reported for applications and originations if the ratio is calculated and relied on by the financial institution in making the credit decision. Other commenters explained that the debt-to-income information should not be reported for loans related to multifamily properties or loans to a trust because financial institutions do not calculate the debt-to-income ratio in making a credit decision on applications for those types of loans. Commenters explained that financial institutions usually consider the cash flow of the property, such as the debt service coverage ratio, rather than the income of the applicant when evaluating a multifamily loan or loan to a non-natural person. The Bureau understands that this cash flow analysis is different from the debt-to-income ratio. However, some commenters expressed uncertainty about whether financial institutions would be required to report the debt service coverage ratio or other cash flow analysis for loans to non-natural persons or for multifamily properties. To eliminate the confusion, the final rule will not require the financial institution to report the debt-to-income ratio for such loans. New comments 4(a)(23)-5 and -6 explain that a financial institution may report that the requirement does not apply if the applicant and co-applicant, if applicable, are not natural persons and for loans secured by, or proposed to be secured by, multifamily dwellings. In addition, the Bureau has excluded purchased covered loans from the requirements of Sec. 1003.4(a)(23). The Bureau does not believe that the debt-to-income ratio information is as valuable for purchased covered loans as for applications and originations. The debt-to-income ratio that the originating financial institution relied on in making the credit decision may no longer be accurate because a borrower's debts and incomes may have changed since origination. In addition, the Bureau believes that purchasing financial institutions may face practical challenges in ascertaining the debt-to-income ratio that the originating financial institution relied on in making the credit decision because it may not be evident on the face of the loan documents. In light of the limited value of the data and these practical challenges, the Bureau is excluding purchased covered loans from the requirements in Sec. 1003.4(a)(23). However, as discussed in comments 4(a)-2 through -4, a financial institution that reviews an application for a covered loan, makes a credit decision on that application prior to closing, and purchases the covered loan after closing will report the covered loan as an origination, not a purchase. In that case, the final rule requires the financial institution to report the debt-to-income ratio that it relied on in making the credit decision. Finally, an industry commenter also asked the Bureau to explain what a financial institution should report if it calculates more than one ratio in making the credit decision. The Bureau is finalizing proposed comment 4(a)(23)-1, which addresses the situation in which more than one ratio is used. If a financial institution calculated an applicant's or borrower's ratio more than one time, the financial institution reports the debt-to-income ratio relied on in making the credit decision. Final Rule Having considered the comments received and for the reasons discussed above, pursuant to its authority under [[Page 66220]] sections 305(a) and 304(b)(6)(J) of HMDA, the Bureau is finalizing Sec. 1003.4(a)(23) as proposed with technical modifications. In addition, the Bureau is finalizing proposed comments 4(a)(23)-1 through -4, as proposed, with the clarifying modifications discussed above and other technical modifications. Finally, the Bureau is finalizing new comments 4(a)(23)-5 through -7 to clarify when a financial institution is not required to report the applicant's or borrower's debt-to-income ratio. In addition, proposed Sec. 1003.4(a)(23) excluded reverse mortgages from the requirement to report the debt-to-income ratio. The Bureau is removing that exclusion from the final rule. The Bureau included that exclusion because it understood that financial institutions historically did not consider income or debt-to-income information when evaluating applications for reverse mortgages. HUD recently changed its guidelines for evaluating reverse mortgages for participation in the Home Equity Conversion Mortgage (HECM) program, which currently accounts for the majority of the reverse mortgage market.\334\ These revised guidelines include consideration of some income information.\335\ Currently, the revised standards do not contemplate calculation of a debt-to-income ratio. However, it is possible that in the future these guidelines or other underwriting standards applicable to reverse mortgages may include the consideration of a debt-to-income ratio. Therefore, the final rule removes the exclusion for reverse mortgages from Sec. 1003.4(a)(23). The Bureau anticipates that this information will not be reported for most reverse mortgages because an institution is only required to report the debt- to-income ratio if it relies on it in making a credit decision and institutions do not typically rely on a debt-to-income ratio in making a credit decision on a reverse mortgage. --------------------------------------------------------------------------- \334\ U.S. Dept. of Housing and Urban Dev., Mortgagee Letter 2014-22, HECM Fin. Assessment and Property Charge Requirements, available at http://portal.hud.gov/hudportal/documents/huddoc?id=14-22ml.pdf. \335\ Id. at 33. --------------------------------------------------------------------------- 4(a)(24) Currently, neither HMDA nor Regulation C contains requirements regarding loan-to-value ratio. Section 304(b) of HMDA permits the disclosure of such other information as the Bureau may require.\336\ The Bureau proposed Sec. 1003.4(a)(24), which requires financial institutions to report the ratio of the total amount of debt secured by the property to the value of the property. The ratio of total amount of secured debt to the value of the property securing the debt is generally referred to as the combined loan-to-value (CLTV) ratio. --------------------------------------------------------------------------- \336\ See Dodd-Frank Act section 1094(3)(A)(iv). --------------------------------------------------------------------------- The Bureau proposed two different calculations for CLTV--one calculation for a covered loan that is a home-equity line of credit and another calculation for a covered loan that is not a home-equity line of credit. Specifically, the Bureau proposed Sec. 1003.4(a)(24)(i), which provides that, for a covered loan that is a home-equity line of credit, the CLTV ratio shall be determined by dividing the sum of the unpaid principal balance of the first mortgage, the full amount of any home-equity line of credit (whether drawn or undrawn), and the balance of any other subordinate financing by the property value identified in proposed Sec. 1003.4(a)(28). As to a covered loan that is not a home- equity line of credit, the Bureau proposed Sec. 1003.4(a)(24)(ii), which provides that the CLTV ratio shall be determined by dividing the combined unpaid principal balance amounts of the first and all subordinate mortgages, excluding undrawn home-equity lines of credit amounts, by the property value identified in proposed Sec. 1003.4(a)(28). In addition, the Bureau proposed instruction 4(a)(24)-1, which directs financial institutions to enter the CLTV ratio applicable to the property to two decimal places, and if the CLTV ratio is a figure with more than two decimal places, directs institutions to truncate the digits beyond two decimal places. The Bureau also proposed instruction 4(a)(24)-2, which provides technical instructions for covered loans in which no combined loan-to-value ratio is calculated. The Bureau also proposed three comments to clarify this reporting requirement. Proposed comment 4(a)(24)-1 clarifies that, if a financial institution makes a credit decision without calculating the combined loan-to-value ratio, the financial institution complies with Sec. 1003.4(a)(24) by reporting that no combined loan-to-value ratio was calculated in connection with the credit decision. Proposed comment 4(a)(24)-2 describes the CLTV calculation for home-equity lines of credit proposed in Sec. 1003.4(a)(24)(i) and provides illustrative examples. Proposed comment 4(a)(24)-3 describes the CLTV calculation for transactions that are not home-equity lines of credit proposed in Sec. 1003.4(a)(24)(ii) and provides illustrative examples. The Bureau solicited feedback regarding whether proposed Sec. 1003.4(a)(24) is appropriate generally. Most commenters that provided feedback on proposed Sec. 1003.4(a)(24) supported the Bureau's proposal. For example, one consumer advocate commenter stated that the CLTV ratio provides the most accurate calculation of borrower equity and is therefore most relevant to assess the credit risk of the loan. Another consumer advocate commenter pointed out that CLTV ratio data provides important information regarding both an individual property's leverage and the general level of leverage in specific geographic locations, and noted that areas in which many properties are highly leveraged are especially vulnerable to changes in economic conditions. Another consumer advocate commenter suggested that CLTV ratio data is vital to determining whether particular financial institutions are making loans with high CLTV ratios on a census tract level. Some industry commenters also supported the Bureau's proposal. For example, as with credit score data, one industry commenter stated that for purposes of fair lending analysis, CLTV is crucial to understanding a financial institution's credit and pricing decision and that without such information, inaccurate conclusions may be reached by users of HMDA data. In contrast, several industry commenters opposed the Bureau's proposal to require reporting of CLTV. For example, some industry commenters stated that the proposed requirement is an unnecessary burden on financial institutions since loan-to-value ratio may be calculated using the Bureau's proposed property value data and the loan amount data that the regulation already requires. These commenters explained that while the proposed CLTV requirement would provide the ratio of the total amount of debt secured by the property to the value of the property, they believe the additional burden placed on financial institutions by this new reporting requirement outweighs any added value to data users. The Bureau has considered this feedback and determined that CLTV ratio data would improve the HMDA data's usefulness. CLTV ratio is a standard underwriting factor regularly calculated by financial institutions, both for a financial institution's own underwriting purposes and to satisfy investor requirements. For a particular transaction in which a CLTV ratio is not calculated or considered during the underwriting process, the Bureau is adopting a new comment, discussed further below, which permits financial institutions to report that the requirement is not applicable if the [[Page 66221]] financial institution did not rely on the CLTV ratio in making the credit decision. The Bureau believes that the CLTV ratio is an important factor both in the determination of whether to extend credit and for the pricing terms upon which credit would be extended. Consequently, the Bureau is adopting proposed Sec. 1003.4(a)(24), modified as discussed further below. The Bureau has determined to exclude purchased covered loans from the requirements of Sec. 1003.4(a)(24). The Bureau does not believe that the combined-loan-to-value ratio information is as valuable for purchased covered loans as for applications and originations. The combined-loan-to-value ratio that the originating financial institution relied on in making the credit decision may no longer be accurate, because the total amount of debt secured by the property to the value of the property likely has changed since origination. In addition, the Bureau believes that purchasing financial institutions may face practical challenges in ascertaining the combined-loan-to-value ratio that the originating financial institution relied on in making the credit decision because it may not be evident on the face of the loan documents. In light of the limited value of the data and these practical challenges, the Bureau is excluding purchased covered loans from the requirements in Sec. 1003.4(a)(24). However, as discussed in comment 4(a)-3, a financial institution that reviews an application for a covered loan, makes a credit decision on that application prior to closing, and purchases the covered loan after closing will report the covered loan that it purchases as an origination, not a purchase. In that case, the final rule requires the financial institution to report the combined-loan-to-value ratio that it relied on in making the credit decision. The Bureau solicited feedback regarding whether the proposed alignment to the MISMO data standards for CLTV is appropriate and whether the text of this proposed requirement should be clarified. Consistent with the Small Business Review Panel's recommendation, the Bureau also solicited feedback regarding whether it would be less burdensome for small financial institutions to report the combined loan-to-value relied on in making the credit decision, or if it would be less burdensome to small financial institutions for the Bureau to adopt a specific combined loan-to-value ratio calculation as proposed under Sec. 1003.4(a)(24). Several commenters did not support the Bureau's proposal to align with the MISMO data standards and require two different CLTV calculations depending on whether or not the transaction is a home- equity line of credit. Both consumer advocates and industry were concerned with the proposed requirement to calculate CLTV ratio one way for home-equity lines of credit but another way for non-home-equity lines of credit. Several commenters did not support the Bureau's proposed CLTV calculations under proposed Sec. 1003.4(a)(24), which requires that the full amount of a home-equity line of credit be included in the CLTV calculation for a covered loan that is a home- equity line of credit, whether it is drawn or not, but that for transactions that are not home-equity lines of credit, only the outstanding amount of any home-equity line of credit should be included. One industry commenter noted that it calculates the CLTV ratio for a covered loan that is not a home-equity line of credit by including the total amount of home-equity lines of credit (and does not exclude ``undrawn'' home-equity lines of credit as required under the Bureau's proposal). One consumer advocate commenter recommended that the transactions should be treated identically by requiring the full amount be included in the CLTV calculation since the entire amount of a home-equity line of credit available to the borrower constitutes potential leverage of the property in either situation. Similarly, another consumer advocate commenter suggested that loan-to-value calculations involving home- equity lines of credit should always use the full amount of credit available to the borrower because the borrower has access to the full line of credit without any additional underwriting by the financial institution and thus a loan-to-value calculation that ignores the undrawn amount will be unreliable for purposes of analysis. This same commenter stated that the Bureau's desire to align with the MISMO data standards does not justify the adoption of inferior CLTV measurements. Lastly, in order to address the burden that results from requiring different CLTV ratio calculations based on the type of transaction, industry commenters also recommended that the Bureau allow for consistent treatment of outstanding lines of credit, regardless of the loan type being originated. The Bureau has considered this feedback and acknowledges that CLTV ratio calculations on home-equity lines of credit may vary between financial institutions. The Bureau has determined that having two different methods of calculating CLTV--one calculation for a covered loan that is a home-equity line of credit and another calculation for a covered loan that is not a home-equity line of credit--is unduly burdensome on financial institutions. The Bureau has also determined that it would be less burdensome for financial institutions to report the CLTV relied on in making the credit decision. Consequently, the Bureau will not adopt Sec. 1003.4(a)(28) as proposed. Instead, the Bureau is adopting a modified Sec. 1003.4(a)(28), which requires a financial institution to report the ratio of the total amount of debt secured by the property to the value of the property relied on in making the credit decision. As discussed in the proposal, the Bureau is generally concerned about the potential burden associated with reporting calculated data fields, such as the CLTV ratio. Some commenters noted that consistency in the rounding method for all relevant HMDA data will lead to more accurate reporting. A few industry commenters stated that the proposal presented a confusing rounding process that is not intuitive and differs depending on the data point being reported. For example, one commenter suggested that rather than the requirement to truncate any digits beyond the first two decimal places, proposed instruction 4(a)(24)-1 should be adjusted to read that a CLTV ratio be rounded up if the third digit behind the decimal is 5 or larger, and rounded down if the digit is 4 or smaller. The commenter stated that current underwriting systems such as Fannie Mae's Desktop Underwriter use this method and that unnecessary errors can be expected if the CLTV instructions are finalized as proposed. The Bureau acknowledges that the CLTV reporting requirement in proposed instruction 4(a)(24)-1 may have posed some challenges for financial institutions. The Bureau has considered the feedback and believes that the proposed CLTV reporting requirement may be unduly burdensome on financial institutions. Consequently, the Bureau is not adopting the proposed CLTV reporting requirement in the final rule. The Bureau is adopting a modified Sec. 1003.4(a)(24), which requires reporting of the CLTV that a financial institution relied on in making the credit decision and excludes reporting of CLTV for purchased covered loans. In order to align with the new reporting requirement, the Bureau will not adopt comments 4(a)(24)-1, -2, and -3 as proposed, and adopts new comments 4(a)(24)-1, -2, -3, -4, and -5. [[Page 66222]] The Bureau is adopting new comment 4(a)(24)-1, which explains that Sec. 1003.4(a)(24) requires a financial institution to report the CLTV ratio relied on in making the credit decision and provides an illustrative example. The example provides that if a financial institution calculated a CLTV ratio twice--once according to the financial institution's own requirements and once according to the requirements of a secondary market investor--and the financial institution relied on the CLTV ratio calculated according to the secondary market investor's requirements in making the credit decision, Sec. 1003.4(a)(24) requires the financial institution to report the CLTV ratio calculated according to the requirements of the secondary market investor. The Bureau is adopting new comment 4(a)(24)-2, which explains that a financial institution relies on the total amount of debt secured by the property to the value of the property (CLTV ratio) in making the credit decision if the CLTV ratio was a factor in the credit decision even if it was not a dispositive factor. For example, if the CLTV ratio is one of multiple factors in a financial institution's credit decision, the financial institution has relied on the CLTV ratio and complies with Sec. 1003.4(a)(24) by reporting the CLTV ratio, even if the financial institution denies the application because one or more underwriting requirements other than the CLTV ratio are not satisfied. The Bureau is adopting new comment 4(a)(24)-3, which explains that a financial institution should report that the requirement is not applicable for transactions in which a credit decision was not made and provides illustrative examples. The comment provides that if a file was closed for incompleteness, or if an application was withdrawn before a credit decision was made, a financial institution complies with Sec. 1003.4(a)(24) by reporting that the requirement is not applicable, even if the financial institution had calculated the CLTV ratio. The Bureau is adopting new comment 4(a)(24)-4, which explains that a financial institution should report that the requirement is not applicable for transactions in which no CLTV ratio was relied on in making the credit decision. The comment provides that Sec. 1003.4(a)(24) does not require a financial institution to calculate the CLTV ratio, nor does it require a financial institution to rely on a CLTV ratio in making a credit decision. The comment clarifies that if a financial institution makes a credit decision without relying on a CLTV ratio, the financial institution complies with Sec. 1003.4(a)(24) by reporting that the requirement is not applicable since no CLTV ratio was relied on in connection with the credit decision. Lastly, the Bureau is adopting new comment 4(a)(24)-5, which explains that a financial institution complies with Sec. 1003.4(a)(24) by reporting that the reporting requirement is not applicable when the covered loan is a purchased covered loan. The Bureau believes that comments 4(a)(24)-1, -2, -3, -4, and -5 will provide clarity regarding the new reporting requirement adopted in Sec. 1003.4(a)(24) and will facilitate HMDA compliance. The Bureau believes that requiring financial institutions to collect information regarding CLTV ratios is necessary to carry out HMDA's purposes, such as helping to ensure that the citizens and public officials of the United States are provided with sufficient information to enable them to determine whether depository institutions are filling their obligations to serve the housing needs of the communities and neighborhoods in which they are located and assist public officials in their determination of the distribution of public sector investments in a manner designed to improve the private investment environment. CLTV ratios are a significant factor in the underwriting process and provide valuable insight into both the stability of community homeownership and the functioning of the mortgage market. Accordingly, pursuant to its authority under sections 305(a) and 304(b)(6)(J) of HMDA, the Bureau is adopting Sec. 1003.4(a)(24), which requires, except for purchased covered loans, reporting of the CLTV that a financial institution relied on in making the credit decision. 4(a)(25) HMDA section 304(b)(6)(D) requires, for loans and completed applications, that financial institutions report the actual or proposed term in months of the mortgage loan.\337\ Currently, Regulation C does not require financial institutions to report information regarding the loan's term. The Bureau proposed to implement HMDA section 304(b)(6)(D) by requiring in Sec. 1003.4(a)(25) that financial institutions collect and report data on the number of months until the legal obligation matures for a covered loan or application. For the reasons discussed below, the Bureau is finalizing Sec. 1003.4(a)(25) substantially as proposed. --------------------------------------------------------------------------- \337\ Dodd-Frank Act section 1094(3)(A)(iv). --------------------------------------------------------------------------- The Bureau solicited feedback on what method of reporting loan term would minimize the burden on small institutions while still meeting the Dodd-Frank Act reporting requirements and purposes of HMDA. Several commenters opposed the Bureau's proposal and suggested that reporting the loan term, along with other proposed data points specific to applicant or borrower and property characteristics, could create privacy risks. One commenter stated that it would be difficult to retain borrower and lender privacy in transactions that involve multifamily loans because there are a limited number of transactions in a geographic area. The Bureau has considered this feedback. See part II.B above for a discussion of the Bureau's approach to protecting applicant and borrower privacy with respect to the public disclosure of the HMDA data. One commenter stated that collecting data on the loan term is appropriate for closed-end loans but would create burdensome programming demands if it became a requirement for open-end credit. As the Bureau explained in the proposal, the length of time a borrower has to repay a loan is an important feature for borrowers and creditors. With this information, borrowers are able to determine the amount due with each payment, which could significantly influence their ability to afford the loan. Creditors, on the other hand, can use loan term as a factor in assessing interest rate risk, which in turns, affects loan pricing. The Bureau believes that the benefit of the information that the loan term could provide, including loan terms on open-end lines of credit, justifies the burden because this information could help explain pricing or any other differences that are indiscernible with current HMDA data. A few commenters suggested that the loan term should be reported consistent with the loan term disclosed under TILA-RESPA, which provides under Regulation Z Sec. 1026.37(a)(8) that the term to maturity should be disclosed in years or months or both.\338\ Although consistency with TILA-RESPA might mitigate burden if the creditor disclosing the loan term under TILA-RESPA elects to disclose term to maturity in months instead of years or years plus the remaining months, the Bureau believes that a reasonable interpretation of HMDA section 304(b)(6)(D) is that financial institutions [[Page 66223]] should report the actual or proposed term for a loan or application in months. Another commenter stated that reporting loan term can be confusing on loans with unusual terms, such as those with terms that are not in whole months. Proposed comment 4(a)(25)-2 clarified that for covered loans with non-monthly repayment schedules, the loan term should be in months and not include any fractional months remaining. This guidance, for which the Bureau did not receive any comments, should facilitate compliance for loans with repayment schedules that are measured in units of time other than months. --------------------------------------------------------------------------- \338\ See 78 FR 79730 (Dec. 31, 2013). The rule is effective on October 3, 2015 and applies to transactions for which the creditor or mortgage broker receives an application on or after that date. --------------------------------------------------------------------------- Several other commenters supported the Bureau's proposal to include the loan term. One commenter that supported the Bureau's proposal stated that it is very useful, particularly given the risk maturity premium for longer term loans. Moreover, researchers would be able to examine whether a concentration of shorter term loans can lead to a more stable housing market. The Bureau concludes that the information that could be provided by loan terms will help determine whether financial institutions are serving the housing needs of their communities and assist in identifying possible discriminatory lending patterns and enforcing antidiscrimination statutes by allowing information about similar loans to be compared and analyzed appropriately. Accordingly, to implement HMDA section 304(b)(6)(D), the Bureau is adopting Sec. 1003.4(a)(25) substantially as proposed with minor wording changes and is also adopting as proposed comments 4(a)(25)-1 and -2. In addition, the Bureau is adopting a few comments that incorporate material contained in proposed appendix A into the commentary to Sec. 1003.4(a)(25) because of the removal of appendix A as discussed in the section-by- section analysis of appendix A below. These comments 4(a)(25)-3 through 4(a)(25)-5 primarily incorporate proposed appendix A instructions that do not contain any substantive changes from the proposed reporting requirements. 4(a)(26) HMDA section 304(b)(6)(B) requires the reporting of the actual or proposed term in months of any introductory period after which the rate of interest may change.\339\ Currently, Regulation C does not require financial institutions to report information regarding the numbers of months until the first interest rate adjustment. The Bureau proposed to implement HMDA section 304(b)(6)(B) by requiring in Sec. 1003.4(a)(26) that financial institutions collect and report data on the number of months until the first date the interest rate may change after loan origination. The Bureau also proposed that Sec. 1003.4(a)(26) would apply regardless of how the interest rate adjustment is characterized by product type, such as adjustable rate, step rate, or another type of product with a ``teaser'' rate. For the reasons discussed below, the Bureau is adopting Sec. 1003.4(a)(26) generally as proposed. --------------------------------------------------------------------------- \339\ Dodd-Frank Act section 1094(3)(A)(iv). --------------------------------------------------------------------------- The Bureau solicited feedback on what method of reporting initial interest rate period would minimize burden on small financial institutions while still meeting the Dodd-Frank Act reporting requirements and purposes of HMDA. Several commenters supported the Bureau's proposal to collect data about introductory terms. One commenter stated that along with other data points, the introductory rate period will enable accurate analyses and a full understanding of the extent of the terms to which residents have access to credit. The Bureau finds these reasons compelling in finalizing Sec. 1003.4(a)(26). As the Bureau explained in the proposal, interest rate variability can be an important feature in affordability. In addition, having information about introductory rates will enable better analyses of loans and applications, which could be used to identify possible discriminatory lending patterns. One commenter pointed out that the Bureau's proposal to report the number of months until the first date the interest rate may change after origination is a measure different from Regulation Z Sec. 1026.43(e)(2)(iv)(A), which measures the interest rate change from the date the first regular periodic payment is due. This commenter suggested that the measure for the introductory term for HMDA reporting should be consistent with the measure prescribed by Regulation Z Sec. 1026.43(e)(2)(iv)(A), which relates to the underwriting of a qualified mortgage adopted under the Bureau's 2013 ATR Final Rule. Section 1026.43(e)(2)(iv)(A) provides that a qualified mortgage under Sec. 1026.43(e)(2) must be underwritten, taking into account any mortgage- related obligations, using the maximum interest rate that may apply during the first five years after the date on which the first regular periodic payment will be due. As stated in the Bureau's 2013 ATR Final Rule, the Bureau believes that the approach of requiring creditors to underwrite a loan based on the maximum interest rate that applies during the first five years after the first regular periodic payment due date provides greater protections to consumers and is also consistent with Regulation Z disclosure requirements for interest rates on adjustable-rate amortizing loans.\340\ The Bureau, however, believes that a reasonable interpretation of HMDA section 304(b)(6)(B) requires the reporting of the number of months after a loan origination until the first instance of an interest rate changes or for a loan application, the proposed number of months until the first instance of an interest rate change. Accordingly, the Bureau is adopting Sec. 1003.4(a)(26) generally as proposed but is modifying the scope of the provision to include applications. The Bureau is also adopting comments 4(a)(26)-1 and -2 generally as proposed, but with minor modifications for clarification. In addition, because appendix A will be deleted as discussed in the section-by-section analysis of appendix A below, the Bureau is adopting new comments 4(a)(26)-3 and -4 to incorporate instructions in proposed appendix A. New comments 4(a)(26)-3 and -4 to incorporate proposed instructions in appendix A. New comment 4(a)(26)-3 specifies that a financial institution reports that the requirement to report the introductory rate period is not applicable when the transaction involves a fixed rate covered loan or an application for a fixed rate covered loan. Similarly, new comment 4(a)(26)-4 specifies that a financial institution reports that the requirement to report the introductory rate period is not applicable if the transaction involves a purchased fixed rate covered loan. --------------------------------------------------------------------------- \340\ 78 FR 6407, 6521 (Jan. 30, 2013). --------------------------------------------------------------------------- 4(a)(27) HMDA section 304(b)(6)(C) requires reporting of the presence of contractual terms or proposed contractual terms that would allow the mortgagor or applicant to make payments other than fully amortizing payments during any portion of the loan term.\341\ Current Regulation C does not require financial institutions to report whether a loan allows or would have allowed the borrower to make payments other than fully amortizing payments. The Bureau believes it is reasonable to interpret HMDA section 304(b)(6)(C) to require reporting non-amortizing features by identifying specific, well-defined non-amortizing loan features. Thus, the Bureau proposed to implement HMDA section 304(b)(6)(C) by requiring the reporting non-amortizing features, including balloon payments, interest only payments, and negative [[Page 66224]] amortizations. Proposed Sec. 1003.4(a)(27) requires reporting balloon payments, as defined by 12 CFR 1026.18(s)(5)(i); interest only payments, as defined by 12 CFR 1026.18(s)(7)(iv); a contractual term that could cause the loan to be a negative amortization loan, as defined by 12 CFR 1026.18(s)(7)(v); or any other contractual term that would allow for payments other than fully amortizing payments, as defined by 12 CFR 1026.43(b)(2). For the reasons discussed below, the Bureau is finalizing Sec. 1003.4(a)(27) as proposed. --------------------------------------------------------------------------- \341\ Dodd-Frank Act section 1094(3)(A)(iv). --------------------------------------------------------------------------- The Bureau solicited feedback on what method of report non- amortizing features would minimize the burden on small financial institutions but still meet the reporting requirements of the Dodd- Frank Act and the purposes of HMDA. Most commenters, however, supported the proposal to collect non-amortizing features without modification. They stated that the data will indicate whether a high incidence of these features, particularly in loans to vulnerable and underserved populations, is a cause for concern that requires intervention. For the same reason, the Bureau believes that the reporting of non-amortizing features is helpful and can provide insight into lending activity that features these loans. It will provide data about the types of loans that are being made and assist in identifying possible discriminatory lending patterns and enforce antidiscrimination statutes. A few commenters did not support the Bureau's proposal to require the reporting of non-amortizing features. A financial institution commenter stated that it does not originate loans with risky features and opined that most small institutions probably do not originate such loans either. The Bureau recognizes that loans with non-amortizing features may be rare today. However, such features that may not be present in certain markets today may arise at a later time. Given the risk of payment shock with such products, the Bureau proposed Sec. 1003.4(a)(27)(iv) to ensure the data includes information about non- amortizing products. Furthermore, during the SBREFA process, small entity representatives informed the Bureau that information regarding non-amortizing features of a loan is currently collected by financial institutions. Based on this information, the Bureau concludes that at least some small institutions originate loans that contain non- amortizing features. Additionally, commenters that opposed the reporting of non- amortizing features reasoned that such information is not helpful and may not even be pertinent to most underwriting and pricing decisions. The Bureau explained in the proposal that non-amortizing features were a rarity but then became more common in the lead-up to the mortgage crisis. These features could be pertinent to underwriting and pricing decisions because of the nature of the risk they pose on the borrower. One commenter stated that HMDA reporters will experience confusion when multiple loan features apply and create difficulties in developing new products. The proposal and the final rule address this concern by aligning the definitions of non-amortizing features for HMDA purposes with existing definitions in Regulation Z. This alignment will facilitate compliance and reduce potential implementation and compliance difficulties. Accordingly, to implement HMDA section 304(b)(6)(C), the Bureau is finalizing Sec. 1003.4(a)(27) as proposed and is making minor technical amendments and wording changes to the commentary to Sec. 1003.4(a)(27). Data about non-amortizing features will help determine whether financial institutions are serving the housing needs of their communities and assist in identifying possible discriminatory lending patterns and enforcing antidiscrimination statutes by allowing information about similar loans to be compared and analyzed appropriately. 4(a)(28) Regulation C does not require financial institutions to report information regarding the value of the property that secures or will secure the loan. HMDA section 304(b)(6)(A) requires the reporting of the value of the real property pledged or proposed to be pledged as collateral.\342\ The Bureau proposed Sec. 1003.4(a)(28), which implements this requirement by requiring financial institutions to report the value of the property securing the covered loan or, in the case of an application, proposed to secure the covered loan relied on in making the credit decision. The Bureau proposed a new technical instruction in appendix A for reporting the property value relied on in dollars. In addition, in order to provide clarity on proposed Sec. 1003.4(a)(28), the Bureau proposed new illustrative comments 4(a)(28)-1 and -2. --------------------------------------------------------------------------- \342\ Dodd-Frank Act section 1094(3)(A)(iv), 12 U.S.C. 2803(b)(6)(A). --------------------------------------------------------------------------- The Bureau solicited feedback on which property value should be reported. Several commenters, including both industry and consumer advocates, supported the Bureau's proposal to implement the Dodd-Frank Act requirement regarding property value by requiring reporting of the value of the property relied on in making the credit decision in dollars. Other commenters suggested different approaches to collecting property value. One consumer advocate commenter suggested that the Bureau require financial institutions to report the purchase price of the property in all circumstances. Another industry commenter suggested that financial institutions be required to report the final property value determined by the loan underwriter and used in the investment decision. The Bureau believes that financial institutions should report the value relied on in making the credit decision. Thus, if the financial institution relied upon the purchase price in making the credit decision, the financial institution would report that value. If the final property value determined by a loan underwriter and used in the financial institution's investment decision is the property value that the institution relied on in making the credit decision, then reporting that property valuation will comply with Sec. 1003.4(a)(28). To this end, comment 4(a)(28)-1 explains, if a financial institution relies on an appraisal or other valuation for the property in calculating the loan-to-value ratio, it reports that value; if the institution relies on the purchase price of the property in calculating the loan-to-value ratio, it reports that value. A national trade association commenter requested that the Bureau clarify that if an application is withdrawn or is closed for incompleteness, a financial institution may report that the requirement is not applicable since there was no reliance on property value in making the credit decision. In order to help facilitate HMDA compliance by providing additional guidance regarding the property value reporting requirement, the Bureau is adopting new comment 4(a)(28)-3, which clarifies how a financial institution complies with Sec. 1003.4(a)(28) by reporting that the requirement is not applicable for transactions for which no credit decision was made. New comment 4(a)(28)-3 clarifies that if a file was closed for incompleteness or the application was withdrawn before a credit decision was made, the financial institution complies with Sec. 1003.4(a)(28) by reporting that the requirement is not applicable, even if the financial institution had obtained a property value. Two State trade association commenters expressed concern that proposed Sec. 1003.4(a)(28) compels a [[Page 66225]] financial institution to obtain an appraisal even when a property valuation is not in fact required for the underwriting process of a particular transaction or is not required per regulations. In order to address this concern, the Bureau is adopting new comment 4(a)(28)-4, which clarifies that Sec. 1003.4(a)(28) does not require a financial institution to obtain a property valuation, nor does it require a financial institution to rely on a property value in making a credit decision. Comment 4(a)(28)-4 explains that if a financial institution makes a credit decision without relying on a property value, the financial institution complies with Sec. 1003.4(a)(28) by reporting that the requirement is not applicable since no property value was relied on in connection with the credit decision. A consumer advocate commenter suggested that the Bureau require reporting of property value if a valuation was performed and even if the property valuation was not relied on in making the credit decision. The Bureau is not adopting this recommendation in the final rule. The Bureau believes that the property value relied on will be more useful in understanding a financial institution's credit decision and other HMDA data, such as pricing information. The proposed standard in Sec. 1003.4(a)(28) requires a financial institution to report the property value relied on in making the credit decision. As explained in new comments 4(a)(28)-3 and -4, if a financial institution has not made a credit decision or has not relied on property value in making the credit decision, the financial institution complies with Sec. 1003.4(a)(28) by reporting that the requirement is not applicable. The Bureau has determined that this is the appropriate approach for purposes of HMDA compliance. One State trade association commenter recommended that property value be reported in ranges rather than the actual value to better protect the privacy of applicants. While reporting property value in ranges may address some of the privacy concerns raised by commenters, the Bureau has determined that requiring reporting of the value of the property relied on in making the credit decision in dollars is the more appropriate approach. When coupled with Sec. 1003.4(a)(7), which requires a financial institution to report the exact loan amount, a requirement to report the property value relied on in dollars under Sec. 1003.4(a)(28) will allow the calculation of loan-to-value ratio, an important underwriting variable. Reporting property value in ranges would render these calculations less precise, undermining their utility for data analysis. A few commenters were concerned that if information regarding property value is made available to the public, such information could be coupled with other publicly available information on property sales and ownership records to compromise a borrower's privacy. The Bureau has considered this feedback. See part II.B above for a discussion of the Bureau's approach to protecting applicant and borrower privacy with respect to the public disclosure of HMDA data. Several commenters, including both industry and consumer advocates, supported the Bureau's proposal to implement the Dodd-Frank Act requirement regarding property value by requiring reporting of the value of the property relied on in making the credit decision in dollars. As discussed above, knowing the property value in addition to loan amount allows HMDA users to estimate the loan-to-value ratio, which measures a borrower's equity in the property and is a key underwriting and pricing criterion. In addition, requiring financial institutions to report information about property value will enhance the utility of HMDA data. Property value data will further HMDA's purposes by providing the public and public officials with data to help determine whether financial institutions are serving the housing needs of their communities by providing information about the values of properties that are being financed; it will also assist public officials in distributing public-sector investment so as to attract private investment by providing information about property values; and it will assist in identifying possible discriminatory lending patterns and enforcing antidiscrimination statutes by allowing information about similar loans to be compared and analyzed appropriately. Moreover, for the reasons given in the section-by-section analysis of Sec. 1003.4(a)(29), the Bureau believes that implementing HMDA through Regulation C to treat mortgage loans secured by all manufactured homes consistently, regardless of legal classification under State law, is reasonable, and is necessary and proper to effectuate HMDA's purposes and facilitate compliance therewith. Accordingly, pursuant to its authority under HMDA sections 305(a) and 304(b)(6)(A), the Bureau is adopting Sec. 1003.4(a)(28) as proposed, with several technical and clarifying modifications to proposed comments 4(a)(28)-1 and -2. In addition, as discussed above, the Bureau is adopting new comments 4(a)(28)-3 and -4, which will help facilitate HMDA compliance by providing additional guidance regarding the property value reporting requirement. 4(a)(29) Section 304(b) of HMDA permits disclosure of such other information as the Bureau may require. The Bureau proposed Sec. 1003.4(a)(29), which required that financial institutions report whether a manufactured home is legally classified as real property or as personal property. For the reasons discussed below, the Bureau is adopting Sec. 1003.4(a)(29) with modifications, to require financial institutions to report whether a covered loan or application is or would have been secured by a manufactured home and land or a manufactured home and not land. Since 1988, Regulation C has required reporting of home purchase and home improvement loans and refinancings related to manufactured homes, whether or not the homes are considered real property under State law.\343\ Manufactured homes serve vital housing needs in communities and neighborhoods throughout the United States. For example, manufactured housing is the largest unsubsidized source of affordable homeownership in the United States.\344\ Manufactured homes also often share certain essential financing features with non- manufactured homes. But classifications of manufactured homes as real or personal property vary significantly among States and can be ambiguous.\345\ --------------------------------------------------------------------------- \343\ 53 FR 31683, 31685 (Aug. 19, 1988). \344\ Ann M. Burkhart, Bringing Manufactured Housing into the Real Estate Finance System, 37 Pepperdine Law Review 427, 428 (2010), available at http://digitalcommons.pepperdine.edu/cgi/viewcontent.cgi?article=1042&context=plr. \345\ See James M. Milano, An Overview and Update on Legal and Regulatory Issues in Manufactured Housing Finance, 60 Consumer Financial Law Quarterly Report 379, 383 (2006); Burkhart, supra note 344, at 430. --------------------------------------------------------------------------- Regulation C's consistent treatment of manufactured housing in HMDA data has proven important to furthering HMDA's purposes and provided communities and public officials with important information about manufactured housing lending.\346\ The [[Page 66226]] Bureau believes that the unique nature of the manufactured home financing market warrants additional information reporting. Although in many respects manufactured and site built housing are similar, manufactured home financing reflects certain key differences as compared to site built home financing. State laws treat site built homes as real property, with financing secured by a mortgage or deed of trust. On the other hand State law may treat manufactured homes as personal property or real property depending on the circumstances.\347\ Manufactured home owners may own or rent the underlying land, which is an additional factor in manufactured home owners' total housing cost and can be relevant to financing.\348\ --------------------------------------------------------------------------- \346\ Adam Rust & Peter Skillern, Community Reinvestment Association of North Carolina, Nine Myths of Manufactured Housing: What 2004 HMDA Data says about a Misunderstood Sector (2006), available at http://www.reinvestmentpartners.org/sites/reinvestmentpartners.org/files/Myths-and-Realities-of-Manufactured-Housing.pdf; Delaware State Housing Authority, Manufactured Housing in Delaware: A Summary of Information and Issues (2008), available at http://www.destatehousing.com/FormsAndInformation/Publications/manu_homes_info.pdf. \347\ Milano, supra note 345 at 380. \348\ William Apgar et al., An Examination of Manufactured Housing Community- and Asset-Building Strategies, at 5 (Neighborhood Reinvestment Corporation Report to the Ford Foundation, Working Paper No. W02-11, 2002), available at http://www.jchs.harvard.edu/research/publications/examination-manufactured-housing-community-and-asset-building-strategy. --------------------------------------------------------------------------- Many consumer advocate commenters supported the proposed requirement. Some argued, however, that additional information about whether the covered loan was secured by both the manufactured home and land or the manufactured home alone would be valuable in addition to the manufactured home's classification under State law, to distinguish covered loans in States where manufactured homes may be classified as real property even if the home is sited on leased land. Many industry commenters opposed the proposed requirement as burdensome. However, one industry commenter supported the requirement and stated that it had been subject to a fair lending review that would have been unnecessary if the HMDA data had differentiated between land-and-home and home-only manufactured home loans. A few industry commenters stated that in some circumstances financial institutions secure loans using multiple methods to perfect a lien under both State real property and personal property law because of secondary market standards or prudence. Other commenters argued that State law can be difficult to understand and that the proposed requirement would therefore be difficult to comply with and create the risk that the financial institution would be cited for incorrectly stating the legal classification. Some commenters noted that the legal classification may change after the closing date of the loan. Some industry commenters argued that the proposed requirement did not accurately reflect pricing distinctions made by manufactured housing lenders because pricing is based primarily on whether the security interest will cover both the land and home or the home only, regardless of State law classification. One commenter stated that the proposed requirement is relevant only to individual manufactured home loans, and not loans secured by manufactured home communities. The Bureau understands that the proposed requirement may pose reporting challenges because of multiple methods of lien perfection and the complexity of and differences among State laws. However, information about manufactured home loan classification is valuable because there are material differences in types of manufactured home financing related to rate, term, origination costs, legal requirements, and consumer protections. These differences are discussed in the Bureau's white paper on Manufactured Housing Consumer Finance in the United States.\349\ Furthermore, capturing the pricing distinction between types of manufactured home loans is important to facilitate fair lending analyses. Section 1003.4(a)(29) will provide necessary insight into this loan data and allow it to be used to help determine whether financial institutions are serving the housing needs of their communities, assist in identifying possible discriminatory lending patterns and enforcing antidiscrimination statutes, and, potentially, assist public officials in public-sector investment determinations.\350\ --------------------------------------------------------------------------- \349\ See Bureau of Consumer Fin. Prot., Manufactured-Housing Consumer Finance in the United States (2014), available at http://www.consumerfinance.gov/reports/manufactured-housing-consumer-finance-in-the-u-s/; see also 79 FR 51731, 51797-98 (Aug. 29, 2014). \350\ U.S. Gov't. Accountability Office, GAO 07-879, Federal Housing Administration: Agency Should Assess the Effects of Proposed Changes to the Manufactured Home Loan Program (2007), available at http://www.gao.gov/new.items/d07879.pdf; See Milano, supra note 345 at 383; Burkhart, supra note 344 at 428; Washington Hearing, supra note 39. --------------------------------------------------------------------------- After considering the comments, pursuant to its authority under HMDA section 305(a) and 304(b)(6)(J), the Bureau is adopting Sec. 1003.4(a)(29) with modifications. Pursuant to its authority under HMDA section 305(a) to provide for adjustments for any class of transactions, the Bureau believes that interpreting HMDA to treat mortgage loans secured by all manufactured homes consistently is necessary and proper to effectuate HMDA's purposes and facilitate compliance therewith.\351\ Final Sec. 1003.4(a)(29) requires financial institutions to report whether the covered loan is secured by a manufactured home and land or a manufactured home and not land instead of whether the manufactured home is legally classified as real or personal property. The Bureau believes that the final rule will facilitate fair lending analyses, and will help to explain pricing data. At the same time, the final rule will avoid the issues associated with reporting classification under State law such as using multiple methods of lien perfection. As adopted, the requirement will also not apply to multifamily dwellings to make clear that covered loans secured by a manufactured home community are not subject to this reporting requirement. --------------------------------------------------------------------------- \351\ See also 79 FR 51732, 51797-98 (Aug. 29, 2014) (explaining basis for treating mortgage loans secured by all manufactured homes consistently). --------------------------------------------------------------------------- The Bureau is adopting new comment 4(a)(29)-1 to specify that even covered loans secured by a manufactured home classified as real property under State law should be reported as secured by a manufactured home and not land if the covered loan is also not secured by land. The Bureau is adopting new comment 4(a)(29)-2 to specify that this reporting requirement does not apply to loans secured by a multifamily dwelling that is a manufactured home community. Proposed comment 4(a)(29)-1 is adopted as comment 4(a)(29)-3. The Bureau is also adopting new comment 4(a)(29)-4 to provide guidance on the scope of the reporting requirement. 4(a)(30) Section 304(b) of HMDA permits disclosure of such other information as the Bureau may require. The Bureau proposed to require financial institutions to collect and report whether the applicant or borrower owns the land on which a manufactured home is or will be located through a direct or indirect ownership interest or leases the land through a paid or unpaid leasehold interest. For the reasons discussed below, the Bureau is finalizing Sec. 1003.4(a)(30) generally as proposed with technical modifications for clarity and to specify that multifamily dwellings are not subject to the reporting requirement. Many consumer advocate commenters supported the proposed requirement and stated that the information would be valuable. In contrast, many industry commenters opposed the proposed requirement for several reasons. Some industry commenters stated that the proposed requirement is information [[Page 66227]] that they currently do not verify for loans secured by a manufactured home and not land. Other industry commenters stated that they do collect some information about the land interest of the borrower for loans secured by a manufactured home and not land, but that the information reported by the applicant is often unreliable. Other industry commenters stated that the information is not a factor in loan pricing and questioned the value of the information. Some industry commenters stated that the proposed requirement would relate only to individual manufactured home loans and not loans secured by manufactured home communities. The Bureau believes that the proposed requirement will provide valuable information about the land interest of manufactured home loan borrowers. The information could aid in determining whether borrowers are obtaining loans secured by a manufactured home and not land when they could qualify for a loan secured by a manufactured home and land. This information could aid policymakers at the local, State, and Federal level and financial institutions in determining how the housing needs of manufactured home borrowers could best be served by loan products relating to manufactured homes and legal requirements relating to such financing or the classification and treatment of manufactured homes under State law.\352\ --------------------------------------------------------------------------- \352\ See Bureau of Consumer Fin. Prot., Manufactured-Housing Consumer Finance in the United States (2014), available at http://www.consumerfinance.gov/reports/manufactured-housing-consumer-finance-in-the-u-s/; Consumers Union Report, Manufactured Housing Appreciation: Stereotypes and Data (2003), available at http://consumersunion.org/pdf/mh/Appreciation.pdf ; Katherine MacTavish et al., Housing Vulnerability Among Rural Trailer-Park Households, 13 Georgetown Journal on Poverty Law and Policy 97 (2006); Sally Ward et al., Carsey Institute, Resident Ownership in New Hampshire's ``Mobile Home Parks:'' A Report on Economic Outcomes, (2010), available at http://www.rocusa.org/uploads/Carsey%20Institute%20Reprint%202010.pdf. --------------------------------------------------------------------------- After considering the comments, the Bureau is finalizing Sec. 1003.4(a)(30) with technical modifications for clarity and to specify that multifamily dwellings are not subject to the reporting requirement. The Bureau is finalizing comments 4(a)(30)-1, -2, and -3 generally as proposed, with technical modifications for clarity. The Bureau is adopting new comment 4(a)(30)-4 to clarify that a loan secured by a multifamily dwelling that is a manufactured home community is not subject to the reporting requirement. The Bureau is adopting new comment 4(a)(30)-5 to provide guidance on direct ownership consistent with proposed appendix A. The Bureau is also adopting new comment 4(a)(30)-6 to provide guidance on the scope of the reporting requirement. The Bureau is adopting Sec. 1003.4(a)(30) pursuant to its authority under section 305(a) and 304(b)(6)(J) of HMDA. The Bureau finds that Sec. 1003.4(a)(30) is necessary to carry out HMDA's purposes, because it will provide necessary insight into loan data and allow it to be used to help determine whether financial institutions are serving the housing needs of their communities, since this information can have important implications for the financing, long- term affordability, and appreciation of the housing at issue. 4(a)(31) Current Regulation C requires financial institutions to identify multifamily dwellings as a property type. The Bureau proposed to add Sec. 1003.4(a)(31), which requires a financial institution to report the number of individual dwelling units related to the property securing the covered loan or, in the case of an application, proposed to secure the covered loan. As discussed above, the Bureau proposed to replace the current property type reporting requirement with construction method and to separate the concept of the number of units from that reporting requirement. For the reasons discussed below, the Bureau is adopting Sec. 1003.4(a)(31) generally as proposed with additional commentary to provide clarity. Some commenters supported the proposed requirement and stated that it would provide valuable information about covered loans related to multifamily housing and covered loans related to one- to four-unit dwellings. Other commenters argued that the number of units should be reported in ranges, such 1, 2-4, and 5 or more. Some commenters stated that ranges would be insufficient as they would not permit distinguishing between small and large multifamily dwellings or among one- to four-unit dwellings. Other commenters argued that no requirement to report number of units should be adopted and the current property type requirement should be retained. Some commenters stated that they currently collect an exact total number of units and the data would therefore be easy to obtain, while other commenters stated that they use ranges and the proposed requirement would be burdensome. Some commenters stated that there would be compliance difficulties in reporting total units for certain types of properties, such as manufactured home communities, condominium developments, and cooperative housing developments. The Bureau believes that reporting the precise number of individual dwelling units would be preferable to ranges. The precise number would permit better comparison among loans related to dwellings with a single dwelling unit, two- to four-unit dwellings, and multifamily dwellings with similar numbers of dwelling units, thus facilitating the analysis of the housing needs served by both small and large multifamily dwellings. Reporting the precise number of units will also facilitate matching HMDA data to other publically available data about multifamily dwellings. After considering the comments, the Bureau is finalizing Sec. 1003.4(a)(31) as proposed pursuant to its authority under sections 305(a) and 304(b)(6)(J) of HMDA. Multifamily housing has always been an essential component of the nation's housing stock. In the wake of the housing crisis, multifamily housing has taken on an increasingly important role in communities, as families have turned to rental housing for a variety of reasons.\353\ The Bureau finds that Sec. 1003.4(a)(31) will further HMDA's purposes by assisting in determinations about whether financial institutions are serving the housing needs of their communities, and it may assist public officials in targeting public investments. --------------------------------------------------------------------------- \353\ See analysis of HMDA data at 79 FR 51731, 51800 (Aug. 29, 2014). See San Francisco Hearing, supra note 42. --------------------------------------------------------------------------- The Bureau received no specific feedback on comment 4(a)(31)-1, which is adopted with modifications for consistency with final comment 4(a)(9)-2. In response to the requests for clarification, the Bureau is adopting three new comments. New comments 4(a)(31)-2, -3, and -4 provide guidance on: Reporting the total units for a manufactured home community; reporting the total units for condominium and cooperative properties; and the information that a financial institution may rely on in complying with the requirement to report total units. 4(a)(32) The Bureau proposed to add Sec. 1003.4(a)(32), which requires financial institutions to collect and report information on the number of individual dwelling units in multifamily dwellings that are income- restricted pursuant to Federal, State, or local affordable housing programs. The Bureau also solicited comment on whether additional information about the program or type of affordable housing would be valuable and serve HMDA's purposes, and about the [[Page 66228]] burdens associated with collecting such information compared with the burdens of the proposal. In addition to soliciting feedback generally about this requirement, the Bureau specifically solicited comment on the following points: Whether the Bureau should require reporting of information concerning programs targeted at specific groups (such as seniors or persons with disabilities); Whether income restrictions above a certain threshold should be excluded for reporting purposes (such as income restrictions above the area median income); Whether it would be appropriate to simplify the requirement and report only whether a multifamily dwelling contains a number of income-restricted units above a certain percentage threshold; Whether financial institutions should be required to report the specific affordable housing program or programs; Whether financial institutions should be required to report the area median income level at which units in the multifamily dwelling are considered affordable; and Whether the burden on financial institutions may be reduced by providing instructions or guidance specifying that institutions only report income-restricted dwelling units that they considered or were aware of in originating, purchasing, or servicing the loan. Many industry commenters opposed the proposed income-restricted units reporting requirement and stated that it would impose new burden on many financial institutions that do not regularly collect this information currently. Many consumer advocate commenters supported the proposed reporting requirement and stated that it would provide valuable information on how financial institutions are serving the housing needs of their communities. However, most consumer advocate commenters argued that the proposed requirement would not provide enough information, and that the Bureau should add additional reporting requirements to gather information about the affordability level of the income-restricted units. Some commenters proposed additional reporting requirements related to multifamily dwellings including the number of bedrooms for the individual dwellings units, whether the housing is targeted at specific populations, the presence and number of commercial tenants, the debt service coverage ratio at the time of origination, and whether the developer or owner of the housing is a mission-driven nonprofit organization. Regarding whether housing is targeted at specific populations, the Bureau notes that it is providing commentary to the definition of dwelling as discussed above in the section-by-section analysis of Sec. 1003.2(f) regarding when housing associated with related services or medical care should be reported. However, the Bureau does not believe it would be appropriate to adopt a reporting requirement regarding housing targeted at specific populations, at this time. The Bureau does not have sufficient information on the costs and benefits associated with such a reporting requirement and the challenges in developing an appropriate reporting scheme given the wide variety of housing designated for specific populations including persons with disabilities and seniors. Similarly, the Bureau is not finalizing reporting requirements on the other specific suggestions for multifamily dwellings at this time because it does not have sufficient information on the costs and benefits associated with such reporting requirements and the Bureau believes it may be likely that the burdens of such reporting would outweigh the benefits. Consumer advocate commenters generally stated that the Bureau should adopt additional data points similar to the data reporting requirements for the GSEs' affordable housing goals.\354\ One commenter stated that income-restricted units at 80, 100, or 120 percent of area median income should not be considered affordable and not reported. Other commenters stated that financial institutions should be permitted to rely on information provided by the applicant or considered during the underwriting process to fulfill this reporting requirement. --------------------------------------------------------------------------- \354\ 12 CFR part 1282, subpart B. --------------------------------------------------------------------------- The Bureau believes that additional information about income- restricted multifamily dwellings would be valuable, but believes any benefits would not justify the burdens for collecting detailed information about the level of affordability for individual dwelling units. The suggestion to align HMDA reporting with the GSE affordable housing goals would require financial institutions to report five data points.\355\ The Bureau believes that the GSE affordable housing goal reporting requirements are sufficiently distinct from HMDA that they should not be adopted for HMDA purposes. For example, the HMDA reporting requirement proposed concerns only income-restricted dwelling units, which would generally be identifiable from information about the property and not require tenant income or rent determinations for HMDA reporting, whereas dwelling units may qualify for the GSE affordable housing goals based on tenant income information compared to area median income or on rent levels and adopting a similar reporting requirement for HMDA would therefore require information related to tenant income or rent levels that a financial institution may not consider in all instances when not required to do so by GSE requirements.\356\ This would be significantly more burdensome than the requirement proposed. Furthermore, for the GSE affordable housing goals the GSEs themselves participate in analyzing the data and making the determinations, and may estimate in the case of missing information.\357\ The Bureau did not propose to participate in making the determinations on affordable housing in a similar way. --------------------------------------------------------------------------- \355\ Financial institutions would have to report the number of dwelling units affordable at moderate-income (not in excess of 100 percent of area median income), low-income (not in excess of 80 percent of area median income), low-income (not in excess of 60 percent of area median income), very low-income (not in excess of 50 percent of area median income), and extremely low-income (not in excess of 30 percent of area median income). See 12 CFR 1282.17, 12 CFR 1282.18. \356\ 12 CFR 1282.15(d)(1), 12 CFR 1282.15(d)(2). \357\ 12 CFR 1282.15(e). --------------------------------------------------------------------------- Some commenters stated that the burden of imposing the GSE affordable housing goal requirements would not be significant because many HMDA reporters would already be following them for covered loans secured by multifamily dwellings sold to the GSEs. However, according to the 2013 HMDA data, of the 39,861 originated loans secured by multifamily dwellings, only 2,388 were sold to the GSEs within the calendar year of origination. The Bureau is concerned that many financial institutions would not be using the GSE affordable housing goal standards for the majority of their HMDA-reportable loans secured by multifamily dwellings. Therefore, the Bureau is not adopting the suggested reporting requirement aligned with the GSE affordable housing goals. The Bureau believes that information about the number of income- restricted units in multifamily dwellings is valuable and will further HMDA's purposes, in part by providing more useful information about these vital public resources, and thereby assisting public officials in distributing public-sector investment so as to attract private investment to areas where it is needed. Presently the need for affordable [[Page 66229]] housing is much greater than the supply.\358\ Although the requirement entails additional burden for some financial institutions, other financial institutions that specialize in lending related to income- restricted multifamily housing may have lesser initial burden associated with this requirement. By limiting the requirement to income-restricted units and excluding some other forms of affordable housing policies and programs, the rule provides a well-defined scope of reporting that should generally be verifiable through property records and other sources. --------------------------------------------------------------------------- \358\ Harvard University Joint Ctr. for Housing Studies, America's Rental Housing: Evolving Market and Needs (2013), available at http://www.jchs.harvard.edu/sites/jchs.harvard.edu/files/jchs_americas_rental_housing_2013_1_0.pdf. --------------------------------------------------------------------------- After considering the comments and conducting additional analysis, pursuant to HMDA sections 305(a) and 304(b)(6)(J), the Bureau is finalizing Sec. 1003.4(a)(32) as proposed. The Bureau is adopting new comment 4(a)(32)-5 to provide guidance on information that a financial institution may rely on in complying with the requirement to report the number of income-restricted units. The Bureau is adopting new comment 4(a)(32)-6 to provide guidance on the scope of the reporting requirement. The Bureau is also finalizing comments 4(a)(32)-1, -2, -3, and -4 generally as proposed, with modifications for clarity. 4(a)(33) The Bureau proposed Sec. 1003.4(a)(33) to implement the Dodd-Frank Act amendment that requires financial institutions to disclose ``the channel through which application was made, including retail, broker, and other relevant categories'' for each covered loan and application.\359\ Proposed Sec. 1003.4(a)(33) provided that, except for purchased covered loans, a financial institution was required to report the following information about the application channel of the covered loan or application: whether the applicant or borrower submitted the application for the covered loan directly to the financial institution; and whether the obligation arising from the covered loan was or would have been initially payable to the financial institution. The Bureau also proposed illustrative commentary. The Bureau is finalizing Sec. 1003.4(a)(33) as proposed and proposed comments 4(a)(33)-1 through -3 with the modifications discussed below. --------------------------------------------------------------------------- \359\ Dodd-Frank Act section 1094(3), 12 U.S.C. 2803(b)(6)(E). --------------------------------------------------------------------------- Comments Several consumer advocate commenters expressed support for the proposed requirement, noting the importance of this information in identifying risks to consumers. On the other hand, some industry commenters expressed concerns about proposed Sec. 1003.4(a)(33). One industry commenter explained that collecting this information would be burdensome because financial institutions do not routinely capture it in the proposed format. Another industry commenter asked the Bureau to exempt multifamily loans from this requirement. In addition, a commenter asked the Bureau to exempt community banks because all of their originations come through the same application channel. Information about the application channel of covered loans and applications will enhance the HMDA data. The loan terms and rates that a financial institution offers an applicant may depend on how the applicant submits the application (i.e., whether through the retail, wholesale, or correspondent channel).\360\ Thus, identifying transactions by channel may help users to interpret loan pricing and other information in the HMDA data. In addition, these data will aid in understanding whether certain channels present particular risks for consumers. --------------------------------------------------------------------------- \360\ See, e.g., Keith Ernst et al., Center for Responsible Lending, Steered Wrong: Brokers, Borrowers, and Subprime Loans (April 2008), available at http://www.responsiblelending.org/mortgage-lending/research-analysis/steered-wrong-brokers-borrowers-and-subprime-loans.pdf. --------------------------------------------------------------------------- While there is some burden associated with collecting this information, the Bureau understands that the burden is minimal because the information is readily available and easily reported in two true- false fields. For the same reasons, the Bureau does not believe that it is appropriate to exclude certain types of institutions or types of loans from the requirement, except the exclusion for purchased loans discussed below. Some commenters suggested different approaches to collect application channel information. One consumer advocate commenter asked the Bureau to collect the loan channel information as defined by the Secure and Fair Enforcement of Mortgage Licensing Act (SAFE Act), Public Law 110-289, to identify the retail, wholesale, and correspondent channels. However, neither the SAFE Act nor its implementing regulations define loan channels, so it is not possible to align with loan channel definitions in that statute. \361\ --------------------------------------------------------------------------- \361\ See 12 U.S.C. 5101 et seq.; 12 CFR part 1007; 12 CFR part 1008. --------------------------------------------------------------------------- In addition, the final rule will collect sufficient information to identify the various loan channels. The application channels in the mortgage market can be identified with three pieces of information: (1) Which institution received the application directly from the applicant, (2) which institution made the credit decision, and (3) the institution to which the obligation initially was payable. For example, the term ``retail channel'' generally refers to situations where the applicant submits the application directly to the financial institution that makes the credit decision on the application and to which the obligation is initially payable. The term ``wholesale channel,'' which is also referred to as the ``broker channel,'' generally refers to situations where the applicant submits the application to a mortgage broker and the broker sends the application to a financial institution that makes the credit decision on the application and to which the obligation is initially payable. The correspondent channel includes correspondent arrangements between two financial institutions. A correspondent with delegated underwriting authority processes an application much like the retail channel described above. The correspondent receives the application directly from the applicant, makes the credit decision, closes the loan in its name, and immediately or within a short period of time sells the loan to another institution. Correspondents with nondelegated authority operate somewhat more like a mortgage broker in the wholesale channel. These correspondents receive the application from the applicant, but prior to closing involve a third-party institution that makes the credit decision. The transaction generally closes in the name of the correspondent, which immediately or within a short period of time sells the loan to the third-party institution that made the credit decision.\362\ --------------------------------------------------------------------------- \362\ See generally 78 FR 11280, 11284 (Feb. 15, 2013); CFPB Examination Procedures on Mortgage Origination (2014), http://files.consumerfinance.gov/f/201401_cfpb_mortgage-origination-exam-procedures.pdf. --------------------------------------------------------------------------- Regulation C requires the institution that makes the credit decision to report the action taken on the application, as discussed above in the section-by-section analysis of Sec. 1003.4(a). Therefore, the application channels described above can be identified with the information required by proposed Sec. 1003.4(a)(33), which included whether [[Page 66230]] the applicant or borrower submitted the application directly to the financial institution that is reporting the loan and whether the obligation was, or would have been, initially payable to the financial institution that is reporting the loan. An industry commenter suggested that the Bureau implement the Dodd- Frank Act amendment by requiring financial institutions to report whether a broker was involved. The Bureau believes the proposal would be less burdensome than the suggested approach, which would require the final rule to define the term ``broker'' solely for the purpose of HMDA reporting. A broker is generally understood to refer applicants to lenders, but a broker may play a different role in a given transaction depending on the business arrangement it has with a lender or investor. In addition, as discussed above, the commenter's suggested approach would not identify other channels, such as the correspondent channel. Therefore, proposed Sec. 1003.4(a)(33) is the preferable approach. An industry commenter also opposed the exclusion of purchase loans from the requirement to report the information required by proposed Sec. 1003.4(a)(33). The commenter reasoned that it is more efficient to collect information from investors than from the originating organization. The commenter also did not believe that the information required by Sec. 1003.4(a)(33) would be the same for all purchased loans reported by a financial institution. The Bureau continues to believe that collecting application channel information for purchased loans is unnecessary. Under Regulation C, if the financial institution reports a loan as a purchase, the reporting institution did not make a credit decision on the loan. See the section-by-section analysis of Sec. 1003.4(a) and comments 4(a)-2 through -4. Thus data users could assume that most, if not all, entries reported as purchases did not involve an application submitted to the purchaser and that the loan did not close in the institution's name. A consumer advocate commenter urged the Bureau to collect a unique identifier for each loan channel in addition to the information required by proposed Sec. 1003.4(a)(33). The final rule will require financial institutions to report the NMLS ID of the loan originator for covered loans and applications. See the section-by-section analysis of Sec. 1003.4(a)(34). The NMLS ID will further help to identify the loan channel. Direct submission of an application. Some commenters sought clarification about proposed Sec. 1003.4(a)(33)(i), which required financial institutions to indicate whether a financial institution submitted an application directly to the financial institution. A commenter suggested referencing the language used in the SAFE Act about loan origination activities to clarify what proposed Sec. 1003.4(a)(33)(i) required. The Bureau's Regulations G and H, which implement the SAFE Act, provide detailed examples of activities that are conducted by loan originators.\363\ If the loan originator that performed loan origination services for the application or loan that the financial institution is reporting was an employee of the reporting financial institution, the applicant likely submitted the application directly to the financial institution. Section 1003.4(a)(34), discussed below, references the definition of loan originator in the SAFE Act, and directs financial institutions to report the NMLS ID of the loan originator that performed origination activities on the covered loan or application. Therefore, the Bureau is modifying proposed comment 4(a)(33)-1, renumbered as comment 4(a)(33)(i)-1 to clarify that an application was submitted directly to the financial institution that is reporting the covered loan or application if the loan originator identified pursuant to Sec. 1003.4(a)(34) was employed by the financial institution when the loan originator performed loan origination activities for the loan or application that the financial institution is reporting. --------------------------------------------------------------------------- \363\ See 12 U.S.C. 5101 et seq.; 12 CFR part 1007; 12 CFR part 1008. --------------------------------------------------------------------------- Another commenter suggested clarifying whether an application is submitted directly to the financial institution if the application is submitted to a credit union service organization (CUSO) hired by the credit union that is reporting the entry to receive applications for covered loans on behalf of a credit union. The Bureau is also modifying proposed comment 4(a)(33)-1, renumbered as comment 4(a)(33)(i)-1, to illustrate how to report whether the application was submitted directly to the financial institution when a CUSO or other similar agent is involved. Another industry commenter raised privacy concerns about releasing to the public the application channel information. The Bureau appreciates this feedback and is carefully considering the privacy implications of the publicly released data. See part II.B above for a discussion of the Bureau's approach to protecting applicant and borrower privacy with respect to the public disclosure of the data. Due to the significant benefits of collecting this information, the Bureau believes it is appropriate to collect application channel information despite the concerns raised by commenters about collecting this information. The Bureau received no comments on proposed comments 4(a)(33)-2 and -3. Final Rule For the reasons discussed above and pursuant to its authority under HMDA sections 304(b)(6)(E) and 305(a), the Bureau is adopting Sec. 1003.4(a)(33) as proposed. This requirement is an appropriate method of implementing HMDA section 304(b)(6)(E) in a manner that carries out HMDA's purposes. To facilitate compliance, pursuant to HMDA 305(a), the Bureau is excepting purchased covered loans from this requirement. The Bureau is also finalizing proposed comments 4(a)(33)-1, -2, and -3, renumbered as comments 4(a)(33)(i)-1, 4(a)(33)(ii)-1, and 4(a)(33)-1, with the modifications discussed above. The Bureau is also adopting new comment 4(a)(33)(ii)-2 to clarify that a financial institution may report that Sec. 1003.4(a)(33)(ii) is not applicable when the institution had not determined whether the covered loan would have been initially payable to the institution reporting the application when the application was withdrawn, denied, or closed for incompleteness. 4(a)(34) Regulation C does not require financial institutions to report information regarding a loan originator identifier. HMDA section 304(b)(6)(F) requires the reporting of, ``as the Bureau may determine to be appropriate, a unique identifier that identifies the loan originator as set forth in section 1503 of the [Secure and Fair Enforcement for] Mortgage Licensing Act of 2008'' (S.A.F.E. Act).\364\ The Bureau proposed Sec. 1003.4(a)(34), which implements this requirement by requiring financial institutions to report, for a covered loan or application, the unique identifier assigned by the Nationwide Mortgage Licensing System and Registry (NMLSR ID) for the mortgage loan originator, as defined in Regulation G Sec. 1007.102 or Regulation H Sec. 1008.23, as applicable. --------------------------------------------------------------------------- \364\ Dodd-Frank Act section 1094(3)(A)(iv), 12 U.S.C. 2803(b)(6)(F). --------------------------------------------------------------------------- In addition, the Bureau proposed three comments. Proposed comment 4(a)(34)-1 discusses the requirement that a financial institution report the NMLSR ID for the mortgage loan [[Page 66231]] originator and describes the NMLSR ID. Proposed comment 4(a)(34)-2, clarifies that, in the event that the mortgage loan originator is not required to obtain and has not been assigned an NMLSR ID, a financial institution complies with Sec. 1003.4(a)(34) by reporting ``NA'' for not applicable. Proposed comment 4(a)(34)-2 also provides an illustrative example to clarify that if a mortgage loan originator has been assigned an NMLSR ID, a financial institution complies with Sec. 1003.4(a)(34) by reporting the mortgage loan originator's NMLSR ID regardless of whether the mortgage loan originator is required to obtain an NMLSR ID for the particular transaction being reported by the financial institution. Lastly, the Bureau proposed comment 4(a)(34)-3, which clarifies that if more than one individual meets the definition of a mortgage loan originator, as defined in Regulation G, 12 CFR 1007.102, or Regulation H, 12 CFR 1008.23, for a covered loan or application, a financial institution complies with Sec. 1003.4(a)(34) by reporting the NMLSR ID of the individual mortgage loan originator with primary responsibility for the transaction. The proposed comment explains that a financial institution that establishes and follows a reasonable, written policy for determining which individual mortgage loan originator has primary responsibility for the reported transaction complies with Sec. 1003.4(a)(34). The vast majority of commenters supported the Bureau's proposed Sec. 1003.4(a)(34). Many consumer advocate commenters supported the Bureau's proposal to include a unique identifier for a mortgage loan originator because this information may help regulatory agencies and the public identify financial institutions and loan originators that are engaged in problematic loan practices. Commenters also supported the Bureau's proposed Sec. 1003.4(a)(34) because they believe the information is critical to understanding the residential mortgage market. Consistent with the Small Business Review Panel's recommendation, the Bureau specifically solicited comment on whether the mortgage loan originator unique identifier should be required for all entries on the loan/application register, including applications that do not result in originations, or only for loan originations and purchases. One industry commenter stated without explanation that the reporting requirement should only apply to originations and purchases. Another national trade association stated, without further explanation, that reporting of the mortgage loan originator unique identifier should not be required on applications that do not result in originations because such data will not provide any value and will impose burden on industry. In contrast, another industry commenter stated that in order for the NMLSR ID to be useful, such data should only be collected and reported if the loan officer has the authority to decide whether to approve or deny the application. This commenter stated that in such cases, the NMLSR ID would need to be collected for both originated and non-originated applications. The Bureau has considered this feedback and determined it will adopt proposed Sec. 1003.4(a)(34), which applies to applications, originations, and purchased loans. The Bureau believes the HMDA data's usefulness will be improved by being able to identify individual mortgage loan originators with primary responsibility over applications, originations, and purchased loans. While the Bureau acknowledged in its proposal that a requirement to collect and report a mortgage loan originator unique identifier may impose some burden on financial institutions, the Bureau did not receive feedback specifically addressing the potential burden. In fact, a State trade association commented that reporting the mortgage loan originator's NMLS ID would not pose an additional burden for its members because it already collects and reports this information for the mortgage Call Report. A government commenter also stated that this data should be readily accessible by HMDA reporters since it will be provided on the TILA-RESPA integrated disclosure form. The Bureau has determined that the benefits gained by the information reported under proposed Sec. 1003.4(a)(34) justify any potential burdens on financial institutions. As discussed in the Bureau's proposal, this information is provided on certain loan documents pursuant to the loan originator compensation requirements under TILA.\365\ As noted by a commenter, this information will also be provided on the TILA-RESPA integrated disclosure form.\366\ As a result, the Bureau has determined that the NMLSR ID for the mortgage loan originator will be readily available to HMDA reporters at little to no ongoing cost. --------------------------------------------------------------------------- \365\ Regulation Z Sec. 1026.36(g). \366\ Regulation Z Sec. 1026.37(k). --------------------------------------------------------------------------- Several commenters did not support the Bureau's proposed Sec. 1003.4(a)(34) for two main reasons. This opposition is based on concerns related to disclosure of this information by the Bureau. First, one State trade association and a few industry commenters suggested that review of a mortgage loan originator's performance should be left up to the individual financial institution and not be subject to public scrutiny. Second, a few commenters stated that requiring financial institutions to report the NMLSR ID of the individual mortgage loan originator would raise concerns regarding the privacy of those mortgage loan originators. For example, a State trade association and another industry commenter opposed the Bureau's proposed Sec. 1003.4(a)(34) because it believes disclosing an NMLSR ID in connection with specific loan transactions has the potential to violate the financial privacy of individual employees of a financial institution. The commenter suggested that making this information publicly available would create privacy concerns for a financial institution's loan originator employees by opening the door to identification of the loan originator by name and address. In addition, the commenter argued that this information, combined with other transaction specific public information, could enable someone to calculate an individual loan originator employee's commission income, sales volume and other private financial information. Another industry commenter suggested that if a mortgage loan originator can be identified in the HMDA data, and the loan originator originated a large volume of loans at a financial institution that subsequently fails for reasons unrelated to underwriting, the loan originator may be unable to find employment. The Bureau has considered this feedback. The Bureau has concluded that it will not withhold from public release the NMLSR ID of mortgage loan originators for the reasons expressed by commenters. As summarized above, the commenters were concerned that the public disclosure of this information may implicate the privacy interests of mortgage loan originators. As discussed in part II.B above, HMDA directs the Bureau to ``modify or require modification of itemized information, for the purpose of protecting the privacy interests of the mortgage applicants or mortgagors, that is or will be available to the public.'' \367\ The Bureau is applying a balancing test to determine whether and how HMDA data should be modified prior to its disclosure to the public in order to protect applicant and borrower privacy while also fulfilling HMDA's public disclosure purposes. The Bureau will consider NMLSR ID [[Page 66232]] under this applicant and borrower privacy balancing test. The Bureau is implementing, in Sec. 1003.4(a)(34), the Dodd-Frank Act amendment to HMDA requiring a unique identifier for mortgage loan originators. Because the Dodd-Frank Act explicitly amended HMDA to add a loan originator identifier, while at the same time directing the Bureau to modify or require modification of itemized information ``for the purpose of protecting the privacy interests of the mortgage applicants or mortgagors,'' the Bureau believes it is reasonable to interpret HMDA as not requiring modifications of itemized information to protect the privacy interests of mortgage loan originators, and that that interpretation best effectuates the purposes of HMDA. --------------------------------------------------------------------------- \367\ HMDA section 304(h)(1)(E), (h)(3)(B); 12 U.S.C. 2803(h)(1)(3), (h)(3)(B). --------------------------------------------------------------------------- The Bureau is finalizing the Dodd-Frank Act requirement for the collection and reporting of a mortgage loan originator unique identifier as proposed in Sec. 1003.4(a)(34). The Bureau believes that this information will improve HMDA data by, for example, identifying an individual who has primary responsibility in the transaction, which will in turn enable new dimensions of analysis, including being able to link individual mortgage loan originators or groups of mortgage loan originators to a financial institution. Accordingly, the Bureau is adopting Sec. 1003.4(a)(34) as proposed, with minor modification for proposed clarity to proposed comment 4(a)(34)-2 and one substantive change to proposed comment 4(a)(34)-3. In order to facilitate compliance with the new reporting requirement when multiple mortgage loan originators are associated with a particular covered loan or transaction, the comment clarifies that a financial institution reports the NMLSR ID of the individual mortgage loan originator with primary responsibility for the transaction as of the date of action taken pursuant to Sec. 1003.4(a)(8)(ii). A financial institution that establishes and follows a reasonable, written policy for determining which individual mortgage loan originator has primary responsibility for the reported transaction as of the date of action taken complies with Sec. 1003.4(a)(34). 4(a)(35) Currently, Regulation C does not require financial institutions to report information regarding results received from automated underwriting systems, and HMDA does not expressly require this itemization. Section 304(b) of HMDA permits the disclosure of ``such other information as the Bureau may require.'' \368\ The Bureau proposed Sec. 1003.4(a)(35)(i), which provides that except for purchased covered loans, a financial institution shall report the name of the automated underwriting system it used to evaluate the application and the recommendation generated by that automated underwriting system. In addition, the Bureau proposed Sec. 1003.4(a)(35)(ii), which defines an automated underwriting system (AUS) as an electronic tool developed by a securitizer, Federal government insurer, or guarantor that provides a recommendation regarding whether the application is eligible to be purchased, insured, or guaranteed by that securitizer, Federal government insurer, or guarantor. The Bureau also proposed three comments to provide clarification on the reporting requirement regarding AUS information under proposed Sec. 1003.4(a)(35). --------------------------------------------------------------------------- \368\ Dodd-Frank Act section 1094(3)(A)(iv), 12 U.S.C. 2803(b)(6)(J). --------------------------------------------------------------------------- In order to facilitate HMDA compliance and address concerns that it could be burdensome for financial institutions that purchase loans to identify automated underwriting system information, the Bureau excluded purchased covered loans from the requirements of proposed Sec. 1003.4(a)(35)(i). The Bureau solicited feedback on whether this exclusion was appropriate and received a few comments. One consumer advocate commenter recommended that unless and until the ULI is successfully implemented, purchased loans should not be excluded from the automated underwriting data reporting requirement. Another consumer advocate commenter provided feedback recommending that there be no exception for reporting of AUS information for purchased loans. This commenter suggested that the official interpretation of the rule should specify that the Bureau considers it reasonable for any institution purchasing covered loans to negotiate a contractual agreement requiring the seller institution to provide all data required by HMDA. The commenter also suggested that if an exception for purchased loans under proposed Sec. 1003.4(a)(35)(i) remains, it should be limited only to instances where the financial institution does not have and cannot reasonably obtain the AUS information. The Bureau has considered this feedback and has determined that it would be burdensome for financial institutions that purchase loans to identify the AUS used by the originating financial institution to evaluate the application and to identify the AUS result generated by that system. Consequently, the Bureau is adopting the exclusion of purchased covered loans proposed under Sec. 1003.4(a)(35)(i). The Bureau is also adopting new comment 4(a)(35)-5, which explains that a financial institution complies with Sec. 1003.4(a)(35) by reporting that the requirement is not applicable when the covered loan is a purchased covered loan. In response to the Bureau's solicitation for feedback regarding whether the proposed AUS requirements are appropriate, a few commenters recommended that the reporting requirement under proposed Sec. 1003.4(a)(35) be optional. For example, one industry commenter stated that reporting AUS data should be optional, not mandatory, since many smaller institutions do not use an automated system to evaluate certain loans. Another commenter stated that financial institutions do not use an AUS to evaluate multifamily and other commercial mortgage finance applications. While the Bureau acknowledges that proposed Sec. 1003.4(a)(35) will contribute to financial institutions' compliance burden, the Bureau has determined that a requirement of optional reporting of AUS data is not the appropriate approach given the value of the data in furthering HMDA's purposes. As discussed above with respect to denial reasons under Sec. 1003.4(a)(16), the statistical value of optionally reported data is lessened because of the lack of standardization across all HMDA reporters. A requirement that all financial institutions report the name of the AUS used to evaluate an application and the result generated by that system is the proper approach for purposes of HMDA. Moreover, as discussed further below, new comment 4(a)(35)-4 clarifies that a financial institution complies with proposed Sec. 1003.4(a)(35) by reporting that the requirement is not applicable if it does not use an AUS to evaluate the application, for example, if it only manually underwrites an application. In addition, as discussed further below, in order to address the concern that an AUS may not be used for all the types of transactions covered by the final rule, new comment 4(a)(35)-6 clarifies that when the applicant and co-applicant, if applicable, are not natural persons, a financial institution complies with Sec. 1003.4(a)(35) by reporting that the requirement is not applicable. In response to the Bureau's solicitation for feedback regarding [[Page 66233]] whether the proposed AUS requirements are appropriate, several commenters expressed concern that the Bureau's use of the term ``recommendation'' when describing the output from an AUS is inaccurate since such systems do not provide a credit decision. For example, one industry commenter stated that AUS recommendations are not a proxy for underwriter discretion and that even though an AUS recommendation can inform the level of underwriting that is appropriate for an application, it is not a credit decision on that application. Similarly, another industry commenter stated that when a financial institution obtains an AUS recommendation, the loan is then typically fully underwritten by in-house underwriters who make the final credit decision. Another commenter noted that the output from an AUS does not reflect the complete underwriting decision of a loan application and that a financial institution may have additional requirements such as credit-related overlays on top of those specified by the AUS used by the institution to evaluate the application. The Bureau considered this feedback and has determined that in order to address the concern that ``AUS recommendation'' incorrectly signals that the recommendation is a credit decision made by the AUS, the Bureau is adopting Sec. 1003.4(35)(i) generally as proposed, but replaces the term ``recommendation'' with ``result.'' Accordingly, the final rule requires a financial institution to report, except for purchased covered loans, the name of the automated underwriting system it used to evaluate the application and the result generated by that automated underwriting system. The Bureau solicited feedback on whether limiting the definition of an automated underwriting system as proposed in Sec. 1003.4(a)(35)(ii) to one that is developed by a securitizer, Federal government insurer, or guarantor is appropriate, and whether commentary is needed to clarify the proposed definition or to facilitate compliance. The Bureau's proposed AUS definition provided that financial institutions would report AUS data regarding the automated underwriting systems of the government-sponsored enterprises (GSEs)--the Federal National Mortgage Association (Fannie Mae) and the Federal Home Loan Mortgage Corporation (Freddie Mac)--other Federal government insurer or guarantor systems, and the proprietary automated underwriting systems of securitizers. The Bureau's proposed AUS definition did not include the proprietary automated underwriting systems developed by financial institutions that are not securitizers, nor the systems of third party vendors. In response to the Bureau's solicitation for feedback, several commenters suggested that the definition of AUS be expanded to include all systems used by financial institutions to evaluate an application. For example, one consumer advocate commenter stated that financial institutions use automated underwriting systems developed and sold by companies that are not securitizers, Federal government insurers or guarantors to determine whether or not loans will be eligible for government guarantee, insurance programs or sale to private investors, and that the Bureau should require financial institutions to report the use of and results from those systems as well. Another industry commenter stated that the Bureau's failure to cover the full range of all platforms used by financial institutions to make a credit decision, including proprietary or third-party AUSs, will necessarily produce incomplete data. Another commenter stated that the Bureau's proposed AUS definition is both under and over inclusive. The commenter argued that the definition is under inclusive because it excludes from HMDA reporting requirements the AUS name and result generated by a system developed by an entity that is not a securitizer, Federal government insurer, or guarantor. The commenter also argued that the definition is over inclusive since it could be interpreted as capturing other electronic tools used by financial institutions that are designed by the secondary market to provide an assessment of credit risk of an applicant or purchase eligibility of a loan, but are not intended to replace the purpose of an AUS. The Bureau considered this feedback and has determined that it will adopt the proposed definition of AUS in Sec. 1003.4(a)(35)(ii), with three modifications. First, the Bureau added the words ``Federal government'' in front of guarantor to the definition of AUS in the final rule to clarify that the definition captures an AUS developed by a Federal government guarantor, but not one developed by a non-Federal government guarantor. Second, the Bureau added the word ``originated'' to the definition of AUS in the final rule to clarify that in order for an electronic tool to meet the definition of an AUS under Sec. 1003.4(a)(35)(ii), the system must provide a result regarding the eligibility of the covered loan to be originated, purchased, insured, or guaranteed by the securitizer, Federal government insurer, or Federal government guarantor that developed the system being used to evaluate the application. Third, the Bureau added the words ``the credit risk of the applicant'' to the definition of AUS in the final rule to clarify that in order for an electronic tool to meet the definition of an AUS under Sec. 1003.4(a)(35)(ii), the system must also provide a result regarding the credit risk of the applicant. In order to facilitate compliance, the Bureau is also adopting new comment 4(a)(35)-2, discussed further below, which explains the definition of AUS and provides illustrative examples of the reporting requirement. In addition, the Bureau recognizes that the Federal Housing Administration's (FHA) Technology Open to Approved Lenders (TOTAL) Scorecard is different than the automated underwriting systems developed by Fannie Mae and Freddie Mac. TOTAL Scorecard is a tool developed by HUD that is used by financial institutions to evaluate the creditworthiness of applicants and determine an associated risk level of a loan's eligibility for insurance by the FHA. Unlike the automated underwriting systems of the GSEs, TOTAL Scorecard works in conjunction with various automated underwriting systems.\369\ However, if a financial institution uses TOTAL Scorecard to evaluate an application, the Bureau has determined that the HMDA data's usefulness will be improved by requiring the financial institution to report that it used that system along with the result generated by that system. --------------------------------------------------------------------------- \369\ See http://portal.hud.gov/hudportal/HUD?src=/program_offices/housing/sfh/total. --------------------------------------------------------------------------- Accordingly, pursuant to its authority under sections 305(a) and 304(b)(6)(J) of HMDA, the Bureau is adopting Sec. 1003.4(a)(35)(ii), which provides that an automated underwriting system means an electronic tool developed by a securitizer, Federal government insurer, or Federal government guarantor that provides a result regarding the credit risk of the applicant and whether the covered loan is eligible to be originated, purchased, insured, or guaranteed by that securitizer, Federal government insurer, or Federal government guarantor. Notwithstanding the concerns associated with collecting and reporting information about automated underwriting systems and results, the Bureau has determined that this information will further HMDA's purposes. This data will assist in understanding a financial institution's underwriting decisionmaking and will provide information that will assist in [[Page 66234]] identifying potentially discriminatory lending patterns and enforcing antidiscrimination statutes. As discussed above, the Bureau solicited feedback on whether commentary is needed to facilitate compliance. Several commenters provided a variety of feedback, including concern that the proposal will result in incomplete or inconsistent data. One commenter noted while the Bureau's proposed commentary recognizes the fact that financial institutions often use multiple AUSs for any given loan application, the proposal leaves open potential inconsistencies in how a lender chooses which AUS to report. For example, a few commenters noted that the ``closest in time'' standard in the proposal for reporting an AUS name and result could result in the HMDA data not capturing AUS data that the financial institution actually considered in making the credit decision. To highlight this concern, one commenter stated that financial institutions may use a ``waterfall strategy'' to evaluate applications by which an institution runs loan applications through one AUS first, then takes the `caution' loans from the first system and runs them through a second AUS. The commenter stated that the first AUS would see a lower risk population, while the second AUS would see a pre-screened higher risk population. The commenter expressed concern that since the Bureau's proposal requires a financial institution to report one AUS it used to evaluate an application and one AUS result generated by that system, the waterfall approach could potentially provide inaccurate HMDA results if not properly understood because it might be possible that such reporting would exclude AUS data that actually played a role in a financial institution's credit decision. Commenters noted if the Bureau is to take a comprehensive approach to collecting AUS data and address the concerns related to incomplete and inconsistent data, it should take into account the sequential decision making processes that financial institutions may use when running applications through multiple AUSs. One commenter suggested that until the Bureau adopts an approach that takes into account the various differences and complexities involved when a loan application is evaluated using multiple AUSs, it should reconsider requiring disclosure of AUS data. Another commenter recommended that the Bureau require financial institutions to report each AUS result (including non-securitizer proprietary and third party systems) that was used in the credit decision, as well as an indication of the relative importance of each result to the credit decision. Lastly, another commenter requested clarification as to whether a financial institution is required to report AUS information in the circumstance when an AUS provides a negative result, but the institution chooses to assume the credit risk and hold the resulting loan in its portfolio, rather than sell the loan to an investor. The Bureau considered this feedback and has determined that revisions to the proposed commentary and additional comments will facilitate compliance with the reporting requirement. For example, comment 4(a)(35)-3, discussed further below, provides additional clarity as to what AUS (or AUSs) and result (or results) a financial institution is required to report in cases when the institution uses one or more AUSs, which generate two or more results. In addition, comment 4(a)(35)-1.ii provides two illustrative examples and explains that a financial institution that uses an AUS, as defined in Sec. 1003.4(a)(35)(ii), to evaluate an application, must report the name of the AUS it used to evaluate the application and the result generated by that system, regardless of whether the financial institution intends to hold the covered loan in its portfolio or sell the covered loan. The Bureau solicited feedback on the proposed requirement that a financial institution enter, in a free-form text field, the name of the AUS used to evaluate the application and the result generated by that system, when ``Other'' is selected. Several industry commenters did not support the proposed requirement for a variety of reasons. A few commenters recommended removal of the free-form text field because it would be impossible to aggregate the data, without further explanation. Another commenter did not support the proposal to include a free-form text field for automated underwriting system information because there is no way to make the text input consistent among staff and financial institutions and as such, suggested that simply requiring a financial institution to report ``Other'' would be appropriate and sufficient. Lastly, another commenter stated that free-form text fields are illogical because they lack the ability of being sorted and reported accurately. This commenter also opined that the additional staff and/or programming that will be needed on a government level to analyze these free text fields is costly and not justified when looking at the minimal impact these fields have on the overall data collection under HMDA. The Bureau has considered the concerns expressed by industry commenters with respect to the proposed requirement that a financial institution enter the name of the AUS used to evaluate the application and the result generated by that system in a free-form text field when ``Other'' is reported but has determined that the utility of this data justifies the potential burden that may be imposed by the reporting requirement. As to the commenters' concern that data reported in the free-form text field would be impossible to aggregate, perhaps due to the variety of potential AUS names and results reported, the Bureau has determined that the data reported in the free-form text field will be useful even if the data cannot be aggregated. Lastly, with respect to a commenter's recommendation that requiring a financial institution to report ``Other'' is appropriate and sufficient and that the Bureau should not also require an institution to enter the name of the AUS used to evaluate the application and the result generated by that system in a free-form text field in these circumstances, the Bureau has determined that such an approach would hinder the utility of the AUS data for purposes of HMDA. As with the other free-form text fields the Bureau is adopting--the name and version of the scoring model when ``Other credit scoring model'' is reported by financial institutions under Sec. 1003.4(a)(15) and the denial reason or reasons when ``Other'' is reported by financial institutions under Sec. 1003.4(a)(16)--the free-form text field for AUS data will provide key information on the automated underwriting systems that are not listed and the results generated by those systems. For example, the AUS data can be used to monitor other automated underwriting systems that may enter the market or to add common, but previously unlisted, AUSs and results to the lists. The Bureau has determined that the HMDA data's usefulness will be improved by requiring financial institutions to report the name of the AUS used to evaluate the application and the result generated by that system in a free-form text field when the institution enters ``Other'' in the loan/ application register. The Bureau has modified proposed comments 4(a)(35)-1, -2, which is renumbered as -3, and -3, which is renumbered as -5. The Bureau is also adopting new comments 4(a)(35)-2, -4, and -6. As discussed below, the Bureau believes these modified and new [[Page 66235]] comments will facilitate compliance with the AUS reporting requirement. The Bureau is adopting proposed comment 4(a)(35)-1, with modifications. Comment 4(a)(35)-1 explains that a financial institution complies with Sec. 1003.4(a)(35) by reporting, except for purchased covered loans, the name of the automated underwriting system used by the financial institution to evaluate the application and the result generated by that automated underwriting system, and provides four scenarios to illustrate when a financial institution reports this information. The Bureau is adopting new comment 4(a)(35)-2, which explains that a financial institution must report the information required by Sec. 1003.4(a)(35)(i) if the financial institution uses an automated underwriting system (AUS), as defined in Sec. 1003.4(a)(35)(ii), to evaluate an application. Comment 4(a)(35)-2 clarifies that in order for an AUS to be covered by the definition in Sec. 1003.4(a)(35)(ii), the system must be an electronic tool that has been developed by a securitizer, Federal government insurer, or a Federal government guarantor, and provides two illustrative examples. In addition, comment 4(a)(35)-2 explains that in order for an AUS to be covered by the definition in Sec. 1003.4(a)(35)(ii), the system must provide a result regarding both the credit risk of the applicant and the eligibility of the covered loan to be originated, purchased, insured, or guaranteed by the securitizer, Federal government insurer, or Federal government guarantor that developed the system being used to evaluate the application, and provides an illustrative example. Comment 4(a)(35)-2 clarifies that a financial institution that uses a system that is not an AUS, as defined in Sec. 1003.4(a)(35)(ii), to evaluate an application does not report the information required by Sec. 1003.4(a)(35)(i). The Bureau is adopting proposed comment 4(a)(35)-2, with modifications, and renumbered as -3. Comment 4(a)(35)-3 sets forth the reporting requirements under Sec. 1003.4(a)(35) when multiple AUS results are generated by one or more AUSs. Comment 4(a)(35)-3 explains that when a financial institution uses one or more AUS to evaluate the application and the system or systems generate two or more results, the financial institution complies with Sec. 1003.4(a)(35) by reporting, except for purchased covered loans, the name of the AUS used by the financial institution to evaluate the application and the result generated by that AUS as determined by the principles set forth in the comment. The comment explains that to determine what AUS (or AUSs) and result (or results) to report under Sec. 1003.4(a)(35), a financial institution must follow each of the principles that is applicable to the application in question, in the order in which they are set forth in comment 4(a)(35)-3. First, comment 4(a)(35)-3.i explains that if a financial institution obtains two or more AUS results and the AUS generating one of those results corresponds to the loan type reported pursuant to Sec. 1003.4(a)(2), the financial institution complies with Sec. 1003.4(a)(35) by reporting that AUS name and result, and provides an illustrative example. Comment 4(a)(35)-3.i also explains that if a financial institution obtains two or more AUS results and more than one of those AUS results is generated by a system that corresponds to the loan type reported pursuant to Sec. 1003.4(a)(2), the financial institution identifies which AUS result should be reported by following the principle set forth in comment 4(a)(35)-3.ii. Second, comment 4(a)(35)-3.ii explains that if a financial institution obtains two or more AUS results and the AUS generating one of those results corresponds to the purchaser, insurer, or guarantor, if any, the financial institution complies with Sec. 1003.4(a)(35) by reporting that AUS name and result, and provides an illustrative example. Comment 4(a)(35)-3.ii also explains that if a financial institution obtains two or more AUS results and more than one of those AUS results is generated by a system that corresponds to the purchaser, insurer, or guarantor, if any, the financial institution identifies which AUS result should be reported by following the principle set forth in comment 4(a)(35)-3.iii. Third, comment 4(a)(35)-3.iii explains that if a financial institution obtains two or more AUS results and none of the systems generating those results correspond to the purchaser, insurer, or guarantor, if any, or the financial institution is following this principle because more than one AUS result is generated by a system that corresponds to either the loan type or the purchaser, insurer, or guarantor, the financial institution complies with Sec. 1003.4(a)(35) by reporting the AUS result generated closest in time to the credit decision and the name of the AUS that generated that result, and provides illustrative examples. Lastly, comment 4(a)(35)-3.iv explains that if a financial institution obtains two or more AUS results at the same time and the principles in comment 4(a)(35)-3.i through .iii do not apply, the financial institution complies with Sec. 1003.4(a)(35) by reporting the name of all of the AUSs used by the financial institution to evaluate the application and the results generated by each of those systems, and provides an illustrative example. In any event, however, comment 4(a)(35)-3.iv explains that a financial institution does not report more than five AUSs and five results. If more than five AUSs and five results meet the criteria in the principle set forth in comment 4(a)(35)-3.iv, the financial institution complies with Sec. 1003.4(a)(35) by choosing any five among them to report. The Bureau believes that it is reasonable to limit the number of AUSs to five and the number of results to five when a financial institution meets the criteria in the principle set forth in comment 4(a)(35)-3.iv. The Bureau believes that the likelihood of a financial institution evaluating an application through more than five AUSs at the same time is low. Moreover, the Bureau believes that requiring financial institutions to report all AUSs and the results of each of those systems, with no limitation, would be unnecessarily burdensome. Accordingly, as discussed above, comment 4(a)(35)-3.iv limits the number of AUSs and results that financial institutions are required to report to five each. The Bureau is adopting proposed comment 4(a)(35)-3, with modifications, and renumbered as -4. Comment 4(a)(35)-4 addresses transactions for which an AUS was not used to evaluate the application and explains that Sec. 1003.4(a)(35) does not require a financial institution to evaluate an application using an AUS, as defined in Sec. 1003.4(a)(35)(ii). For example, if a financial institution only manually underwrites an application and does not use an AUS to evaluate the application, the financial institution complies with Sec. 1003.4(a)(35) by reporting that the requirement is not applicable since an AUS was not used to evaluate the application. Proposed comment 4(a)(35)-3 also addressed transactions for which no credit decision was made by a financial institution by explaining that if a file was closed for incompleteness, or if an application was withdrawn before a credit decision was made, a financial institution complies with Sec. 1003.4(a)(35) by reporting that the requirement is not applicable. However, the Bureau has determined that it is not adopting this portion of proposed comment 4(a)(35)-3. The Bureau believes that if a financial institution uses an AUS to evaluate an application, regardless of whether the [[Page 66236]] file is closed for incompleteness or the application is withdrawn before a credit decision is made, the AUS data will assist in understanding the financial institution's underwriting decisionmaking and will provide information that will assist in identifying potentially discriminatory lending patterns and enforcing antidiscrimination statutes. Consequently, if a financial institution uses an AUS to evaluate an application and the file is closed for incompleteness and is so reported in accordance with Sec. 1003.4(a)(8), a financial institution complies with Sec. 1003.4(a)(35) by reporting the AUS information. Similarly, if a financial institution uses an AUS to evaluate an application and the application was withdrawn by the applicant before a credit decision was made and is so reported in accordance with Sec. 1003.4(a)(8), the financial institution complies with Sec. 1003.4(a)(35) by reporting the AUS information. As discussed above, the Bureau is adopting new comment 4(a)(35)-5, which explains that a financial institution complies with Sec. 1003.4(a)(35) by reporting that the requirement is not applicable when the covered loan is a purchased covered loan. Lastly, the Bureau is adopting new comment 4(a)(35)-6, which explains that a financial institution complies with Sec. 1003.4(a)(35) by reporting that the requirement is not applicable when the applicant and co-applicant, if applicable, are not natural persons. The Bureau believes that comments 4(a)(35)-1 through -6 will provide clarity regarding the new reporting requirement adopted in Sec. 1003.4(a)(35) and will facilitate HMDA compliance. In response to the Bureau's solicitation for feedback regarding whether the proposed AUS requirements are appropriate, a few commenters expressed concern about potential privacy implications for applicants or borrowers if the Bureau were to release AUS data to the public. One commenter did not support the proposal to include AUS results because it opined that such disclosure is in direct conflict with laws and rules designed to protect a consumer's non-public personal information. This commenter suggested that if AUS results were available to the public, such disclosure would make it easier for hackers around the world to gain access to personal financial data and place the safety and welfare of citizens in jeopardy. A national trade association commented that unless the Bureau establishes the appropriate safeguards against the misuse of sensitive consumer financial data, adding more sensitive and non-public information to HMDA disclosure, such as creditworthiness, creates considerable privacy concerns. Lastly, another commenter stated that the release of AUS data, either alone or when combined with other publicly available sources (including loan- level data associated with mortgage-backed securities issuances) could increase the risk to borrower privacy by facilitating re-identification of borrowers. A few commenters also expressed concern about the disclosure of confidential, proprietary information if the Bureau were to release AUS data to the public. One commenter did not support proposed Sec. 1003.4(a)(35) because, it argued, lenders would be required to disclose proprietary information. Another commenter expressed concern that competitor financial institutions could use public HMDA data to reverse engineer its proprietary underwriting systems, thereby harming its competitive position in the mortgage marketplace. Similarly, another commenter stated that to the extent that AUS data are available to persons outside government, such disclosure may pose serious risks that persons would seek to reverse engineer proprietary and confidential information about how an AUS is designed and risks significant competitive disadvantages for such entities whose AUS information would be collected. The commenter explained that persons may seek to reverse engineer the decision-making and purchase-process used by an AUS by analyzing the recommendations in connection with the other HMDA data that is disclosed to the public. The commenter reasoned that as a result of the volume of loan-level data reported pursuant to HMDA, disclosure of AUS data may well enable competitors and other parties to seek to recreate the criteria used by an AUS to reach recommendations on loans. The commenter urged the Bureau to ensure that if AUS data are to be reported by financial institutions, that only regulators of financial institutions and other government agencies responsible for fair lending enforcement have access to such data, and that it not be made available to financial institutions or others. Lastly, another commenter also expressed concern that the release of AUS data could facilitate reverse engineering to reveal proprietary information about an AUS and the profile of loans sold to a particular entity. The commenter stated that this could have a significant impact on an entity that developed an AUS by revealing proprietary information about the design of the AUS as well as the entity's loan purchases, security performance, and portfolio management. On the other hand, several commenters recommended that AUS data be released to the public and supported the proposal primarily based on the argument that such data will assist in fair lending analyses as well as in understanding access to credit. For example, one consumer advocate commenter stated that the collection and public dissemination of AUS information will help regulators, policymakers, and the public to more precisely investigate discriminatory mortgage lending. Another consumer advocate commenter stated that AUS data will identify which lenders rely on AUSs heavily as opposed to which lenders use manual underwriting, which it argued, can result in responsible lending being more accessible to populations that may have thin credit files or less than perfect credit. Lastly, another commenter stated that AUS data provides important insight into the modern underwriting process that will help policymakers better understand credit constraints and the challenges to maintaining broad access to credit. The Bureau has considered this feedback. It anticipates that, because public disclosure of itemized AUS data may raise concerns, such release may not be warranted. However, at this time the Bureau is not making determinations about what HMDA data will be publicly disclosed or the forms of such disclosures. 4(a)(36) Currently, neither HMDA nor Regulation C requires a financial institution to report whether a reportable transaction is a reverse mortgage. Although reverse mortgages that are home purchase loans, home improvement loans, or refinancings are reported under Regulation C currently, financial institutions are not required to separately identify if a reported transaction is a reverse mortgage.\370\ Proposed Sec. 1003.4(a)(36) provided that a financial institution must record whether the covered loan is, or the application is for, a reverse mortgage, and whether the reverse mortgage is an open- or closed-end transaction. The [[Page 66237]] Bureau solicited feedback regarding whether this proposed requirement is appropriate, whether commentary would help clarify or illustrate the requirement, and any costs and burdens associated with the proposed requirement.\371\ For the reasons discussed below, the Bureau is finalizing in Sec. 1003.4(a)(36) a requirement to identify whether the covered loan is, or the application is for, a reverse mortgage. --------------------------------------------------------------------------- \370\ The Bureau received a number of comments from consumer advocacy groups and industry commenters about including a reverse mortgage transaction as a type of covered loan that must be reported. The Bureau addresses those comments in the section-by- section analysis of Sec. 1003.2(q), which defines ``reverse mortgage.'' \371\ Commenters did not address the cost of finalizing the requirement to identify whether a transaction involves a reverse mortgage. However, the costs and benefits of all of the new and revised data points are discussed elsewhere in the Supplementary Information. --------------------------------------------------------------------------- Industry commenters opposed the requirement to report whether a loan or application is for a reverse mortgage because reverse mortgages are a small portion of the market. Consumer advocates supported the requirement, noting that data users currently cannot identify the populations taking out reverse mortgages. Consumer advocates generally stated that identifying which reported loans and lines of credit are reverse mortgages will help illuminate patterns of equity extraction by older consumers. It is important that the public and regulators be able to identify easily which transactions covered by Regulation C involve reverse mortgages. Reverse mortgages are substantively different from other mortgages and are subject to different underwriting criteria.\372\ Including in the dataset an indicator that readily identifies the transaction as a reverse mortgage will provide necessary context on the other data reported for the same transaction. For example, identification of a transaction as a reverse mortgage may help explain why certain data points are reported as not applicable to the transaction. As a result, financial institutions will need to spend less time verifying submitted data and users will have a better context in which to consider the data submitted, both for that transaction and in comparison with other transactions. --------------------------------------------------------------------------- \372\ See generally CFPB Report to Congress on Reverse Mortgages (2012), available at http://files.consumerfinance.gov/a/assets/documents/201206_cfpb_Reverse_Mortgage_Report.pdf. --------------------------------------------------------------------------- Pursuant to its authority under sections 305(a) and 304(b)(6)(J) of HMDA, the Bureau is finalizing in Sec. 1003.4(a)(36) a requirement to identify whether the covered loan is, or the application is for, a reverse mortgage. However, because the Bureau is also adopting Sec. 1003.4(a)(37), which will require financial institutions to identify whether the transaction involves an open-end line of credit, it is not necessary to require financial institutions to separately identify whether the reverse mortgage is a closed-end or open-end transaction. Instead, the final rule simplifies the reporting requirement in Sec. 1003.4(a)(36) to indicate only whether the transaction involves a reverse mortgage. Data users can use the reverse mortgage and open-end line of credit indicators in combination to determine whether a transaction involves a reverse mortgage and, if so, the type of reverse mortgage. This simplification also addresses the request of one consumer group to clarify potentially confusing terminology used in the proposed rule for different types of open-end lines of credit. 4(a)(37) Currently, neither HMDA nor Regulation C requires a financial institution to identify whether a reportable transaction is an open-end line of credit. Although dwelling-secured lines of credit currently may be reported as home purchase loans or home improvement loans, users of the HMDA data cannot identify which reported transactions involve open- end lines of credit. Proposed Sec. 1003.4(a)(37) provided that a financial institution must record whether the covered loan is, or the application is for, an open-end line of credit, and whether the covered loan is, or the application is for, a home-equity line of credit. The proposed rule defined ``open-end line of credit'' as a new term in Regulation C, and did not revise the current definition of home-equity line of credit. As discussed in the section-by-section analyses of Sec. 1003.2(h) and (o), the final rule deletes the definition of ``home-equity line of credit'' and modifies the proposed definition of ``open-end line of credit.'' The modified definition of open-end line of credit subsumes the current definition of home-equity line of credit. For the reasons discussed below, the Bureau is finalizing in Sec. 1003.4(a)(37) a requirement that financial institutions identify whether the covered loan is, or the application is for, an open-end line of credit, as that term is defined in the final rule. The Bureau solicited feedback regarding whether the proposed requirement to identify whether the transaction involved an open-end credit plan is appropriate and whether commentary would help clarify the requirement. Most commenters who addressed dwelling-secured open- end credit plans did not address this solicitation for comment. A number of industry participants recommended modifying the proposal to identify the transaction as either involving a home-equity line of credit, or not.\373\ Similarly, a consumer advocacy group commented that distinguishing between open-end lines of credit that are home- equity lines of credit and those that are not is confusing. --------------------------------------------------------------------------- \373\ These commenters generally also favored eliminating commercial loans from coverage under Regulation C, which they stated would eliminate reporting of most open-end lines of credit that are not home-equity lines of credit under the current definition in Regulation C. The coverage of commercial and business loans is discussed in the section-by-section analysis of Sec. 1003.3(c)(10). --------------------------------------------------------------------------- Some of the concerns that commenters raised about reporting HMDA data on dwelling-secured open-end credit plans will be mitigated by also requiring financial institutions to indicate whether the transaction being reported involves an open-end line of credit.\374\ Specifically, a number of industry commenters stated that a requirement to report data on open-end lines of credit would likely result in skewed data, including data that may create an inaccurate appearance of subprime lending. Industry trade groups stated that commingling data on open-end lines of credit with HMDA data on closed-end mortgage loans will produce misleading information. However, consumer advocates commented that having additional information about dwelling-secured open-end credit plans will enable communities to more fully understand the mortgage market and better serve vulnerable populations. One consumer advocate commented that open-end lines of credit should be identified in the data, given the difference in their underwriting relative to closed-end loans. Another consumer advocate commented that, without an indication that the transaction involves open-end credit, information on loan term and price is less meaningful. --------------------------------------------------------------------------- \374\ ``Open-end line of credit'' is defined in Sec. 1003.2(o) of the final rule. --------------------------------------------------------------------------- It is important that the public and public officials be able to identify easily which transactions covered by Regulation C involve open-end lines of credit. Open-end lines of credit are a different credit product than closed-end mortgage loans. Including in the dataset an indicator that readily identifies the transaction as an open-end line of credit will provide the public and public officials more context for the other data reported for the same transaction and will facilitate more-effective data analysis. For example, identification of a transaction as an open-end line of credit may help explain why the financial institution has reported certain data points as being not applicable to the transaction. As a result, financial [[Page 66238]] institutions will need to spend less time verifying submitted data and the public will have a better context in which to consider the data submitted, both for that transaction and in comparison with other transactions. Therefore, pursuant to its authority under sections 305(a) and 304(b)(6)(J) of HMDA, the Bureau is finalizing Sec. 1003.4(a)(37), which requires that financial institutions identify whether covered loans are, or applications are for, an open-end line of credit. The Bureau, however, is not finalizing the proposal that financial institutions also identify whether the covered loan is, or the application is for, a home-equity line of credit. As discussed in the section-by-section analysis of Sec. 1003.4(a)(38), the final rule also requires financial institutions to identify whether the covered loan is, or the application is for, a covered loan that is, primarily for a business or commercial purpose. In combination, the open-end line of credit indicator and the business- or commercial-purpose indicator can be used to identify whether open-end credit is for a consumer or business purpose.\375\ Therefore, a separate indicator for a consumer- purpose open-end credit plan secured by a dwelling is not necessary.\376\ The final rule simplifies the reporting requirement in Sec. 1003.4(a)(37) to indicate only whether the transaction involves an open-end line of credit. Simplifying the data point that indicates an open-end line of credit also addresses the request of one consumer group to clarify potentially confusing terminology used in the proposed rule for several types of open-end credit. --------------------------------------------------------------------------- \375\ See Sec. 1003.2(o) for additional discussion of consumer- and business-purpose open-end credit. \376\ In addition, because open-end line of credit is defined to be more comprehensive than home-equity line of credit, retaining both terms in Regulation C could result in inconsistencies in reporting the information. --------------------------------------------------------------------------- The Bureau did not propose any comment to accompany proposed Sec. 1003.4(a)(37) and commenters did not request clarifying commentary. For consistency and convenience, however, the final rule adds new comment 4(a)(37)-1, which references comments 2(o)-1 and -2 for guidance on determining whether a covered loan is, or an application is for, an open-end line of credit. 4(a)(38) Qualified Mortgage Indicator Currently, neither HMDA nor Regulation C contains requirements related to whether a loan would be considered a qualified mortgage under Regulation Z. Proposed Sec. 1003.4(a)(38) provided that a financial institution must record whether the covered loan is subject to the ability-to-repay provisions of Regulation Z and whether the covered loan is a qualified mortgage, as described under Regulation Z.\377\ The proposed rule also specified that financial institutions report the qualified mortgage information using a code to indicate which type of qualified mortgage described the covered loan. The Bureau solicited feedback regarding whether the proposed requirement was appropriate, would result in more useful data, and would impose additional burdens or result in additional challenges that the Bureau had not considered in making the proposal. In addition, the Bureau requested feedback regarding whether modifications to the proposed requirement would minimize the burden of collecting information related to a covered loan's qualified mortgage status. For the reasons discussed below, the Bureau is not finalizing proposed Sec. 1003.4(a)(38). --------------------------------------------------------------------------- \377\ The ability-to-repay provisions are in 12 CFR 1026.43. The proposed rule invoked the provisions on qualified mortgage in Sec. 1026.43(e) and (f). --------------------------------------------------------------------------- The Bureau received a significant number of comments from consumer advocacy groups, researchers, financial institutions, State and national trade associations, and other industry participants concerning proposed Sec. 1003.4(a)(38). Consumer advocates and researchers supported reporting whether a covered loan is a qualified mortgage. Some of these commenters also noted that a covered loan may fit into more than one category of qualified mortgage and that, if finalized, the reporting requirements should be structured to accommodate changes in underlying regulations (such as sunsetting categories for qualified mortgages). Some also recommended that financial institutions should report all of the categories under which a loan can be characterized as a qualified mortgage. A consumer advocacy group stated that qualified mortgage status limits liability for lenders, so these loans should be monitored closely to determine if that status results in more sustainable loan terms and better loan performance. Several consumer advocacy and research organization commenters identified the qualified mortgage data as one of the most important additions proposed and stated that understanding exactly how the Bureau's qualified mortgage regulation is affecting mortgage credit is critical to ensuring that the Bureau's joint goals of access to credit and consumer protection are both achieved. Industry commenters recommended against requiring reporting of qualified mortgage status. Some noted the same issues as consumer advocates and researchers had noted. In addition, industry commenters questioned the HMDA purpose for this data point and asserted a potentially stigmatizing effect for loans that are not qualified mortgages that would be inconsistent with Federal banking agencies' joint guidance and oral statements preserving a role for non-qualified mortgage loans.\378\ Financial institutions and industry trade groups commented that whether a loan is a qualified mortgage is often not known at origination. For example, one industry commenter reported that it does not limit its lending to qualified mortgages, so it would be burdensome and expensive to implement systems to track and report qualified mortgage status. Other industry commenters stated that whether a loan could be a qualified mortgage may be revealed by other data points, when considered together, including points and fees; rate spread; existence of features such as negative amortization, balloon payments, and prepayment penalties; whether a loan is backed by a government-sponsored enterprise or Federal agency; automated underwriting system results; high-cost status; and debt-to-income ratio. A number of industry commenters expressed concern about the consequences of misreporting a loan as either a qualified mortgage or not a qualified mortgage. Industry commenters requested that if the Bureau requires reporting the qualified mortgage status of loans, it should also add options to indicate whether a loan is exempt from the ability-to-repay requirements and whether the qualified mortgage status was relevant to the credit decision, and clarify reporting responsibilities for repurchases of loans misreported as qualified mortgages and for small-creditor loans sold to a buyer that is not a small creditor. --------------------------------------------------------------------------- \378\ CFPB, OCC, Bd. of Governors of the Fed. Reserve System, FDIC, and NCUA, Interagency Statement on Fair Lending Compliance and the Ability-to-Repay and Qualified Mortgage Standards Rule at 2 (2013), http://files.consumerfinance.gov/f/201310_cfpb_guidance_qualified-mortgage-fair-lending-risks.pdf. In part, the statement explains: [C]onsistent with the statutory framework, there are several ways to satisfy the Ability-to-Repay Rule, including making responsibly underwritten loans that are not Qualified Mortgages. The Bureau does not believe that it is possible to define by rule every instance in which a mortgage is affordable for the borrower. --------------------------------------------------------------------------- Coverage conditions and exemptions applicable to the ability-to- repay requirements mean that the reporting requirements in the proposed rule did [[Page 66239]] not apply to applications or open-end lines of credit, reverse mortgages, extensions of credit pursuant to certain programs, multifamily loans, or business-purpose loans. At the time of the proposed rule, the Bureau believed that financial institutions would be in a position to report the qualified mortgage status of each covered loan in a manner that is consistent with the regular business practices of financial institutions, and that such a reporting requirement would not be unduly burdensome. The Bureau has been persuaded, however, that reporting the qualified mortgage status and, as applicable, the type of qualified mortgage for each loan will impose burdens identified by industry commenters that were not intended and would not be justified by the benefits of this additional reporting requirement in the HMDA data. The final rule includes other new data that might be used to approximate the borrower's ability to repay and the loan's qualified mortgage status with sufficient accuracy to serve HMDA's purposes. Financial institutions should be able to provide this other data readily, without having to develop new collection mechanisms as might be necessary to report qualified mortgage status. In addition, the Bureau has not changed its position that non-qualified mortgages can satisfy ability-to-repay standards. The Bureau had not intended that a financial institution reporting under HMDA its reasonable belief about the qualified mortgage status of its loans at a point in time should be susceptible to increased public or regulatory scrutiny based on that classification. Therefore, the Bureau is not finalizing proposed Sec. 1003.4(a)(38). Business- or Commercial-Purpose Indicator Currently, neither HMDA nor Regulation C requires a financial institution to report whether a reportable transaction has a business or commercial purpose. Although business- and commercial-purpose transactions that are home purchase loans, home improvement loans, or refinancings are reported under Regulation C currently, financial institutions are not required to separately identify if a reported transaction has a business or commercial purpose.\379\ In the proposed rule, the Bureau expanded coverage of business and commercial transactions, but it did not separately propose a specific requirement for financial institutions to differentiate those transactions in their reported HMDA data.\380\ As discussed in the section-by-section analysis of Sec. 1003.3(c)(10), the final rule maintains the current requirement that financial institutions must report business- and commercial-purpose transactions that are home purchase loans, home improvement loans, or refinancings. To make the data collected on business- and commercial-purpose transactions more useful, Sec. 1003.4(a)(38) of the final rule requires financial institutions to report whether the covered loan or application is or will be made primarily for a business or commercial purpose. --------------------------------------------------------------------------- \379\ The Bureau received many comments about the coverage of business- and commercial-purpose loans in HMDA and Regulation C. The Bureau addresses those comments in the section-by-section analysis of Sec. 1003.3(c)(10), which provides an exclusion for some business- and commercial-purpose transactions. \380\ In the proposed rule, the Bureau invited feedback regarding whether, if commercial loans were not exempted in the final rule, it would be appropriate to add a loan purpose requirement applicable to commercial loans or some other method of uniquely identifying commercial loans in the HMDA data. 79 FR 51731, 51767 (Aug. 29, 2014). --------------------------------------------------------------------------- Even though the final rule does not expand the scope of coverage of business- and commercial-purpose loans, some of the concerns that commenters raised about reporting HMDA data on all business- and commercial-purpose loans are relevant to the current, more limited reporting requirements.\381\ For example, some industry commenters stated that mixing data about dwelling-secured, commercial-purpose transactions with traditional mortgage loans would skew the HMDA dataset and impair its integrity for users of the data. These concerns will be mitigated by also requiring financial institutions to indicate whether the transaction being reported involves business- or commercial-purpose credit. Including in the dataset an indicator that readily identifies the transaction as business- or commercial-purpose credit will provide the public and public officials more context for the other data reported for the same transaction and will facilitate more-effective data analysis. The public and public officials will be able to use this information to improve their understanding of how financial institutions may be meeting the housing needs of their communities and public-sector funds are being distributed. These HMDA purposes are served by gathering data not only about transactions to individual consumers for consumer purposes, but also, for example, about the available stock of multifamily rental housing in particular communities. --------------------------------------------------------------------------- \381\ See section-by-section analysis of Sec. 1003.3(c)(10). --------------------------------------------------------------------------- For the reasons discussed above and pursuant to the Bureau's authority under sections 305(a) and 304(b)(6)(J) of HMDA, Sec. 1003.4(a)(38) of the final rule provides that a financial institution must identify whether the covered loan or application is or will be made primarily for a business or commercial purpose. Proposed 4(a)(39) Section 304(b) of HMDA permits the disclosure of such other information as the Bureau may require.\382\ Pursuant to HMDA sections 305(a) and 304(b)(5)(D), the Bureau proposed to require financial institutions to report, for a home-equity line of credit and an open- end reverse mortgage, the amount of the draw on the covered loan, if any, made at account opening. For the reasons given below, the Bureau is not finalizing proposed Sec. 1003.4(a)(39). --------------------------------------------------------------------------- \382\ Section 1094(3)(A)(iv) of the Dodd-Frank Act amended section 304(b) of HMDA. --------------------------------------------------------------------------- Several consumer advocates supported the proposed requirement to report the initial draw for an open-end line of credit. One consumer advocate said that such information would assist in identifying loans where the borrower draws an amount at or close to the maximum amount available for the line of credit. The commenter believed that these loans were more properly characterized as closed-end credit. Another consumer advocate stated that, for reverse mortgages, large initial draws may be predictive of future financial difficulties. Information regarding the initial draw on an open-end line of credit might provide important information about the behavior and degree of leverage of borrowers with such loans. Industry commenters, however, almost universally opposed the initial draw reporting requirement. Many of these commenters believed that the amount of the initial draw would provide no valuable data. A few commenters stated that the first draw played an insignificant role in underwriting or pricing decisions, and other commenters noted that the amount reflected the choice of the borrower. Several commenters were generally skeptical of the utility of the information or asserted that it offered little value for purposes of fair lending analysis or determining whether financial institutions were meeting the housing needs of their communities. The amount of the initial draw on a home-equity line of credit or an open-end reverse mortgage would provide [[Page 66240]] information about the leverage of borrowers with open-end lines of credit. The extent of leverage is important for evaluating the potential overextension of credit and the risk of default faced by borrowers in certain communities. Such information may also be used to detect structural problems in the mortgage market. However, the initial draw often consists only of an amount necessary to cover fees or charges associated with opening the account, or to satisfy the requirements of a particular promotion. The Bureau believes that these data would fail to provide the information about borrower leverage or use of open-end lines of credit that the proposal intended to capture. Industry commenters also stated that proposed Sec. 1003.4(a)(39) would distort the HMDA data. The Bureau understands that many initial draws do not occur at account opening for a variety of reasons. For example, consumers might wait days or even months before drawing on the line of credit. By requiring reporting of the draw at account opening, proposed Sec. 1003.4(a)(39) would omit these draws and therefore fail to serve its intended purpose. The Bureau could extend the reporting period applicable to proposed Sec. 1003.4(a)(39) in an attempt to capture information about these loans. However, the Bureau understands that the necessary information often exists in separate loan servicing systems rather than the loan origination system. As detailed in the section 1022 discussion below, the Bureau recognizes that mandatory open-end line of credit reporting will impose a significant operational burden on financial institutions, largely because open-end lines of credit are originated and maintained on different computer systems than traditional mortgages. Upgrading or integrating the separate systems used to originate and service open-end lines of credit would represent a similar operational burden. Forcing such a systems change for the purpose of collecting a single data point would impose an unjustified burden on financial institutions. For the reasons provided above, the Bureau is not finalizing proposed Sec. 1003.4(a)(39). 4(b) Collection of Data on Ethnicity, Race, Sex, Age, and Income Section 1003.4(b)(1) of current Regulation C requires that a financial institution collect data about the ethnicity, race, and sex of the applicant or borrower as prescribed in appendix B. Section 1003.4(b)(2) provides that the ethnicity, race, sex, and income of an applicant or borrower may but need not be collected for loans purchased by the financial institution. The Bureau proposed to add age to Sec. 1003.4(b)(1) and (b)(2), and proposed to amend Sec. 1003.4(b)(1) by requiring a financial institution to collect data about the ethnicity, race, sex, and age of the applicant or borrower as prescribed in both appendices A and B. The Bureau also proposed minor wording changes to Sec. 1003.4(b)(1) and (b)(2). Consistent with the current requirement under the regulation, proposed Sec. 1003.4(b)(2) provided that ethnicity, race, sex, and income data may but need not be collected for loans purchased by a financial institution. While the proposed reporting requirement does not require reporting of ethnicity, race, sex, age, and income for loans purchased by a financial institution, the Bureau solicited feedback on whether this exclusion is appropriate. In particular, the Bureau specifically solicited feedback on the general utility of ethnicity, race, sex, age, and income data on purchased loans and on the unique costs and burdens associated with collecting and reporting the data that financial institutions may face if the reporting requirement were modified to no longer permit optional reporting but instead require reporting of this applicant and borrower information for purchased loans. A few commenters opposed the proposed optional reporting of ethnicity, race, sex, age, and income for loans purchased by a financial institution. For example, one consumer advocate stated that the proposal creates a significant gap in the data that is reported under HMDA and such data is important to achieving HMDA's goals. The commenter noted that while it may be possible to close this gap by using the proposed ULI to match a purchased loan with the data on the ethnicity, race, sex, age, and income reported by the originating financial institution, doing so will be time consuming and would require a significant effort from users of the data. The commenter recommended that the Bureau clarify in commentary that the Bureau considers it reasonable for any institution purchasing covered loans to negotiate a contractual agreement requiring the seller institution to provide all data required by HMDA. The commenter also suggested that if the optional reporting of the ethnicity, race, sex, age, and income for purchased loans under proposed Sec. 1003.4(b)(2) remains, it should be limited only to instances where the financial institution does not have and cannot reasonably obtain the information. Another consumer advocate suggested that reporting of demographic information on purchased loans be required to enhance its understanding of trends in the mortgage market and how well financial institutions are or are not serving the communities which it represents. Similarly, another commenter expressed concern that an increase in the depository institution threshold and any delay in establishing a unique ULI will enable the nonreporting of critical demographic data with respect to large numbers of purchased loans and as such, recommended that the Bureau extend the mandatory reporting of ethnicity, race, sex, age, and income to purchased loans. Lastly, another commenter recommended that unless and until the ULI is successfully implemented, purchased loans should not be excluded from this reporting requirement. On the other hand, the industry commenters who addressed this aspect of the proposal supported the current optional reporting of ethnicity, race, sex, age, and income data on purchased loans. For example, one industry commenter recommended that reporting of this data should only be optional because it would be an enormous regulatory burden for community banks to collect and report. Another commenter stated that purchased loans should not be subject to HMDA reporting overall. The Bureau is adopting Sec. 1003.4(b)(1) as proposed, with a few changes. First, the Bureau deleted reference to appendix A in Sec. 1003.4(b)(1) since the instructions in the final rule requiring a financial institution to collect data about the ethnicity, race, and sex of the applicant or borrower are located in appendix B. Second, the Bureau removed age from Sec. 1003.4(b)(1) since, as discussed above, the instructions in the final rule requiring a financial institution to collect the age of an applicant or borrower are found in comments 4(a)(10)(ii)-1, -2, -3, -4, and -5. The Bureau has considered the feedback and determined that the final rule will continue to allow for optional reporting of ethnicity, race, sex, and income for loans purchased by a financial institution. In addition, as proposed, the final rule will also allow optional reporting of age for loans purchased by a financial institution. While the Bureau recognizes the potential utility of ethnicity, race, sex, age, and income data on purchased loans, it is concerned with the costs and burdens associated with collecting and reporting the data that financial institutions will face if the reporting [[Page 66241]] requirement is mandatory. Consequently, the Bureau is adopting Sec. 1003.4(b)(2) as proposed, which provides a financial institution with the option to collect the ethnicity, race, sex, age, and income data for covered loans it purchased. 4(c) Optional Data 4(c)(1) Current Sec. 1003.4(c)(1) provides that a financial institution may report the reasons it denied a loan application but is not required to do so. As discussed in the section-by-section analysis of Sec. 1003.4(a)(16), the final rule makes reporting of denial reasons mandatory instead of optional. To conform to that requirement, the final rule deletes Sec. 1003.4(c)(1). 4(c)(2) Current Sec. 1003.4(c)(2) provides that a financial institution may report requests for preapproval that are approved by the institution but not accepted by the applicant but is not required to do so. The Bureau proposed to make reporting of requests for preapprovals approved by the financial institution but not accepted by the applicant mandatory under Sec. 1003.4(a) instead of optional under Sec. 1003.4(c)(2). Few commenters addressed this proposal specifically, though as discussed above in the section-by-section analysis of section 2(b)(2) some commenters addressed other aspects of preapproval programs. A few commenters questioned the value of mandatory reporting for preapprovals approved but not accepted. The Bureau is finalizing the requirement to report preapprovals approved by the financial institution but not accepted by the applicant because it believes that reporting of preapprovals approved by the financial institution but not accepted by the applicant provides context for denials of preapproval requests, and improves fair lending analysis because it allows denials to be compared to a more complete set of approved preapproval requests.\383\ To conform to that requirement, the final rule deletes Sec. 1003.4(c)(2). --------------------------------------------------------------------------- \383\ The Bureau incorporates and relies on its prior description of the importance and usefulness of this data. See 79 FR 51731, 51809-10 (Aug. 29, 2014). --------------------------------------------------------------------------- 4(c)(3) Section 1003.4(c)(3) of Regulation C currently provides that a financial institution may report, but is not required to report, home- equity lines of credit made in whole or in part for the purpose of home improvement or home purchase. As discussed in the section-by-section analysis of Sec. 1003.2(o), the final rule makes reporting of open-end lines of credit (which include home-equity lines of credit) mandatory, rather than optional. To conform to that modification, the final rule deletes Sec. 1003.4(c)(3) and comment 4(c)(3)-1. 4(d) Section 1003.4(d) of Regulation C currently provides exclusions for certain data. As discussed in the section-by-section analysis of Sec. 1003.3(c), the Bureau is moving those exclusions to Sec. 1003.3(c). To conform to this modification, the final rule removes and reserves Sec. 1003.4(d). 4(e) For ease of reference, the Bureau is republishing Sec. 1003.4(e) and making technical modifications. No substantive change is intended. 4(f) Quarterly Recording of Data The Bureau proposed to move the data recording requirement in Sec. 1003.4(a) to proposed Sec. 1003.4(f) and to make technical modifications to the requirement. Proposed Sec. 1003.4(f) provided that a financial institution was required to record \384\ the data collected pursuant to Sec. 1003.4 on a loan/application register within 30 calendar days after the end of the calendar quarter in which final action was taken (such as origination or purchase of a covered loan, or denial or withdrawal of an application). The Bureau received no comments on proposed Sec. 1003.4(f) and is finalizing it with technical amendments. The Bureau is renumbering proposed comment 4(a)- 1.iv as comment 4(f)-1 and existing comments 4(a)-2 and -3 as comments 4(f)-2 and -3, respectively. The Bureau is also making technical modifications to these comments to clarify a financial institution's obligation to record data on a quarterly basis. --------------------------------------------------------------------------- \384\ A financial institution's obligation to report data is addressed below in the section-by-section analysis of Sec. 1003.5(a). --------------------------------------------------------------------------- Section 1003.5 Disclosure and Reporting 5(a) Reporting to Agency 5(a)(1) HMDA section 304(h)(1) provides that a financial institution shall submit its HMDA data to the Bureau or to the appropriate agency for the institution in accordance with rules prescribed by the Bureau. HMDA section 304(h)(1) also directs the Bureau to develop regulations, in consultation with other appropriate agencies, that prescribe the format for disclosures required under HMDA section 304(b), the method for submission of the data to the appropriate agency, and the procedures for disclosing the information to the public. HMDA section 304(n) also requires that the data required to be disclosed under HMDA section 304(b) shall be submitted to the Bureau or to the appropriate agency for any institution reporting under HMDA, in accordance with regulations prescribed by the Bureau. HMDA section 304(c) requires that information required to be compiled and made available under HMDA section 304, other than loan/application register information under section 304(j), must be maintained and made available for a period of five years.\385\ --------------------------------------------------------------------------- \385\ HMDA section 304(j)(6) requires that loan/application register information described in HMDA section 304(j)(1) for any year shall be maintained and made available, upon request, for three years. --------------------------------------------------------------------------- Currently, Sec. 1003.5(a)(1) of Regulation C requires that, by March 1 following the calendar year for which data are compiled, a financial institution must submit its complete loan/application register to the agency office specified in appendix A. Section 1003.5(a)(1) also provides that a financial institution shall retain a copy of its complete loan/application register for its records for at least three years. Part II of appendix A to Regulation C provides information concerning where financial institutions should submit their complete loan/application registers. Additional information concerning submission of the loan/application register is found in comments 4(a)- 1.vi and -1.vii, 5(a)-1 and -2, and 5(a)-5 through -8. Comment 5(a)-2 provides that a financial institution that reports 25 or fewer entries on its loan/application register may submit the register in paper form. The Bureau proposed several changes to Sec. 1003.5(a)(1). Quarterly Reporting The Bureau proposed that a financial institution with a high transaction volume report its HMDA data to the Bureau or appropriate agency on a quarterly, rather than an annual, basis. Proposed Sec. 1003.5(a)(1)(ii) required that, within 60 calendar days after the end of each calendar quarter, a financial institution that reported at least 75,000 covered loans, applications, and purchased covered loans, combined, for the preceding calendar year would submit its loan/ application register containing all data required to be recorded pursuant to Sec. 1003.4(f).\386\ The [[Page 66242]] Bureau's proposal allowed for a delay in the effective date of proposed Sec. 1003.5(a)(1)(ii) and stated that the Bureau was considering a delay of at least one year from the effective date of the other proposed amendments to Regulation C. --------------------------------------------------------------------------- \386\ Currently, Sec. 1003.4(a) requires that ``all reportable transactions shall be recorded, within thirty calendar days after the end of the calendar quarter in which final action is taken (such as origination or purchase of a loan, or denial or withdrawal of an application), on a register in the format prescribed in Appendix A of this part.'' The Bureau's proposal moved this requirement, with some revisions, to proposed Sec. 1003.4(f). The Bureau is finalizing Sec. 1003.4(f) as proposed with technical amendments. --------------------------------------------------------------------------- The Bureau received several comments on proposed Sec. 1003.5(a)(1)(ii), including comments on the threshold for coverage under the provision and its effective date. For the reasons discussed below, the Bureau is adopting Sec. 1003.5(a)(1)(ii) as proposed with several modifications and with an effective date of January 1, 2020. The Bureau also is adopting new Sec. 1003.6(c)(2) to provide a safe harbor to protect financial institutions that satisfy certain conditions from liability for HMDA and Regulation C violations for errors and omissions in data submitted pursuant to Sec. 1003.5(a)(1)(ii). The requirement to submit data on a quarterly basis. Consumer advocate and researcher commenters supported the proposal to require quarterly reporting insofar as quarterly reporting would not adversely impact the accuracy of annual HMDA data released to the public and would expedite the FFIEC's annual release of HMDA data.\387\ All but a few industry commenters opposed the proposal, with most comments questioning the benefits of quarterly reporting and raising concerns about burdens on financial institutions subject to the proposed quarterly reporting requirement, the accuracy of data submitted on a quarterly basis, and error thresholds applicable to quarterly submissions. --------------------------------------------------------------------------- \387\ As discussed above in part II.B, the FFIEC currently makes available on its Web site aggregate and loan-level HMDA data. Currently, these data are made available in September of the year following the calendar year in which the data were collected. --------------------------------------------------------------------------- Most industry commenters asserted that institutions subject to the proposed quarterly reporting requirement would expend significant additional resources to comply with the requirement. These comments clearly conveyed that the need to ``clean'' HMDA data to maximize its accuracy before submission to regulators would be a significant driver of the increased operational burden associated with quarterly reporting. Although commenters suggested that most financial institutions currently review and correct their HMDA data throughout the year the data are collected, several stated that rigorous scrubbing typically is performed before the data are submitted to regulators by March 1 of the following year. A few commenters stated that performing this level of review four times each year instead of one would significantly increase costs to financial institutions and noted that these costs could change from quarter to quarter, depending on volume. Several industry commenters also stated that HMDA data reported on a quarterly basis would be less accurate than data reported on an annual basis. A few commenters argued that systemic errors can take months to resolve and that the current annual reporting cycle maximizes opportunities to address systemic issues before the HMDA data are submitted to regulators. A few commenters noted that the need to ``update'' quarterly data previously submitted, whether to reflect the sale of a loan or to correct errors or omissions, would complicate submission for quarterly reporters and would introduce inaccuracies. Several commenters stated that, even with increased resources devoted to preparing quarterly submissions, 60 days after the close of the quarter would not provide sufficient time to properly scrub quarterly data prior to submission, especially if the Bureau were to finalize its proposal to require reporting of additional transactions and data. A few commenters expressed concern that errors or omissions in quarterly submissions would expose financial institutions subject to proposed Sec. 1003.5(a)(1)(ii) to increased risk of violations under the agencies' accuracy requirements in determining HMDA compliance. Industry commenters also argued that the significant burden of quarterly reporting would outweigh any benefits it might provide. Several commenters stated that annual reporting of HMDA data is sufficient to satisfy the purposes of HMDA. A few commenters stated that useful analyses cannot be performed with quarterly data, especially for purposes of fair lending enforcement. One commenter argued that, because only the largest lenders would be reporting quarterly, quarterly data would not provide a good ``community lending'' picture. One commenter noted that, with each quarter, the reduction in delay between a reportable event and the date it is reported that exists under the annual reporting scheme is decreased, and so the corresponding benefit of quarterly reporting is decreased. As discussed above, several commenters stated that quarterly reporting would decrease the accuracy of HMDA data submitted, not improve it as the Bureau suggested in the proposal. A few commenters expressed skepticism that quarterly reporting would significantly hasten the FFIEC's release of annual HMDA data, and several commenters asserted that quarterly reporting would provide limited or no benefit to the public and public officials, who would continue to have access to HMDA data on an annual basis only under the proposal. The Bureau has considered the comments received and has determined that the benefits of quarterly reporting by large-volume financial institutions justify some degree of additional burden on these financial institutions. Quarterly reporting will provide regulators with more timely data, which will be of significant value for HMDA and market monitoring purposes. Currently, HMDA data may be reported as many as 14 months after final action is taken on an application or loan.\388\ Although this delay decreases as the year progresses (e.g., a loan originated in December is currently reported by March 1 of the following year), increasing the timeliness of HMDA data will provide meaningful benefits to various analyses by regulators. Timelier data will allow regulators to determine, in much closer to ``real time,'' whether financial institutions are fulfilling their obligations to serve the housing needs of communities in which they are located. Timelier identification of risks to local housing markets and troublesome trends by regulators will allow for more effective interventions or other actions by the agencies and other public officials. Quarterly data will allow for deeper and timelier analyses of the lending activities of large volume lenders. For example, in fair lending examinations, quarterly reporting will permit comparisons of recent data from the subjects of examinations and similar lenders. Further, timelier HMDA data will allow the agencies to not only better understand the market and identify trends and shifts that may warrant interventions, but also will provide data that will allow the agencies to sooner understand the impacts of prior interventions. For example, although the Bureau's Ability-to-Repay and Qualified Mortgage provisions went into effect in January 2014, data on loans subject to these provisions were not reported until March 2015. Timelier HMDA data would have enhanced the Bureau's understanding of the effects of those protections. --------------------------------------------------------------------------- \388\ A loan originated on January 2, 2015 may not be reported until March 1, 2016. --------------------------------------------------------------------------- [[Page 66243]] Further, quarterly reporting would allow for the release of timelier data and analysis to the public. In its proposal, the Bureau noted that, although based on its analysis to date it believed that releasing HMDA data to the public on a quarterly basis may create risks to applicant and borrower privacy that would not be justified by the benefits of such release, it would evaluate options for the agencies' release of data or analysis more frequently than annually. Upon further consideration, the Bureau has determined that useful analyses of data submitted on a quarterly basis, or aggregated data, could be provided to the public in a manner that appropriately protects applicant and borrower privacy.\389\ The Bureau intends to release analyses of HMDA data or aggregated HMDA data to the public more frequently than annually in such a privacy-protective manner. As aggregates of HMDA data collected by all reporting institutions during a given calendar year currently are not publicly available until September of the following year, the release of aggregate quarterly data or analysis would further the statute's purposes and deliver a direct disclosure benefit to the public. --------------------------------------------------------------------------- \389\ At this time, the Bureau believes that loan-level data should not be released to the public more frequently than annually due to privacy concerns. Currently, dates are redacted from the modified loan/application register and the agencies' annual loan- level data release to reduce re-identification risk created by the disclosure of loan-level data. See 55 FR 27886, 27888 (July 6, 1990) (concerning the agencies' decision to release loan-level data to the public and stating that ``[a]n unedited form of the data would contain information that could be used to identify individual loan applicants'' and that the data would be edited prior to public release to remove the application identification number, the date of application, and the date of final action). Based on its analysis to date, the Bureau believes that disclosure of loan-level data with more granular date information than year of final action would create risks to applicant and borrower privacy that are not outweighed by the benefits of such disclosure. --------------------------------------------------------------------------- The Bureau acknowledges the concerns industry commenters raised about burdens that could be imposed by the proposed quarterly reporting requirement. Based on the comments, the Bureau understands that these burdens would result mainly from a requirement that quarterly submissions achieve the degree of data accuracy the regulators currently require in annual submissions. To address this concern, the Bureau is adopting a quarterly reporting requirement, but is finalizing Sec. 1003.5(a)(1)(i) and Sec. 1003.5(a)(1)(ii) with modifications and adopting new Sec. 1003.6(c)(2) to provide that quarterly submissions are considered preliminary submissions and to provide a safe harbor that protects a financial institution that satisfies certain conditions from being cited for violations of HMDA or Regulation C for errors and omissions in its quarterly submissions. Under the final rule, within 60 calendar days after the end of each calendar quarter except the fourth quarter,\390\ financial institutions subject to Sec. 1003.5(a)(1)(ii) will submit the HMDA data that they are already required to record on their loan/application registers within 30 days after the end of each calendar quarter. Pursuant to new Sec. 1003.6(c)(2), errors and omissions in the data submitted pursuant to Sec. 1003.5(a)(1)(ii) will not be considered HMDA or Regulation C violations assuming the conditions that currently provide a safe harbor for errors and omissions in quarterly recorded data are satisfied.\391\ By March 1 of the following year, quarterly reporters will submit their final annual HMDA data pursuant to Sec. 1003.5(a)(1)(i), which will be subject to examination for HMDA and Regulation C compliance and required to satisfy the agencies' error thresholds. This annual submission will contain all reportable data for the preceding calendar year. --------------------------------------------------------------------------- \390\ Sixty days after end of the fourth calendar quarter coincides with March 1, the date by which all financial institutions must submit their annual HMDA data pursuant to Sec. 1003.5(a)(1)(i) as finalized. Financial institutions subject to Sec. 1003.5(a)(1)(ii) will report their fourth quarter data as part of their annual submission. In its annual submission, a quarterly reporter will resubmit the data previously submitted for the first three calendar quarters of the year, including any corrections to the data, as well as its fourth quarter data. \391\ Currently, Sec. 1003.6(b)(3) provides that ``[i]f an institution makes a good-faith effort to record all data concerning covered transactions fully and accurately within thirty calendar days after the end of each calendar quarter, and some data are nevertheless inaccurate or incomplete, the error or omission is not a violation of the act or this part provided that the institution corrects or completes the information prior to submitting the loan/ application register to its regulatory agency.'' Modifications to this provision and new Sec. 1003.6(c)(2) are discussed below in the section-by-section analysis of Sec. 1003.6(c). --------------------------------------------------------------------------- The Bureau is moving the certification requirement from proposed Sec. 1003.5(a)(1)(iii) into adopted Sec. 1003.5(a)(1)(i) to clarify that such certification is only required in connection with a financial institution's annual data submission, and is making other technical and conforming changes to Sec. 1003.5(a)(1)(i) and Sec. 1003.5(a)(1)(ii).\392\ The final rule thus preserves the annual reporting structure of current Regulation C for all financial institutions reporting under HMDA and imposes an additional, quarterly submission requirement on large-volume institutions only. These additional submissions need only consist of the data a large-volume institution is already required to maintain, however, significantly limiting the burden imposed by the quarterly reporting requirement.\393\ --------------------------------------------------------------------------- \392\ As discussed below, the Bureau also is modifying the certification provision in the final rule to clarify who may certify on behalf of a financial institution and to provide that the institution must certify to the completeness of the submission as well as to its accuracy. \393\ This approach also addresses concerns raised by a few industry commenters that sixty days is insufficient time after the close of the quarter for a financial institution to submit its quarterly data. Financial institutions must already record the data to be submitted under Sec. 1003.5(a)(1)(ii) within thirty days after the calendar quarter. --------------------------------------------------------------------------- The final rule provides the benefits of timelier data to the regulators without requiring quarterly reporters to apply to each quarterly submission the rigorous scrubbing typically performed on annual HMDA submissions. The Bureau has considered that potential inaccuracies in quarterly data submitted under the final rule may decrease the data's utility and reliability. Although a quarterly reporting requirement would ideally yield timelier and highly accurate data, the Bureau recognizes that minimizing burdens to financial institutions associated with quarterly reporting may require a tradeoff between these goals. Based on its examination experience, the Bureau believes that the typical degree of accuracy in quarterly recorded HMDA data maintained by most financial institutions will be sufficient for the kinds of analyses for which the Bureau anticipates quarterly data may be used.\394\ The Bureau further believes that edit checks it is building into the HMDA data submission tool it is developing will decrease some types of inaccuracies in submissions. --------------------------------------------------------------------------- \394\ The Bureau believes that the accuracy levels typically found in quarterly recorded data likely result from the good-faith requirement set forth in current Sec. 1003.6(b)(3) and the data review that many financial institutions perform year-round. --------------------------------------------------------------------------- As an alternative to the adopted approach, the Bureau considered requiring semiannual reporting rather than quarterly reporting. Under this approach, large volume reporters would submit their final HMDA data for the first and second quarters of the calendar year within 60 days after the end of the second quarter, and their final HMDA data for the third and fourth quarters by March 1 of the following year. These submissions would be subject to examination for HMDA compliance and the agencies' error thresholds. This approach would require financial institutions subject to Sec. 1003.5(a)(1)(ii) to perform the more rigorous data review described by industry commenters only twice each year, rather than four times, reducing burden on [[Page 66244]] these institutions compared to the Bureau's proposal. Further, industry comments suggest that data submitted on a semiannual basis may contain fewer inaccuracies than data submitted on a quarterly basis. This alternative approach would not provide as timely data to the agencies as the quarterly reporting approach discussed above, however, reducing the utility of the data to the agencies as well as the disclosure benefit to the public. To the extent that quarterly data contain errors and omissions, the Bureau believes these inaccuracies are unlikely to be significant enough to have a negative impact on the analyses the data will allow and that the risks of inaccurate data are outweighed by the benefits of timelier data. Although the approach adopted in the final rule reduces the likelihood that the quarterly reporting requirement will expedite the agencies' release of annual HMDA data as compared to the proposal,\395\ it will nonetheless allow the Bureau to provide a direct disclosure benefit to the public in the form of periodic aggregate data or analysis, as described above. Based on the comments received, the Bureau has determined that the approach adopted in the final rule would limit burden on financial institutions subject to Sec. 1003.5(a)(1)(ii) and that it best balances any burden with the benefits of more frequent HMDA reporting. --------------------------------------------------------------------------- \395\ As explained in the proposal, the Bureau believed that the proposed quarterly reporting requirement would reduce reporting errors and allow it to process data throughout the year. See 79 FR 51731, 51811 (Aug. 29, 2014). The Bureau believed that these benefits of quarterly reporting would reduce the time currently required to edit and process annual HMDA data, which would expedite the release of the annual data to the public. Because the final rule provides that data submitted quarterly need only be preliminary data and a quarterly reporter will resubmit all previously submitted quarterly data with its annual submission, the Bureau now believes that the quarterly reporting requirement may not significantly reduce the time needed to process the annual data. The Bureau notes, however, that it believes improvements to the submission process, including a requirement that edit checks currently performed by the processor after submission are performed by the financial institution prior to submission, will reduce the time needed to process the annual HMDA data and will thus expedite the release of the annual data to the public. --------------------------------------------------------------------------- A few commenters raised operational questions concerning quarterly reporting, including how financial institutions reporting on a quarterly basis would report updates and corrections to previously- submitted quarterly data and whether they would be required to update and correct previously-submitted data with each quarterly submission. For example, these commenters suggested that quarterly reporters may be required to report the same loan repeatedly throughout the calendar year in order to correct errors in a previous quarterly submission or reflect the sale or repurchase of the loan. A quarterly reporter is required to update a previously reported transaction in a subsequent quarterly submission if the new information is required to be recorded on the loan/application register pursuant to Sec. 1003.4(f). Under the final rule, a financial institution required to comply with Sec. 1003.5(a)(1)(ii) must submit, within 60 calendar days after the end of each calendar quarter except the fourth quarter, its quarterly loan/application register containing all data required to be recorded for that quarter pursuant to Sec. 1003.4(f). Pursuant to Sec. 1003.4(f), data must be recorded on the quarterly loan/ application register within 30 calendar days after the end of the calendar quarter in which final action is taken (such as origination or purchase of a covered loan, sale of a covered loan in the same calendar year it is originated or purchased, or denial or withdrawal of an application). The sale or repurchase of a loan, if occurring in the first three quarters of the calendar year, must be reflected in the quarterly submission for the quarter in which the action was taken because it must be recorded on the quarterly loan/application register for that quarter pursuant to Sec. 1003.4(f).\396\ --------------------------------------------------------------------------- \396\ See Sec. 1003.4(f); comment 4(a)(11)-9 (where a financial institution originates a covered loan in one quarter and sells it in a subsequent quarter of the same calendar year, the institution must record the purchaser on the loan/application register for the quarter in which the covered loan was sold); comment 4(a)-6 (clarifying that a repurchase is reported as a purchase). --------------------------------------------------------------------------- Final Sec. 1003.6(c)(2) provides that, if a quarterly reporter makes a good faith effort to report all data required to be reported pursuant to Sec. 1003.5(a)(1)(ii) fully and accurately within 60 calendar days after the end of each calendar quarter, inaccuracies or omissions in quarterly data submitted need not be corrected or completed until the financial institution submits it annual loan/ application register pursuant to Sec. 1003.5(a)(1)(i). Thus, for example, if a quarterly reporter makes a good faith effort to report income for a particular transaction accurately in its quarterly submission and discovers in a subsequent quarter that the reported amount was incorrect, it is not required to update the record for the transaction until it submits its annual loan/application register pursuant to Sec. 1003.5(a)(1)(i). The Bureau received no comments on proposed comment 5(a)-1. The Bureau is adopting comment 5(a)-1 as proposed, modified to conform to Sec. 1003.5(a)(1)(ii) as finalized and to add two new subsections clarifying how a surviving or newly formed financial institution's obligation to report on a quarterly basis under Sec. 1003.5(a)(1)(ii) is determined for the calendar year of the merger or acquisition and the calendar year after the merger or acquisition. The Bureau received no comments on proposed comment 5(a)-2. The Bureau is adopting proposed comment 5(a)-2 as modified in two ways. First, comment 5(a)-2 as adopted requires that, if the appropriate Federal agency for a financial institution subject to Sec. 1003.5(a)(1)(ii) changes, the financial institution must identify the new appropriate Federal agency in its quarterly submission pursuant to Sec. 1003.5(a)(1)(ii) beginning with its submission for the quarter of the change, unless the change occurs during the fourth quarter, in which case the financial institution must identify the new agency in its annual submission pursuant to Sec. 1003.5(a)(1)(i). This change aligns the requirement for quarterly submissions with the requirement for annual submissions and conforms to Sec. 1003.5(a)(1)(ii) as adopted. The Bureau has also modified comment 5(a)-2 to provide illustrative examples. The threshold for coverage under Sec. 1003.5(a)(1)(ii). The Bureau proposed that the quarterly reporting requirement under proposed Sec. 1003.5(a)(1)(ii) apply to a financial institution that reported at least 75,000 covered loans, applications, and purchased covered loans, combined, for the preceding calendar year. The Bureau received no comments from consumer advocates on the proposed threshold for quarterly reporting. The Bureau received a few industry comments on the proposed threshold. One industry commenter suggested that the Bureau should impose a $10 billion asset threshold, instead of a transaction-based threshold, to align the quarterly reporting requirement with the Bureau's supervisory authority. Another industry commenter suggested that the threshold should be lowered to 50,000 transactions in the preceding calendar year so as to increase the amount of quarterly data available for analysis, and yet another suggested that all HMDA reporters should be required to report on a quarterly basis to facilitate the earlier release of the annual HMDA data by the agencies. One industry commenter suggested that the threshold should include originated covered loans only (not applications or purchased loans), though offered no rationale for [[Page 66245]] this recommendation. Two industry comments stated that the Bureau's estimate of the number of institutions that would be covered by the proposed threshold was inaccurate because it did not take into account the Bureau's proposal to expand transactional coverage to include open- end lines of credit and commercial-purpose loans. One of these comments, submitted by several national trade associations, stated that the associations' members reported that mandatory open-end line of credit reporting would double or triple the number of reportable transactions. For the reasons described below, the Bureau is adopting Sec. 1003.5(a)(1)(ii) with modifications to the proposed threshold to exclude purchased covered loans from the threshold calculation and to lower the threshold from at least 75,000 transactions in the preceding calendar year to at least 60,000 transactions in the preceding calendar year. The Bureau has determined that it is appropriate to exclude purchased covered loans from the quarterly reporting threshold due to changes to the currently-applicable FFIEC guidance concerning reporting of repurchased loans that it is adopting herein.\397\ The Bureau understands that loans are repurchased under a variety of circumstances and arrangements, some of which are very common. The Bureau lacks data concerning repurchase activity sufficient to allow it to estimate the impact of a quarterly reporting threshold that takes repurchases into consideration, however, and is concerned that inclusion of repurchases in the quarterly reporting threshold calculation could conceivably significantly increase the number of financial institutions that would be required to comply with Sec. 1003.5(a)(1)(ii). Rather than excluding only repurchased loans from the threshold calculation, which would require financial institutions to identify repurchased loans in their HMDA data and would thus add burden, the final rule excludes all purchases from the threshold. Institutions that are required to submit their HMDA data on a quarterly basis under Sec. 1003.5(a)(1)(ii) will include purchased covered loans in the quarterly data they submit, but purchased covered loans will not be considered in determining whether a financial institution must comply with Sec. 1003.5(a)(1)(ii). --------------------------------------------------------------------------- \397\ The Bureau is adopting comment 4(a)-6 to require the reporting of most repurchases as purchased loans regardless of when the repurchase occurs. As adopted, comment 4(a)-6 eliminates the exception for reporting repurchases occurring in the same calendar year as origination that currently exists under FFIEC guidance. --------------------------------------------------------------------------- Based on 2013 HMDA data, a threshold of at least 60,000 transactions, excluding purchases, would have required 29 financial institutions to report on a quarterly basis in 2014. In 2013, these 29 institutions reported approximately 49 percent of all transactions reported under HMDA.\398\ The Bureau notes that market fluctuations may influence the number of financial institutions that are required to comply with Sec. 1003.5(a)(1)(ii) from year to year. For example, based on preliminary HMDA data submitted for 2014, a threshold of at least 60,000 transactions, excluding purchases, would have required only approximately 19 financial institutions to report on a quarterly basis in 2015. The preliminary data suggest that these institutions reported approximately 37 percent of all transactions reported under HMDA for 2014. The Bureau recognizes that the percentage of the market reflected in quarterly reported data may vary from year to year and has determined that a 60,000 transaction volume threshold should result in data sufficient to realize the benefits of a quarterly reporting requirement. --------------------------------------------------------------------------- \398\ These numbers align with those based on 2012 HMDA data and the proposed 75,000 transaction threshold included in the Bureau's proposal. See 79 FR 51731, 51811 (Aug. 29, 2014) (noting that, based on 2012 HMDA data, the 75,000 transaction threshold proposed would have required 28 financial institutions to report on a quarterly basis in 2013 and that, in 2012, these 28 institutions reported approximately 50 percent of all transactions reported under HMDA). --------------------------------------------------------------------------- The Bureau believes that the requirement to report open-end lines of credit under the final rule is unlikely to have a significant impact on the number of financial institutions that must comply with Sec. 1003.5(a)(1)(ii). As discussed elsewhere in this document, the Bureau has faced challenges in analyzing the impact of the mandatory reporting of open-end lines of credit required under the final rule on financial institutions' HMDA-reportable transaction volume.\399\ Using estimates of the number of consumer-purpose open-end line of credit originations and applications in 2013,\400\ the Bureau's analysis suggests that, had these originations and applications been required to be reported for 2013, one additional financial institution would have become a quarterly reporter in 2014, as compared to the number of institutions that would have become quarterly reporters without mandatory reporting of open-end line of credit originations and applications.\401\ Based on these estimates as applied to 2013 HMDA data, the Bureau believes that, although mandatory reporting of consumer-purpose open-end lines of credit and applications will increase HMDA-reportable transaction volumes for many financial institutions, and may increase these volumes significantly for some financial institutions, this increase is unlikely to significantly increase the number of financial institutions required to comply with Sec. 1003.5(a)(1)(ii). Further, the Bureau believes that relatively few dwelling-secured, commercial-purpose open- end lines of credit are used for home purchase, home improvement, or refinancing purposes.\402\ The Bureau thus expects that reporting these transactions will not significantly increase the number of transactions reported by financial institutions and, accordingly, will not significantly increase the number of financial institutions that must comply with Sec. 1003.5(a)(1)(ii). --------------------------------------------------------------------------- \399\ See section-by-section analyses for Sec. 1003.2(g), (o), Sec. 1003.3(c)(10), and part VII. \400\ As discussed in part VII, these estimates are based on 2013 HMDA data, 2013 Call Report data, and Consumer Credit Panel data. Due to the limited data available, these estimates rely on several assumptions. \401\ This analysis assumes that these institutions did not voluntarily report open-end line of credit originations and applications in 2013. \402\ As discussed in the section-by-section analysis of Sec. 1003.3(c)(10), the final rule maintains coverage of commercial- purpose transactions generally at its existing level. Section 1003.3(c)(10) does expand coverage of dwelling-secured commercial- purpose lines of credit, which are not currently reported, by requiring them to be reported if they primarily are for home purchase, home improvement, or refinancing purposes, however. As discussed above, the Bureau has faced challenges estimating institutions' open-end lending volume given limitations in publicly available data sources. For example, it is difficult to estimate commercial-purpose open-end lending volume because available data sources do not distinguish between consumer- and commercial-purpose lines of credit. --------------------------------------------------------------------------- The final rule does not base the threshold for quarterly reporting on a financial institution's asset size, as recommended by a commenter. The central goal of the quarterly reporting requirement is to provide the agencies with timelier HMDA data in a quantity sufficient to perform meaningful analyses. A transaction-based threshold limits the imposition of costs associated with quarterly reporting to those institutions with the largest transaction volumes in order to minimize the number of financial institutions subject to the requirement while maximizing the volume of data reported on a quarterly basis. An asset- based threshold cannot guarantee such a relationship between the number of affected institutions and the quantity of data submitted on a quarterly basis. [[Page 66246]] Effective date of Sec. 1003.5(a)(1)(ii). The Bureau received no consumer advocate comments and very few industry comments on its request for comment as to whether and how long it should delay the effective date of proposed Sec. 1003.5(a)(1)(ii). Industry commenters recommended a delay of either one or two years from the effective date of the other amendments to Regulation C. The Bureau is adopting an effective date of January 1, 2020 for Sec. 1003.5(a)(1)(ii). This delay is to permit financial institutions subject to the quarterly reporting requirement time to implement amended Regulation C and to allow for two annual reporting cycles under the amended rule before quarterly submissions are required. Financial institutions that report for 2019 at least 60,000 covered loans and applications, combined, excluding purchased covered loans, must comply with Sec. 1003.5(a)(1)(ii) in 2020. Financial institutions subject to Sec. 1003.5(a)(1)(ii) in 2020 will first report quarterly data under this provision by May 30, 2020. Elimination of Paper Reporting The Bureau proposed to delete comment 5(a)-2, which allows a financial institution that reports 25 or fewer entries on its loan/ application register to submit the register in paper form, and to clarify in proposed Sec. 1003.5(a)(1) that the register must be submitted in electronic format in accordance with instructions in appendix A. The Bureau received no comments from consumer advocates on this proposal and very few comments from industry. One industry commenter supported the proposal. A few industry commenters opposed the proposal. The majority of these commenters suggested that the option to report on paper should be available until the Bureau builds an improved data submission tool. One industry commenter argued that it would be cost prohibitive for a financial institution to purchase new software to report a few transactions per month. For the reasons described below, the Bureau is finalizing its proposal to delete comment 5(a)-2. In recent years, very few financial institutions have submitted their loan/application registers in paper form. Further, the Bureau is finalizing its proposal to exclude from the definition of financial institution any institution that originated less than 25 closed-end mortgages loans and less than 100 open-end lines of credit,\403\ so only a financial institution that originated exactly 25 closed-end mortgage loans and received no other applications would be eligible to submit its register in paper form under amended Regulation C were this option to remain available. The Bureau is developing an improved HMDA data submission system and tools to assist smaller financial institutions with data entry. The Bureau is confident that these developments will reduce even further any need for a financial institution to submit its HMDA data in paper form. --------------------------------------------------------------------------- \403\ See Sec. 1003.2(g). --------------------------------------------------------------------------- As discussed in part VI below, most of Sec. 1003.5(a) is effective January 1, 2019 and applies to data collected and recorded in 2018 pursuant to this final rule.\404\ However, the Bureau will intake and process HMDA data on behalf of the agencies using the improved web- based submission tool it is developing beginning with financial institutions' 2017 HMDA data submission. Data collected and recorded in 2017 pursuant to current Regulation C will be reported by March 1, 2018 pursuant to current Sec. 1003.5(a). The final rule's amendments to supplement I effective January 1, 2018 generally maintain the current commentary to Sec. 1003.5(a) with respect to the reporting of data collected in 2017 and reported in 2018 but, because the improved submission tool that financial institutions will use to submit their 2017 HMDA data will not accept loan/application registers in paper form, the Bureau is deleting comment 5(a)-2 effective January 1, 2018. --------------------------------------------------------------------------- \404\ Section 1003.5(a)(1)(ii) is effective January 1, 2020. --------------------------------------------------------------------------- Retention of Annual Loan/Application Register in Electronic Format Section 1003.5(a)(1) requires that a financial institution shall retain a copy of its complete loan/application register for three years, but current Regulation C is silent concerning the formats in which the complete loan/application register may be retained. The Bureau proposed comment 5(a)-4 to clarify that retention of the loan/ application register in electronic format is sufficient to satisfy the requirements of Sec. 1003.5(a)(1). The Bureau received no consumer advocate comments concerning proposed comment 5(a)-4. The Bureau received very few industry comments concerning proposed comment 5(a)-4, but all supported the proposal. The Bureau adopts comment 5(a)-4 as proposed, modified to clarify that the obligation to retain the loan/application register applies only to a financial institution's annual data submitted pursuant to Sec. 1003.5(a)(1)(i). Submission Procedures As stated in its proposal, as part of its efforts to improve and modernize HMDA operations, the Bureau is developing improvements to the HMDA data submission process. The Bureau proposed to reorganize parts I and II of appendix A and portions of the commentary so that instructions relating to data submission are found in one place in the regulation. Specifically, the Bureau proposed to: Delete the content of part II of appendix A and comment 5(a)-1; move the portion of comment 4(a)-1.vi concerning certification to proposed Sec. 1003.5(a)(1)(iii); and incorporate the pertinent remaining portion of comment 4(a)-1.vi and comments 4(a)-1.vii and 5(a)-7 and -8 into proposed instructions 5(a)-2 and -3 in appendix A and delete the remaining portions of these comments. The Bureau proposed new instruction 5(a)-1 in appendix A to provide procedural and technical information concerning data submission. The Bureau did not receive comment on these proposals. The Bureau noted in its proposal that, as part of its efforts to improve and modernize HMDA operations, it was considering various improvements to the HMDA data submission process. The Bureau received a few industry comments concerning data submission. A few commenters urged the Bureau to adopt a web-based submission tool that is accessible by multiple work stations and users within a financial institution, rather than a downloadable tool that would reside on a single work station. Commenters also suggested that the tool automatically identify and code inapplicable fields so that, for example, if a loan is identified on the loan/application register as a commercial-purpose loan, all data fields not required to be reported for commercial-purpose loans would automatically be populated with the code for ``not applicable.'' Finally, a few commenters stated that the tool should be secure and should not allow regulators access to any data until the data is submitted by the financial institution. As will be described in more detail in separately published procedures, the Bureau is developing a Web-based submission tool that financial institutions will use to submit their HMDA data to their regulators. The Bureau anticipates that this submission tool will be accessible from multiple work stations and will perform edit checks on HMDA data prior to submission. The Bureau believes that this submission tool will significantly improve the data submission process. The Bureau does not anticipate that this [[Page 66247]] submission tool will include a data entry function, and therefore it would not have capacity to automatically identify and code inapplicable fields, as recommended by some commenters. The Bureau believes that, at this time, the costs of a Web-based data entry tool outweigh the benefits such a tool could provide. The Bureau is developing a tool to assist smaller financial institutions with data entry, but the Bureau anticipates that it will not be Web-based. Effective January 1, 2019, the Bureau is deleting appendix A from Regulation C and is instead separately publishing procedures for the submission of HMDA data.\405\ The Bureau is adopting modifications to Sec. 1003.5(a)(1)(i) and (ii) and new Sec. 1003.5(a)(4) to clarify that financial institutions submit HMDA data to the appropriate Federal agency for the financial institution. The Bureau is also adopting modifications to the certification requirement in Sec. 1003.5(a)(1)(i).\406\ These modifications require that a financial institution certify to the completeness of the HMDA data submitted as well as to their accuracy in order to reflect the obligation to report both accurate and complete data, and clarify who may certify on behalf of a financial institution in order to align the requirement with current practice. --------------------------------------------------------------------------- \405\ See final Sec. 1003.5(a)(5) (providing that procedures for the submission of data pursuant to Sec. 1003.5(a) are published on the Bureau's Web site). \406\ The Bureau proposed to move the certification requirement from the transmittal sheet to proposed Sec. 1003.5(a)(1)(iii). As discussed above, in the final rule, the Bureau is moving the certification requirement to Sec. 1003.5(a)(1)(i) to clarify that the certification is only required in connection with a financial institution's annual data submission pursuant to that paragraph. --------------------------------------------------------------------------- As discussed in part VI below, most of Sec. 1003.5(a) is effective January 1, 2019 and applies to data collected and recorded in 2018 pursuant to this final rule.\407\ However, the Bureau will intake and process HMDA data on behalf of the agencies using the improved Web- based submission tool it is developing beginning with financial institutions' 2017 HMDA data submission. Data collected and recorded in 2017 pursuant to current Regulation C will be reported by March 1, 2018 pursuant to current Sec. 1003.5(a). The final rule's amendments to supplement I effective January 1, 2018 generally maintain the current commentary to Sec. 1003.5(a) with respect to the reporting of data collected in 2017 and reported in 2018, but operation of this improved submission tool requires that current comment 5(a)-1 is deleted effective January 1, 2018.\408\ Current comments 5(a)-3 and -4 have been incorporated elsewhere in the final rule as appropriate and are also deleted from supplement I effective January 1, 2018. In addition, part II of appendix A to current Regulation C is revised effective January 1, 2018 to provide updated instructions relating to the reporting of 2017 HMDA data. --------------------------------------------------------------------------- \407\ Section 1003.5(a)(1)(ii) is effective January 1, 2020. \408\ As discussed above, comment 5(a)-2 also is deleted effective January 1, 2018. --------------------------------------------------------------------------- Finally, the Bureau received several identical comments from employees of one financial institution suggesting that the Bureau change the date by which annual HMDA data must be submitted pursuant to Sec. 1003.5(a)(1)(i) to allow financial institutions additional time to prepare HMDA data for submission. The final rule retains the March 1 deadline for submitting annual HMDA data pursuant to Sec. 1003.5(a)(1)(i). Postponing this deadline would necessarily delay the release of annual HMDA data to the public. The Bureau has determined that any benefits to financial institutions that would result from additional time to prepare HMDA data for submission are outweighed by the costs of such an approach to the public disclosure goals of the statute. 5(a)(1)(iii) The Bureau is adopting new Sec. 1003.5(a)(1)(iii) to provide that, when the last day for submission of data prescribed under Sec. 1003.5(a)(1) falls on a Saturday or Sunday, a submission shall be considered timely if it is submitted on the next succeeding Monday.\409\ This is consistent with the approach taken by the agencies when this situation has arisen in the past.\410\ --------------------------------------------------------------------------- \409\ As discussed above, the certification requirement set forth in proposed Sec. 1003.5(a)(1)(iii) is moved into final Sec. 1003.5(a)(1)(i). \410\ For example, in 2015, March 1 fell on a Sunday and the reporting deadline for 2014 HMDA data was moved to March 2. Fed. Fin. Insts. Examination Council, CRA/HMDA Reporter, Calendar Year 2014 Initial Submission Deadline, at 1 (Jan. 2015), available at http://www.ffiec.gov/hmda/pdf/15news.pdf. --------------------------------------------------------------------------- 5(a)(2) The Bureau did not propose changes or solicit feedback regarding Sec. 1003.5(a)(2) in the proposal. Current Sec. 1003.5(a)(2) provides that a subsidiary of a bank or savings association shall complete a separate loan/application register and submit it directly or through its parent to the agency of its parent. The Bureau is making non- substantive changes to Sec. 1003.5(a)(2) to clarify that a financial institution that is a subsidiary of a bank or savings association shall complete a separate loan/application register and submit it directly or through its parent to the appropriate Federal agency for its parent at the address identified by the agency. 5(a)(3) The Bureau proposed Sec. 1003.5(a)(3) to require that when an institution reports its data, the institution shall provide with each covered loan or application its Legal Entity Identifier (LEI) issued by a utility endorsed by the LEI Regulatory Oversight Committee or a utility endorsed or otherwise governed by the Global LEI Foundation (GLEIF) (or any successor of the GLEIF) after the GLEIF assumes operational governance of the global LEI system. Regulation C currently requires financial institutions to provide a Reporter's Identification Number (HMDA RID) in their transmittal sheet and loan/application register. The HMDA RID consists of an entity identifier specified by the financial institution's appropriate Federal agency combined with a code that designates the agency. Each Federal agency chooses the entity identifier that its financial institutions would use in reporting their HMDA data. Currently, the Research Statistics Supervision and Discount (RSSD) number is used by institutions supervised by the Board and depository institutions supervised by the Bureau; the Federal Tax Identification number is used by nondepository institutions supervised by agencies other than the Board; the charter number is used by depository institutions supervised by the National Credit Union Administration (NCUA) and the OCC; and the certificate number is used by depository institutions supervised by the FDIC. For the reasons discussed below, the Bureau is adopting Sec. 1003.5(a)(3) as proposed. The Bureau is also incorporating material from proposed Sec. 1003.5(a)(2) in appendix A, as discussed below. The Bureau solicited feedback on whether the LEI would be a more appropriate entity identifier than the current HMDA RID and also whether other identifiers, such as the RSSD number or Nationwide Mortgage Licensing System & Registry identifier (NMLSR ID), would be an appropriate alternative to the proposed LEI. Several commenters opposed the requirement for financial institutions to obtain an LEI, mostly citing the cost associated with obtaining an LEI and the availability of alternative identifiers. The Bureau acknowledged in the proposal that requiring financial institutions to obtain an LEI would impose some costs. However, because the LEI system is based on a cost-recovery model, the cost associated with obtaining an LEI could decrease as the [[Page 66248]] LEI identifier is used more widely. Despite the cost, the Bureau believes that the benefit of all HMDA reporters using an LEI may justify the associated costs. An LEI could improve the ability to identify financial institution reporting the data and link it to its corporate family. Facilitating identification of a financial institution's corporate family could help data users identify possible discriminatory lending patterns and assist in identifying market activity and risks by related companies. Some commenters suggested that instead of the proposed LEI, the Bureau should consider requiring either the current HMDA RID, NMLSR ID, Federal Tax Identification number, or a Bureau-created unique identifier for entities. These suggested alternatives may have some merit, but they pose concerns that would make data aggregation, validation, and analyses difficult for users. The current HMDA RID varies across each Federal agency and there is a lack of consistency in the availability of the financial institutions corporate information when researching a financial institution's corporate information using the HMDA RID. For example, a search using the FDIC certificate number may only provide the bank holding company and financial institution affiliates, but may not provide other corporate information. The NMLSR ID would not pose much additional burden on industry because most institutions that originate loans are already assigned unique identifier by the NMLS. However, the NMLSR does not contain consistent information regarding corporate information. For example, parent company and affiliate information are not readily available in the NMLS. The Federal Tax Identification Number would also not pose additional burden on industry because financial institutions would already have one. However, as the Bureau explained in the proposal, there is no mechanism to link nondepository institutions identified by a Federal Tax Identification Number to related companies. All of the suggested alternatives above would still result in a lack of information to enable users to link corporate information to the financial institution reporting HMDA data. Accordingly, the Bureau is adopting Sec. 1003.5(a)(3) to require an institution to provide its LEI with its submission. As mentioned in the section-by-section analysis of Sec. 1003.4(a)(1)(i), the Bureau is making a technical change and moving proposed Sec. 1003.5(a)(3)(i) and (ii) to Sec. 1003.4(a)(1)(i)(A)(1) and (2) for ease of reference. The Bureau concludes that requiring use of the LEI will improve the ability to identify the legal entity that is reporting data and to link it to its corporate family. For these reasons, pursuant to HMDA section 305(a), the Bureau is adopting Sec. 1003.5(a)(3) as proposed. This requirement is necessary and proper to effectuate HMDA's purposes and facilitate compliance therewith. By facilitating identification, this requirement will help data users achieve HMDA's objectives of identifying whether financial institutions are serving the housing needs of their communities, as well as identifying possible discriminatory lending patterns. This requirement could also assist in identifying market activity and risks by related companies. The Bureau proposed Sec. 1003.5(a)(4) to require a financial institution to report its parent company, if any, when reporting its data. Currently, Regulation C requires financial institutions to report their parent company, if any, in the transmittal sheet as provided in appendix A. Information about a financial institution's parent company helps ensure that the financial institution's submission can be linked with that of its corporate parent. One commenter suggested that the name and LEI of the parent company should be provided by the financial institution reporting data because financial institutions that submit HMDA data may be affiliated with large financial institutions. This commenter stated that the lack of information around parent company affiliations can make it difficult to accurately analyze lending patterns. The Bureau has determined that requiring the parent company of a financial institution to obtain an LEI would not be appropriate. Requiring the parent company to obtain an LEI specifically for HMDA purposes, except if the parent company is also HMDA reporter, and requiring the financial institution to submit its parent company's LEI with its HMDA data submission would be an unnecessary additional burden because, once the LEI is fully implemented, information regarding parent company is expected to become available.\411\ Therefore, the Bureau does not believe that the benefit of requiring parent information justifies the burden since information about parent company most likely will be available through an alternative source. Accordingly, the Bureau will not require a financial institution to provide its parent information, including the parent's LEI, and therefore is withdrawing the requirement in proposed Sec. 1003.5(a)(4) that a financial institution shall identify its parent company, if any. --------------------------------------------------------------------------- \411\ See generally Fin. Stability Bd., A Global Legal Entity Identifier for Financial Markets 38-39 (June 8, 2012), http://www.financialstabilityboard.org/wp-content/uploads/r_120608.pdf?page_moved=1 (including a recommendation on LEI reference data relating to ownership; Fin. Stability Bd., LEI Implementation Group, Fourth Progress Notes on the Global LEI Initiative 4 (Dec. 11, 2012), http://www.financialstabilityboard.org/wp-content/uploads/r_121211.pdf?page_moved=1 (noting that the LEI Implementation Group is developing proposals for additional reference data on the direct and ultimate parent(s) of legal entities and on relationship data more generally). --------------------------------------------------------------------------- The Bureau also proposed comment 5(a)-3 to explain that the parent company to be identified by the financial institution pursuant to Sec. 1003.5(a)(3) is the entity that holds or controls an ownership interest in the financial institution that is greater than 50 percent. One industry commenter suggested that the Bureau should explain which parent should be identified by the financial institution. This commenter added, however, that they do not see the benefit that information about the parent company would provide. As mentioned above, once the LEI is fully implemented, information about parent company is expected to become available and therefore, the Bureau will not require a financial institution to identify its parent. Consequently, the Bureau is modifying comment 5(a)-3 to remove parent company. Additionally, the Bureau is moving the instructions to 5(a)(2) in proposed appendix A and is incorporating it into Sec. 1003.5(a)(3) because of the removal of appendix A from the final rule, as explained in the section-by-section analysis of appendix A below. Pursuant to its authority under section HMDA 305(a), the Bureau is also adding certain information related to the data submission that is currently provided on an institution's transmittal sheet, as illustrated in current appendix A, to Sec. 1003.5(a)(3). The Bureau believes this will aid in the analyses of HMDA data and assist agencies in the supervision of financial institutions. 5(a)(4) As discussed in the section-by-section analysis of Sec. 1003.5(a)(3) above, the Bureau is withdrawing proposed Sec. 1003.5(a)(4). In its place, the Bureau is adopting new Sec. 1003.5(a)(4) to clarify that, for purposes of Sec. 1003.5(a), ``appropriate Federal agency'' means the appropriate agency for the financial institution as determined pursuant to HMDA section 304(h)(2) or, with respect to a financial institution subject to the [[Page 66249]] Bureau's supervisory authority under section 1025(a) of the Consumer Financial Protection Act of 2010 (12 U.S.C. 5515(a)), the Bureau. This paragraph reflects the regulatory structure in place since the Dodd- Frank Act became effective, as first described in the FFIEC's January 2012 CRA/HMDA Bulletin.\412\ --------------------------------------------------------------------------- \412\ Fed. Fin. Insts. Examination Council, CRA/HMDA Reporter, 2011 HMDA Panel Changes Resulting from Dodd-Frank Act, at 1-3 (Jan. 2012), available at http://www.ffiec.gov/hmda/pdf/11news.pdf. --------------------------------------------------------------------------- 5(a)(5) As described above,\413\ effective January 1, 2019, the Bureau is deleting appendix A from Regulation C and is instead separately publishing procedures for the submission of HMDA data. The Bureau is adopting new Sec. 1003.5(a)(5) to identify where these procedures will be published. --------------------------------------------------------------------------- \413\ See section-by-section analysis of Sec. 1003.5(a)(1). See also section-by-section analysis of appendix A. --------------------------------------------------------------------------- 5(b) Public Disclosure of Statement Under Regulation C as originally promulgated, the disclosure statement was the means by which financial institutions made available to the public the aggregate data required to be disclosed under HMDA section 304.\414\ At present, the FFIEC prepares an individual disclosure statement for each financial institution using the HMDA data submitted by the institution for the preceding calendar year. --------------------------------------------------------------------------- \414\ 41 FR 23931, 23937-38 (June 14, 1976). --------------------------------------------------------------------------- 5(b)(1) HMDA section 304(k) requires the FFIEC to make available a disclosure statement for each financial institution required to make disclosures under HMDA section 304.\415\ Section 1003.5(b)(1) of Regulation C requires that the FFIEC prepare a disclosure statement for each financial institution based on the data each financial institution submits on its loan/application register. The Bureau proposed to modify Sec. 1003.5(b)(1) to clarify that, although some financial institutions would report on a quarterly basis under proposed Sec. 1003.5(a)(1)(ii), disclosure statements for these financial institutions would be based on all data submitted by each institution for the preceding calendar year. The Bureau also proposed to replace the word ``prepare'' with ``make available'' in Sec. 1003.5(b)(1). --------------------------------------------------------------------------- \415\ HMDA section 304(k)(1)(A) provides that a financial institution ``shall make a disclosure statement available, upon request, to the public no later than 3 business days after the institution receives the statement from the Federal Financial Institutions Examination Council.'' --------------------------------------------------------------------------- The Bureau received no comments on proposed Sec. 1003.5(b)(1). Therefore, the Bureau adopts this provision generally as proposed, with one modification to clarify that disclosure statements made available in 2018 are based on a financial institution's annual 2017 data submitted pursuant to current Sec. 1003.5(a), and that disclosure statements made available beginning in 2019 are based on a financial institution's annual data submitted pursuant to Sec. 1003.5(a)(1)(i), not data submitted on a quarterly basis pursuant to Sec. 1003.5(a)(1)(ii). As discussed in its proposal,\416\ the Bureau believes that advances in technology may permit, for example, the FFIEC to produce an online tool that would allow users of the tool to generate disclosure statements. It is the Bureau's interpretation that the FFIEC's obligation under HMDA section 304(k) would be satisfied if the FFIEC produced such a tool, which in turn would produce disclosure statements upon request. Further, pursuant to its authority under HMDA section 305(a), the Bureau believes that permitting the FFIEC to produce a tool that allows members of the public to generate disclosure statements is necessary and proper to effectuate the purposes of HMDA and to facilitate compliance therewith. --------------------------------------------------------------------------- \416\ 79 FR 51731, 51841 (Aug. 29, 2014). --------------------------------------------------------------------------- 5(b)(2) HMDA section 304(k)(1) requires that, in accordance with procedures established by the Bureau, a financial institution shall make its disclosure statement available to the public upon request no later than three business days after it receives the statement from the FFIEC. HMDA section 304(m) provides that a financial institution shall be deemed to have satisfied the public availability requirements of section 304(a) if it compiles the information required at the home office of the institution and provides notice at the branch locations specified in HMDA section 304(a) that such information is available from the home office upon written request. Section 1003.5(b)(2) of Regulation C requires that each financial institution make its disclosure statement available to the public in its home office within three business days of receiving it. In addition, Sec. 1003.5(b)(3) requires that a financial institution must either (1) make the statement available to the public in at least one branch office in each other MSA and each other MD where the institution has offices or (2) post the address for sending written requests for the disclosure statement in the lobby of each branch office in each other MSA and each other MD and provide a copy of the disclosure statement within 15 calendar days of receiving a written request. The Bureau proposed to require a financial institution to make its disclosure statement available to the public by making available a notice that clearly conveys that the disclosure statement may be obtained on the FFIEC Web site and that includes the FFIEC's Web site address. The Bureau proposed a new comment 5(b)-3 to provide an example of notice content that would satisfy the requirements of proposed Sec. 1003.5(b)(2). The Bureau also proposed to modify comment 5(b)-2 to conform to proposed Sec. 1003.5(b)(2) and to allow a financial institution to provide the proposed notice in paper or electronic form. For the reasons discussed below, the Bureau is adopting Sec. 1003.5(b)(2) as proposed with clarifying modifications. The Bureau received several comments from industry concerning proposed Sec. 1003.5(b)(2). Most of these comments supported the proposal. Many industry commenters stated that they had never or rarely received a request for their disclosure statements. The one consumer advocate that commented on proposed Sec. 1003.5(b)(2) also supported the proposal. Two industry commenters suggested that, because disclosure statements are available on the FFIEC Web site, requiring financial institutions to provide members of the public seeking HMDA data with the notice under proposed Sec. 1003.5(b)(2) was unnecessary and duplicative. One of these commenters suggested that, as an alternative to the notice required under proposed Sec. 1003.5(b)(2), the Bureau should revise the posted lobby notice required pursuant to Sec. 1003.5(e) to include text referring members of the public to the FFIEC Web site to obtain the institution's HMDA data. Although the final rule relieves financial institutions of the obligation to provide the disclosure statement directly to the public, the Bureau has determined that provision of the notice required under Sec. 1003.5(b)(2) to a member of the public seeking a financial institution's disclosure statement is necessary to ensure that she is clearly informed of where to obtain it. Currently, a member of the public seeking a disclosure statement from a financial institution would leave the institution with the data in hand. As amended, Sec. 1003.5(b)(2) requires that the individual take an additional step to [[Page 66250]] obtain the data--visit the Bureau's Web site--but provides that she leaves the institution with the specific information needed to do so. Another industry commenter opposed the maintenance of disclosure statements on a government Web site, stating that it is an inefficient use of government resources. The Bureau disagrees. The government has played a critical role in disseminating HMDA data to fulfill the purposes of the statute since 1980, when Congress amended HMDA to require the FFIEC to implement a system to facilitate access to HMDA data required to be disclosed under HMDA section 304.\417\ For the reasons given in the proposal, the Bureau concludes that the FFIEC's use of a Web site to publish HMDA data satisfies this statutory obligation and that this means of providing access to HMDA data is necessary and proper to effectuate HMDA's purposes and facilitate compliance therewith.\418\ The Bureau believes that a significant portion of HMDA data used by the public and public officials is obtained from the FFIEC's Web site, rather than directly from financial institutions. --------------------------------------------------------------------------- \417\ HMDA section 304(f), added by Housing and Community Development Act of 1980, Public Law 96-399, section 340, 94 Stat. 1614, 1657-58 (1980). \418\ 79 FR 51731, 51818 (Aug. 29, 2014). --------------------------------------------------------------------------- One other industry commenter opposed the proposal, arguing that eliminating the option to obtain data directly from a financial institution, and instead requiring a member of the public seeking a financial institution's disclosure statement to obtain it online, would impose undue burden on some members of the public. This commenter argued that a substantial portion of the public does not have access to the internet or does not know how to use it. The commenter suggested that this population is likely largely comprised of low-income minorities, some middle-aged women, and seniors, with the result that the Bureau's proposal may disproportionately impact vulnerable groups. The commenter also asserted that it is significantly more inconvenient and expensive for a member of the public seeking a disclosure statement to locate it online, download it, and print it than it is to obtain a copy of a printed disclosure statement at a financial institution's home or branch office. Available data suggests that approximately 99 percent of Americans have access to broadband internet.\419\ Although the Bureau recognizes that accessing data online is not without barriers for some members of the public and that broadband speeds vary,\420\ the Bureau believes that the vast majority of members of the public seeking HMDA data should be able to readily access HMDA disclosure statements online with minimum inconvenience, if any. As discussed in the Bureau's proposal, such inconvenience is not greater than, and is likely less than, the potential inconvenience of receiving a disclosure statement on a floppy disc or other electronic data storage medium which may be used with a personal computer, as is expressly contemplated by HMDA section 304(k)(1)(b). In fact, the Bureau believes that, for most HMDA users, accessing disclosure statements online will be much more convenient than contacting individual financial institutions to request the data. Further, because members of the public are not currently entitled to printed disclosure statements free of charge, Sec. 1003.5(b)(2) as adopted should not increase monetary costs to members of the public desiring a disclosure statement in printed form.\421\ Although there may be members of the public that are adversely affected by the elimination of the right to obtain a disclosure statement directly from a financial institution,\422\ the Bureau has determined that the burden to financial institutions associated with the provision of disclosure statements directly to members of the public upon request is not justified by any benefit to the current disclosure statement dissemination scheme. --------------------------------------------------------------------------- \419\ Anne Neville, Nat'l. Broadband Map has Helped Chart Broadband Evolution, Nat'l. Telecomms. & Info. Admin. Blog (Mar. 23, 2015), http://www.ntia.doc.gov/blog/2015/national-broadband-map-has-helped-chart-broadband-evolution. \420\ Id. (noting the gap between urban and rural areas with respect to broadband at higher speeds). \421\ Under current Sec. 1003.5(d), financial institutions may charge a reasonable fee for any costs incurred in providing or reproducing their HMDA data. This provision is retained in the final rule. \422\ The Bureau notes that, under final Sec. 1003.5(d)(2), a financial institution may make its disclosure statement available to the public in addition to, but not in lieu of, the notice required by Sec. 1003.5(b)(2). --------------------------------------------------------------------------- The Bureau is adopting Sec. 1003.5(b)(2) as proposed with three modifications. Reference to making the disclosure statement available to the public is eliminated in order to clarify that a financial institution must only make the notice described available to the public. This paragraph is also modified to clarify that the notice must only be made available in branch offices physically located in a MSA or MD. Finally, this paragraph is modified to reflect that the Bureau will publish the disclosure statements on the Bureau's Web site. The Bureau believes it is reasonable to deem that financial institutions make disclosure statements available, pursuant to HMDA sections 304(k)(1) and 304(m), by referring members of the public seeking disclosure statements to the Bureau's Web site, as provided under Sec. 1003.5(b)(2) as adopted. Section 1003.5(b)(2) is also adopted pursuant to the Bureau's authority under HMDA 305(a); Sec. 1003.5(b)(2) is necessary and proper to effectuate the purposes of HMDA and facilitate compliance therewith. The Bureau received no comments on proposed comment 5(b)-2. Therefore, the Bureau adopts this comment as proposed. The Bureau received no comments on proposed comment 5(b)-3, and adopts this comment as proposed with modifications to reflect that HMDA data will be made available on the Bureau's Web site and that HMDA data for other financial institutions is also available. The Bureau did not propose changes to current comment 5(b)-1, but is adopting a modification to this requirement to clarify the paragraph to which it applies. Finally, the Bureau adopts new comment 5(b)-4 to clarify that a financial institution may use the same notice to satisfy the requirements of both Sec. 1003.5(b)(2) and Sec. 1003.5(c).\423\ --------------------------------------------------------------------------- \423\ As discussed below, the Bureau is adopting modifications to proposed Sec. 1003.5(c) to require that a financial institution make available to the public a notice that clearly conveys that its modified loan/application register may be obtained on the Bureau's Web site and that includes the Bureau's Web site address. --------------------------------------------------------------------------- The Bureau notes that Sec. 1003.5(b) is effective January 1, 2018 and thus applies to the disclosure of 2017 HMDA data. Current Regulation C applies to requests received by financial institutions for HMDA data for calendar years prior to 2017. 5(c) Modified Loan/Application Register HMDA section 304(j)(1) requires that financial institutions make available to the public, upon request, ``loan application register information'' as defined by the Bureau and in the form required under regulations prescribed by the Bureau. HMDA section 304(j)(2) provides that the Bureau shall require such deletions from the loan application register information made available to the public as the Bureau may determine to be appropriate to protect any privacy interest of any applicant and to protect financial institutions from liability under any Federal or State privacy law, and identifies three fields in particular as appropriate for deletion.\424\ HMDA [[Page 66251]] section 304(j)(5) requires that the loan application register information described in section 304(j)(1) must be made available as early as March 31 following the calendar year for which the information was compiled. HMDA section 304(j)(7) provides that the Bureau shall make every effort to minimize costs incurred by financial institutions in complying with section 304(j). --------------------------------------------------------------------------- \424\ The fields identified in the statute as appropriate for deletion are ``the applicant's name and identification number, the date of the application, and the date of any determination by the institution with respect to such application.'' HMDA section 304(j)(2)(B). --------------------------------------------------------------------------- Section 1003.5(c) of Regulation C requires a financial institution to make its loan/application register available to the public after removing three fields to protect applicant and borrower privacy: the application or loan number, the date that the application was received, and the date action was taken. An institution must make this ``modified'' loan/application register publicly available following the calendar year for which the data are compiled by March 31 for a request received on or before March 1, and within 30 calendar days for a request received after March 1. The Bureau proposed to modify Sec. 1003.5(c) to require that a financial institution make available to the public a modified loan/ application register showing only the data fields that currently are released on the modified loan/application register. For the reasons described below, the Bureau is not finalizing Sec. 1003.5(c) as proposed, and instead is adopting a requirement that a financial institution shall make available to the public at its home office, and each branch office physically located in each MSA and each MD, a notice that clearly conveys that the institution's modified loan/application register may be obtained on the Bureau's Web site. The Bureau received several comments concerning proposed Sec. 1003.5(c). A large majority of industry commenters recommended that the agencies make the modified loan/application register available to the public on a public Web site, such as the FFIEC's Web site. Many industry commenters specifically suggested that Regulation C require financial institutions to make their modified loan/application registers available in the same way the Bureau proposed to require institutions to make their disclosure statements available, i.e., by making available a notice that clearly conveys that the modified loan/ application register may be obtained on the FFIEC Web site and that includes the FFIEC's Web site address. Commenters argued that this approach would reduce burden to financial institutions, eliminate risk to financial institutions associated with deadlines by which they must make available their modified loan/application registers, increase public access to modified loan/application registers, and allow the Bureau to modify or redact the data as it determines necessary to protect applicant and borrower privacy. One industry commenter stated that, because the modified loan/application register is already available on the FFIEC Web site, the requirement that financial institutions make their modified loan/application registers available should be eliminated as duplicative. A few other industry commenters stated that financial institutions should be permitted to post their modified loan/application registers on their own Web sites instead of providing them to members of the public upon request. With respect to the content of the modified loan/application register, a few industry commenters stated that some data currently disclosed on the modified loan/application register create risk that individual applicants and borrowers could be identified in the data. A few other industry commenters stated that public disclosure of many of the proposed new data fields would create risks of potential harm to applicant and borrower privacy. A handful of industry commenters misunderstood the Bureau's proposal concerning the modified loan/ application register to provide that the proposed new data points would never be disclosed to the public, and some of these commenters supported such an approach. Virtually all of the consumer advocate and researcher commenters opposed the proposal to exclude the proposed new data fields from the modified loan/application register. These commenters stated that many or most of the new data fields proposed were not likely to create risks to applicant or borrower privacy and should be released by March 31, not delayed until the agencies' later release of loan-level data.\425\ Most of these commenters also argued that, at a minimum, the currently- released data fields should continue to be released. Several consumer advocate and researcher commenters articulated the benefits to HMDA purposes of many currently-released and proposed new data fields in arguing for the disclosure of these data on the modified loan/ application register. --------------------------------------------------------------------------- \425\ The Bureau's proposal provided that the Bureau would include the proposed new data fields, modified as appropriate to protect applicant and borrower privacy, in the loan-level data release that the FFIEC makes available on its Web site on behalf of the agencies. See 79 FR 51731, 51816 (Aug. 29, 2014). As explained in the proposal, whereas a financial institution must make available its modified loan/application register as early as March 31, the regulators' loan-level HMDA data currently are not released until almost six months later, in September. Id. --------------------------------------------------------------------------- For the reasons described below, final Sec. 1003.5(c) requires that a financial institution shall make available to the public at its home office, and each branch office physically located in each MSA and each MD, a notice that clearly conveys that the institution's modified loan/application register may be obtained on the Bureau's Web site. This approach fulfills the goals of the Bureau's proposal \426\ and has several additional advantages. The final rule reduces costs to financial institutions associated with preparing and making available to the public the modified loan/application register, including costs associated with the application of privacy protections to the data before disclosure, and eliminates a financial institution's risk of missing the deadline to make the modified loan/application register available. It also eliminates the risks to financial institutions associated with errors in preparing the modified loan/application register that could result in the unintended disclosure of data. In addition, this approach aligns Regulation C's treatment of the modified loan/application register and the disclosure statement, which are the only HMDA data that the statute and Regulation C require financial institutions to make available to the public. --------------------------------------------------------------------------- \426\ As explained in its proposal, the Bureau believed that its proposed approach ``would avoid creating new privacy risks or liabilities for financial institutions in connection with the release of loan-level data via the modified loan/application register. It would also minimize the burden to institutions associated with preparing their modified loan/application registers to implement amendments to Regulation C. The proposed approach would allow the Bureau and the other agencies flexibility in disclosing new data points in the agencies' data release, including flexibility to adjust any privacy protections as risks evolve, without unduly burdening financial institutions or creating opportunities for the modified loan/application register and the agencies' data release to interact in ways that might increase privacy risk.'' Id. --------------------------------------------------------------------------- The approach adopted in the final rule also increases the availability of the modified loan/application register. The Bureau's Web site provides one, easily accessible location where members of the public will be able to access all modified loan/application registers for all financial institutions required to report under the statute, which furthers the disclosure goals of the statute.\427\ As [[Page 66252]] discussed above with respect to the disclosure statement,\428\ although there may be members of the public that are adversely affected by the elimination of the right to obtain a modified loan/application register directly from a financial institution, the Bureau has determined that the burden to financial institutions associated with the provision of these data directly to members of the public upon request is not justified by any benefit to the current dissemination scheme. --------------------------------------------------------------------------- \427\ Under proposed Sec. 1003.5(c), as under current Sec. 1003.5(c), for example, a member of the public that requests a financial institution's modified loan/application register need only be provided with a modified loan/application register containing data relating to the MSA or MD for which the request is made. Referral to the Bureau Web site would allow that member of the public to easily view the financial institution's modified loan/ application registers for all available MSAs and MDs. Also, to the extent a member of the public wanted to compare the lending activities of financial institutions in a particular MSA or MD, the Bureau Web site allows her to do so all in one place, rather than requiring her to obtain a modified loan/application register from multiple institutions. \428\ See section-by-section analysis of Sec. 1003.5(b)(2). --------------------------------------------------------------------------- Finally, the approach in the final rule allows the Bureau and the other agencies increased flexibility in disclosing new data fields in a manner that appropriately protects applicant and borrower privacy. As discussed above,\429\ the Bureau's assessment under its balancing test of the risks to privacy interests created by the disclosure of HMDA data and the benefits of such disclosure is ongoing and includes consideration of currently-released data points. Section 1003.5(c) as adopted will allow decisions with respect to what to include on the modified loan/application register to be made in conjunction with decisions regarding the agencies' loan-level data release, providing flexibility with respect to the agencies' release and flexibility to include on the modified loan/application register the new data fields that do not raise privacy concerns. This approach also will allow for easier adjustment of privacy protections applied to disclosures of HMDA data as risks evolve. The Bureau plans to provide a process for the public to provide input on the application of the balancing test to determine the HMDA data to be publicly disclosed both on the modified loan/application register and in the agencies' release. --------------------------------------------------------------------------- \429\ See part II.B above. --------------------------------------------------------------------------- The final rule imposes fewer burdens on financial institutions than a requirement that the modified loan/application register be made available on financial institutions' Web sites, as suggested by some industry commenters.\430\ The Bureau also declines to eliminate Sec. 1003.5(c) altogether. As discussed above with respect to the disclosure statement,\431\ although the final rule relieves financial institutions of the obligation to provide the modified loan/application register directly to the public, the Bureau has determined that provision of the notice required under Sec. 1003.5(c) to members of the public seeking a financial institution's modified loan/application register is necessary to ensure that they are clearly informed of where to obtain it. --------------------------------------------------------------------------- \430\ The Bureau notes that the final rule permits a financial institution to make available on its Web site a copy of the institution's modified loan/application register obtained from the Bureau's Web site. See Sec. 1003.5(d)(2). \431\ See section-by-section analysis of Sec. 1003.5(b)(2). --------------------------------------------------------------------------- The final rule eliminates the 30-day period between a financial institution's receipt of a request for its modified loan/application register and its obligation to provide in response the notice required pursuant to Sec. 1003.5(c). Rather than preparing a modified loan/ application register in response to a request, as required under the current regulation, under the final rule a financial institution will only need to provide a member of the public seeking a modified loan/ application register with a simple notice. The Bureau has determined that 30 days to provide such a notice is unnecessary and conflicts with the disclosure purposes of the statute. Further, as a financial institution's ability to provide the notice required under the final rule in response to a request is not dependent on the financial institution's possession of the data, as is its ability to provide the modified loan/application register under the current regulation, a financial institution does not need to wait until March 31 to provide a notice in response to a request for its modified loan/application register. The Bureau believes it is reasonable to deem that financial institutions make available to the public loan application register information, pursuant to HMDA section 304(j), by referring members of the public seeking loan application register information to the Bureau Web site, as provided under Sec. 1003.5(c). Section 1003.5(c) is also authorized pursuant to the Bureau's authority under HMDA section 305(a). For the reasons given above, the Bureau concludes that Sec. 1003.5(c) as adopted is necessary and proper to effectuate HMDA's purposes and facilitate compliance therewith. The Bureau did not propose changes to current comment 5(c)-1 but is adopting modifications to this comment to conform to Sec. 1003.5(c) as finalized. Proposed comment 5(c)-2 is adopted as modified to provide an example of notice content that would satisfy the requirements of Sec. 1003.5(c). Proposed comment 5(c)-3 is adopted as modified to clarify that a financial institution may use the same notice to satisfy the requirements of both Sec. 1003.5(b)(2) and Sec. 1003.5(c). The Bureau notes that Sec. 1003.5(c) is effective January 1, 2018 and thus applies to the disclosure of 2017 HMDA data. Current Regulation C applies to requests received by financial institutions for HMDA data for calendar years prior to 2017. 5(d) Availability of Written Notice HMDA sections 304(c) and 304(j)(6) set forth the time periods for which financial institutions must maintain and make available information required to be disclosed under the statute. HMDA sections 304(j)(4) and 304(k)(3) permit a financial institution that provides its loan/application register information or its disclosure statement to a member of the public to impose a reasonable fee for any cost incurred in reproducing the information or statement. Section 1003.5(d) of Regulation C requires that a financial institution must make its modified loan/application register available to the public for a period of three years and its disclosure statement available to the public for a period of five years. This section also provides that an institution must make these disclosures available to the public for inspection and copying during the hours the office is normally open to the public for business and may impose a reasonable fee for any cost incurred in providing or reproducing the data. The Bureau proposed to delete the requirement that a financial institution make its HMDA data available for inspection and copying and to make additional technical modifications to Sec. 1003.5(d). The Bureau is adopting Sec. 1003.5(d) as proposed with clarifying modifications. The Bureau received very few comments on proposed Sec. 1003.5(d). One industry commenter supported the proposal to delete the requirement that a financial institution make its data available for inspection and copying. Another industry commenter misunderstood the proposal to require that financial institutions retain their disclosure statements and modified loan/application registers for the requisite periods, and stated that the availability of these data on the FFIEC Web site made these requirements duplicative and unnecessary. The Bureau is adopting Sec. 1003.5(d)(1) generally as proposed, with modifications to clarify that it requires [[Page 66253]] a financial institution to retain the notices concerning its disclosure statements and modified loan/application registers required pursuant to Sec. 1003.5(b)(2) and (c), not the disclosure statements and modified loan/application registers themselves. The Bureau adopts Sec. 1003.5(d)(2) as modified to clarify that a financial institution may make its disclosure statement and its modified loan/application register available to the public in addition to, but not in lieu of, the notices required by Sec. 1003.5(b)(2) and (c), and may impose a reasonable fee for any cost associated with providing or reproducing its disclosure statement or modified loan/application register. The Bureau notes that Sec. 1003.5(d) is effective January 1, 2018 and thus applies to the disclosure of 2017 HMDA data. Current Regulation C applies to requests received by financial institutions for HMDA data for calendar years prior to 2017. 5(e) Posted Notice of Availability of Data HMDA section 304(m) provides that a financial institution shall be deemed to have satisfied the public availability requirements of HMDA section 304(a) if it compiles its HMDA data at its home office and provides notice at certain branch locations that its information is available upon written request. Section 1003.5(e) of Regulation C requires that a financial institution post a notice concerning the availability of its HMDA data in the lobby of its home office and of each branch office located in an MSA and MD. Section 1003.5(e) also requires that a financial institution must provide, or the posted notice must include, the location of the institution's office where its disclosure statement is available for inspection and copying. Comment 5(e)-1 suggests text for the posted notice required under Sec. 1003.5(e). Comment 5(e)-2 suggests text concerning disclosure statements that may be included in the posted notice to satisfy Sec. 1003.5(b)(3)(ii). The Bureau proposed clarifying and technical modifications to Sec. 1003.5(e) and related comments and modifications to conform to proposed Sec. 1003.5(b)(2). The Bureau received very few comments on proposed Sec. 1003.5(e). One industry commenter supported deleting language from Sec. 1003.5(e) concerning the location of the institution's office where its disclosure statement is available for inspection and copying. The Bureau adopts Sec. 1003.5(e) as proposed with one modification to clarify that the required lobby notice must clearly convey that the institution's HMDA data may be obtained on the Bureau's Web site. One industry commenter opposed the proposed changes to comment 5(e)-1 concerning the suggested notice text, stating that it was a waste of financial institution resources to update the posted notice to reflect that the HMDA data include age. The addition of language concerning age was not the only proposed change to the suggested notice text, however. The proposed suggested text also updated the posted notice to provide information about where HMDA data could be found online. The Bureau has determined that inclusion of information concerning where HMDA data can be found online is necessary to ensure access to HMDA data, especially as financial institutions will no longer be required to provide either their disclosure statements or their modified loan/application registers directly to the public under amended Regulation C. The Bureau adopts comment 5(e)-1 as proposed with technical modifications. 5(f) Aggregation HMDA section 310 requires the FFIEC to compile aggregate data by census tract for all financial institutions reporting under HMDA and to produce tables indicating aggregate lending patterns for various categories of census tracts grouped according to location, age of housing stock, income level, and racial characteristics. HMDA section 304(f) requires the FFIEC to implement a system to facilitate access to data required to be disclosed under HMDA section 304, including arrangements for central depositories where such data are made available for inspection and copying. Section 1003.5(f) of Regulation C provides that the FFIEC will produce reports for individual institutions and reports of aggregate data for each MSA and MD, showing lending patterns by property location, age of housing stock, and income level, sex, ethnicity, and race, and will make these reports available at central depositories. Section 1003.5(f) also contains information concerning how to obtain a list of central depositories from the FFIEC. The Bureau proposed to modify Sec. 1003.5(f) to replace the word ``produce'' with ``make available'' for clarity and to delete reference to central depositories. The Bureau is adopting Sec. 1003.5(f) as proposed with minor modifications. The Bureau received one comment concerning proposed Sec. 1003.5(f). This commenter stated that disclosure of automated underwriting system name and result in the aggregated data, could reveal proprietary information concerning these systems. As discussed above,\432\ at this time the Bureau is not making determinations about what HMDA data will be publicly disclosed or the forms of such disclosures. --------------------------------------------------------------------------- \432\ See section-by-section analysis of Sec. 1003.4(a)(35). --------------------------------------------------------------------------- The Bureau is adopting proposed Sec. 1003.5(f) with three modifications. The final rule clarifies that the aggregates described in this paragraph and made available in 2018 are based on 2017 data submitted pursuant to current Sec. 1003.5(a), and that the aggregates made available beginning in 2019 are based on data submitted on an annual basis pursuant to Sec. 1003.5(a)(1)(i), not data submitted on a quarterly basis pursuant to Sec. 1003.5(a)(1)(ii). The Bureau has determined that reference to reports for individual institutions in this paragraph is no longer necessary \433\ and is eliminating this reference in the final rule. Finally, the Bureau has determined that reference to the location where the aggregate data described in this paragraph will be made available is unnecessary and is eliminating this reference in the final rule. --------------------------------------------------------------------------- \433\ The FFIEC's obligation to make available the disclosure statements is set forth in final Sec. 1003.5(b)(1). --------------------------------------------------------------------------- As discussed in its proposal,\434\ the Bureau believes that advances in technology may permit, for example, the FFIEC to produce an online tool, such as a tabular engine, that would allow public officials and members of the public to generate the tables described in HMDA section 310. It is the Bureau's interpretation that the obligation to ``produce tables'' set forth in HMDA section 310 would be satisfied if the FFIEC produced such a tool, which in turn would produce the tables described in HMDA section 310 on request. Further, pursuant to HMDA section 305(a), the Bureau believes that permitting the FFIEC to produce a tool that allows members of the public to generate tables described in HMDA section 310 is necessary and proper to effectuate the purposes of HMDA and facilitate compliance therewith. --------------------------------------------------------------------------- \434\ 79 FR 51731, 51818 (Aug. 29, 2014). --------------------------------------------------------------------------- Section 1003.6 Enforcement 6(b) Bona Fide Errors The Bureau did not propose to amend Sec. 1003.6. HMDA section 305(b) provides that compliance with HMDA is enforced by the Board, FDIC, OCC, the Bureau, NCUA, and HUD.\435\ Each of these Federal agencies can rely on its own authorities to enforce compliance with [[Page 66254]] HMDA, including the authority conferred in HMDA section 305(b).\436\ Section 1003.6(a) of Regulation C provides that a violation of HMDA or Regulation C is subject to administrative sanctions as provided in HMDA section 305, including the imposition of civil money penalties.\437\ Regulation C Sec. 1003.6(b) provides authority to find that ``bona fide errors'' are not violations of HMDA and Regulation C. Section 1003.6(b)(1) provides that an error in compiling or recording loan data is not a violation if the error was unintentional and occurred despite the maintenance of procedures reasonably adapted to avoid such errors. Section 1003.6(b)(2) provides that an incorrect entry for a census tract number is deemed a bona fide error, and is not a violation of HMDA or Regulation C, if the financial institution maintains procedures reasonably adapted to avoid such errors. Currently, Sec. 1003.6(b)(3) addresses and provides some latitude for inaccurate or incomplete quarterly recording of data. --------------------------------------------------------------------------- \435\ 12 U.S.C. 2804(b). Most commenters who addressed the enforcement and examination practices of the Federal agencies did not specify the particular agency to which the commenters submit their data. \436\ HMDA section 305(c); 12 U.S.C. 2804(c). \437\ See CFPB Bulletin 2013-11 (2013), http://files.consumerfinance.gov/f/201310_cfpb_hmda_compliance-bulletin_fair-lending.pdf, which, among other things, sets out factors the Bureau will consider in determining any civil money penalty for violations of HMDA and Regulation C. --------------------------------------------------------------------------- Although the Bureau did not propose specific changes to Sec. 1003.6, it sought feedback generally about concerns raised by the small entity representatives during the Small Business Review Panel process regarding whether, in light of new reporting requirements, it would be appropriate to add new provisions to Sec. 1003.6 to clarify compliance expectations and address compliance burdens or operational challenges.\438\ The Bureau specifically sought feedback on whether a more precise definition of what constitutes an error would be helpful, whether there are ways to improve the current methods of calculating error rates, and whether tolerance levels for error rates would be appropriate. For the reasons discussed below, the Bureau is revising current Sec. 1003.6(a), (b)(1), and (b)(2), and comment 6(b)-1, only by making technical, nonsubstantive edits. The Bureau is moving Sec. 1003.6(b)(3) to new Sec. 1003.6(c)(1), as discussed below. --------------------------------------------------------------------------- \438\ The comments of the small entity representatives were summarized in the proposed rule. See 79 FR 51731, 51818 (Aug. 29, 2014). --------------------------------------------------------------------------- Comments on Enforcement Approximately one-third of the commenters addressed enforcement, data errors, and administrative resubmission requirements related to Regulation C. Nonindustry commenters generally did not comment on enforcement policies and error rates. Most industry commenters that addressed the topic identified what they viewed as unrealistic tolerance levels as being an issue with Regulation C compliance and enforcement. Many industry commenters stated that the compliance and enforcement concerns would likely be exacerbated by additional data points in the final rule. Some industry commenters expressly recognized the importance of the submission of accurate data, affirmed that reporting entities are concerned with the integrity of their data, and acknowledged that they would understand reasonable and fair requirements relating to errors. Many of the commenters stated that despite the implementation of appropriate systems and controls and efforts to comply with the spirit of Regulation C, innocent errors and human judgment errors in interpretation and data input are impossible to eliminate completely. A common theme among industry commenters was that additional data collection and reporting requirements mean there is a greater likelihood of errors. A number of commenters echoed a request that the Bureau reconsider examination procedures and guidelines and make adjustments to acceptable error rates, especially in light of the significant increase in the amount of data that reporting entities will be required to compile, audit, and report. Many commenters suggested that tolerances for errors be increased if the final rule includes additional data points in Regulation C. One commenter urged the Bureau not to discount the burden of reporting accurate data. Others stated that data is not easy to get right because of the number of people involved in loan production, and that manual audits conducted on the additional data by compliance staff will take significantly more time and force reporting institutions to shift resources or add staff. A few commenters noted exposure to reputational risks, as well as to administrative enforcement, that could be associated with increased reporting errors. A trade association commented that reasonable tolerances are necessary to minimize compliance costs. A few commenters observed that a demonstrated pattern of these types of errors could suggest that the errors are not inadvertent. A number of commenters requested relief from responsibility for errors based on: good faith efforts; technical, de minimis errors; distinguishing critical and noncritical errors; inadvertent errors; bona fide errors; immaterial errors; distinguishing random and systemic errors; and distinguishing key and non-key errors. Multiple commenters suggested specific data points that, in addition to institutional and transaction coverage changes, might contribute to a need for increasing the current error tolerances, including: age; income, as proposed; denial reasons; universal loan identifier; debt-to-income ratio; loan-to-value ratio; AUS information; points and fees; and data points that contain dates, dollar amounts, and percentages. Similarly, some commenters advocated that the Bureau establish acceptable ranges for the values reported for certain data points, for reasons that include the potential for rounding numbers incorrectly and making errors in calculations, and allow latitude for entering the wrong text in data fields, such as ``N/A'' instead of ``none.'' Other specific recommendations included: preclude resubmissions of data on loans that do not constitute a material percentage of all loans in a reporting year in the associated metropolitan statistical area; limit punitive actions for reporting errors that do not lead to findings of discrimination; adopt a tiered evaluation of errors that is dependent on the reasons for the errors; excuse errors resulting from reliance on third-party information; apply more-lenient standards to new data points initially; develop guidance and interagency exam procedures that support compliance; and provide a sufficient implementation period to adjust to new requirements. One industry commenter acknowledged that the Bureau may not want to address clarifications of error rates and tolerances through rulemaking, at the same time expressing concern about potential compliance burdens for accuracy in a significantly larger data submission. Another commenter suggested that Regulation C include a statement that a bona fide unintentional error is not a violation. A few commenters predicted that the proposed reporting changes would cause more financial institutions to exit mortgage lending, with the exiting institutions skewing small, and would discourage new entrants to the market, significantly decreasing the availability of credit. Final Rule After considering the comments, the Bureau has concluded that there are more effective ways to address the issues raised by the commenters than by making substantive changes to Sec. 1003.6(b). In reaching this conclusion, the Bureau accepts that some errors in data compilation and reporting are [[Page 66255]] difficult to avoid altogether. HMDA data are important for the public and public officials, therefore the final rule seeks to balance the need for accurate data and the challenge of generating that data. The Bureau believes that many of the error-related issues raised by commenters would be best addressed through supervisory policy, rather than regulatory language. Most of the comments specifically or implicitly addressed current administrative examination procedures and guidelines for required resubmission of data when error levels exceed established thresholds. Decisions regarding when to pursue an enforcement action or other solution for noncompliance with HMDA or Regulation C are a matter of agency discretion. Each of the agencies that has authority to enforce HMDA can develop internal procedures and guidelines for citing a financial institution for inaccurate data. For example, the Bureau makes its HMDA examination guidelines available publicly, so that financial institutions understand, and can develop internal processes to meet, expectations for HMDA data accuracy.\439\ The use of guidelines, which provide a measure for application of enforcement principles, coupled with language in Sec. 1003.6(b) that deems certain errors to be excused, benefits examiners and financial institutions, alike. In particular, as the agencies and financial institutions gain experience with the new definitions, requirements, increased number of data points, reporting instructions, and technology, the guidelines can be tailored, adjusted, and applied as appropriate. --------------------------------------------------------------------------- \439\ See CFPB Supervision and Examination Manual, HMDA Resubmission Schedule and Guidelines (2013), http://files.consumerfinance.gov/f/201310_cfpb_hmda_resubmission-guidelines_fair-lending.pdf. --------------------------------------------------------------------------- In addition, however, the final rule addresses some of the commenters' particular areas of concern in stating the requirements and providing commentary for individual data points. For example, financial institutions are permitted to report the information they relied on for several data points and have some flexibility in the format they use to report certain data points. The final rule provides further guidance and examples of acceptable values in commentary and, more generally, addresses many common issues with the current regulation by clarifying various provisions in the regulations and commentary. The Bureau also plans to expand data submission edit checks to improve the ability of financial institutions to identify and fix mistaken data before final submission to the agencies, which could also benefit the financial institutions in their internal audit processes. Finally, the Bureau will develop additional guidance materials to help financial institutions understand the final rule and avoid errors in interpreting its requirements. Public officials rely on the data reported by financial institutions to further HMDA's purposes. In addition, the data disclosed under HMDA provide the public with information on the mortgage activities of particular reporting financial institutions and in communities. Because HMDA data serve these important purposes, accurate data is essential. The accuracy of HMDA data depends on good operational and validation processes. Financial institutions have primary responsibility for these processes; the institutions must develop and maintain appropriate compliance management systems that are reasonably designed to ensure the accuracy of the data. Examination procedures used by the Federal regulators further assure appropriate validation of the HMDA data, by assessing a financial institution's policies, procedures, monitoring, and corrective-action processes. The Bureau has concluded that it should not establish in Regulation C global thresholds for the number or percentage of errors in a financial institution's data submission that would trigger compliance or enforcement action. Establishing regulatory thresholds for errors or adding resubmission requirements to the regulation are not likely to lead to a satisfactory outcome for industry or the regulators. The current provision on bona fide errors in Sec. 1003.6(b), in conjunction with agency guidelines, provides appropriate flexibility for regulators to exercise judgment in assessing compliance violations. The Bureau anticipates that the Federal agencies enforcing HMDA will review their enforcement approaches in light of the significant regulatory changes included in the final rule and consult on any appropriate adjustments to their policies, both during the final rule's implementation period and beyond. Currently, some errors are found and addressed in the data submission process, using edits developed through the FFIEC coordination agreement, while other errors can be identified only in subsequent audits or examinations by comparing HMDA data submitted to loan files. As the Bureau collaborates with the other HMDA enforcement agencies on future administrative examination and review procedures, it will consider, and bring to the attention of those agencies, the numerous comments and suggestions received on this topic during the public comment process on the proposed rule. The final rule makes technical, nonsubstantive edits to current Sec. 1003.6(a), (b)(1), and (b)(2) and comment 6(b)-1, for purposes of clarity and consistency. 6(c) Quarterly Recording and Reporting The Bureau did not propose changes to Sec. 1003.6(b)(3), but is adopting changes to this provision in connection with the quarterly reporting requirement finalized in Sec. 1003.5(a)(1)(ii). Under Sec. 1003.5(a)(1)(ii) as adopted, within 60 calendar days after the end of each calendar quarter except the fourth quarter, financial institutions subject to Sec. 1003.5(a)(1)(ii) will submit the HMDA data that they are required to record on their loan/application registers for that calendar quarter pursuant to Sec. 1003.4(f). Pursuant to new Sec. 1003.6(c)(2), errors and omissions in the data submitted pursuant to Sec. 1003.5(a)(1)(ii) will not be considered HMDA or Regulation C violations assuming the conditions that currently provide a safe harbor for errors and omissions in quarterly recorded data are satisfied. Currently, Sec. 1003.6(b)(3) provides that errors and omissions in data that a financial institution records on its loan/application register on a quarterly basis as required under Sec. 1003.4(a) are not violations of HMDA or Regulation C if the institution makes a good- faith effort to record all required data fully and accurately within thirty calendar days after the end of each calendar quarter and corrects or completes the data prior to reporting the data to its regulator. That is, Sec. 1003.6(b)(3) provides a safe harbor that protects a financial institution that satisfies certain conditions from being cited for violations of HMDA or Regulation C for errors and omissions on its quarterly recorded loan/application register. The Bureau is moving Sec. 1003.6(b)(3) to new paragraph Sec. 1003.6(c)(1) and adding paragraph (c)(2) to provide that a similar safe harbor applies to data reported on a quarterly basis pursuant to Sec. 1003.5(a)(1)(ii). The Bureau adopts Sec. 1003.6(c). Section 1003.6(c)(1) applies to data that an institution records on its loan/application register on a quarterly basis as required under Sec. 1003.4(f), as finalized herein. It provides that, if a financial institution makes a good-faith effort to record all data required to be recorded pursuant to Sec. 1003.4(f) fully and accurately within 30 calendar days after the end of each calendar quarter, and some data are nevertheless [[Page 66256]] inaccurate or incomplete, the inaccuracy or omission is not a violation of HMDA or Regulation C provided that the institution corrects or completes the data prior to submitting its annual loan/application register pursuant to Sec. 1003.5(a)(1)(i). Section 1003.6(c)(2) applies to data that an institution reports on a quarterly basis pursuant to Sec. 1003.5(a)(1)(ii). It provides that, if an institution subject to Sec. 1003.5(a)(1)(ii) makes a good-faith effort to report all data required to be reported pursuant to Sec. 1003.5(a)(1)(ii) fully and accurately within 60 calendar days after the end of each calendar quarter, and some data are nevertheless inaccurate or incomplete, the inaccuracy or omission is not a violation of HMDA or Regulation C provided that the institution corrects or completes the data prior to submitting its annual loan/application register pursuant to Sec. 1003.5(a)(1)(i). The Bureau is adopting an effective date of January 1, 2019 for Sec. 1003.6. Accordingly, this section applies to HMDA data reported beginning in 2019. For example, compliance is enforced pursuant to this final rule with respect to 2018 data reported in 2019. Section 1003.6 of current Regulation C applies to the collection and recording of HMDA data in 2018. Appendix A to Part 1003 Form and Instructions for Completion of HMDA Loan/Application Register Part I of appendix A to Regulation C currently provides instructions for the Loan/Application Register. Part II of appendix A contains instructions related to reporting HMDA data, including instructions for sending HMDA data via U.S. mail. Appendix A also contains a form for the transmittal sheet, a form for the loan/ application register, and a technical code sheet for completing the loan/application register. As discussed in many of the section-by- section analyses above, the Bureau is expanding the regulation text and commentary to address the requirements currently provided in part I of appendix A and in the form for the transmittal sheet. As discussed in the section-by-section analysis of Sec. 1003.5(a)(1) above, the Bureau is eliminating paper reporting. Furthermore, the Bureau intends to publish procedures related to the submission of the data required to be reported under Regulation C, which will replace the existing form for the loan/application register and technical code sheet for completing it. Thus, the requirements and other information currently provided in appendix A are no longer necessary, and the final rule deletes appendix A. To accomplish the transition from reporting current to amended data, the final rule deletes appendix A in two stages. First, effective January 1, 2018, the final rule adds to appendix A a new paragraph explaining the transition requirements for data collected in 2017 and reported in 2018. Also effective January 1, 2018, part II of appendix A is revised to provide updated instructions relating to the reporting of 2017 HMDA data. Then, effective January 1, 2019, appendix A is deleted in its entirety, when instructions relating to the reporting of 2017 HMDA data will no longer be necessary. I. Effective Date A. Comments In response to the proposed rule, the Bureau received roughly a few dozen comments concerning effective date and implementation period. Industry commenters, including banks and credit unions; software providers; and trade associations provided recommendations on the timing for implementation. The recommendations for the implementation period ranged from a minimum of at least one full calendar year to several years. Most commenters recommended 18 to 24 months while several other commenters advocated for 24 to 36 months. A couple of commenters did not suggest a specific timing period but urged the Bureau to allow as much time as possible. Many commenters cited operational challenges as a reason why ample time is needed for implementation. These commenters stated that systems will need to be redesigned or replaced to accommodate the new rules. A couple of commenters pointed out that not all business areas of a bank use the same system to capture HMDA data. One commenter, in particular, stated that if all the proposed data fields are finalized, then it may require data from two or more systems. This commenter cited the possibility of the need to integrate data from several systems designed for origination and servicing for consumer, real estate, and business transactions. One software provider that advocated for a 36 month implementation period stated that software providers need time to design, develop, and distribute software to financial institution clients. These clients will then need to test need the software, implement procedural changes, and train staff. Several commenters indicated that policies and procedures will need to be developed and staff will need to be trained on those policies and procedures. One commenter asked that the Bureau consider the time it takes to interpret the final regulation. Several commenters pointed out that the industry is currently focusing on implementing the TILA-RESPA and other mortgage rules and staff is fully engaged in implementing those rules or enhancing compliance programs. One commenter stated that forcing industry to shift or split resources between TILA-RESPA and HMDA may affect the ability to implement one or both rules by their effective date. While many commenters suggested a specific number of months or years, a few commenters specified January 1 as the day that data collection should begin regardless of the year of the effective date. One commenter suggested that the Bureau specify that the effective date applies to applications taken on or after the date the Bureau designates. Another commenter argued that implementing the final rule any day of the year other than January 1 would cause confusion for financial institutions collection and reporting the data, and may even possibly affect data quality. Several commenters noted that the Dodd-Frank Act does not provide a deadline for implementing amendments to the HMDA rule, so they urged the Bureau to use its discretionary authority to provide adequate additional time for compliance. One trade association suggested that the Bureau should use its discretionary authority and consider the burden on small entities by providing an extended effective date for certain groups of entities One trade association asked the Bureau to provide transition rules for applications received before the effective date but where final action is taken on the application after the effective date. The Bureau has considered the comments, including the potential issues that could arise as a result of an inadequate implementation period and industry's focus on other recent mortgage rulemakings, and believes that the effective date described below achieves the right balance between ample time for implementation and the need for useful HMDA data that reflects the current housing finance market. B. The Effective Date and Implementation Period In consideration of the comments and recommendations suggested by commenters, the final rule is effective [[Page 66257]] January 1, 2018,\440\ except that: Sec. 1003.2(g)(1)(v)(A) is effective January 1, 2017; Sec. 1003.5(a)(1)(i), (a)(1)(iii), and (a)(2) through (5) are effective January 1, 2019; Sec. 1003.6 is effective January 1, 2019; and Sec. 1003.5(a)(1)(ii) is effective January 1, 2020. Section 1003.5(b) and (f), as revised effective January 1, 2018, are revised again on January 1, 2019. Appendix A is revised effective January 1, 2018 and then deleted effective January 1, 2019. Commentary to Sec. 1003.5(a) and Sec. 1003.6 in supplement I, as revised effective January 1, 2018, are revised again effective January 1, 2019. These exceptions to the general effective date of January 1, 2018 are described in further detail below. --------------------------------------------------------------------------- \440\ HMDA section 304(n) provides that institutions shall not be required to report new data under HMDA section 304(b)(5) and (6) before the first January 1 that occurs after the end of the 9-month period beginning on the date on which regulations are issued by the Bureau in final form with respect to such disclosures. Although the statute permits a shorter period than the effective date the Bureau is finalizing, the Bureau believes that a longer period will help reduce implementation burden on industry. --------------------------------------------------------------------------- This final rule applies to covered loans and applications with respect to which final action is taken beginning on January 1, 2018. Data on these covered loans and applications are submitted to the appropriate Federal agency pursuant to Sec. 1003.5(a) beginning on January 1, 2019. For example, if a financial institution described in 2(g) of this part receives an application on January 1, 2018 and takes final action on that application on March 1, 2018, data about that application will be collected and recorded pursuant to Sec. 1003.4, and submitted to the appropriate Federal agency by March 1, 2019 pursuant to Sec. 1003.5(a). Similarly, if a financial institution described in 2(g) of this part receives an application on December 1, 2017 and does not take final action on that application until January 1, 2018, data about that application would be collected and recorded pursuant to Sec. 1003.4 and submitted to the appropriate Federal agency by March 1, 2019 pursuant to Sec. 1003.5(a).\441\ The final rule also applies to purchases that occur on or after January 1, 2018. For example, a financial institution described in 2(g) of this part that purchases a HMDA reportable loan on February 1, 2018 would collect and record data about that purchase pursuant to Sec. 1003.4, and submit the data to the appropriate Federal agency by March 1, 2019 pursuant to Sec. 1003.5(a). --------------------------------------------------------------------------- \441\ The Bureau understands that final action taken on an application may not occur until a few months after the application date. A financial institution may receive an application at the end of a calendar year but may not determine the final disposition of the application until the following calendar year. --------------------------------------------------------------------------- Lower-Volume Depository Institutions The Bureau is adopting an effective date of January 1, 2017 for Sec. 1003.2(g)(1)(v)(A), which is one of the prongs of the institutional coverage test for depository institutions. Specifically, this prong provides that a depository institution must originate at least 25 closed-end mortgage loans in each of the preceding two calendar years. Therefore, a depository institution that originates at least 25 closed-end mortgage loans in each of two calendars years and that otherwise meets all the other criteria specified in Sec. 1003.2(g)(1) would be required to report HMDA data for 2017. However, if the depository institution originated less than 25 closed-end mortgage loans in each of two calendars years, then it would not be required to report HMDA data even if it meets all other reporting criteria specified in Sec. 1003.2(g)(1). Similarly, if the depository institution originated 25 closed-end mortgage loans in one calendar year and then originated less than 25 closed-end mortgage loans in the subsequent calendar year, the depository institution would not be required to report HMDA data for 2017. Reporting Data to the Appropriate Federal Agency and Disclosing Data to the Public The Bureau is adopting an effective date of January 1, 2019 for Sec. 1003.5(a)(1)(i), (a)(1)(iii), and (a)(2) through (a)(5), and related commentary, which concern the submission of data collected and recorded pursuant to this final rule. Financial institutions will submit data on covered loans and applications with respect to which final action is taken in 2018 to the appropriate Federal agency pursuant to these provisions by March 1, 2019.\442\ --------------------------------------------------------------------------- \442\ Appendix A is deleted effective January 1, 2019, so will not apply to the submission of data on covered loans and applications with respect to which final action is taken in 2018. --------------------------------------------------------------------------- Data collected and recorded in 2017 pursuant to current Regulation C will be reported by March 1, 2018 pursuant to current Sec. 1003.5(a). The final rule's amendments to supplement I effective January 1, 2018 generally maintain the current commentary to Sec. 1003.5(a) with respect to the reporting of data collected in 2017 and reported in 2018.\443\ Effective January 1, 2019, commentary to Sec. 1003.5(a) is revised to address the reporting of data beginning in 2019. The final rule adds to appendix A a new paragraph explaining the transition requirements for data collected in 2017 and reported in 2018, effective January 1, 2018. On that date, part II of appendix A is also revised to provide updated instructions relating to the reporting of 2017 HMDA data. Then, effective January 1, 2019, appendix A is deleted in its entirety, when instructions relating to the reporting of 2017 HMDA data will no longer be necessary. --------------------------------------------------------------------------- \443\ As discussed further above in the section-by-section analysis of Sec. 1003.5(a), some of the current comments to Sec. 1003.5(a) are removed and reserved effective January 1, 2018. --------------------------------------------------------------------------- Financial institutions will make available to the public their 2017 HMDA data pursuant to Sec. 1003.5(b) through (e) of this final rule. Financial institutions make available to the public their HMDA data for calendar years prior to 2017 pursuant to current Regulation C. Quarterly Reporting The Bureau is adopting an effective date of January 1, 2020 for Sec. 1003.5(a)(1)(ii), which concerns quarterly reporting. This delay is to permit financial institutions subject to the quarterly reporting requirement time to implement the final rule and complete two annual reporting cycles under the final rule before being required to submit quarterly data. A financial institution required to comply with Sec. 1003.5(a)(1)(ii) will submit its first quarterly data to the appropriate Federal agency by May 30, 2020. For example, a financial institution that reports at least 60,000 covered loans and applications, not including purchased covered loans, in its 2019 HMDA data submission is required to report its 2020 HMDA data on a quarterly basis pursuant to Sec. 1003.5(a)(1)(ii), beginning with the first quarterly submission due on May 30, 2020. Enforcement The Bureau is adopting an effective date of January 1, 2019 for Sec. 1003.6, which concerns enforcement of HMDA and Regulation C. The amendments to Sec. 1003.6 adopted in this final rule apply to HMDA data reported beginning in 2019. Thus, current Sec. 1003.6 applies to data collected in 2017 and reported in 2018, and amended Sec. 1003.6 applies to 2018 data reported in 2019. Implementation Period The Bureau believes that these effective dates, which provide an extended implementation period of over two years, is appropriate and will provide industry with sufficient time to revise and update policies and procedures; implement comprehensive systems change; and train staff. In addition, the implementation period will assist in facilitating updates to the processes of the Federal regulatory [[Page 66258]] agencies responsible for supervising financial institutions for compliance with the HMDA rule. In order to assist industry with an efficient and effective implementation of the rule, the Bureau intends to provide guidance in the form of plain language compliance guides and aids, such as videos and reference charts; technical specifications and documentation; and in conducting meetings with stakeholders to discuss the rule and implementation issues. VII. Section 1022(b)(2) of the Dodd-Frank Act The Bureau has considered the potential benefits, costs, and impacts of the final rule.\444\ In developing the final rule, the Bureau has consulted with or offered to consult with the prudential regulators (the Board of Governors of the Federal Reserve System, the Federal Deposit Insurance Corporation, the National Credit Union Administration, and the Office of the Comptroller of the Currency), the Department of Justice, the Department of Housing and Urban Development, the Federal Housing Finance Agency, the Securities and Exchange Commission, and the Federal Trade Commission regarding, among other things, consistency with any prudential, market, or systemic objectives administered by such agencies. --------------------------------------------------------------------------- \444\ Specifically, section 1022(b)(2)(A) of the Dodd-Frank Act calls for the Bureau to consider the potential benefits and costs of a regulation to consumers and covered persons, including the potential reduction of access by consumers to consumer financial products or services; the impact on depository institutions and credit unions with $10 billion or less in total assets as described in section 1026 of the Dodd-Frank Act; and the impact on consumers in rural areas. --------------------------------------------------------------------------- As discussed in greater detail elsewhere throughout this supplementary information, in this rulemaking the Bureau is amending Regulation C, which implements HMDA, and the official commentary to the regulation, as part of the Bureau's implementation of the Dodd-Frank Act amendments to HMDA regarding the reporting and disclosure of mortgage loan information. The amendments to Regulation C implement section 1094 of the Dodd-Frank Act, which made certain amendments to HMDA.\445\ --------------------------------------------------------------------------- \445\ These amendments, among other things, require financial institutions to itemize their HMDA data by: The age of mortgagors and mortgage applicants; points and fees payable at origination in connection with a mortgage; the difference between the annual percentage rate associated with a loan and a benchmark rate or rates for all loans; the term in months of any prepayment penalty or other fee or charge payable on repayment of some portion of principal or the entire principal in advance of scheduled payments; the value of the real property pledged or proposed to be pledged as collateral; the actual or proposed term in months of any introductory period after which the rates of interest may change; the presence of contractual terms or proposed contractual terms that would allow the applicant or borrower to make payments other than fully amortizing payments during any portion of the loan term; the actual or proposed term in months of the mortgage; the channel through which the mortgage application was made, including retail, broker, and other relevant categories; and the credit score of mortgage applicants and borrowers. --------------------------------------------------------------------------- The final rule includes additional amendments to Regulation C to implement the Dodd-Frank Act's provisions permitting reporting of, as the Bureau may determine to be appropriate, a unique identifier that identifies the loan originator, a universal loan identifier, and the parcel number that corresponds to the property pledged or proposed to be pledged as collateral. The final rule also requires financial institutions to report additional information pursuant to authority under sections 304(b)(5)(D) and 304(b)(6)(J) of HMDA, which permit the disclosure of such other information as the Bureau may require, and section 305(a) of HMDA, which, among other things, broadly authorizes the Bureau to prescribe such regulations as may be necessary to carry out HMDA's purposes. Certain additional data points included in the final rule are not specifically identified by the Dodd-Frank Act amendments to HMDA.\446\ --------------------------------------------------------------------------- \446\ These additional data include: The construction method for the dwelling related to the subject property; mandatory reporting of the reasons for denial of a loan application; the total origination charges associated with the loan; the total points paid to the lender to reduce the interest rate of the loan; the total amount of any general credits provided to the borrower by the lender; the interest rate applicable at closing or account opening; the applicant's or borrower's debt-to-income ratio; the ratio of the total amount of debt secured by the property to the value of the property; for transactions involving manufactured homes, whether the loan or application is or would have been secured by a manufactured home and land, or by a manufactured home and not land; the land property interest for loans or applications related to manufactured housing; the total number of individual dwelling units contained in the dwelling related to the loan; the number of individual dwellings units that are income-restricted pursuant to Federal, State, or local affordable housing programs; information related to the automated underwriting system used in evaluating an application; whether the loan is a reverse mortgage; whether the loan is an open- end line of credit; and whether the loan is primarily for a business or commercial purpose. --------------------------------------------------------------------------- The final rule also modifies the regulation's transactional and institutional coverage. Regarding transactional coverage, the final rule requires financial institutions to report activity for consumer- purpose dwelling-secured loans and lines of credit, regardless of whether the loans or credit lines are for home purchase, home improvement, or refinancing.\447\ The final rule adjusts institutional coverage to adopt loan-volume thresholds of 25 closed-end mortgage loans or 100 open-end lines of credit for all financial institutions. --------------------------------------------------------------------------- \447\ The final rule retains reporting of commercial-purpose transactions only if they are for the purpose of home improvement, home purchase, or refinancing. --------------------------------------------------------------------------- Furthermore, the Bureau is modifying the frequency of reporting for certain financial institutions with large numbers of transactions, and the requirements regarding the public availability of the HMDA disclosure statement and the modified loan/application register. Financial institutions that reported at least 60,000 covered loans and applications, excluding purchased covered loans, for the preceding calendar year, are required to report data quarterly to the appropriate Federal agency for the first three quarters of each calendar year. Financial institutions are required to make available to the public notices that clearly convey that the institution's disclosure statement and modified loan/application register may be obtained on the Bureau's Web site and that includes the Web site address. The Bureau is also separately implementing several operational enhancements and modifications designed to reduce the burden of reporting HMDA data. The Bureau is working to improve the geocoding process, creating a web-based HMDA data submission and edit-check system, developing a data-entry tool for small financial institutions that currently use Data Entry Software, and otherwise streamlining the submission and editing process to make it more efficient. The Bureau is also adopting definitions of many data points that are consistent with existing regulations and with the MISMO data standards for residential mortgages. A. Provisions To Be Analyzed The discussion below considers the benefits, costs, and impacts of the following major provisions of the final rule: 1. The scope of the institutional coverage of the final rule. 2. The scope of the transactional coverage of the final rule. 3. The data that financial institutions are required to report about each covered loan or application. 4. The modifications to disclosure and reporting requirements. For each major provision in the final rule, the discussion considers the benefits, costs, and impacts to consumers and covered persons, and addresses certain alternative provisions that the Bureau considered. The [[Page 66259]] discussion also addresses comments the Bureau received on the proposed Dodd-Frank Act section 1022 analysis as well as certain other comments on the benefits or costs of provisions of the proposed rule when doing so is helpful to understanding the Dodd-Frank Act section 1022 analysis. Comments that mentioned the benefits or costs of a provision of the proposed rule in the context of commenting on the merits of that provision are addressed in the relevant section-by-section analysis, above. In this respect, the Bureau's discussion under Dodd-Frank Act section 1022 is not limited to this discussion in part VII of the final notice. B. Statement of Need 1. HMDA's Purposes and the Current Deficiencies in Regulation C Congress intended HMDA to provide the public and public officials with information to help determine whether financial institutions are serving the housing needs of their communities, to target public investment to attract private investment in communities, and to identify possible discriminatory lending patterns and enforce antidiscrimination statutes. Today, HMDA data are the preeminent data source for regulators, researchers, economists, industry, and advocates analyzing the mortgage market both for the three stated purposes of HMDA and for general market monitoring. For example, HMDA data are used by bank supervisors to evaluate depository institutions for purposes of the Community Reinvestment Act (CRA); by local community groups as the basis for discussions with lenders about local community needs; and by regulators, community groups, and researchers to identify disparities in mortgage lending that may provide evidence of prohibited discrimination. In addition, HMDA data provide a broadly representative, national picture of home lending that is unavailable from any other data source. This information permits users to monitor market conditions and trends, such as the supply and demand of applications and originations. For example, industry uses HMDA data to identify and meet the needs of underserved markets through potentially profitable lending and investment opportunities. HMDA data include records regarding both applications by mortgage borrowers and the flow of funding from lenders to borrowers. Together, these records form a near-census of the home mortgage market for covered loans and applications, with rich geographical detail (down to census tract level) and identification of the specific financial institution for each transaction. Therefore, HMDA allows users to draw a detailed picture of the supply and demand of mortgage credit at various levels of geography and lender aggregation. Despite its extensive benefits, serious inadequacies exist in the information currently collected under Regulation C. Although HMDA data can generally be used to calculate underwriting and pricing disparities across various protected classes and at various levels of analysis, the data lack key fields that explain legitimate underwriting and pricing decisions for mortgage loans. Therefore, in most cases, HMDA data alone cannot demonstrate whether borrowers and applicants have received nondiscriminatory treatment by financial institutions. Additional data points, such as credit score, AUS results, combined loan to value ratio (CLTV), and debt-to-income ratio (DTI), will help users better understand the reasons for approvals and denials of applications and for pricing decisions regarding originations. Similarly, current HMDA data provide certain information about borrowers (race, ethnicity, sex, and income) and loans (loan amount, purpose, loan type, occupancy, lien status, and property type), but they do not fully characterize the types of loans for which consumers are applying and do not explain why some applications are denied. The additional data points, such as non- amortizing features, prepayment penalties, and loan terms, will help fill these important information gaps. Additionally, analysis of the cost of credit to mortgage borrowers is incomplete without the inclusion of key pricing information. The current rate spread data point requires financial institutions to report rate spread only for higher-priced mortgage loans. Currently, such loans comprise roughly 5 percent of total originations. These limited data restrict analysis of the cost of credit to a small segment of total mortgage originations and create severe selection bias as changes in the market lead to shifts in the average spread between APR values and APOR. Adding new pricing data fields, such as discount points, lender credits, origination charges, interest rate, and total loan costs will allow users to better understand the price that consumers pay for mortgages and more effectively analyze the tradeoffs between rates, points, and fees. HMDA also currently provides limited information about the property that secures or will secure the loan. Despite being one of the most important characteristics for underwriting and pricing decisions, the value of the property securing the loan has not been collected under the current HMDA reporting requirements. The final rule addresses this deficiency by providing for reporting of the value of the property securing the covered loan or application. Current HMDA data also lack certain information about the manufactured housing segment of the mortgage market. Manufactured housing is an important source of housing for many borrowers, such as low-income and elderly borrowers, that are often financially fragile and possibly more vulnerable to unfair and predatory practices.\448\ Multifamily financing for both institutional and individual borrowers serves the housing needs of multifamily unit dwellers who are mostly renters and many of whom face challenges related to housing affordability. The Bureau's final rule provides for reporting of the construction method, number of multifamily affordable units, whether a loan is or would have been secured by a manufactured home and land or by a manufactured home and not land, and the land property interest for loans or applications for manufactured housing. The improved data will help users to better understand the properties for which borrowers are receiving or being denied credit or receiving different loan pricing. --------------------------------------------------------------------------- \448\ See Mark Duda & Eric S. Belsky, The Anatomy of the Low- Income Homeownership Boom in the 1990s (Joint Ctr. for Hous. Studies of Harvard Univ., Low-Income Homeownership Working Paper Series 01- 1, 2001) (providing evidence that manufactured housing was an important driver of the homeownership boom for the low-income population in the 1990s). Manufactured housing is also an important source of housing for the elderly. See Robert W. Wilden, Comment on Affordable Housing and Health Facility Needs for Seniors in the 21st Century, Manufactured Housing and Its Impact on Seniors (2002). For additional information on manufactured housing, including the market and regulatory environment, see the Bureau's 2014 white paper, Manufactured-housing Consumer Finance in the U.S, available at http://files.consumerfinance.gov/f/201409_cfpb_report_manufactured-housing.pdf. --------------------------------------------------------------------------- Finally, Regulation C's current transactional coverage criteria omit a large proportion of dwelling-secured loan products, including large segments of the home-equity line of credit market. In the lead-up to the financial crisis between 2000 and 2008, the total balance of closed- and open-end home-equity loans and lines of credit increased by approximately 16.8 percent annually, growing from a total of $275.5 billion to $953.5 billion. Recent research has shown that this growth in home-equity lending was correlated with subsequent home price depreciation, as [[Page 66260]] well as high default and foreclosure rates among first mortgages.\449\ These correlations were driven in part by borrowers using home-equity lines of credit to fund investment properties, which impacted default rates when housing prices began to fall. By identifying home-equity lines of credit and loan purposes, industry, members of the public, and public officials will be better able to identify and respond to similar patterns in the future. --------------------------------------------------------------------------- \449\ Michael LaCour-Little et al., The Role of Home Equity Lending in the Recent Mortgage Crisis, 42 Real Estate Economics 153 (2014). --------------------------------------------------------------------------- Congress recognized current deficiencies in HMDA and responded with the Dodd-Frank Act, which amended HMDA and provided broader reforms to the financial system. The Dodd-Frank Act's amendments to HMDA require the collection and reporting of several new data points, including information about borrowers (age and credit score), information about loan features and pricing, and, as the Bureau may determine to be appropriate, unique identifiers for loans, properties, and loan originators. It also authorizes the Bureau to require financial institutions to collect and report ``such other information as the Bureau may require.'' In doing so, Congress sought to ensure that HMDA data continue to be useful for determining whether institutions are serving the housing needs of their communities, for identifying potentially discriminatory lending patterns, and for helping public officials target public investment to attract private investment where it is needed. 2. Improving HMDA Data To Address Market Failures HMDA is not principally focused on regulating the interactions between lenders and borrowers. Instead, HMDA requires financial institutions to report detailed information to their Federal supervisory agencies and to the public about mortgage applications, originations, and purchases at the transaction level. Such information provides an important public good that illuminates the lending activities of financial institutions and the mortgage market in general. This increased transparency allows members of the public, community groups, and public officials to better assess compliance with various Federal laws and regulations. In doing so, HMDA data help correct the potential market failures that those laws and regulations were designed to address. From an economics perspective, the final rule's improvements to HMDA data address two market failures: (1) The under-production of public mortgage data by the private sector, and (2) the information asymmetries in credit markets. First, HMDA data is a public good in that it is both non-rival, meaning that it may be used without reducing the amount available for others, and non-excludable, meaning that it cannot be withheld from consumers who do not pay for it. As with other public goods, standard microeconomic principles dictate that public mortgage data will be under-produced by the private sector, creating an outcome that is not socially optimal. Not surprisingly, no privately produced loan-level mortgage databases with comprehensive national coverage exist that are easily accessible by the public. Private data vendors offer a few large databases for sale that typically contain data collected from either the largest servicers or securitizers. However, none of these databases match the near-universal coverage of the HMDA data.\450\ Furthermore, commercial datasets are costly for subscribers, creating a substantial hurdle for community groups, government agencies, and researchers that wish to obtain access. Importantly, these commercially available datasets typically do not identify individual lenders and therefore cannot be used to study whether specific lenders are meeting community needs or making nondiscriminatory credit decisions. In addition, all of the privately produced, commercially available mortgage databases that the Bureau is aware of cover only originated loans and exclude applications that do not result in originations. A crucial feature of the HMDA data is that they include information about applications in addition to originations and purchases. In other words, in economic terms, private mortgage databases only provide information about the market outcome resulting from the intersection of supply and demand, while HMDA data provide information about both the market outcome and the demand for credit. Thus, users can examine both supply and demand regarding mortgage credit and understand the reasons for discrepancies between supply and demand at various levels of analysis, including by lender, geographic region, type of product or feature, credit risk, income, and race or ethnicity. --------------------------------------------------------------------------- \450\ Although limited transactions and institutions are excluded from HMDA, these are also typically excluded from commercial datasets. --------------------------------------------------------------------------- Second, it is well-accepted that credit markets are characterized by information asymmetries. Mortgage products and transactions are highly complex, and lenders have a significant information advantage. Such information asymmetry affects price and quantity allocations and can contribute to types of lender behavior, such as discrimination or predatory lending, that conflict with the best interests of borrowers. In addition to disadvantaging individual consumers, information failure may also lead to herding behavior by both lenders and consumers, creating substantial systemic risk to the mortgage market and the nation's overall financial system. The recent mortgage crisis provides a vivid demonstration of such a threat to the overall safety and stability of the housing market. These market failures are intertwined. Following the financial crisis, the Bureau and other government regulators have attempted to address misallocation of credit, enhance consumer protection, and stem systemic risk in the mortgage market through rules that regulate the business practices of financial institutions. The final rule provides an additional approach to solving failures in the mortgage market: Correcting the informational market failure. Enhanced mortgage data provide greater transparency about the mortgage market, weakening the information advantage that lenders possess relative to borrowers, community groups, and public officials. Greater information enables these groups to advocate for financial institutions to adopt fairer practices and increases the prospect that self-correction by financial institutions will be rewarded. Additional information also helps to reduce the herding behavior of both lenders and borrowers, reducing the systemic risk that has been so detrimental to the nation. In general, more information leads to more efficient outcomes. Thus, as a public good that reduces information asymmetry in the mortgage market, HMDA data are irreplaceable. In addition to addressing the two market failures, the final rule also meets the compelling public need for improved efficiency in government operations. The new data will allow government agencies to more effectively assess financial institutions' compliance with antidiscrimination statutes, including the Equal Credit Opportunity Act and the Fair Housing Act. The new data will also help to assess certain financial institutions' performance under the CRA. Improved HMDA data will also provide valuable information that supports future market analyses and optimal policy-making. [[Page 66261]] C. Baseline for Consideration of Costs and Benefits As stated in the proposal, the Bureau has discretion in any rulemaking to choose an appropriate scope of consideration for potential benefits and costs and an appropriate baseline. The Bureau does not believe the amendments to HMDA in section 1094 of the Dodd- Frank Act would take effect automatically without implementing rules. Financial institutions are not required to report additional data required by section 304(b)(5) and (6) of HMDA, as amended, ``before the first January 1 that occurs after the end of the 9-month period beginning on the date on which regulations are issued by the Bureau in final form with respect to such disclosures.'' \451\ Therefore, the Bureau believes that the requirements to report all of the new data elements under HMDA section 304(b)(4)-(6) cannot become effective until the Bureau completes a rulemaking with respect to the reporting of such data. Accordingly, this analysis considers the benefits, costs, and impacts of the major provisions of the final rule against a pre-Dodd- Frank Act baseline, i.e., the current state of the world before the provisions of the Dodd-Frank Act that amended HMDA are implemented by an amended Regulation C. The Bureau believes that such a baseline will also provide the public with better information about the benefits and costs of the statutory amendments to HMDA. The Bureau did not receive any comments on the baseline used. --------------------------------------------------------------------------- \451\ HMDA section 304(n). --------------------------------------------------------------------------- D. Coverage of the Final Rule Each provision of the final rule applies to certain financial institutions and requires them to report data regarding covered loans secured by a dwelling that they originate or purchase, or for which they receive applications. The final rule also requires financial institutions to make these data available to the public by making available brief notices referring members of the public seeking these data to the Bureau's Web site to obtain them. The provisions for which financial institutions must report, and what information they must report, are described further in each section below. E. Basic Approach of the Bureau's Consideration of Benefits and Costs and Data Limitations This discussion relies on data that the Bureau obtained from industry, other regulatory agencies, and publicly available sources, as well as public comments contained in the record established by the proposed rule. As discussed in detail below, the Bureau's ability to fully quantify the potential costs, benefits, and impacts of the final rule is limited in some instances by a scarcity of necessary data. 1. Costs to Covered Persons The final rule generally establishes which financial institutions, transactions, and data points are covered under HMDA's reporting requirements. In order to precisely quantify the costs to covered persons, the Bureau would need, for both current and future HMDA reporters, representative data on: (1) The ongoing operational costs that financial institutions incur to gather and report HMDA data; (2) one-time costs for financial institutions to update reporting infrastructure in response to the final rule; and (3) the level of complexity of financial institutions' business models and compliance systems. As stated in the proposal, the Bureau does not believe that data on HMDA reporting costs with this level of granularity is systematically available from any source. However, the Bureau has made reasonable efforts to gather as much relevant data on HMDA reporting costs as possible. Through review of the public comments and outreach efforts with industry, community groups, and other regulatory agencies, the Bureau has obtained some information about ongoing operational and one-time compliance costs, and the discussion below uses this information to quantify certain costs of the final rule. The Bureau believes that the discussion constitutes the most comprehensive assessment to date of the costs of HMDA reporting by financial institutions. However, the Bureau recognizes that these calculations may not fully quantify all costs to covered persons. The Bureau also recognizes that these calculations may not accurately represent the costs of each specific reporter, especially given the wide variation of HMDA reporting costs across financial institutions. The Bureau's process for estimating the impact of the final rule on the cost of compliance to covered persons proceeds in three general stages. First, the Bureau attempted to understand and estimate the current cost of reporting for financial institutions, i.e., the baseline cost at the institution level. Second, the Bureau evaluated the one-time costs and ongoing operational costs that financial institutions would incur in response to the final rule. Part VII.F.2, below, provides details on the Bureau's approach in performing these institution-level analyses. The Bureau realizes that costs vary by institution due to many factors, such as size, operational structure, and product complexity, and that this variance exists on a continuum that is impossible to fully represent. To conduct a cost consideration that is both practical and meaningful, the Bureau chose an approach that focuses on three representative tiers of financial institutions: Low-complexity, moderate-complexity, and high-complexity. For each tier, the Bureau produced a reasonable estimate of the cost of compliance given the limitations of the available data. Part VII.F.2, below, provides additional details on this approach. More elaboration of the Bureau's basic approach is available in the notice accompanying the proposal, the Small Business Review Panel Outline of Proposals, and the Small Business Review Panel Report.\452\ --------------------------------------------------------------------------- \452\ See 79 FR 51731 (Aug. 29, 2014); Bureau of Consumer Fin. Prot., Small Business Review Panel for Home Mortgage Disclosure Act Rulemaking: Outline of Proposals Under Consideration and Alternative Considered (Feb. 7, 2014) (Outline of Proposals), available at http://files.consumerfinance.gov/f/201402_cfpb_hmda_outline-of-proposals.pdf. Certain basic assumptions, such as wage rate and number of data fields, were updated after the proposed rule to reflect changes adopted by the final rule and more recent wage data. The Bureau also modified the tier designations for the estimated open-end reporters as a result of a separate open-end reporting threshold that was not in the proposal. --------------------------------------------------------------------------- The third stage of the Bureau's consideration of costs involved aggregating up to the market-level the institution-level cost estimates from the first two stages. This aggregation required an estimate of the total number of potentially impacted financial institutions and a mapping of these institutions to the three tiers described above. The Bureau used a wide range of data in conducting these tasks, including current HMDA data, Call Reports, NMLSR data and Consumer Credit Panel data.\453\ These analyses were challenging, because no single data source provided complete coverage of all the financial institutions that could be impacted, and the data quality of some sources was less than perfect. For example, estimating the number of HMDA reporters of closed-end mortgage loans that will be removed from coverage under the final rule was relatively easier than estimating the number of HMDA reporters that will be added. Similarly, the Bureau faced certain challenges in mapping the financial institutions to the three representative tiers, because data on the operational complexity of each financial [[Page 66262]] institution was very limited. Where the Bureau is uncertain about the aggregate impacts, it has generally provided range estimates. --------------------------------------------------------------------------- \453\ NMLSR is a national registry of nondepository financial institutions, including mortgage loan originators. --------------------------------------------------------------------------- As described in greater detail below, the Bureau received many public comments on estimating the costs of certain components of the HMDA reporting process for individual financial institutions. These comments have been considered in revising the estimates contained in this part. In general, however, the comments did not provide representative data for all current and future HMDA reporters. 2. Costs to Consumers In addition to estimating the cost impact on covered persons, the Bureau also estimated the costs to consumers. Following standard economic theory, in a perfectly competitive market where financial institutions are profit maximizers, the affected financial institutions would pass on to consumers the marginal, i.e., variable, cost per application or origination, and absorb the one-time and increased fixed costs of complying with the rule. Based on this theory, the Bureau used estimates of changes in variable costs to assess the impact of the rule on consumers. The Bureau received feedback through the Small Business Review Panel process and public comments that, if the market permitted, some lenders would attempt to pass on to consumers the entire amount of the increased cost of compliance and not just the increase in variable costs. To the extent that this were to occur, the impact of the rule on consumers would be higher than the Bureau's estimates based on variable costs. No data were available to determine whether lenders would pass on the entire increase in compliance costs. 3. Benefits to Consumers and Covered Persons The Bureau also assessed the benefits of the final rule both to consumers and covered persons. In general, the Bureau relied on qualitative discussions of benefits as opposed to quantitative estimates. The Bureau cannot readily quantify many of the benefits to consumers and covered persons with precision, both because the Bureau does not have the data to quantify all benefits and because the Bureau is not able to assess completely how effective the Dodd-Frank amendments to HMDA will be in achieving those benefits. Congress intended for HMDA, including the Dodd-Frank Act amendments to the Act and the Bureau's rules implementing HMDA, to achieve compelling social benefits. As explained elsewhere in this supplementary information, the Bureau believes that the final rule appropriately implements the statutory amendments and is necessary and proper to effectuate HMDA's purposes. For consumers, the Bureau believes that the benefit of enhanced transparency will be substantial. For example, the final rule will facilitate the detection and remediation of discrimination; promote public and private investment in certain under-served markets, potentially increasing access to mortgage credit; and promote more stable and competitive markets. As a sunshine rule regarding data reporting and disclosure, most of the benefits of the enhanced rule on consumers will be realized indirectly. Quantifying and monetizing these benefits, however, would require identifying all possible uses of HMDA data, establishing causal links to the resulting public benefits, and then quantifying the magnitude of these benefits. For instance, quantification would require measuring the impact of increased transparency on financial institution behavior, the need for public and private investment, the housing needs of communities, the number of lenders potentially engaging in discriminatory or predatory behavior, and the number of consumers currently being unfairly disadvantaged and the level of quantifiable damage from such disadvantage. The Bureau is unaware of data that would enable reliable quantitative estimates of all of these effects. Similar issues arose in attempting to quantify the benefits to covered persons. For example, the Bureau believes that the enhanced HMDA data will facilitate improved monitoring of mortgage markets in order to prevent major disruptions to the financial system, which in turn will benefit financial institutions over the long run. Such effects, however, are hard to quantify because they are largely related to future events that the final rule itself is designed to prevent. Similarly, the Bureau believes that the enhanced HMDA data will provide a better analytical basis for financial regulators and community groups to screen and monitor lenders for possible discrimination. Because of limitations in the current HMDA data fields, the potential for false positives has been widely cited by financial institutions in various HMDA-related fair lending examinations, complaints, and lawsuits. The final rule will greatly reduce the rate of false positives and the associated compliance burden on financial institutions. The Bureau believes that such benefits to financial institutions could be substantial. Nevertheless, quantifying them would require data that are currently unavailable. In light of these data limitations, the discussion below generally provides a qualitative consideration of the benefits, costs, and impacts of the final rule. These qualitative insights into the benefits are based on general economic principles, together with the limited data available. The Bureau has made quantitative estimates where possible. F. Potential Benefits and Costs to Consumers and Covered Persons 1. Overall Summary In this part VII.F.1, the Bureau presents a concise, high-level overview of the benefits and costs of the final rule. This is not intended to capture all details and nuances that are provided both in the rest of the analysis and in the section-by-section analyses above but rather to provide an overview. Major benefits of the rule. The final rule has a number of major benefits. First, the amendments will improve the usefulness of HMDA data in identifying possible discriminatory lending patterns and enforcing antidiscrimination statutes. By expanding the institutional and transactional coverage, the final rule expands the scope of the market that community groups and government agencies can include in fair lending analyses. The addition of pricing data fields such as interest rate, discount points, lender credits, and origination charges improves understanding of disparities in pricing outcomes beyond that permitted by the current rate spread data field. The addition of data fields such as CLTV, credit score, DTI, and AUS results allows for a more refined analysis and understanding of disparities in both underwriting and pricing outcomes. Overall, the changes adopted make fair lending analyses more comprehensive and accurate. This is especially important for the prioritization and peer analysis or redlining reviews that regulatory agencies conduct for fair lending supervision and enforcement purposes because a consistent and clean dataset will be available for all financial institutions subject to HMDA reporting. Second, the final rule will help determine whether financial institutions are serving the housing needs of their communities and help public officials target public investment to better attract private investment, two of HMDA's stated purposes. The expansion of institutional and transactional coverage will provide additional data helpful to the public, industry, and government in [[Page 66263]] identifying profitable lending and investment opportunities in underserved communities. Similarly, the data points related to multifamily dwellings and manufactured housing will reveal more information about these segments of the market. Borrowers who seek financing for manufactured housing are typically more financially vulnerable than borrowers financing site-built homes, and may deserve closer attention from government agencies and community groups. Although financing involving multifamily dwellings reported under HMDA is typically offered to institutional borrowers, the ultimate constituents these loans serve are mostly low- to mid-income renters who live in these financed units. Advocacy groups and government agencies have raised concerns over affordability issues faced by individuals living in multifamily dwellings, who also tend to be more financially vulnerable than individuals living in single-family dwellings. Overall, by permitting a better and more comprehensive understanding of these markets, the rule will improve the usefulness of HMDA data for assessing the supply and demand of credit, and financial institutions' treatment of applicants and borrowers, in these communities. Third, the final rule will assist in earlier identification of trends in the mortgage market, including the cyclical loosening and tightening of credit. Expanded transactional coverage, principally through reporting of most dwelling-secured consumer-purpose transactions, including open-end lines of credit, closed-end home- equity loans, and reverse mortgages, and additional data fields, such as amortization type, prepayment penalty, and occupancy type, will improve understanding of the types of products and product characteristics received by consumers. Recent research has indicated that certain product types and characteristics may have increased the likelihood of default and exacerbated declines in housing values during the recent financial crisis. These risk factors could similarly play important roles in future credit cycles. Therefore, the additional transactions and data points will improve research efforts to understand mortgage markets, help identify new risk factors that might increase systemic risk to the overall economy, and provide early warning signals of worrisome market trends. In particular, quarterly reporting will provide regulators with more timely data, which will be of significant value for HMDA and market monitoring purposes. Timelier data will improve the identification of risks to local housing markets, the analyses of the lending activities of large volume lenders, and the effectiveness of interventions or other actions by the agencies and other public officials. Fourth, the rule will improve the effectiveness of policy-making efforts. In response to the recent financial crisis, the government has generated a number of rules and implemented a wide array of public policy measures to address market failures and protect consumers. Additional data, timelier data, and increased institutional and transactional coverage will allow for more informed decisions by policy makers and will improve the consideration of benefits, costs, and impacts for future policy efforts, resulting in more effective policy. Quantifying these benefits is difficult because the size of each particular effect cannot be known in advance. Given the number of mortgage transactions and the size of the mortgage market, however, small changes in behavior can have substantial aggregate effects. Major costs of the rule. The final rule will increase ongoing operational costs and impose one-time costs on financial institutions. Financial institutions conduct a variety of operational tasks to collect the necessary data, prepare the data for submission, conduct compliance and audit checks, and prepare for HMDA-related exams. These ongoing operational costs are driven primarily by the time spent on each task and the wage of the relevant employee. The Bureau estimates that current annual operational costs of reporting under HMDA are approximately $2,500 for a representative low-complexity financial institution with a loan/application register size of 50 records; $35,600 for a representative moderate-complexity financial institution with a loan/application register size of 1,000 records; and $313,000 for a representative high-complexity financial institution with loan/ application register size of 50,000 records. This translates into an estimated per-application cost of approximately $51, $36, and $6 for representative low-, moderate-, and high-complexity financial institutions, respectively. Using recent survey estimates of net income from the Mortgage Bankers Association (MBA) \454\ as a frame of reference for these ongoing operational costs, the average net income per origination is approximately $2,900 for small/mid-size banks, $3,900 for medium banks, and $2,100 for large banks; and approximately $2,300 for small/mid-size independent mortgage companies, $3,000 for medium independent mortgage companies, and $1,900 for large independent mortgage companies.\455\ --------------------------------------------------------------------------- \454\ These estimates come from an annual survey conducted by the Mortgage Bankers Association and the STRATMOR group as part of the Peer Group Program. \455\ The Bureau notes that these net income estimates were reported by the Mortgage Bankers Association and the STRATMOR group on a per-origination basis. The Bureau estimates the HMDA operational cost per application, not per origination. --------------------------------------------------------------------------- The final rule will affect the operational tasks associated with collecting and reporting HMDA data. More time will be required for tasks such as transcribing and checking data, and more resources will need to be devoted to tasks such as internal and external audits. The Bureau estimates that, absent the mitigation efforts discussed below, covered persons' ongoing operational costs will increase by approximately $2,600 for a representative low-complexity financial institution; $17,500 for a representative moderate-complexity financial institution; and $35,700 for a representative high-complexity financial institution, per year. These estimates do not include the increases in ongoing operational costs for financial institutions that will be required to report quarterly data or open-end lines of credit. This translates into a market-level impact of approximately $50,600,000 to $88,500,000 per year. Using a 7 percent discount rate, the net present value of this impact over five years across the entire market is an increase in costs of approximately $207,400,000 to $362,900,000. For financial institutions that will be required to report HMDA data quarterly, which the Bureau estimates are all high-complexity financial institutions, the additional ongoing operational costs will be approximately $41,000 per year.\456\ This translates into a market- level impact of approximately $1,200,000 per year. Using a 7 percent discount rate, the net present value of this impact over five years across the entire market is an increase in costs of approximately $4,900,000. --------------------------------------------------------------------------- \456\ The Bureau estimates there will be 29 financial institutions that will be required to report HMDA data quarterly and that they will be high-complexity institutions. Note that this estimate refers to increased ongoing costs due to quarterly reporting beyond the costs already mentioned. --------------------------------------------------------------------------- For financial institutions that originated at least 100 open-end lines of credit in each of the two preceding years and will be required to report information about open-end lines of credit, the additional ongoing operational costs from open-end [[Page 66264]] reporting will be approximately $9,500 per year for a representative low-complexity financial institution, $53,000 per year for a representative moderate-complexity financial institution, and $288,000 per year for a representative high-complexity financial institution. This translates into a market-level impact of approximately $30,900,000 per year. Using a 7 percent discount rate, the net present value of this impact over five years across the entire market is an increase in costs of approximately $126,600,000. Combined, the impact on ongoing operational costs to reporters of closed-end mortgage loans, open-end lines of credit, and quarterly reporting translates into a market-level impact of approximately $82,600,000 to $120,600,000 per year, without accounting for any operational improvements. Using a 7 percent discount rate, the net present value of this impact over five years across the entire market is an increase in costs of approximately $338,900,000 to $494,400,000.\457\ --------------------------------------------------------------------------- \457\ The market-level estimates provide lower and upper bounds of the impact of the final rule on the market as a whole. To convey differences in impacts across the three representative tiers of financial institutions, the Bureau presents institution-level estimates for each tier and does not aggregate up to market-level estimates for each tier. The institution-level estimates for each tier provide more useful and accurate estimates of differences in impacts across the three representative financial institutions, because they do not require the additional assumptions used to map HMDA reporters to tiers. See part VII.F.2, below. --------------------------------------------------------------------------- Accounting for operational improvements undertaken by the Bureau, the estimated net increase in ongoing operational costs will be smaller than the above estimates. The Bureau's initial outreach efforts, as well as information gathered during the Small Business Review Panel process, indicated that reportability questions, regulatory clarity, geocoding, and submission processes and edits were significant concerns to financial institutions. Along with modifying the reporting requirements, the Bureau is making operational enhancements and modifications to address these concerns. For example, the Bureau is working to consolidate the outlets for assistance; provide implementation support similar to the support provided for the title XIV and the TILA-RESPA Integrated Disclosure rules; improve points of contact for help inquiries; modify the types of edits and when edits are approved; develop a Web-based HMDA data submission and edit-check system, create a data entry tool for small financial institutions that use Data Entry Software; and develop approaches to reduce geocoding burdens. All of these enhancements will improve the submission and processing of data, increase clarity, and reduce reporting burden. Accounting for these operational improvements, the estimated net impact of the final rule on ongoing operational costs for closed-end reporters will be approximately $1,900, $7,800, and $20,000 per year, for representative low-, moderate-, and high-complexity financial institutions, respectively. This translates into a market-level impact of approximately $26,700,000 to $41,400,000 per year. Using a 7 percent discount rate, the net present value of this impact over five years across the entire market is an increase in costs of approximately $109,500,000 to $169,800,000. For quarterly reporters, which the Bureau assumes are all high-complexity financial institutions, the estimated net impact of the final rule on ongoing operational costs will be approximately an additional $31,200 per year. This translates into an additional market-level impact of approximately $900,000 per year. Using a 7 percent discount rate, the net present value of this impact over five years across the entire market is an increase in costs of approximately $3,700,000. For open-end reporters, the estimated net impact of the final rule on ongoing operational costs will be approximately $8,600, $43,400, and $273,000 per year, for representative low-, moderate-, and high-complexity financial institutions respectively. This translates into a market-level impact of approximately $26,000,000 per year. Using a 7 percent discount rate, the net present value of this impact over five years across the entire market is an increase in costs of approximately $106,600,000. Combined, with the inclusion of the operational improvements, the impact on ongoing operational costs to reporters of closed-end mortgage loans, open-end lines of credit, and quarterly reporting translates into a market-level impact of approximately $53,600,000 to $68,300,000 per year. Using a 7 percent discount rate, the net present value of this impact over five years across the entire market is an increase in costs of approximately $219,800,000 to $280,100,000. In addition to impacting ongoing operational costs, the final rule will impose one-time costs necessary to modify processes in response to the new regulatory requirements. These one-time costs are driven primarily by updating software systems, training staff, updating compliance procedures and manuals, and overall planning and preparation time. The Bureau estimates that these one-time costs due to reporting of closed-end mortgage loans will be approximately $3,000 for low- complexity financial institutions, $250,000 for moderate-complexity financial institutions, and $800,000 for high-complexity financial institutions. These estimates include the impact on financial institutions that will be required to report quarterly data, but exclude the impact of expanding transactional coverage to include mandatory reporting of open-end lines of credit for financial institutions that meet the open-end reporting threshold.\458\ --------------------------------------------------------------------------- \458\ The Bureau realizes that the impact to one-time costs varies by institution due to many factors, such as size, operational structure, and product complexity, and that this variance exists on a continuum that is impossible to fully capture. As a result, the one-time cost estimates will be high for some financial institutions and low for others. --------------------------------------------------------------------------- Industry commenters indicated that many financial institutions, especially larger and more complex institutions, process applications for open-end lines of credit in their consumer lending departments using separate procedures, policies, and data systems. In addition, because most financial institutions do not currently report open-end lines of credit, many financial institutions will have to develop completely new reporting infrastructures to comply with the switch to mandatory reporting. As a result, there will be one-time costs to create processes and systems for open-end lines of credit in addition to the one-time costs summarized above to modify processes and systems for other mortgage products. The Bureau recognizes that the one-time cost of reporting open-end lines of credit could be substantial for many financial institutions, but lacks the data necessary to accurately quantify it. Although some commenters provided feedback on the additional burden of reporting data on these products, no commenter provided specific estimates of the potential one-time costs of reporting open-end lines of credit. The closest information was provided by one commenter that estimated that HELOC reporting would increase system fees by $117,000, which is similar to the Bureau's estimate of a $125,000 one-time cost related to reporting open-end lines of credit for moderately complex financial institutions.\459\ --------------------------------------------------------------------------- \459\ It is not clear from this comment whether the estimate excludes open-end lines of credit for commercial or business purposes other than purchase, home improvement, or refinancing, which financial institutions will not have to report. --------------------------------------------------------------------------- For this discussion, the Bureau assumes that if a lender will report both closed-end mortgage loans and open- [[Page 66265]] end lines of credit, the one-time cost of integrating open-end lines of credit into HMDA reporting processes will be roughly equal to 50 percent of the one-time cost absent mandatory reporting of such products. This estimate accounts for the fact that reporting open-end lines of credit will require some new systems, extra start-up training, and new compliance procedures and manuals, but that some fixed, one- time costs could be shared with closed-end lines of business subject to Regulation C because both have to undergo systemic changes. This assumption is consistent with the Bureau's estimate that, under the open-end reporting threshold, an overwhelming majority of open-end reporters would also be reporting closed-end mortgage loans and applications simultaneously, as will be discussed below in parts VII.F.3 and VII.F.4. The Bureau therefore estimates that high- and moderate-complexity financial institutions that will be required to report open-end lines of credit while also reporting closed-end mortgage loans will incur additional one-time costs of $400,000 and $125,000, respectively, due to open-end reporting. The Bureau believes that the additional one-time costs of open-end reporting will be relatively low for low-complexity financial institutions. The Bureau believes that these institutions are less reliant on information technology systems for HMDA reporting and that they may process open- end lines of credit on the same system and in the same business unit as closed-end mortgage loans. Therefore, for low-complexity financial institutions, the Bureau estimates that the additional one-time cost created by open-end reporting is minimal and is derived mostly from new training and procedures adopted for the overall changes in the final rule. For the estimated 24 lenders that would only report open-end lines of credit but not closed-end mortgage loans, because there would be no cost sharing between open-end and closed-end reporting, the Bureau adopts the one-time cost estimates for similar-sized closed-end reporters and hence conservatively estimates that the one-time costs for these open-end reporters will be approximately $3,000 for low- complexity financial institutions and $250,000 for moderate-complexity financial institutions.\460\ --------------------------------------------------------------------------- \460\ The Bureau estimates that none of the open-end-only reporters will fall into the high-complexity category. The Bureau also estimates that these open-end-only reporters previously would have been reporting under HMDA as they are depository institutions that have closed-end mortgage loan/application register sizes between 1 and 24 records. Therefore the Bureau believes that they will be able to repurpose and modify the existing HMDA reporting process for open-end reporting. --------------------------------------------------------------------------- The specific approach used to estimate one-time costs is based on the Bureau's outreach efforts prior to the proposal. Specifically, for low-complexity financial institutions, these outreach efforts indicated that the cost to update information technology systems will be minimal, because the processes involved in reporting are highly manual. The estimate of one-time training costs for low-complexity financial institutions is based on estimated ongoing training costs of $300 per year for staff directly responsible for data reporting. In response to the final rule, additional staff will require one-time training, but the intensity of this training will be lower than ongoing training. To capture this additional, less-intensive training, the Bureau used five times the annual training cost as the estimated one-time training cost ($1,500). Training costs provide the best-available proxy for the one- time cost to update compliance procedures and manuals, so the Bureau used $1,500 as an estimate of these costs as well. Therefore, the total one-time cost estimate for low-complexity financial institutions is approximately $3,000 (= $0 + $1,500 + $1,500). This estimate varies little regardless of whether the financial institution reports open-end lines of credit. For moderate-complexity financial institutions, outreach efforts indicated that representative costs to update information technology, excluding possible open-end reporting, will be approximately $225,000. The estimate of one-time training costs for moderate-complexity financial institutions, excluding possible open-end reporting, is based on the estimated ongoing training costs of $2,500 per year. Again, the Bureau used five times the annual training cost as the estimated one- time training cost ($12,500). Training costs provide the best-available proxy for the one-time cost to update compliance procedures and manuals, so the Bureau used $12,500 as an estimate of these costs as well. The one-time cost estimate for a representative moderate- complexity financial institution is therefore approximately $250,000 (= $225,000 + $12,500 + $12,500), excluding the costs of reporting open- end lines of credit. By including the 50 percent multiplier discussed above, the Bureau assumes that the one-time cost of open-end reporting by moderate-complexity financial institutions is $125,000. Therefore, for a representative moderate-complexity financial institution that meets both the open-end and closed-end reporting thresholds, the total one-time cost estimate is $375,000. For high-complexity financial institutions, outreach efforts indicated that representative costs to update information technology, excluding open-end reporting, will be approximately $500,000. The estimate of one-time training costs for high-complexity financial institutions, excluding open-end reporting, is based on the estimate of ongoing training costs of $30,000 per year. Again, the Bureau used five times the annual training cost as the estimated one-time training cost ($150,000). Training costs provide the best available proxy for the one-time cost to update compliance procedures and manuals, so the Bureau used $150,000 as an estimate of these costs as well. The one- time cost estimate for a representative high-complexity financial institution is therefore approximately $800,000 (= $500,000 + $150,000 + $150,000), excluding the costs of reporting open-end lines of credit. By including the 50 percent multiplier discussed above, the Bureau assumes that the one-time cost of open-end reporting by high-complexity financial institutions is $400,000. Therefore, for a representative high-complexity financial institution that meets both the open-end and closed-end reporting thresholds, the total one-time cost estimate is $1,200,000. Based on outreach discussions with financial institutions prior to the proposal, the Bureau also believes that additional nondepository institutions that currently do not report under HMDA but will have to report closed-end mortgage loans under the final rule will incur start- up costs to develop policies and procedures, infrastructure, and training. These start-up costs for closed-end reporting will be approximately $25,000 for these financial institutions, which the Bureau assumes to be all tier 3 institutions. This startup cost differs from the one-time costs presented above, because the one-time costs mostly involve the costs from modifying existing reporting systems for existing HMDA reporters that will continue to report, while the startup cost is the cost incurred from building an entirely new reporting system for a new HMDA reporter. The Bureau estimates the overall market impact on one-time costs for closed-end reporting to be between $650,000,000 and $1,263,200,000; the overall market impact on one-time costs for open-end reporting by financial institutions that are also closed-end reporters to be approximately [[Page 66266]] $61,600,000; the overall market impact on one-time costs for open-end reporting alone to be approximately $3,000,000; and the start-up cost for nondepository institutions that will become new closed-end reporters to be approximately $11,300,000. With these four sets of numbers together, the Bureau estimates the combined overall market impact on one-time and start-up costs of the final rule is between $725,900,000 and $1,339,100,000. As a frame of reference for all of these market-level, one-time cost estimates, the total non-interest expenses for current HMDA reporters were approximately $420 billion in 2012. The upper-bound estimate of around $1,339,100,000 is approximately 0.3 percent of the total annual non-interest expenses.\461\ Because these costs are one-time investments, financial institutions are expected to amortize these costs over a period of years. In this analysis, the Bureau amortizes all costs over five years, using a simple straight-line amortization. Using a 7 percent discount rate and a five-year amortization window, the annualized one- time and start-up costs estimate is approximately between $177,000,000 and $326,600,000 per year. --------------------------------------------------------------------------- \461\ The Bureau estimated the total non-interest expense for banks, thrifts, and credit unions that reported under HMDA based on Call Report data for depository institutions and credit unions and NMLSR data for nondepository institutions, all matched with 2012 HMDA reporters. --------------------------------------------------------------------------- Comments on the impact analysis in the proposed rulemaking. Throughout the Dodd-Frank Act section 1022 discussion in the proposal, the Bureau solicited feedback about data or methodologies that would enable it to more precisely estimate the benefits, costs, and impacts of the proposed changes. For example, the Bureau solicited data on the operational activities and distribution of financial institutions across the three tiers used to estimate costs, and on the one-time cost of reporting dwelling-secured home-equity products. The Bureau also invited feedback on possible ways to quantify the benefits of the proposal. The Bureau also sought information on what data points are applicable to specific products, and on whether there are any alternatives to or adjustment in each data point that would reduce burden on covered persons while still meeting the purposes of HMDA. In general, industry commenters offered various estimates of the burden associated with the proposal for the particular financial institution represented by the commenter. For example, commenters representing different financial institutions provided estimates of the increased burden on a per-loan basis that ranged from $3 to over $73.42, 30 minutes to 60 minutes, and 70 to 100 percent. Other industry commenters framed their estimated increases in burden in terms of additional full-time employees, and provided estimates ranging from one to 15 employees. Other industry commenters attempted to estimate the overall increased cost of all aspects of the proposal, which ranged from $40,000 to $1,000,000. Other commenters framed their estimates of the overall increased costs of all aspects of the proposal on an annual basis, which ranged from $7,500 to $75,000 per year. One national trade association commenter surveyed its members and reported that implementing the data points required by the Dodd-Frank Act would represent one-time costs of $9,591 and ongoing costs of $3,842 per year, and implementing the Bureau's discretionary data points would represent one-time costs of $13,955 and ongoing costs of $4,842 per year. Finally, several industry commenters offered general estimates that the burden of reporting would double, triple, or increase exponentially. The Bureau has reviewed these estimates and considered the information reported by the commenters. Many industry commenters criticized aspects of the proposal's Dodd- Frank Act section 1022 discussion. The most common criticism was disagreement with the accuracy of the cost estimates contained in the proposal. Several industry commenters pointed out that the proposal's cost estimates were considerably different than the actual costs involved in HMDA reporting by the individual financial institution represented by the commenter. For example, one industry commenter specifically questioned the $1,600 estimate for operational costs for low-complexity financial institutions in the proposal. As a second example, another commenter suggested that the estimated cost per transaction could not be accurate, because a small entity representative reported that it spent an average of three hours just on following up with loan officers regarding missing government monitoring information. The Bureau notes that the current costs of reporting data under HMDA, as well as the impact of the final rule, are all institution- specific. For the purpose of the section 1022 discussion, however, it is not possible to generate separate estimates for each HMDA reporter. As a meaningful alternative, the Bureau constructed benefits, costs, and impacts for three representative institutions. As a result, estimates from specific commenters often deviated from the Bureau's estimates as expected. Sometimes, however, the cost estimates of the representative financial institution and the cost estimates of a particular commenter aligned. For example, one industry commenter described the Bureau's estimated one-time implementation costs for moderate-complexity financial institution as potentially correct. Although the estimated impacts of the proposed rule on many institutions deviated from the estimates the Bureau constructed for three representative institutions, these commenters, in general, did not disagree with the Bureau's methodology or assumptions. Other industry commenters cited flaws with the data used to estimate the costs and benefits of the proposal. For example, one commenter explained that the discussion was based on data from current HMDA reporters and therefore may not allow accurate estimates of the impact on newly reporting nondepository institutions. Another commenter generally stated that the discussion used insufficient quantitative data. Scarcity of data in general, and of quality data in particular, posed a challenge when estimating the benefits and costs of the final rule. This was especially true when constructing estimates for newly reporting financial institutions, because it is difficult to identify exactly which institutions would have to report, and data on these institutions are limited. To the extent possible, the Bureau utilized the best and most current data from what it knew to be the relevant and available data sources. No commenter identified any additional data sources that would have improved the Bureau's estimates. Nevertheless, in response to those comments, the Bureau reanalyzed currently available data sources to better understand the impacts of the final rule. For example, following the proposal and comment period, the Bureau thoroughly analyzed Call Reports and Consumer Credit Panel data to better understand the open-end line of credit market and the impacts of requiring reporting of these products. Details of this analysis are included in the discussions on institutional and transactional coverage below. Some industry commenters believed that the cost estimates were internally inconsistent or inconsistent with other parts of the proposal. For example, one commenter doubted that variable costs would increase by only $0.30 per application if the number of fields were essentially doubling. This comment highlights one of the many nuances of the analysis in the proposal. The $0.30 [[Page 66267]] estimate is for a representative moderate-complexity institution, and captures the estimated impact on variable operational costs of having to report 37 additional data fields.\462\ As indicated in Tables 2-4 below, the Bureau designated five of the 18 operational tasks as variable-cost tasks, so the $0.30 estimate only captures part of the overall impact of increasing the number of fields financial institutions must report. When assessing the impact to consumers, the Bureau focused on the variable costs based on standard economic theory that, under perfect competition, institutions will pass on increases in variable costs to consumers but will absorb the one-time costs and increases to fixed costs. No commenters disagreed with the Bureau's designation of tasks as variable-cost or fixed-cost, and no commenters suggested improvements to the formulations or assumptions the Bureau used to construct estimates for each operational task. Therefore, although the representative institution estimates may not precisely match the projected impact for a particular institution, the Bureau continues to believe that the representative estimates are a meaningful alternative to a particularized estimate for each institution, and has decided not to modify its basic methodological approach in response to this comment.\463\ --------------------------------------------------------------------------- \462\ The 37 additional data fields were contained in the proposed rule. The final rule increases the total number of additional data fields. That change has been reflected in the Bureau's updated impact analyses in this final rule. \463\ However, the Bureau did update some of its basic assumptions, including wage rate and number of data fields after the proposal to reflect the final rule and more recent wage data. The Bureau also modified the tier designations for estimated open-end reporters as a result of a separate open-end reporting threshold that the Bureau instituted in the final rule in response to the public comments. --------------------------------------------------------------------------- Many industry commenters believed that the Bureau had not considered certain costs associated with reporting HMDA data. A few commenters believed that the methodology used to estimate costs omitted certain tasks connected to reporting, such as the increased time spent on examinations and scrubbing and re-scrubbing the data. As noted in Tables 2-4 below, the Bureau included standard annual edits and internal checks, as well as examination preparation and examination assistance as three of the 18 operational steps institutions use when preparing and reporting HMDA data. The Bureau discussed all 18 operational steps with small entity representatives during the Small Business Review Panel process and solicited feedback on these steps, along with formulations for estimating their costs, in the proposed rule. Although some institutions indicated that they used slightly different tasks, in general, all feedback received indicated that these 18 operational tasks generally reflect the steps most financial institutions take when gathering and reporting HMDA data. Other commenters cited other elements of cost that they believed should have been included in the discussion. One industry commenter stated that the Bureau should consider the opportunity cost of time spent reporting HMDA data. Although not explicitly stated, the current estimates do consider the opportunity cost of the impact of the final rule. In response to the final rule, some current employees will trade off profit-related activities for HMDA-related activities. The opportunity cost of the final rule is the lost profit from this reallocation of staff time. Wages are typically used as a proxy for opportunity cost, and this is the measure the Bureau uses to estimate the cost of financial institutions having to reallocate employee time to HMDA-related activities in response to the final rule. Two other commenters suggested that the Bureau include the privacy costs of the proposed rule, such as the cost associated with data breaches. These commenters provided no information that would enable accurate estimates of such costs. Because any potential data breach is an inherent part of lenders' operational risk associated with any data operation, the Bureau cannot precisely estimate its cost for the representative institutions in its three-tier approach. Financial institutions collect and maintain significant amounts of highly sensitive, personally identifiable information concerning customers in the ordinary course of business. The Bureau understands that substantially all of the new data to be compiled under the final rule either are data that HMDA reporters compile for reasons other than HMDA or are calculations that derive from such data, and must be retained by financial institutions to comply with other applicable laws. Therefore, the Bureau does not believe that costs related to the risk of data breaches substantially affect the estimates contained in this section 1022 discussion. Several other industry commenters stated that the Bureau did not discuss potential competitive disadvantages that small financial institutions might suffer as a result of the rule, because they would be unable to distribute the cost of compliance among as large a transaction base as large financial institutions. Several industry commenters cited reports from Goldman Sachs and Banking Compliance Index figures to support claims that regulatory burdens were disproportionally affecting small financial institutions and preventing low-income consumers from accessing certain financial products. Another industry commenter cited the decline in HMDA reporters from 2012 to 2013 as evidence that small financial institutions have left the market. The Bureau presented separate impact estimates for low-, moderate-, and high-complexity institutions, broadly reflecting differences in impact across institutions of different size. For low- complexity institutions, which best represent small institutions, the estimated impact on ongoing operational costs from reporting closed-end mortgage loans, after the operational modifications the Bureau is making, is approximately $1,900 under the final rule. This translates into approximately a $38 increase in per-application costs. Based on recent survey estimates of net income from the MBA, this impact represents approximately 1.3 percent ($38/$2,900) of net income per origination for small/mid-size banks.\464\ The Bureau views that amount as relatively small. In addition, the Bureau has increased the closed- end mortgage loan reporting threshold for depository institutions from one to 25, and instituted an open-end line of credit reporting threshold of 100 to alleviate burden on small financial institutions while still maintaining the benefits of HMDA data. Therefore, the Bureau concludes that the final rule is unlikely to competitively disadvantage small institutions. --------------------------------------------------------------------------- \464\ According to a recent annual survey on mortgage originators by the Mortgage Bankers Association and the STRATMOR group as part of the Peer Group Program, the average net income per origination is approximately $2,900 for small/mid-size banks, $3,900 for medium banks, and $2,100 for large banks; and approximately $2,300 for small/mid-size independent mortgage companies, $3,000 for medium independent mortgage companies, and $1,900 for large independent mortgage companies. --------------------------------------------------------------------------- A few industry commenters stated that the Dodd-Frank Act section 1022 discussion did not address the proposal's expanded coverage of commercial loans. As explained above, based on these comments and subsequent analysis, the Bureau has decided to maintain Regulation C's existing transactional coverage scheme for commercial-purpose transactions. The final rule will only require reporting of applications for, and originations of, dwelling-secured [[Page 66268]] commercial-purpose loans and lines of credit if they are for home purchase, home improvement, or refinancing purposes. The Bureau believes the volume of such transactions is fairly small and that, as a result, it is unnecessary to account separately for the costs, benefits, and impacts of commercial-purpose reporting under the final rule. Many industry commenters argued that the degree of alignment to the MISMO data standards would increase burden. Several financial institutions reported that they would need to train their staff members in order to understand the MISMO definitions. One commenter suggested that use of the MISMO data standards should be optional because it would be burdensome for small financial institutions. A national trade association commenter reported that only 22 percent of its members reported using MISMO. These commenters have misunderstood the implications of the proposed MISMO utilization. The Bureau did not propose to, and the final rule does not require, any financial institution to use or become familiar with the MISMO data standards. Rather, the rule merely recognizes that many financial institutions are already using the MISMO standard for collecting and transmitting mortgage data and uses similar definitions for certain data points in order to reduce burden. Thus, the rule decreases costs for those institutions that already maintain data points with the same definitions and values as MISMO. Financial institutions that are unfamiliar with MISMO may not realize a similar reduction in cost, and will have to report data points not required under the current rule, but they will not experience any increased burden from reporting those HMDA data points that the Bureau has defined consistently with MISMO definitions. These institutions will not need to learn anything about MISMO because the final rule itself and the associated materials contain all the necessary definitions and instructions for reporting HMDA data. One industry commenter believed that the cost estimates should not be amortized over five years because financial institutions may not recover these costs over that time period. The Bureau presented both non-amortized market-level estimates and market-level estimates amortized over five years. As noted earlier, it is not feasible to tailor the analysis to each financial institution subject to the rule. The Bureau believes that these results effectively provide a general picture of the impact of the final rule on costs. Many industry commenters believed that the proposal would likely increase the cost of credit for consumers. Several of these commenters cited the cost of system modifications associated with reporting open- end lines of credit. A few commenters claimed that certain small financial institutions, such as small credit unions, small farm credit lenders, or small banks, would be faced with difficult choices, such as merging, raising prices, originating fewer loans, or exiting the market. A small number of industry commenters stated that they would double their origination fees as a result of the proposed rule. A national trade association commenter cited, among other things, a study from several individuals at the Mercatus Center at George Mason University and a survey of its members showing that small financial institutions were decreasing their mortgage lending activity in response to increased regulatory burdens. Similarly, other industry commenters pointed to a report from Goldman Sachs showing that higher regulatory costs had priced some low-income consumers out of the credit card and mortgage markets. Following standard economic theory, in a perfectly competitive market where financial institutions are profit maximizers, the affected financial institutions would pass on to consumers the marginal, i.e., variable, cost per application or origination and would absorb the one-time and increased fixed costs of complying with the rule. Overall, the Bureau estimates that the final rule will increase variable costs by $23 per closed-end mortgage application for representative low-complexity institutions, $0.20 per closed-end mortgage application for representative moderate-complexity institutions, and $0.10 per closed-end mortgage application for representative high-complexity institutions. The Bureau estimates that the final rule will increase variable costs by $41.50 per open-end line of credit application for representative low-complexity institutions, $6.20 per open-end line of credit application for representative moderate-complexity institutions, and $3 per open-end line of credit application for representative high-complexity institutions. These expenses will be amortized over the life of a loan and represent a negligible increase in the cost of a mortgage loan. Therefore, the Bureau does not anticipate any material adverse effect on credit access in the long or short term even if financial institutions pass on these costs to consumers. One national trade association commenter asked the Bureau to consider the indirect impact on rural consumers and to analyze the effect of the proposed rule combined with the other recent mortgage rules. This commenter noted that most of its members lend in rural areas and cited the Mercatus Center study mentioned above, which explained that small financial institutions in rural markets were particularly burdened by recent regulatory changes. Part VII.G.2 of the proposed rule considered the impact of the proposed rule on rural consumers. Following standard economic principles suggesting that institutions will pass on increases in variable costs, the Bureau estimated that the impact on consumers in rural areas will be small. Although some commenters suggested considering these impacts further, no commenters provided any specific estimates or suggested changes to methodology that could alter that conclusion. Many commenters suggested that the Bureau provide an analysis of the costs and benefits of different alternatives, such as additional possible loan-volume thresholds. The Bureau has considered several alternatives and has described the costs and benefits of these alternatives, to the extent permitted by available data, in greater detail elsewhere in this final notice. As one example, Tables 5-7 in part VII.F.3 summarize the numbers of institutions and applications that would be excluded under closed-end reporting thresholds of 25, 50, 100, 250, and 500 loans. Similarly, in response to comments received, the Bureau conducted additional analyses and subsequently constructed analogous tables showing the impact of the rule on reporting of open- end lines of credit at various thresholds. These estimates are shown in Table 8. One industry commenter claimed that the Bureau improperly discussed benefits outside of the statutory purposes of HMDA. Section 1022 of the Dodd-Frank Act, however, contains no such limitation. Instead, the statute directs the Bureau to consider, among other things, the ``potential benefits and costs to consumers and covered persons.''\465\ Although the discussion of benefits is focused on the statutory purposes of HMDA, improved information about the mortgage market will have other benefits that may fall outside of a narrow reading of the statutory purposes. The Bureau believes that failing to consider these benefits would deprive the public of important [[Page 66269]] information about the potential impacts of the final rule. --------------------------------------------------------------------------- \465\ 12 U.S.C. 5512(b)(2)(A)(i). --------------------------------------------------------------------------- Finally, one commenter urged the Bureau to gather data and define clear metrics for evaluating the success of the rule for retrospective review. This commenter offered several means of evaluation, including whether changes occur in antidiscrimination enforcement, redlining activity, false positive rates, access to credit, public and private investment, or costs to consumers. Section 1022(d) of the Dodd-Frank Act requires the Bureau to assess each ``significant'' rule or order adopted by the Bureau under Federal consumer financial law.\466\ This assessment must consider the effectiveness of the rule in meeting the purposes and objectives of the Consumer Financial Protection Act of 2010 and the specific goals stated by the Bureau, and the Bureau must publish a report of its assessment within five years of the effective date of the rule.\467\ Before publishing the report of its assessment, the Bureau must also invite public comment regarding the modification, expansion, or elimination of the significant rule.\468\ The Bureau believes that this rule will almost certainly constitute a significant rule that warrants assessment under section 1022(d) of the Dodd-Frank Act. Therefore, it will be evaluating the effectiveness of the rule along dimensions similar to those proposed by the commenter and will provide the public with an opportunity for public comment. --------------------------------------------------------------------------- \466\ 12 U.S.C. 5512(d). \467\ 12 U.S.C. 5512(d)(1)-(2). \468\ 12 U.S.C. 5512(d)(3). --------------------------------------------------------------------------- 2. Methodology for Generating Cost Estimates Prior to the proposal, the Bureau reviewed the current HMDA compliance systems and activities of financial institutions. The review used a cost-accounting, case-study methodology consisting, in part, of interviews with 20 financial institutions of various sizes, nine vendors, and 15 governmental agency representatives.\469\ These interviews provided the Bureau with detailed information about current HMDA compliance processes and costs.\470\ This information showed how financial institutions gather and report HMDA data and provided the foundation for the approach the Bureau took to considering the benefits, costs, and impacts of the final rule. The Bureau augmented this information through the Small Business Review Panel process and through relevant academic literature, publicly available information and data sources available through the Internet,\471\ historical HMDA data, Call Report Data, NMLSR Data, public comments contained in the rulemaking docket established by the proposal, and the Bureau's expertise. --------------------------------------------------------------------------- \469\ For a discussion of this methodology in the analysis of the costs of regulatory compliance, see Gregory Elliehausen, Bd. of Governors of the Fed. Reserve Sys., Staff Studies Series No. 171, The Cost of Bank Regulation: A Review of the Evidence, (April 1998), available at http://www.federalreserve.gov/pubs/staffstudies/1990-99/ss171.pdf. In addition, the Bureau recently conducted a Compliance Cost Study as an independent analysis of the costs of regulatory compliance. See U.S. Consumer Fin. Prot. Bureau, Understanding the Effects of Certain Deposit Regulations on Financial Institution's Operations: Findings on Relative Costs for Systems, Personnel, and Processes at Seven Institutions (2013), available at http://files.consumerfinance.gov/f/201311_cfpb_report_findings-relative-costs.pdf. \470\ The financial institutions interviewed were selected to provide variation in key characteristics like institution type (bank, credit union, independent mortgage bank), regulator, record count, submission mechanism, number of resubmissions, and other designations like whether the financial institution was a multifamily or rural lender. However the Bureau recognizes that this does not constitute a random survey of financial intuitions and the sample size might not be large enough to capture all variations among financial institutions. \471\ Internet resources included, among others, sites such as Jstor.org, which provides information on published research articles; FFIEC.gov, which provides information about HMDA, CRA, and the financial industry in general; university Web sites, which provide information on current research related to mortgages, HMDA, and the financial industry; community group Web sites, which provide the perspective of community groups; and trade group Web sites, which provide the perspective of industry. --------------------------------------------------------------------------- Based on the outreach described above, the Bureau classified the operational activities that financial institutions currently use for HMDA data collection and reporting into discrete compliance ``tasks.'' This classification consists of 18 ``component tasks,'' which can be grouped into four ``primary tasks.'' The level of detail of the classification is intended to facilitate estimation of baseline costs and to enable rigorous analysis of the impact of the final rule across a wide range of financial institutions. The four primary tasks are described briefly below. 1. Data collection: transcribing data, resolving reportability questions, and transferring data to HMDA Management System (HMS). 2. Reporting and resubmission: Geocoding, standard annual edit and internal checks, researching questions, resolving question responses, checking post-submission edits, filing post-submission documents, creating modified loan/application register, distributing modified loan/application register, distributing disclosure statement, and using vendor HMS software. 3. Compliance and internal audits: Training, internal audits, and external audits. 4. HMDA-related exams: Examination preparation and examination assistance. In addition to collecting information about operational activities and costs, the Bureau also used outreach efforts and the Small Business Review Panel process to better understand the potential one-time costs that HMDA reporters will incur in response to the proposed rule. Management, legal, and compliance personnel will likely require time to learn new reporting requirements and assess legal and compliance risks. Financial institutions that use vendors for HMDA compliance will incur one-time costs associated with software installation, troubleshooting, and testing. The Bureau is aware that these activities will take time and that the costs may vary depending on the time available. Financial institutions that maintain their own reporting systems will incur one- time costs to develop, prepare, and implement necessary modifications to those systems. In all cases, financial institutions will need to update training materials to reflect new requirements and activities and may have certain one-time costs for providing initial training to current employees. The Bureau recognizes that the cost per loan of complying with the current requirements of HMDA, as well as the operational and one-time impact of the final rule, will differ by financial institution. During the Bureau's outreach with financial institutions, the Bureau identified seven key dimensions of compliance operations that were significant drivers of compliance costs. These seven dimensions are: The reporting system used; the degree of system integration; the degree of system automation; the compliance program; and the tools for geocoding, performing completeness checks, and editing. The Bureau found that financial institutions tended to have similar levels of complexity in compliance operations across all seven dimensions. For example, if a given financial institution had less system integration, then it tended to use less automation and less-complex tools for geocoding. Financial institutions generally did not use less-complex approaches on one dimension and more-complex approaches on another. The small entity representatives validated this perspective during the Small Business Review Panel meeting. To capture the relationships between operational complexity and compliance [[Page 66270]] cost, the Bureau used these seven dimensions to define three broadly representative financial institutions according to the overall level of complexity of their compliance operations. Tier 1 denotes a representative financial institution with the highest level of complexity, tier 2 denotes a representative financial institution with a moderate level of complexity, and tier 3 denotes a representative financial institution with the lowest level of complexity. For each tier, the Bureau developed a separate set of assumptions and cost estimates. All of these assumptions and cost estimates apply at the institutional level.\472\ In the Outline of Proposals prepared for the Small Business Review Panel, the Bureau provided a detailed exposition of the analytical approach used for the three tiers.\473\ Small business representatives attending the Small Business Review Panel did not raise substantial objections to this three-tier approach. --------------------------------------------------------------------------- \472\ The Bureau assumes that, for closed-end reporters, the tier 1 representative financial institution has 50,000 records, the tier 2 representative has 1,000 records, and the tier 3 representative has 50 records on the HMDA loan/application register. All cost estimates reflect the assumptions defining the three representative financial institutions and reflect general characteristics and patterns, including man-hours spent on each of the 18 component tasks and salaries of the personnel involved. To the extent that an individual financial institution specializes in a given product, or reports different numbers of records on its loan/ application register, these representative estimates will differ from the actual cost to that particular financial institution. \473\ See U.S. Consumer Fin. Prot. Bureau, Small Business Review Panel for Home Mortgage Disclosure Act Rulemaking: Outline of Proposals Under Consideration and Alternative Considered (Feb. 7, 2014) (Outline of Proposals), available at http://files.consumerfinance.gov/f/201402_cfpb_hmda_outline-of-proposals.pdf. --------------------------------------------------------------------------- Table 1 below provides an overview of all three representative tiers across the seven dimensions of compliance operations: [GRAPHIC] [TIFF OMITTED] TR28OC15.001 Tables 2-4 convey the baseline estimates of annual ongoing operational costs as well as the underlying formulas used to calculate these estimates for the 18 operational tasks for the three representative financial institutions. The wage rate is $33 per hour, which is the national average wage for compliance officers based on the most recent National Compensation Survey from the Bureau of Labor Statistics (May 2014).\474\ The number of applications for tier 3, tier 2, and tier 1 financial institutions is 50, 1,000, and 50,000, respectively. The Bureau used similar breakdowns of the 18 operational tasks for each representative financial institution to estimate the impact of the final rule on ongoing operational costs. The Bureau notes that with the assumed wage rate, number of applications, and other key assumptions provided in the notes following each table, readers of this discussion may back out all elements in the formulas provided below using the baseline estimates for each task in each tier. --------------------------------------------------------------------------- \474\ The Bureau has updated the wage rate used throughout the impact analyses accompanying this final rule to $33 per hour, up from $28 used in the proposal, in order to reflect the most recent ongoing labor costs for financial institutions. Consequently, the baseline cost estimates in this final rule are higher than what the Bureau presented in the proposal. --------------------------------------------------------------------------- BILLING CODE 4810-AM-P [[Page 66271]] [GRAPHIC] [TIFF OMITTED] TR28OC15.002 [[Page 66272]] [GRAPHIC] [TIFF OMITTED] TR28OC15.003 [[Page 66273]] [GRAPHIC] [TIFF OMITTED] TR28OC15.004 BILLING CODE 4810-AM-C The baseline cost assumptions and cost estimates presented above reflect the current world in which most open-end lines of credit are not reported under HMDA. In the final rule, reporting of open-end lines of credit becomes mandatory for those institutions that meet all the other criteria for a ``financial institution'' in final Sec. 1003.2(g) and originated at least 100 open-end lines of credit. The Bureau estimated that currently only about 1 percent of total open-end lines of credit secured by dwellings were reported under HMDA. Hence, the Bureau has assumed that the baseline costs for open-end reporting in the current rule are zero. The Bureau believes that the HMDA reporting process and ongoing operational cost structure for reporting open-end lines of credit under the final rule will be fundamentally similar to closed-end reporting. Therefore, for open-end reporting the Bureau adopted the three-tier approach and most of the key assumptions used for closed-end reporting above, with two modifications. First, for the representative low-complexity open-end reporter, the Bureau assumed that the number of open-end lines of credit applications would be 150. This was set to both accommodate the threshold of 100 open-end lines of credit and to reasonably reflect the likely distribution among the smallest open-end reporters based on the Bureau's estimated number [[Page 66274]] of likely open-end reporters and their volumes. Second, for the representative high-complexity open-end reporter, the Bureau assumed that the number of open-end line of credit applications would be 30,000. This reflects a reasonable distribution among the largest open- end reporters based on the Bureau's estimated number of likely open-end reporters and their volumes. The Bureau assumed that the number of open-end line of credit applications for the representative moderate- complexity open-end reporter would still be 1,000, just as for the moderate-complexity closed-end reporter. The sections on transactional and institutional coverage discuss the Bureau's approach regarding the cost of open-end line of credit reporting in more detail. To this point, all estimates apply at the level of the institution. To aggregate institution-level information to generate cost estimates at the market level, the Bureau developed an approach to map all HMDA closed-end reporters to one of the three tiers. Because financial institutions are arrayed along a continuum of compliance costs that cannot be precisely mapped to the three representative tiers, the Bureau has adopted a conservative strategy based on a possible range of the number of financial institutions in each tier. To identify these distributions, the Bureau relied on the Bureau's best estimate of the total number of closed-end reporters and the number of total closed-end loan/application register records under the final rule. In particular, the Bureau used the total number of reporters (7,197) and the total number of loan/application register records (16,698,000) in the 2013 HMDA data. As a first step, the Bureau identified all possible tier distributions among closed-end reporters that were consistent with the reporter and record counts, using the same loan/application register sizes adopted in the institutional-level analysis (50,000 for tier 1 institutions; 1,000 for tier 2 institutions; and 50 for tier 3 institutions). Specifically, the Bureau set the following two constraints: (1) The total number of HMDA reporters in all three tiers must sum to 7,197; and (2) using the assumed loan/application register size in each tier, the total number of loan/application register records by all reporters in all three tiers must sum to 16,698,000. Additionally, the Bureau imposed two constraints. First, the Bureau classified all 184 HMDA reporters with over 10,000 records as tier 1, because the Bureau's investigation led it to believe that these large financial institutions all possess a high level of complexity in HMDA reporting. Second, the Bureau assumed that at least 20 percent of financial institutions were tier 2 and at least 20 percent were tier 3. These assumptions helped to narrow the range of possible combinations. The Bureau also substituted the actual loan/application register size of the 184 largest HMDA reporters into the constraint for the loan/ application register size of a tier 1 financial institution, further narrowing the range of possible combinations. The Bureau notes that all distributions identified are mathematically possible based on the Bureau's assumptions. Second, for the subset of tier distributions satisfying these closed-end reporter and count constraints, the Bureau then estimated market-level costs associated with closed-end reporting based on the tier-specific assumptions and cost estimates. That is, for a given distribution derived in the first step, the Bureau multiplied the institutional-level cost estimate associated with closed-end reporting for each tier by the number of institutions in that tier, and then summed across all three tiers. The distributions with the lowest- and highest-estimated market-level costs provided the lower and upper bounds for the market-level closed-end cost estimates throughout the consideration of the benefits and costs. Specifically, the Bureau arrived at two distributions for all closed-end reporters: (1) The first distribution has 3 percent of financial institutions in tier 1, 71 percent of financial institutions in tier 2, and 26 percent of financial institutions in tier 3; and (2) the second distribution has 4 percent of financial institutions in tier 1, 28 percent of financial institutions in tier 2, and 68 percent of financial institutions in tier 3. These two distributions likely do not match the state of the world exactly. Nevertheless, for the set of assumptions described above, these distributions provide upper and lower bounds for the market-level estimates of closed-end reporting. The Bureau recognizes that this range estimate does not permit perfect precision in estimating the impact of the final rule, but rather provides ranges. The Bureau adopted a different strategy in assigning open-end reporters to the 3 tiers that will be discussed in detail in the sections on transactional and institutional coverage. Initial outreach efforts, as well as information gathered during the Small Business Review Panel process, indicated that compliance costs for financial institutions were impacted by the complexity of the data field specifications and the process of submitting and editing HMDA data. The public comments the Bureau received for the proposed rule did not present information contrary to that conclusion. As part of implementing the final rule, the Bureau will be implementing several operational improvements. For example, the Bureau is working to consolidate the outlets for assistance, provide implementation support similar to the support provided for title XIV and the TILA-RESPA Integrated Disclosure rules; and improving points of contact for help inquiries. In addition, the Bureau is improving the geocoding process, creating a web-based submission tool, developing a data-entry tool for small financial institutions that currently use Data Entry Software, and otherwise streamlining the submission and editing process to make it more efficient. All of these enhancements will clarify the data field specifications and reduce burden. The consideration of benefits and costs discusses how these enhancements will affect the impact of the final rule. 3. The Scope of the Institutional Coverage of the Final Rule The final rule revises the threshold that determines which financial institutions are required to report data under HMDA. Specifically, depository and nondepository institutions that meet all the other criteria for a ``financial institution'' in final Sec. 1003.2(g) \475\ will only be required to report HMDA data if they originated at least 25 closed-end mortgage loans or at least 100 open- end lines of credit in each of the two preceding calendar years. Also, certain nondepository institutions that currently are exempt will become HMDA reporters under the final rule. --------------------------------------------------------------------------- \475\ Under Sec. 1003.2, a bank, savings association, or credit union meets the definition of financial institution if it satisfies all of the following criteria in addition to the loan-volume test described above: (1) on the preceding December 31, it had assets in excess of the asset threshold established and published annually by the Bureau for coverage by the Act; (2) on the preceding December 31, it had a home or branch office in a MSA; (3) during the previous calendar year, it originated at least one home purchase loan or refinancing of a home purchase loan secured by a first-lien on a one- to four-unit dwelling; and (4) the institution is federally insured or regulated, or the mortgage loan referred to in item (3) was insured, guaranteed, or supplemented by a Federal agency or intended for sale to the Federal National Mortgage Association or the Federal Home Loan Mortgage Corporation. A nondepository institution meets the definition of financial institution if it (1) had a home or branch office in an MSA in the preceding calendar year and (2) satisfies the loan-volume test discussed above. --------------------------------------------------------------------------- Based on data from Call Reports, HMDA, and the NMLSR, the Bureau estimates that the new threshold of 25 closed-end mortgage loans will reduce the number of reporting depository [[Page 66275]] institutions by approximately 1,400 (eliminating approximately 51,000 loan/application register records) and will increase the number of reporting nondepository institutions by approximately 75-450 (adding approximately 30,000 loan/application register records), for a net reduction of 950 institutions and 21,000 records.\476\ Based on data from Call Reports, HMDA, and Consumer Credit Panel data, the Bureau estimates that the new separate threshold of 100 open-end lines of credit will not require reporting by any financial institutions that are not currently reporting. The open-end threshold will require a small number of depository institutions, approximately 24, that will not be required to report HMDA data on their closed-end lending to report HMDA data on their open-end lending. These 24 financial institutions are current HMDA reporters but would have been excluded under the proposal's coverage test because they originate fewer than 25 closed-end mortgage loans annually. Combined, these 24 financial institutions will account for approximately 60,000 loan/application register records regarding open-end lines of credit. The vast majority of loan/application register records related to open-end lines of credit, approximately 900,000 loan/application register entries, will come from financial institutions that are both open- and closed-end reporters, because most financial institutions that will be required to report open-end lines of credit also will report closed-end mortgage loans.\477\ --------------------------------------------------------------------------- \476\ Estimates of the number of depository institutions that will no longer be required to report closed-end mortgage loans under HMDA, as well as the reduction in loan/application register volume associated with the 25 closed-end mortgage loan threshold can be obtained directly from current HMDA data and are therefore relatively reliable. It is difficult to estimate how many nondepository institutions will become HMDA reporters under the final rule's closed-end reporting threshold. These institutions are not currently HMDA reporters, so estimating how the final rule will affect them requires gathering, and making assumptions about, data and information from other sources. There are various data quality issues related to these sources, so the estimates for nondepository institutions should be viewed as the best-effort estimates given the data limitations. To avoid underestimating the costs of the final rule, the Bureau's quantitative estimates are based on the assumption that 450 nondepository institutions will become HMDA reporters, which is the high end of the range. \477\ The Bureau believes that few nondepositories engage in open-end lending. Determining the exact number of depository institutions that will be required to report under HMDA because of the open-end-line-of-credit reporting threshold requires information that is not readily available. As discussed further below, the Bureau had to rely on certain assumptions to derive the estimated number of depository institutions that will report open-end lines of credit. Based on recent HMDA data, Call Reports, credit union Call Reports, and Consumer Credit Panel data, the Bureau estimates there will be approximately 749 financial institutions that will report open-lines of credit, including approximately 725 depositories that will also report closed-end mortgage loans. In total they likely will report approximately 900,000 loan/application register records. Much of that detail is discussed in the section on transactional coverage. Expansions or contractions of the number of financial institutions, or changes in product offerings and demands between now and implementation of the final rule may alter these estimated impacts. --------------------------------------------------------------------------- Because the final rule includes both open-end and closed-end reporting thresholds, it is difficult to discuss the impact on institutional coverage without also discussing the impact on transactional coverage. Given that the Bureau estimates that adopting a threshold of 100 open-end lines of credit will affect the number of reportable transactions more significantly than the number of reporting institutions, much of the discussion relating to the open-end reporting threshold is found in the discussion of transactional coverage in part VII.F.4, below. The discussion in this part primarily addresses the changes to institutional coverage resulting from the closed-end reporting threshold and open-end-only reporters resulting from the separate open-end reporting threshold. Benefits to consumers. The institutional coverage threshold related to closed-end mortgage loans will have several benefits to consumers. First, the final rule will expand the coverage among nondepository institutions for HMDA reporting by removing the 100-loan threshold applicable to nondepository institutions in the existing rule. Traditionally, nondepository institutions have been subject to less scrutiny by regulators than depository institutions, and little is known about the mortgage lending behavior of nondepository institutions that fall below the current reporting thresholds. By illuminating this part of the mortgage market, the final rule will provide regulators, public officials, and members of the public with important information. For example, it is possible that small nondepository institutions are serving particular market segments or populations that would benefit from more oversight by public officials and community groups. This oversight can be enhanced only if more information is revealed about the segments, and the change in institutional coverage in the final rule is designed to fill this vacuum. To the extent that such increased data and transparency enhances social welfare, consumers served by these nondepository institutions will benefit. Similarly, expanding coverage among nondepository institutions could improve the processes used to identify possible discriminatory lending patterns and enforce antidiscrimination statutes. Financial regulators and enforcement agencies use HMDA data in their initial prioritization and screening processes to select institutions for examination or investigation. HMDA data also provide information that is used in fair lending reviews of mortgage lenders for potential violations of antidiscrimination statutes, including ECOA and the Fair Housing Act. This is especially true for redlining analyses, which compare lending patterns across lenders within given markets. Current deficiencies in HMDA's institutional coverage leave gaps in the data used by regulators for conducting fair lending prioritization and redlining analyses to compare lenders or markets. Because many depository and nondepository institutions with similar loan volumes are similar in other respects, excluding some nondepository institutions with fewer than 100 loans may weaken the understanding of markets needed for prioritization and redlining analyses. Consequently, increased reporting among nondepository institutions may increase the ability to identify fair lending risk. The final rule will also improve the ability to determine whether financial institutions are serving the housing needs of their communities. Information from data sources such as the United States Census, Call Reports, and the NMLSR can be used to help identify the housing needs of the communities that lenders serve. HMDA data provide a supply-side picture of how well each financial institution is meeting these housing needs. Indeed, HMDA data may be analogized to a census of mortgage demand and supply for covered financial institutions. However, such data currently paints only a partial picture of the market served by financial institutions with 25 to 99 closed-end mortgage loans. The addition of nondepository institutions with between 25 and 99 closed- end mortgage loan originations will provide an improved understanding of the mortgage markets where these financial institutions operate, thereby enhancing efforts to assess whether these institutions, and financial institutions overall, are serving the housing needs of their communities. Costs to consumers. The revised threshold will not impose any direct costs on consumers. Consumers may bear some indirect costs if financial institutions that will be required to report under the final rule pass on some or all of their costs to consumers. Following standard microeconomic [[Page 66276]] principles, the Bureau believes that these institutions will pass on increased variable costs to future mortgage applicants but will absorb start-up costs, one-time costs, and increased fixed costs if financial institutions are profit maximizers and the market is perfectly competitive.\478\ --------------------------------------------------------------------------- \478\ If markets are not perfectly competitive or financial institutions are not profit maximizers, then the costs that a financial institutions may pass on may differ. For example, financial institutions may attempt to pass on one-time costs and increases in fixed costs, or they may not be able to pass on variable costs. --------------------------------------------------------------------------- The Bureau defines variable costs as costs that depend on the number of applications received. Based on initial outreach efforts, the following five operational steps affect variable costs: Transcribing data, resolving reportability questions, transferring data to an HMS, geocoding, and researching questions. The primary impact of the final rule on these operational steps is an increase in time spent per task. Overall, the Bureau estimates that the impact of the final rule on variable costs per closed-end application is approximately $25 for a representative tier 3 financial institution, $0.40 for a representative tier 2 financial institution, and $0.10 for a representative tier 1 financial institution.\479\ The 75-450 nondepository institutions that will now be required to report closed-end mortgage loans and applications have small origination volumes, so the Bureau expects most of them to be tier 3 financial institutions. Hence, based on microeconomics principles, the Bureau expects that a representative nondepository financial institution affected by this final rule will pass on to mortgage borrowers costs of approximately $25 per application. This expense will be amortized over the life of the loan and represents a negligible increase in the cost of a mortgage loan. Therefore, the Bureau does not anticipate any material adverse effect on credit access in the long or short term even if the additional nondepository institutions that must begin reporting pass on these costs to consumers. --------------------------------------------------------------------------- \479\ These cost estimates do not incorporate the impact of operational improvements. Incorporating these additional operational changes will reduce the estimated impact on variable costs. Therefore, the estimates we provided are upper bound estimates of the increase in variable costs that financial institutions will pass on to consumers. These estimates of the impact of the final rule on variable costs per application show the combined impact of all components of the final rule and therefore differ from estimates of the impact on variable costs presented below, which show the impact of specific components of the final rule. In addition, these estimates focus only on the variable-cost tasks, while other estimates incorporate both variable- and fixed-cost tasks. --------------------------------------------------------------------------- During the Small Business Review Panel process, some small entity representatives noted that they would attempt to pass on all increased compliance costs associated with the proposed rule, but that whether these costs were passed on would depend on the competiveness of the market in which they operate, especially for smaller financial institutions. In addition, some small entity representatives noted that they would attempt to pass on costs through higher fees on other products, leave geographic or product markets, or spend less time on customer service. Many industry commenters echoed similar sentiments that the proposal would likely increase the cost of credit for consumers. A few commenters noted that small financial institutions in general would be required to merge, raise prices, make fewer loans, or exit markets. To the extent that the market is less than perfectly competitive and financial institutions are able to pass on a greater amount of these compliance costs, the cost to consumers will be slightly larger than the estimates described above. Even so, the Bureau believes that the potential costs that will be passed on to consumers are small. The final rule may impose additional costs on consumers as well. Reducing the number of depository institutions required to report will reduce HMDA's overall coverage of the mortgage market. This reduction will reduce the usefulness of HMDA data for assessing whether lenders are meeting the housing needs of their communities and highlighting opportunities for public and private investment. This reduction may also affect the usefulness of HMDA for identifying possible discriminatory lending patterns--especially for redlining analyses, which focus on market-level data and data on competitors. To better understand these potential costs, the Bureau analyzed the characteristics of the depository institutions that would be excluded from reporting closed-end mortgage loans by the 25-loan threshold, and compared these characteristics to depository institutions that currently report and would not be excluded. This type of analysis is possible because the final rule reduces both the number of closed-end reporting depository institutions and the closed-end mortgage loans that they report, and the total universe reported under the current regulation is known. For this exercise, the Bureau excluded purchased loans from its comparisons. Overall, the Bureau found that, relative to depository institutions that will continue to report under the final rule (i.e., reporting depositories), applications for closed-end mortgage loans at excluded depository institutions were more likely to be (1) made to the depository institutions supervised by the FDIC or NCUA (over 42 and 41 percent, respectively, compared to 13.74 percent and 10.21 percent at reporting depositories); (2) second-lien (over 9 percent, compared to 2.96 percent at reporting depositories); (3) home improvement (over 23 percent, compared to 6.83 percent at reporting depositories); \480\ (4) non-owner-occupied (over 22 percent, compared to 11.86 at reporting depositories); (5) manufactured housing or multifamily (slightly less than 4 and 5 percent, respectively, compared to 1.83 percent and 0.42 percent at reporting depositories); (6) portfolio loans (approximately 88 percent, compared to roughly 33 percent at reporting depositories); and (7) higher-priced (nearly 13 percent, compared to 2.92 percent at reporting depositories). To the extent that these excluded loans are different from those that remain and these loans serve a somewhat different group of consumers that are more disadvantaged, the loss of those records will impose a cost on this group of consumers as less information may be available to the government, community groups, and researchers to serve their unique needs. --------------------------------------------------------------------------- \480\ These totals include applications for both secured and non-dwelling-secured home improvement loans, even though non- dwelling-secured home improvement loans will not be reported under the final rule. To the extent that excluded depository institutions engage in more non-dwelling-secured home improvement lending than reporting depositories, these numbers will overestimate the difference in reportable home improvement applications by the two types of institutions under the final rule. --------------------------------------------------------------------------- Excluding small-volume depository institutions currently reporting under HMDA also impacts the volume of records available for analysis at the market level. The geographic data fields currently in the HMDA data provide four possible market levels: State, MSA, county, and census tract. Overall, analysis of these markets shows that for most markets, a small percentage of loan/application register records would be lost by excluding small-volume depository institutions for closed-end mortgage loan reporting.\481\ But the lost records are more likely to be in certain States, territories, and MSAs. The percentage excluded is greater than 1 percent for Alaska and Puerto Rico, which showed the highest percentage of [[Page 66277]] excluded records at 1.93 percent and 7.32 percent, respectively. Ranked by the percentage of loan/application register records that would be excluded for each MSA, the 75th percentile was 0.35 percent, suggesting that for 75 percent of MSAs, excluding small depository institutions would exclude less than 0.35 percent of total loan/application register records. The 95th percentile was 1.05 percent, suggesting that for 5 percent of MSAs, excluding small depository institutions would exclude more than 1.05 percent of total loan/application register records. The five MSAs with the most excluded records were all in Puerto Rico. Census tracts have smaller loan volumes than States and MSAs, so the variation in percentages is naturally expected to be higher. Ranked by the percentage of loan/application register records that would be excluded, the 75th and 95th percentiles for census tracts were 0.47 percent and 2.65 percent, respectively. To the extent that government, community groups, and researchers rely on HMDA data relevant to these particular markets to further social goals, the loss of this information will impose a cost on the consumers in these markets. --------------------------------------------------------------------------- \481\ This analysis includes purchased loans. --------------------------------------------------------------------------- Benefits to covered persons. The final rule will provide some cost savings to depository institutions that will be excluded under the revised closed-end mortgage loan-volume threshold. The Bureau estimated 1,400 depository institutions will be excluded from reporting closed- end mortgage loans and applications under the closed-end reporting threshold in the final rule. The Bureau also believes that these 1,400 depository institutions most likely would not be subject to open-end reporting under the open-end reporting threshold. Therefore, these depository institutions will no longer incur current operational costs associated with gathering and reporting HMDA data. The Bureau expects most of these depository institutions to be tier 3 financial institutions, given the small volume of home purchase, refinance, and home improvement mortgages they originate. The Bureau estimates that the current annual operational costs of reporting under HMDA are approximately $2,500 for representative tier 3 financial institutions with a loan/application register size of 50 records. This translates into a market-level benefit of approximately $3,500,000 (= $2,500 * 1,400) per year. Using a 7 percent discount rate, the net present value of this impact savings over five years is approximately $14,400,000.\482\ --------------------------------------------------------------------------- \482\ Note that the figures above refer to cost savings by the newly excluded small-volume depository institutions, assuming costs based on the current Regulation C reporting system. With the changes in the final rule, along with the operational improvements that the Bureau is making, the impact of the final rule on operational costs will be approximately $1,900 per year for a representative tier 3 financial institution. This translates into a market-level savings of approximately $2,660,000 (= $1,900 * 1,400) per year. Using a 7 percent discount rate, the net present value of this savings over five years is approximately $10,900,00. --------------------------------------------------------------------------- In addition to avoiding ongoing costs, the 1,400 excluded depository institutions will not incur the one-time costs necessary to modify processes in response to the final rule. The Bureau estimates that these one-time costs from reporting closed-end mortgage loans are, on average, $3,000 for tier 3 financial institutions. Assuming that all 1,400 depository institutions are tier 3 institutions, this yields an overall market savings of $4,200,000. Using a 7 percent discount rate and a five-year amortization window, the annualized one-time savings is approximately $17,200,000. One-time costs to covered persons. The estimated additional 75-450 nondepository institutions that will have to report closed-end mortgage loans under the final rule will incur start-up costs to develop policies and procedures, infrastructure, and training. Given the relatively small origination volume by these nondepository institutions, the Bureau expects most of them to be tier 3 financial institutions. Based on outreach discussions with financial institutions prior to the proposal, the Bureau believes that these start-up costs for closed-end reporting will be approximately $25,000 for tier 3 financial institutions.\483\ This yields an overall market cost of approximately $11,250,000 (= 450 * $25,000). Using a 7 percent discount rate and a five-year amortization window, the annualized start-up cost is $46,100,000. --------------------------------------------------------------------------- \483\ Note this start-up cost differs from the one-time cost presented previously, because the one-time cost mostly involves the costs from modifying an existing reporting system for an existing reporter, while the startup cost is the cost incurred from building an entirely new reporting system for a new HMDA reporter. --------------------------------------------------------------------------- The estimated 24 financial institutions meeting the open-end reporting threshold but falling below the closed-end reporting threshold will incur one-time costs from building reporting systems, including developing policies and procedures, infrastructure, and training for reporting open-end lines of credit.\484\ The Bureau has estimated that these one-time costs will be approximately $3,000 for low-complexity financial institutions, $250,000 for moderate-complexity financial institutions, and $800,000 for high-complexity financial institutions. The Bureau assumes 12 of these institutions are tier 3 institutions and 12 are tier 2 institutions. This yields an overall one-time cost of approximately $3,000,000. Using a 7 percent discount rate and a five-year amortization window, the annualized one-time cost is approximately $740,000 per year. --------------------------------------------------------------------------- \484\ The Bureau estimates that these open-end-only reporters are not technically new HMDA reporters in the sense that they previously would have been reporting under HMDA because they are depository institutions that have closed-end mortgage loan/ application register sizes between 1 and 24. Therefore, the Bureau believes they will be able to repurpose and modify the existing HMDA reporting process for open-end reporting. The Bureau estimates none of these open-end-only reporters will be high-complexity financial institutions. --------------------------------------------------------------------------- Ongoing costs to covered persons. The estimated 75-450 nondepository institutions that will have to report closed-end mortgage loans under the final rule will incur the operational costs of gathering and reporting data. Including both current operational costs and the impact of the final rule, the Bureau estimates that these operational costs will total approximately $5,100 for a representative tier 3 financial institution per year, without incorporating the Bureau's operational improvements. This yields an overall market impact of approximately $2,300,000 (= 450 * $5,100). Using a 7 percent discount rate, the net present value of this cost over five years is approximately $9,400,000. With operational improvements, the Bureau estimates that these operational costs will total approximately $4,400 for a representative tier 3 financial institution per year. This yields an overall market impact of approximately $2,000,000 (= 450 * $4,400). Using a 7 percent discount rate, the net present value of this cost over five years is approximately $8,100,000. The estimated 24 depository institutions that will have to report open-end lines of credit under the final rule but not closed-end mortgage loans will incur the operational costs of gathering and reporting data for open-end lines of credit. The Bureau estimates that the operational costs for depository institutions will total approximately $8,600 per year for a representative tier 3 open-end reporter and $43,400 per year for a representative tier 2 open-end reporter, and assumes current operational cost is equal to zero for open-end reporting. Assuming 12 of these 24 financial institutions are tier 3 open-end reporters [[Page 66278]] and the rest are tier 2 open-end reporters, this yields an overall market impact of approximately $620,000 (= 12 * $8,600 + 12 * $43,400). Using a 7 percent discount rate, the net present value of this cost over five years is approximately $2,600,000. These estimates incorporate all of the Bureau's operational improvements. Alternatives considered. Regarding closed-end mortgage loans, the Bureau considered several reporting thresholds higher than 25 loans. The Bureau sought to exclude financial institutions whose data are of limited value in the HMDA dataset, thus ensuring that the institutional coverage criteria do not impair HMDA's ability to achieve its purposes, while also minimizing the burden for financial institutions. Specifically, these alternative thresholds were evaluated according to the extent to which they balanced several important factors, including simplifying the reporting regime by establishing a uniform loan-volume threshold applicable to both depository and nondepository institutions; eliminating the burden of reporting from low-volume depository institutions while maintaining sufficient data for analysis at the national, local, and institutional levels; and increasing visibility into the home mortgage lending practices of nondepository institutions. Table 5, below, provides estimates of the coverage among depository institutions at various closed-end reporting thresholds. Table 6 provides estimates of the loss of HMDA data for certain geographic markets. Table 7 provides estimates of the coverage among nondepository institutions at various closed-end reporting thresholds. BILLING CODE 4810-AM-P [[Page 66279]] [GRAPHIC] [TIFF OMITTED] TR28OC15.005 BILLING CODE 4810-AM-C [[Page 66280]] [GRAPHIC] [TIFF OMITTED] TR28OC15.006 The Bureau believes that a threshold of 25 closed-end mortgage loans reduces burden on small depository institutions while preserving important data about communities and improving visibility into the lending practices of nondepository institutions. As shown above in Table 5, the 25-loan threshold will achieve a significant reduction in burden by eliminating reporting by more than 20 percent of depository institutions that are currently reporting. As described in greater detail throughout this discussion, the Bureau estimates that the most significant driver of costs under HMDA is the requirement to report, rather than any specific aspect of reporting, such as the number or complexity of required data fields or the number of entries. For example, the estimated annual ongoing costs of reporting closed-end mortgage loans under the final rule are estimated to be approximately $4,400 for a representative tier 3 financial institution, accounting for the Bureau's operational improvements. About $2,300 of this annual ongoing cost is comprised of fixed costs. As a comparison, each required data field accounts for approximately $42 of this annual ongoing cost. Thus, a threshold of 25 closed-end mortgage loans provides a meaningful reduction in burden by reducing the number of depository institution reporters. Higher thresholds would further reduce burden but would produce data losses that would undermine the benefits provided by HMDA data. One of the most substantial impacts of any low loan-volume threshold is that it reduces information about lending at the community level, including information about vulnerable consumers and the origination activities of smaller lenders. Public officials, community advocates, and researchers rely on HMDA data to analyze access to credit at the neighborhood level and to target programs to assist underserved communities and consumers. For example, Lawrence, Massachusetts identified a need for homebuyer counseling and education based on HMDA data, which showed a high percentage of high-cost loans in the area compared to surrounding communities.\485\ Similarly, HMDA data helped bring to light discriminatory lending patterns in Chicago neighborhoods, resulting in a large discriminatory lending settlement.\486\ In addition, researchers and consumer advocates analyze HMDA data at the census-tract level to identify patterns of discrimination at a national level.\487\ Higher closed-end loan-volume thresholds would eliminate data about more communities and consumers. At a closed-end reporting threshold of 100, according to 2013 HMDA data, the number of census tracts that would lose 20 percent of reported data would increase by almost eight times over the number under a closed-end reporting threshold of 25 loans. The number of affected low- to-moderate-income tracts would increase six times over the number at the 25-loan level. --------------------------------------------------------------------------- \485\ See City of Lawrence, Massachusetts, HUD Consolidated Plan 2010-2015, at 68 (2010), http://www.cityoflawrence.com/Data/Sites/1/documents/cd/Lawrence_Consolidated_Plan_Final.pdf. Similarly, in 2008 the City of Albuquerque used HMDA data to characterize neighborhoods as ``stable,'' ``prone to gentrification,'' or ``prone to disinvestment'' for purposes of determining the most effective use of housing grants. See City of Albuquerque, Five Year Consolidated Housing Plan and Workforce Housing Plan 100 (2008), available at http://www.cabq.gov/family/documents/ConsolidatedWorkforceHousingPlan20082012final.pdf. As another example, Antioch, California, monitors HMDA data, reviews it when selecting financial institutions for contracts and participation in local programs, and supports home purchase programs targeted to households purchasing homes in Census Tracts with low loan origination rates based on HMDA data. See City of Antioch, California, Fiscal Year 2012-2013 Action Plan 29 (2012), http://www.ci.antioch.ca.us/CitySvcs/CDBGdocs/Action%20Plan%20FY12-13.pdf. See, e.g., Dara D. Mendez et al., Institutional Racism and Pregnancy Health: Using Home Mortgage Disclosure Act Data to Develop an Index for Mortgage Discrimination at the Community Level, 126 Pub. Health Reports (1974-) Supp. 3, 102-114 (Sept./Oct. 2011) (using HMDA data to analyze discrimination against pregnant women in redlined neighborhoods), available at http://www.publichealthreports.org/issueopen.cfm?articleID=2732. \486\ See, e.g., Yana Kunichoff, Lisa Madigan Credits Reporter with Initiating Largest Discriminatory Lending Settlements in U.S. History (June 14, 2013), http://www.chicagonow.com/chicago-muckrakers/2013/06/lisa-madigan-credits-reporter-with-initiating-largest-discriminatory-lending-settlements-in-u-s-history/ (``During our ongoing litigation . . . the Chicago Reporter study looking at the HMDA data for the City of Chicago came out . . . It was such a startling statistic that I said . . . we have to investigate, we have to find out if this is true . . . We did an analysis of that data that substantiated what the Reporter had already found . . . [W]e ultimately resolved those two lawsuits. They are the largest fair-lending settlements in our nation's history.'') \487\ See, e.g., California Reinvestment Coalition, et al., Paying More for the American Dream VI: Racial Disparities in FHA/VA Lending, at http://www.woodstockinst.org/research/paying-more-american-dream-vi-racial-disparities-fhava-lending. Likewise, researchers have analyzed GSE purchases in census tracts designated as underserved by HUD using HMDA data. James E. Pearce, Fannie Mae and Freddie Mac Mortgage Purchases in Low-Income and High-Minority Neighborhoods: 1994-96, Cityscape: A Journal of Policy Development and Research (2001), available at http://www.huduser.org/periodicals/cityscpe/vol5num3/pearce.pdf. --------------------------------------------------------------------------- [[Page 66281]] The Bureau also believes that it is important to increase visibility into nondepository institutions' activity given the lack of available data about lower-volume nondepository institutions' mortgage lending practices. A uniform closed-end reporting threshold of fewer than 100 loans annually will expand nondepository institution coverage, because the current test requires reporting by all nondepository institutions that meet the other applicable criteria and originate 100 loans annually.\488\ Any closed-end reporting threshold set at 100 loans would not provide any enhanced insight into nondepository institution lending, and a threshold above 100 closed-end mortgage loans would decrease visibility into nondepository institutions' practices and hamper the ability of HMDA users to monitor risks posed to consumers by those institutions. The threshold of 25 closed-end mortgage loans, however, achieves a significant expansion of nondepository institution coverage, with up to a 40 percent increase in the number of reporting institutions. --------------------------------------------------------------------------- \488\ In addition, nondepository institutions that originate fewer than 100 applicable loans annually are required to report if they have assets of at least $10 million and meet the other criteria. See 12 CFR 1003.2 (definition of financial institution). --------------------------------------------------------------------------- The Bureau's proposal did not include an open-end line of credit threshold for institutional coverage. Under the Bureau's proposal, an institution that met the 25 closed-end mortgage loan threshold (and the other criteria for institutional coverage) would have been required to report all of its open-end lines of credit, even if its open-end lending volume was very low. On the other hand, institutions that did not meet the 25 closed-end mortgage loan threshold but that had significant open-end lending volume would not have been HMDA reporters. As noted, the Bureau received a large number of comments expressing concerns related to the burden of reporting under this threshold. In response to these concerns and in an attempt to reduce reporting by financial institutions that have originated at least 25 closed-end mortgage loans but only a very small number of open-end lines of credit, the final rule adopts a separate open-end reporting threshold. A financial institution will be required to report open-end lines of credit only if its open-end origination volume exceeds this threshold. When setting this separate threshold, the Bureau considered several alternatives to the final threshold of 100 open-end lines of credit. In doing so, the Bureau sought to exclude financial institutions whose data are of limited value while ensuring that the institutional coverage criteria for mandatory reporting of open-end lines of credit do not impair HMDA's ability to achieve its purposes. Specifically, these alternative thresholds were evaluated according to the extent to which they balanced several important factors, including limiting the number of open-end reporters in general, limiting the number of small- volume open-end reporters whose data are of limited use in particular, and limiting the number of open-end reporters that would not have reported closed-end mortgage loans under HMDA, while maintaining sufficient data for analysis with adequate market coverage. Table 8, below, provides estimates of the coverage among depository institutions at various open-end reporting thresholds. It is the Bureau's belief that most nondepository institutions do not originate dwelling-secured open-end lines of credit. The Bureau notes that no single data source accurately reports the number of originations of open-end lines of credit, as that term is defined in the final rule. The Bureau had to use multiple data sources, including credit union Call Reports, Call Reports for banks and thrifts, HMDA data, and Consumer Credit Panel data, in order to develop estimates about open- end originations for currently reporting depository institutions.\489\ --------------------------------------------------------------------------- \489\ For this exercise, the Bureau limits its analysis to current HMDA reporters, because it believes that those depository institutions would be the ones who would have met all other HMDA reporting requirements, such as location and asset tests, as well as origination of at least one home purchase loan or refinancing of a home purchase loan, secured by a first lien on a one- to four-unit dwelling. In general, credit union Call Reports provide the number of originations of open-end lines of credit secured by real estate but exclude lines of credit in the first lien status and may have included business loans that will be excluded from HMDA reporting according to the final rule. Call Reports for banks and thrifts report only the balance of the home-equity lines of credit at the end of reporting period but not the number of originations in the period. [GRAPHIC] [TIFF OMITTED] TR28OC15.007 The first row under the heading corresponds to the estimated coverage under the proposed rule where any financial institution that satisfied the proposed 25-closed-end mortgage loan threshold \490\ would have reported open-end lines of credit. The other rows correspond to various other thresholds the Bureau considered for an independent open-end reporting threshold. --------------------------------------------------------------------------- \490\ For this analysis, the Bureau has not considered reverse mortgages that are structured as open-end lines of credit. Reverse mortgages cannot be identified within the current HMDA data. It is the Bureau's belief that most reverse mortgages currently are not reported under HMDA. --------------------------------------------------------------------------- The Bureau believes that a threshold of 100 open-end lines of credit reduces burden on financial institutions while preserving important coverage and visibility into the market for dwelling-secured lines of credit. As shown above [[Page 66282]] in Table 8, compared to the proposal, the open-end reporting threshold reduces the number of open-end reporters by almost 3,400, while reducing the market coverage by only about 6 percent. Other thresholds may have more imbalanced effects on either reporting burden or market coverage. For example, at a threshold of 25 open-end lines of credit, the projected market coverage by reporting institutions will only increase by 5 percent compared to the coverage level at a threshold of 100 open-end lines of credit, but almost 1,000 additional institutions would be burdened by reporting requirements. On the other hand, while a threshold of 1,000 open-end lines of credit would substantially reduce the number of reporting institutions, it would only cover about two- thirds of the total market. It is also worth noting that, at a threshold of 100 open-end lines of credit, almost all open-end reporters will also report closed-end mortgage loans.\491\ The Bureau believes that sharing of reporting and compliance resources within the same financial institution for both closed-end and open-end reporting will help reduce reporting costs. --------------------------------------------------------------------------- \491\ Note that, while the Bureau estimates there will be 24 financial institutions that will report open-end lines of credit but not report closed-end mortgage loans, that number (24) is well within the margin of error and thus may be close to zero due to the uncertainty of the raw estimation. --------------------------------------------------------------------------- The Bureau also considered exempting certain small financial institutions, such as those defined as ``small entities'' as described in part VIII, below, from the reporting requirements of the final rule. As described above, however, excluding small financial institutions would undermine both the utility of HMDA data for analysis at the local level and the benefits that HMDA provides to communities. Thus, removing these institutions would deprive users of important data about communities and vulnerable consumers. Finally, the Bureau considered a tiered reporting regime under which smaller financial institutions would be exempt from reporting some or all of the data points not identified by the Dodd-Frank Act. Tiered reporting would preserve some information about availability of credit in particular communities and to vulnerable consumers while relieving some burden. Tiered reporting presents a number of problems, however. First, because under a tiered reporting regime smaller financial institutions would not report all or some of the HMDA data points, tiered reporting would prevent communities and users of HMDA data from learning important information about the lending and underwriting practices of smaller financial institutions, which may differ from those of larger institutions. Second, as discussed above, the primary driver of HMDA costs is establishing and maintaining systems to collect and report data, not the costs associated with collecting and reporting a particular data field. Therefore, tiered reporting would reduce the costs of low-volume depository institutions somewhat, but not significantly. 4. The Scope of the Transactional Coverage of the Final Rule The final rule requires financial institutions generally to report all dwelling-secured, consumer-purpose closed-end loans and open-end lines of credit, as well as commercial-purpose loans and lines of credit made for home purchase, home improvement, or refinancing purposes.\492\ The final rule eliminates home improvement loans not secured by a dwelling from the reporting requirements, while consumer- purpose closed-end mortgage loans, open-end lines of credit, and reverse mortgages will now be reported regardless of whether they were for home purchase, home improvement, or refinancing. Commercial-purpose closed-end loans will continue to be reported only if the purpose is for home purchase, home improvement, or refinancing. Commercial-purpose open-end lines of credit with home purchase, home improvement, or refinancing purposes must now be reported. Finally, for preapproval requests that are approved but not accepted, reporting will change from optional to mandatory. --------------------------------------------------------------------------- \492\ A financial institution reports data on dwelling-secured, closed-end mortgage loans only if it originated at least 25 closed- end mortgage loans in each of the two preceding calendar years and also met all the other reporting criteria. Similarly, a financial institution reports data on dwelling-secured, open-end lines of credit only if it originated at least 100 open-end lines of credit in each of the two preceding calendar years and also met all the other reporting criteria. --------------------------------------------------------------------------- Benefits to consumers. The revisions to Regulation C's transactional coverage will benefit consumers by providing a more complete picture of the dwelling-secured lending market. The additional transactions required to be reported will improve market monitoring, and will potentially aid in identifying and tempering future financial crises. Using open-end lines of credit and closed-end home-equity loans as an example, in the lead up to the financial crisis between 2000 and 2008, the balance of home-equity lending increased by approximately 16.8 percent annually, moving from $275.5 billion to $953.5 billion in total.\493\ Various researchers have pointed out that rapidly expanding lending activities in home-equity lines of credit and home-equity loans contributed to the housing bubble as borrowers and lenders both vigorously took on high leverage. Additional research has shown that the growth in home-equity lending was correlated with subsequent home price depreciation, as well as high default and foreclosure rates among first mortgages.\494\ Researchers have argued that these correlations were driven in part by consumers using open-end lines of credit to fund investment properties, which impacted default rates when housing prices began to fall. Researchers have also shown evidence that distressed homeowners with closed-end subordinate-lien mortgage loans encountered several challenges when seeking assistance from public and private mortgage relief programs.\495\ Data on these loans might have helped public officials improve the effectiveness of these relief programs. However, because HMDA does not currently cover all home-equity loans, and most financial institutions choose not to report home-equity lines of credit, this substantial market is almost completely missing from the HMDA data. Based on information from HUD and Moody's Analytics (May 2013), HMDA data currently include only approximately 1 percent of all open-end lines of credit and 35 percent of closed-end home-equity loan originations. Data identifying the presence and purpose of home-equity lending may enable government, industry, and the public to avert similar scenarios in the future. --------------------------------------------------------------------------- \493\ Michael LaCour-Little et al., The Role of Home Equity Lending in the Recent Mortgage Crisis, 42 Real Estate Economics 153 (2014). \494\ See Atif Mian & Amir Sufi, House Prices, Home Equity-Based Borrowing, and the U.S. Household Leverage Crisis, 101 Am. Econ. Rev. 2132, 2154 (Aug. 2011); Donghoon Lee et al., Fed. Reserve Bank of New York, Staff Report No. 569, A New Look at Second Liens, at 11 (Aug. 2012); Michael LaCour-Little et al., The Role of Home Equity Lending in the Recent Mortgage Crisis, 42 Real Estate Economics 153 (2014). \495\ See Vicki Been et al., Furman Ctr. for Real Estate and Urban Policy, Essay: Sticky Seconds--The Problems Second Liens Pose to the Resolution of Distressed Mortgages, at 13-18 (Aug. 2012). --------------------------------------------------------------------------- Changes to transactional coverage will also improve the ability of government, researchers, and community groups to determine whether financial institutions are serving the housing needs of their communities. Home equity has long been the most important form of household savings and consumers often resort to tapping their home equity for various purposes. The optional reporting of open-end lines of credit, and limited coverage of closed-end [[Page 66283]] home-equity lending and reverse mortgages under the current Regulation C, provide an incomplete picture of whether financial institutions are serving the housing needs of their communities. The changes to transactional coverage will significantly close this gap. Additionally, the changes to transactional coverage in the final rule will benefit consumers by improving fair lending analyses. Regulators, community groups, and researchers use HMDA data to identify disparities in mortgage lending based on race, ethnicity, and sex. These analyses are used for prioritization and scoping purposes to select the institutions, and parts of institutions, to review. As discussed above, a substantial amount of open-end lines of credit and closed-end home-equity loans are not reported. The extent of reverse mortgage reporting under HMDA is unknown because the existing data provide no way to distinguish reverse mortgages from other loans, but the Bureau believes that a substantial number of reverse mortgages are not reported. Because a substantial amount of these transactions are not reported, it is not possible during prioritization analyses to develop a clear assessment of the fair lending risk to consumers of these specific products. In addition, all of these products may have unique underwriting and pricing guidelines that would merit separate analyses. It is not currently possible to identify these products in HMDA, however, so most fair lending analyses that use HMDA data combine these products and other products with potentially different underwriting and pricing standards. These shortcomings reduce the reliability of risk assessment analyses, limiting the ability to identify consumers that might have been subjected to illegal discrimination. Requiring reporting of all reverse mortgages also benefits consumers through improved fair lending analysis focused on age discrimination. Reverse mortgages are a special mortgage product designed to satisfy the later-life consumption needs of seniors by leveraging their home equity while permitting them to maintain homeownership. During its 2013 fiscal year, HUD endorsed 60,091 home- equity conversion mortgages (HECMs), which counted for almost all of the reverse mortgage market. Various stakeholders and advocates have called for better data about the reverse mortgage market based on concerns about potential abuse of vulnerable seniors. Mandatory reporting of reverse mortgages will provide public officials, community organizations, and members of the public with more information to assist consumers age 62 or older. This change is consistent with Congress's decision to include age as a new data point in the Dodd- Frank Act, which the Bureau believes signaled an intention to strengthen protections for seniors. Mandatory reporting of preapproval requests that are approved but not accepted will also benefit consumers through improved fair lending analyses. Currently, data about preapproval requests that are approved but not accepted are optionally reported. Thus, these data are largely absent from the HMDA data that regulators and community groups analyze. Including these preapproval requests will improve fair lending analyses by providing for a more accurate comparison between those applications that satisfy a financial institution's underwriting criteria and those that are reported as either originated or approved but not accepted, and those that are reported as denials. The changes to transactional coverage in the final rule also improve the ability of public officials to distribute public-sector investment so as to attract private investment to areas where it is needed. HMDA data provide a broadly representative picture of home lending in the nation unavailable from any other data source. Open-end lines of credit and closed-end home-equity loans are important forms of lending that are considered in evaluations under the CRA. Expanded reporting of open-end lines of credit, closed-end home-equity loans, and reverse mortgages will improve HMDA's coverage of mortgage markets, which in turn will enhance the HMDA data's usefulness in identifying areas in need of public and private investment and thereby benefit consumers. Finally, expanded reporting of home-equity lending will reduce the chance of regulatory gaming by financial institutions. To the extent that open-end lines of credit and closed-end home-equity loans are largely interchangeable for customers applying for credit for a given purpose, lenders could, under current Regulation C reporting requirements, intentionally recommend consumer-purpose open-end lines of credit as substitutes for closed-end home-equity loans to avoid reporting of home-equity loans. Expanded reporting of both closed-end home-equity loans and open-end lines of credit will mitigate such misaligned incentives and ultimately benefit consumers by closing the data reporting gap. Costs to consumers. The final rule eliminates reporting of home improvement loans not secured by a dwelling (i.e., whether unsecured or secured by non-dwelling collateral), which reduces the data available to analysts. This, in turn, imposes a cost on consumers. The Bureau estimates that financial institutions reported approximately 340,000 non-dwelling-secured home-improvement loans under HMDA during 2013. This comprised 2.4 percent of the total record volume. Under the final rule, regulators, community groups, and researchers will not be able to use HMDA data to assess fair lending risks for this product, which will reduce the likelihood of identifying consumers who are potentially disadvantaged when taking out non-dwelling-secured home improvement loans. In addition, it is possible that the general loss of data may negatively affect research in other unexpected ways and thus negatively impact consumers. However, commenters did not state that they or others have used HMDA data about non-dwelling-secured home-improvement loans to further HMDA's purposes, and the Bureau does not believe HMDA data on such loans is widely used for those purposes. The increased transactional coverage will not impose any direct costs on consumers. However, consumers may bear some indirect costs of increased transactional coverage if financial institutions pass on some or all of the costs imposed on them by reporting additional transactions. Following microeconomic principles, the Bureau believes that financial institutions will absorb one-time costs and increased fixed costs but will pass on increased variable costs to future mortgage applicants. The Bureau estimates that the final rule's changes to transactional coverage regarding open-end lines of credit will increase variable costs per open-end line of credit application by approximately $41.50 for a representative tier 3 open-end reporter, $6.20 for a representative tier 2 open-end reporter, and $3 for a representative tier 1 open-end reporter.\496\ Thus, the Bureau expects that a representative tier 3 financial institution covered by the final rule will pass on to borrowers of open-end lines of credit $41.50 per [[Page 66284]] application; a representative tier 2 financial institution will pass on $6.20 per open-end application; and a representative tier 1 financial institution will pass on $3 per open-end application. This expense will be amortized over the life of the loan and represents a negligible increase in the cost of a mortgage loan. Therefore, the Bureau does not anticipate a material adverse effect on credit access in the long or short term if financial institutions pass on to consumers the costs of reporting open-end lines of credit under the transactional coverage adopted in the final rule. --------------------------------------------------------------------------- \496\ These cost estimates incorporate all the required data fields in the final rule and the operational improvements the Bureau is developing. This differs from cost impacts regarding data points presented in part VII.F.5, which normally isolate one change by, for example, not counting operational improvements. This is because the Bureau assumes that the overwhelming majority of open-end-line-of- credit reporting will be new and hence the baseline cost would be zero and the number of data fields as well as operational details in the baseline scenarios for open-end reporting would be inapplicable. --------------------------------------------------------------------------- During the Small Business Review Panel process, some small entity representatives noted that they would attempt to pass on all increased compliance costs associated with the proposed rule, but that this would be difficult in the current market where profit margins for mortgages are tight, especially for smaller financial institutions. In addition, some small entity representatives noted that they would attempt to pass on costs through higher fees on other products offered, leave geographic or product markets, or spend less time on customer service. Similarly, several industry commenters stated that the rule would increase costs to consumers or force small financial institutions to consider merging, raising prices, originating fewer loans, or exiting the market. As discussed above, the Bureau believes that any costs passed on to consumers will be amortized over the life of a loan and represent a negligible increase in the cost of a mortgage loan. Therefore, the Bureau does not anticipate any material adverse effect on credit access in the long or short term even if financial institutions pass on these costs to consumers. Benefits to covered persons. The final rule eliminates reporting of non-dwelling-secured home improvement loans, which will reduce costs to covered persons. Using HMDA data, as well as information from interviews of financial institutions, the Bureau estimates that each year, on average, tier 3, tier 2, and tier 1 financial institutions receive approximately 1, 20, and 900 applications for non-dwelling- secured home improvement loans, respectively. Excluding those average numbers of non-dwelling-secured home improvement loans from reporting will reduce annual operational costs by approximately $43 for a representative tier 3 financial institution, $128 for a representative tier 2 financial institution, and $2,740 for a representative tier 1 financial institution.\497\ This translates into a market-level savings of approximately $1,090,000 to $1,150,000 per year. Using a 7 percent discount rate, the net present value of this impact over five years will be a reduction in cost of approximately $4,500,000 to $4,700,000. --------------------------------------------------------------------------- \497\ These estimates do not include potential cost savings from operational improvements. --------------------------------------------------------------------------- The final rule's expanded transactional coverage will improve the prioritization process used to identify institutions at higher risk of fair lending violations. This will reduce the false positives that occur when inadequate information causes lenders with low fair lending risk to be initially misidentified as high risk. Additional information on these products will explain some of these false positives, so that examination resources are used more efficiently and that lenders with low fair lending risk receive a reduced level of regulatory scrutiny. One-time costs to covered persons. The Bureau believes that the greatest one-time cost to covered persons from the final rule's changes to transactional coverage will come from the requirement to report open-end lines of credit. Based on outreach efforts and comments received, the Bureau believes that many financial institutions process applications for open-end lines of credit on separate data platforms and data systems in different business units than home-purchase and refinance mortgages. Financial institutions not currently reporting open-end lines of credit will incur one-time costs to develop reporting capabilities for these business lines and products. Financial institutions, whether they use vendors for HMDA compliance or develop software internally, will incur one-time costs to prepare, develop, implement, integrate, troubleshoot, and test new systems for open-end reporting. Management, operations, legal, and compliance personnel in these business lines will likely require time to learn the new reporting requirements and to assess legal and compliance risks. Financial institutions will need to update training materials to reflect new requirements and may incur certain one-time costs for providing initial training to current employees. The Bureau is aware that these activities will take time and that the costs may be sensitive to the time available for them. The Bureau also believes that financial institutions that will report both open-end lines of credit and closed-end mortgage loans, which comprise the overwhelming majority of open-end reporters, could share one-time costs related to open-end and closed-end reporting. The degree of such cost sharing likely will vary based on operational complexities. The Bureau expects these one-time costs to be smaller for financial institutions that are less complex and less likely to have separate business lines with separate data platforms and data systems for open- end lines of credit. These entities use less complex reporting processes, so more tasks are manual rather than automated, and new requirements may involve greater use of established processes. As a result, compliance will likely require straightforward changes in systems and workplace practices and therefore impose relatively low one-time costs. In estimating the impact of the transactional coverage changes for representative tier 3 open-end reporters that will also report closed-end mortgage loans, the Bureau assumes that the one-time cost of open-end reporting is minimal and already absorbed into the one-time cost of closed-end reporting because most of these straightforward changes would have occurred anyway due to the modified closed-end reporting requirements. For representative tier 3 open-end reporters that will not report closed-end mortgage loans, because the one-time cost from open-end reporting cannot be absorbed into the one- time costs of closed-end reporting, the Bureau believes that such costs can be proxied by the overall estimate of the one-time costs that the tier 3 closed-end reporters will incur, absent expanded reporting of open-end lines of credit. Thus, the Bureau estimates that the changes to transactional coverage in the final rule will impose average one- time costs of $3,000 for tier 3 open-end reporters. For more complex financial institutions that meet the open-end reporting threshold, the Bureau expects the one-time costs imposed by the change in transactional coverage in the final rule to be relatively large. To estimate these one-time costs, the Bureau views the business lines responsible for open-end lines of credit in moderate-to-high complexity institutions as a second business line that has to modify its reporting infrastructure in response to the final rule. Industry stated this view of additional costs in comments on the proposed rule. However, very few financial institutions or trade associations provided the Bureau with specific estimates of the one-time cost associated with this change. In outreach conducted before the proposed rule, some industry participants generally stated that the one-time cost of reporting open-end lines of credit could be twice [[Page 66285]] as much as the one-time cost of adapting to other parts of the final rule, but did not provide any further detail. One commenter stated that the Bureau's estimated one-time implementation costs for moderate- complexity financial institutions were potentially correct. The Bureau estimates that, excluding open-end-line-of-credit reporting, the final rule will impose average one-time costs of $250,000 for tier 2 financial institutions and $800,000 for tier 1 financial institutions. The Bureau assumes that for tier 1 and tier 2 open-end reporters that will also report closed-end mortgage loans, which form the majority of the projected open-end reporting tier 1 and tier 2 institutions, the one-time cost of integrating open-end lines of credit into HMDA reporting processes will be roughly equal to 50 percent of the one-time costs absent expanded reporting of such products. This estimate accounts for the fact that some new systems may have to be built to facilitate reporting for these lines of business but that some fixed, one-time costs could be shared with lines of business currently subject to Regulation C, because both have to undergo systemic changes. Using these general estimates for open-end reporting tier 1 and tier 2 institutions that will also report closed-end mortgage loans, therefore, the Bureau estimates one-time costs of $125,000 and $400,000 for business lines responsible for open-end lines of credit. On the other hand, for representative tier 2 open-end reporters that will not report closed-end mortgage loans, because such cost sharing between open-end and closed-end reporting is not possible, the Bureau proxies for the one-time costs associated with open-end reporting by using the overall estimate of the one-time costs that the tier 2 closed-end reporter will incur in response to the final rule absent expanded reporting of open-end lines of credit. Thus, the Bureau estimates that the changes to transactional coverage in the final rule will impose average one-time costs of $250,000 for tier 2 open-end reporters that will not report closed-end mortgage loans under the final rule. The Bureau does not project any tier 1 financial institutions that will report open-end lines of credit but not closed- end mortgage loans under the final rule. Under the final rule, the open-end reporting threshold is set separately from the closed-end reporting threshold. A financial institution can report open-end lines of credit only, closed-end mortgage loans only, or both. For open-end reporters, the Bureau estimates that 749 financial institutions will meet the threshold for reporting data on open-end lines of credit, including 24 that will report open-end lines of credit only but not closed-end mortgage loans and 725 that will report open-end and closed-end simultaneously. Coupled with the fact that lenders often process open-end lines of credit in business lines separate from closed-end mortgage loans, for the purpose of transactional and institutional coverage analyses, the Bureau has adopted an approach that treats these open-end reporters as if they were separate entities distinct from their closed-end mortgage units.\498\ --------------------------------------------------------------------------- \498\ The Bureau estimates that under the final rule almost all open-end reporters would have some business activity in closed- mortgage arena, even if a handful of them will not be reporting closed-end mortgage loans under the final rule due to their low closed-end mortgage origination volume (below 25 but greater than zero). --------------------------------------------------------------------------- As with closed-end mortgage loan reporting, the Bureau realizes that costs for open-end reporting vary by institutions due to many factors, such as size, operational structure, and product complexity, and that this variance exists on a continuum that is impossible to fully represent. Nevertheless, the Bureau believes that the HMDA reporting process and ongoing operational cost structure for open-end reporters will be fundamentally similar to closed-end reporting. To conduct a cost consideration that is both practical and meaningful for open-end reporting, the Bureau therefore adopts the same three-tier approach and most of the key assumptions used for closed-end reporting, with two modifications. First, for representative low-complexity open- end reporters, the Bureau assumed that the number of open-end line of credit applications would be 150. This was set to both accommodate the open-end reporting threshold of 100 open-end lines of credit and to reflect a reasonable distribution among the smallest open-end reporters, based on the Bureau's estimated number of likely open-end reporters and their volumes. Second, for representative high-complexity open-end reporters, the Bureau assumed that the number of open-end line of credit applications would be 30,000. This reflects a reasonable distribution among the largest open-end lines of credit based on the Bureau's estimated number of likely open-end reporters and their volumes. The Bureau assumed that the number of open-end line of credit applications for the representative moderate-complexity open-end reporter would still be 1,000, just as for the moderate-complexity closed-end reporter. For open-end reporters, the Bureau has adopted 2 cutoffs based on the estimated open-end line of credit volume. Specifically, the Bureau assumes the lenders that originate fewer than 200 but more than 100 open-end lines of credit are tier 3 (low-complexity) open-end reporters; lenders that originate between 200 and 7,000 open-lines of credit are tier 2 (moderate-complexity) open-end reporters; and lenders that originate more than 7,000 open-end lines of credit are tier 1 (high-complexity) open-end reporters. These cutoffs were chosen to match the overall market size in terms of the estimated number of open- end reporters (724) and the estimated number of records (approximately 900,000). Under such assumptions, the Bureau assigns 13 of the possible open-end reporters to tier 1, 463 to tier 2, and 273 to tier 3. Roughly 2 percent of these institutions are in tier 1, 62 percent are in tier 2, and 36 percent are in tier 3. This is close to the high-end distribution of closed-end reporters in which 3 percent are in tier 1, 71 percent are in tier 2, and 26 percent are in tier 3. Dividing open- end-only reporters from open-end reporters that will also report closed-end mortgage loans, the Bureau estimates that among 24 likely reporters that will report only open-end lines of credit, there are 12 tier 2 open-end reporters, 12 tier 3 open-end reporters, and no tier 1 open-end reporters; among 725 likely reporters that will report both open-end lines of credit and closed-end mortgage loans, there are 13 tier 1 open-end reporters, 451 tier 2 open-end reporters, and 261 tier 3 open-end reporters. The baseline cost assumptions and cost estimates presented above reflect the current world in which most open-end lines of credit are not reported under HMDA. In the final rule, reporting open-end lines of credit becomes mandatory for those institutions that meet all the other criteria for a ``financial institution'' in final Sec. 1003.2(g) and originate at least 100 open-end lines of credit. The Bureau estimated that currently only about 1 percent of total open-end lines of credit secured by dwellings were reported under HMDA. Hence the Bureau has assumed that the baseline cost for open-end-line-of-credit reporting in the current rule is zero. By using the one-time cost estimates due to open-end reporting for representative open-end reporters that are in different tiers and that either report only open-end lines of credit or both open-end lines of credit and closed-end mortgage loans, multiplied by the number of open- end reporters of each corresponding type, the Bureau estimates that the total one-time cost due to open-end reporting for open-end reporters that will report both open-end [[Page 66286]] lines of credit and closed-end mortgage loans is approximately $61,600,000 (that is: Tier 1 $400,000 * 13 + Tier 2 $125,000 * 451 + Tier 3 $0 * 261); the total one-time cost due to open-end reporting for open-end reporters that will report only open-end lines of credit is approximately $3,000,000 (that is: Tier 1 $400,000 * 0 + Tier 2 $250,000 * 12 + Tier 3 $3,000 * 12). Combined, the one-time costs due to open-end reporting for all open-end reporters are estimated to be approximately $64,600,000. Using a 7 percent discount rate and a five- year amortization window, the annualized one-time cost due to changes in transactional coverage is approximately $15,800,000 per year. As a frame of reference for these market-level, one-time cost estimates due to open-end reporting, the total non-interest expenses of current HMDA reporters were approximately $420 billion in 2012. The one-time cost estimate of $64,600,000 is about 0.15 percent of the total annual non- interest expenses.\499\ Because these costs are one-time investments, financial institutions are expected to amortize these costs over a period of years. --------------------------------------------------------------------------- \499\ The Bureau estimated the total non-interest expense for banks, thrifts and credit unions that reported to HMDA based on Call Report data for depository institutions and credit unions, and NMLSR data for nondepository institutions, all matched with 2012 HMDA reporters. --------------------------------------------------------------------------- For mandatory reporting of preapproval requests that are approved but not accepted, the Bureau believes that the primary impact will be on ongoing operational costs rather than on one-time costs. Financial institutions are currently required to report whether a preapproval was requested for home purchase loans, and whether the preapproval was approved (if accepted) or denied, so the infrastructure to report preapproval information is already in place. Expanding mandatory reporting to all outcomes of the preapproval process therefore primarily impacts the ongoing, operational tasks required to gather information and data on additional reportable transactions. Ongoing costs to covered persons. The changes to transactional coverage in the final rule will require financial institutions that meet the open-end threshold and other criteria to report open-end lines of credit, thereby increasing the ongoing operational costs of those financial institutions for HMDA reporting. As stated above, for the purpose of transactional coverage analyses, the Bureau treats these open-end reporters as if they were separate entities distinct from their closed-end mortgage units. The Bureau assumes that the operational costs of open-end reporting vary across 3 different open- end reporting complexity tiers, but whether an open-end reporter also reports closed-end mortgage loans does not affect its operational costs on the open-end side. The Bureau estimates that for a representative tier 1 open-end reporter with 30,000 open-end loan/application register records, the ongoing operational cost of open-end reporting is about $273,000 per year, or approximately $9 per record per year. For a representative tier 2 open-end reporter with 1,000 open-end loan/ application register records, the ongoing operational cost of open-end reporting is about $43,400 per year, or approximately $43 per record per year. For a representative tier 3 open-end reporter with 150 open- end loan/application register records, the ongoing operational cost of open-end reporting is about $8,600 per year, or approximately $57 per record per year. Based on information from HUD and Moody's Analytics (May 2013), HMDA data currently include only approximately 1 percent of all open-end lines of credit. Therefore, the Bureau assumes that the ongoing operational cost associated with open-end reporting is practically zero. Therefore, the estimated ongoing operational costs for open-end reporting under the final rule represent the entire impact on operational costs due to the open-end transactional coverage change. These cost estimates incorporate all the required data fields in the final rule and the Bureau's operational improvements. Based on the estimate that 13 open-end reporters are in tier 1, 463 are in tier 2, and 273 are in tier 3, the Bureau estimates that the total impact on ongoing operational costs due to open-end reporting is approximately $26,000,000 per year ($273,000 * 13 + $43,400 * 463 + $8,600 * 273). Using a 7 percent discount rate, the net present value of this cost over five years is approximately $106,600,000. The final rule also modifies transactional coverage by requiring reporting of closed-end home-equity loans, reverse mortgages, and preapproval requests that have been approved but not accepted. To estimate the impact on ongoing operational costs due to these changes, the Bureau allocates these transactions among the three representative closed-end lenders proportionately to the lender's loan/application register size. The Bureau estimated that, on average, tier 3 financial institutions with 50 records receive approximately one application for closed-end home-equity loans; no applications for reverse mortgages; and no preapproval requests that were approved but not accepted. The Bureau estimated that, on average, tier 2 financial institutions with 1,000 records receive an estimated 15 applications for closed-end home- equity loans; no applications for reverse mortgages; and five preapproval requests that were approved but not accepted. And the Bureau estimated that, on average, tier 1 financial institutions with 50,000 records receive an estimated 700 applications for closed-end home-equity loans; five applications for reverse mortgages; and 245 preapproval requests that were approved but not accepted. Reporting data for these additional loans will increase operational costs by approximately $43, $128, and $2,890 per year for representative tier 3, tier 2, and tier 1 financial institutions, respectively, without accounting for operational improvements. Using the two tier distributions discussed previously, this translates into a market-level cost of approximately $1,130,000 to $1,180,000 per year. Using a 7 percent discount rate, the net present value of this cost over five years is approximately $4,600,000 to $4,800,000. Considering operational improvements, operational costs will increase by approximately $42, $125, and $2,880 per year, for the representative entities in tier 3, tier 2, and tier 1, respectively. This translates into a market-level cost of approximately $1,120,000 to $1,160,000 per year. Using a 7 percent discount rate, the net present value of this cost over five years is approximately $4,600,000 to $4,800,000. Alternatives considered. The Bureau considered excluding preapprovals from reporting requirements. Based on a review of historical HMDA data, the Bureau estimates that on average tier 3 financial institutions receive one request for a preapproval per year, tier 2 financial institutions receive 15 requests per year, and tier 1 financial institutions receive 700 requests per year. The estimated reduction in the operational cost of reporting data for these preapprovals is approximately $43, $96, and $2,100 per year, for representative tier 3, tier 2, and tier 1 financial institutions, respectively, without accounting for savings from operational improvements. This translates into a market-level impact of approximately $880,000 to $890,000 per year. Using a 7 percent discount rate, the net present value of this savings over five years is approximately $3,600,000 to $3,700,000. Including the operational improvements reduces the estimated operational costs of reporting data for preapprovals by approximately $41, $94, and $2,100 per year for [[Page 66287]] representative tier 3, tier 2, and tier 1 financial institutions, respectively. This translates into a market-level savings of approximately $870,000 to $880,000 per year. Using a 7 percent discount rate, the net present value of this savings over five years is $3,560,000 to $3,610,000. 5. The Data That Financial Institutions are Required to Report About Each Loan or Application For each application, originated loan, or purchased loan submitted as part of a financial institution's loan/application register, Regulation C currently requires reporting of 35 separate pieces of information, and allows for optional reporting of three denial reasons.\500\ Throughout this part VII.F.5, the Bureau uses the term ``data point'' to refer to each piece of information to be reported and ``data field'' to refer to the actual entries on the loan/application register necessary to report the required data points. For example, currently race is one data point with ten data fields (five for primary applicant race and five for co-applicant race). The Dodd-Frank Act amended HMDA by enhancing two existing data points (rate spread and application ID) and identifying 11 new data points.\501\ As part of this rulemaking, the Bureau comprehensively reviewed all current data points in Regulation C, carefully examined each data point specifically mentioned in the Dodd-Frank Act, and considered proposals to collect other appropriate data points to fill gaps where additional information could be useful to better understand the HMDA data.\502\ --------------------------------------------------------------------------- \500\ The 35 pieces of information are respondent ID, agency code, application number, application date, loan type, property type, purpose, occupancy, loan amount, preapprovals, action, action date, MSA, State, county, census tract, applicant ethnicity, applicant sex, five applicant race data fields, co-applicant ethnicity, co-applicant sex, five co-applicant race data fields, income, purchaser, rate spread, HOEPA status, and lien status. \501\ These 11 data points consist of total points and fees, prepayment penalty term, introductory interest rate term, non- amortizing features, loan term, application channel, loan originator ID, property value, parcel number, age, and credit score. \502\ A financial institution's loan/application register is also accompanied by a transmittal sheet that contains data about the submission, such as the number of entries, the address of the financial institution, and the appropriate Federal agency. The final rule does not change these requirements, except that financial institutions that report data quarterly will identify the relevant quarter and year, and the reporter's identification number is being replaced by the Legal Entity Identifier, discussed below. --------------------------------------------------------------------------- The revisions include improvements and technical revisions to current Regulation C data requirements; the implementation as required or appropriate of the categories of information specifically identified in the Dodd-Frank Act; and the addition of other data points that fill existing informational gaps and will further the purposes of HMDA. One important consideration during the Bureau's rulemaking process that informs this discussion of benefits, costs, and impacts was alignment of data fields to existing regulations or industry data standards. In order to develop this alignment, the Bureau analyzed each data point currently included in Regulation C, each new data point identified in the Dodd- Frank Act, and each additional data point the Bureau considered during the rulemaking process, to determine whether analogous data existed in the Uniform Loan Delivery Dataset (ULDD) (first preference) or the larger Mortgage Industry Standards Maintenance Organization (MISMO) data dictionary (second preference). In each instance, before the Bureau considered aligning to one of these external data standards, the MISMO/ULDD definition needed to be adequate to meet the objectives of HMDA and Regulation C. In some instances, even when analogous data existed in ULDD or MISMO, the Bureau decided to adopt data point definitions different than ULDD or MISMO when other considerations outweighed the benefit of alignment. For data points that could not be aligned with MISMO/ULDD, the Bureau aligned these data points with definitions provided by other regulations if appropriate, or used completely new definitions. Current HMDA data points. Currently, financial institutions are required to collect and report information for 35 data fields, and have the option of reporting three additional fields conveying denial reasons. Considering only the current 35 mandatory fields, the final rule will increase the number of required fields by 12. Reporting of denial reasons is changing from optional to mandatory and reporters will have the option of reporting four denial reasons instead of three. This change will add four required data fields. A fifth additional data field captures number of total units, which along with construction method is replacing property type, as the current ``property type'' data field will be replaced by two fields (number of units and construction method), both of which are in MISMO and ULDD. Disaggregation of ethnicity increases the total number of ethnicity data fields that are reportable by eight, from two to ten. Currently, applicants and co-applicants each choose either Hispanic/Latino or not Hispanic/Latino. Going forward, applicants and co-applicants will continue to have the option of choosing Hispanic/Latino or not Hispanic/Latino, but will also have the option of choosing Mexican, Puerto Rican, Cuban, or Other Hispanic/Latino. Applicants will not be limited on the number of ethnic groups they can choose, and HMDA reporters must report all ethnicities applicants report. Therefore, both the primary applicant and co-applicant can choose up to five ethnicities, for a total of ten data fields, or a net increase of eight data fields. On the other hand, disaggregation of race will not increase the total number of race data fields, because the final rule limits the total number of race fields that can be reported for each applicant/co-applicant to five, the same as the current level. Specifically, currently applicants and co-applicants can each choose up to five racial groups (American Indian or Alaska Native, Asian, Black or African American, Native Hawaiian or Other Pacific Islander, and White). Going forward, the list that applicants and co-applicants can choose from will be expanded to include Asian Indian, Chinese, Filipino, Japanese, Korean, Vietnamese, Other Asian, Native Hawaiian, Guamanian or Chamorro, Samoan, or Other Pacific Islander. Finally, financial institutions will no longer have to report MSA/MD, because these data can be easily obtained from information already provided about the relevant State and county. Adding 13 data fields and losing one yields a net increase of 12 data fields. In addition to adding 12 data fields, the final rule will also change the information reported for 19 current HMDA data fields. These revisions address changes required by the Dodd-Frank Act, align current HMDA fields with industry data standards, and close information gaps. Specifically, to address changes required by the Dodd-Frank Act, the financial institution's identifier will be replaced by a Legal Entity Identifier, application ID will be replaced by a unique, robust ID number, and rate spread will be required for most covered loans subject to Regulation Z. Occupancy will be revised to convey principal residence, second residence, or investment property, and property type will be replaced by number of total units and construction method. Finally, to close information gaps, loan amount will be reported in dollars instead of thousands of dollars; additional ``other'' and ``cash-out refinance'' categories will be added to loan purpose; and the current ethnicity [[Page 66288]] and race fields will contain more granular ethnicity and racial categories. Current HMDA data points--benefits to consumers. The Bureau believes that the revisions to the current HMDA data fields, which increase the amount of information included in HMDA, will improve current processes used to identify possible discriminatory lending patterns and enforce antidiscrimination statutes. The following discussion provides several examples of how the revised existing data fields will ultimately benefit consumers by facilitating enhanced fair lending analyses. The section-by-section analyses in part V, above, provide more detailed exposition on each of the enhanced data points. As one example, the reason for denial is an important data point used to understand underwriting decisions and focus fair lending reviews. Currently, Regulation C permits optional reporting of the reasons for denial of a loan application. Mandatory reporting of this information, combined with enhanced or additional data points commonly used to make underwriting decisions, will provide more consistent and meaningful data. These improved data can improve the ability to identify both discriminatory lending patterns in underwriting decisions and consumers who have been unfairly disadvantaged. In addition, denial reasons, combined with careful analysis of key underwriting data fields, could help reduce the false positive rate of fair lending prioritization analyses, leading to better targeting of fair lending reviews. This will further improve the likelihood of identifying customers who were truly unfairly disadvantaged and merit restitution. Additionally, rate spread is currently the only quantitative pricing measure in HMDA, and is only available for originated loans meeting or exceeding the higher-priced mortgage loan thresholds for first- and subordinate-lien loans. Expanding reporting of rate spread to all covered loans subject to Regulation Z, except assumptions, purchased loans and reverse mortgage transactions, greatly enhances HMDA's usefulness for analyzing fair lending risk in pricing decisions. This change will also reduce the false positive rate observed during fair lending prioritization analyses so that the resources of regulators and financial institutions are used more efficiently. Together with additional pricing measures included in the final rule, this information will also greatly enhance the understanding of the costs of credit that consumers face. The disaggregated racial and ethnic categories will provide meaningful data for advancing HMDA's purposes. In particular, a significant benefit of disaggregated HMDA data is that it could allow non-regulators, such as researchers and community groups, the opportunity to augment the fair lending work that regulatory agencies conduct. These groups could focus on areas and risks that regulatory agencies may not choose to examine. The revisions to the occupancy and property type data fields provide a fourth example of benefit for fair lending analyses. The final rule revises data regarding occupancy status by requiring separate itemization of second residences and investment properties, and revises data regarding property type by replacing this field with construction method and the number of units. These revisions will allow more accurate accounting of the differences in underwriting and pricing policies that financial institutions apply. This will improve analyses of outcomes and hence reduce false positive rates in current fair lending prioritization processes used by regulatory agencies. Improved prioritization will further improve the likelihood of identifying customers who were truly unfairly disadvantaged and merit restitution. The Bureau also believes that the revisions to the current HMDA data fields, which increase the amount of information included in the HMDA dataset, will improve the ability to assess whether financial institutions are meeting the housing needs of their communities and assist public officials in making decisions about public-sector investments. The denial reason data fields will provide greater understanding of why credit is denied to specific applicants, the expanded rate spread data point will provide additional information about the affordability of the credit offered, and the revised occupancy and property type data fields will provide additional insight into more detailed property and product markets. Additionally, the revisions to the occupancy status data field will provide finer gradients by separately identifying second homes and investment properties, which will help identify trends involving potentially speculative purchases of housing units similar to those that contributed to the recent financial crisis. Recent research suggests that speculative purchases by investors were one driver of the recent housing bubble and subsequent financial crisis.\503\ These impacts may be especially relevant for areas that are experiencing sharp increases in investor purchases. Thus, information related to second homes and investment properties may help communities and local officials develop policies tailored to the unique characteristics associated with these separate segments of the mortgage market. --------------------------------------------------------------------------- \503\ See Andrew Haughwout et al., Fed. Reserve Bank of New York, Staff Report No. 514, Real Estate Investors, the Leverage Cycle, and the Housing Market Crisis, (Sept. 2011). --------------------------------------------------------------------------- Finally, revisions to the property type data field will be of particular interest in the wake of the housing crisis as families have increasingly turned to rental housing. Greater detail about multifamily housing finance may provide additional information about whether financial institutions are serving the housing needs of their communities. Current HMDA data points--costs to consumers. The revisions to the current HMDA data fields will not impose any direct costs on consumers. However, consumers may bear some indirect costs if financial institutions pass on some or all of the costs imposed on them by the final rule. Following microeconomic principles, the Bureau believes that financial institutions will pass on increased variable costs to future mortgage applicants but will absorb one-time costs and increased fixed costs if markets are perfectly competitive and financial institutions are profit maximizers. The impact of the changes in the final rule to the 19 current HMDA data fields will affect only one-time costs and fixed costs, as financial institutions modify their infrastructure to incorporate the final data field specifications. The revision to current HMDA data fields that impacts variable cost is the net addition of 12 data fields. To estimate the impact on variable cost of a net increase of 12 additional data fields, the Bureau treated the four denial reason data fields as new data fields, the additional property type field as a new data field that aligns with MISMO/ULDD, the 8 additional ethnicity fields as new data fields, and the MSA/MD data field as an existing data field to be dropped that aligns with MISMO/ULDD. The Bureau estimates that the impact of this component of the final rule on variable costs per application is approximately $10 for a representative tier 3 financial institution, $0.31 for a representative tier 2 financial institution, and $0.03 for a representative tier 1 financial institution.\504\ This expense will be [[Page 66289]] amortized over the life of the loan and represents a negligible increase in the cost of a mortgage loan. Therefore, the Bureau does not anticipate any material adverse effect on credit access in the long or short term if financial institutions pass on these costs to consumers. --------------------------------------------------------------------------- \504\ These estimates are for financial institutions that meet the threshold for reporting closed-end mortgage loans, but not for reporting of open-end lines of credit or quarterly reporting. --------------------------------------------------------------------------- During the Small Business Review Panel process, some small entity representatives noted that they would attempt to pass on all increased compliance costs associated with the final rule, but that this would be difficult in the current market where profit margins for mortgages are tight. In addition, some small entity representatives noted that they would attempt to pass on costs through higher fees on other products offered, leave geographic or product markets, or spend less time on customer service. Many comments to the proposed rule echoed similar sentiments that the proposal would likely increase the cost of credit for consumers. Several commenters cited increased costs associated with reporting additional data fields. A few commenters noted that small financial institutions in general would be required to merge, raise prices, originate fewer loans, or exit markets. As discussed above, the Bureau believes that any costs passed on to consumers will be amortized over the life of a loan and represent a negligible increase in the cost of a mortgage loan. Therefore, the Bureau does not anticipate any material adverse effect on credit access in the long or short term even if financial institutions pass on these costs to consumers. Current HMDA data points--benefits to covered persons. One primary benefit of the revisions to the current HMDA data points in the final rule is the improved alignment between the HMDA data standards and the data standards that many financial institutions already maintain.\505\ For example, the current HMDA definitions for occupancy status and property type are not directly compatible with the records of mortgage loan applications that most financial institutions store in their loan origination systems. This may have created extra burden on the financial institutions that had to use additional software to modify data in existing systems in order to record and submit HMDA data. --------------------------------------------------------------------------- \505\ The final rule eliminates required reporting of the MSA/MD data field. Although the exclusion of this data field creates a benefit to covered persons, it is not considered explicitly here, because on net, the revisions to current HMDA fields in the final rule add 12 data fields. --------------------------------------------------------------------------- The Bureau believes that aligning the requirements of Regulation C to existing industry standards for collecting and transmitting data on mortgage loans and applications will reduce the burden associated with Regulation C compliance and data submission for some institutions. In addition, promoting consistent data standards for both industry and regulatory use has benefits for market efficiency, market understanding, and market oversight. The efficiencies achieved by such alignment should grow over time, as the industry moves toward common data standard platforms. For example, many financial institutions already separately identify second residence and investment properties in their underwriting process and loan origination system (LOS). Separate enumeration of these occupancy types is also present in MISMO/ULDD. Therefore, aligning to industry standards will reduce burden for financial institutions by maintaining the same definition for HMDA reporting that financial institutions use in the ordinary course of business. Smaller, less-complex financial institutions will experience fewer potential benefits, because these institutions rely more on manual reporting processes and are more likely to originate portfolio loans where MISMO/ULDD may have not been adopted. Among current HMDA data fields, property type and occupancy will be modified to align with MISMO/ULDD. The primary benefit of this alignment will be to reduce costs for training and researching questions. The Bureau estimates that this alignment will reduce operational costs by approximately $120, $1,100, and $10,200 per year for representative tier 3, 2, and 1 financial institutions, respectively.\506\ This translates into a market-level impact of $5,700,000 to $7,900,000 per year. Using a 7 percent discount rate, the net present value of this savings over five years is $23,300,000 to $32,200,000. With the inclusion of operational improvements, the estimated reduction in operational costs is approximately $120, $1,000, and $10,100 per year for representative tier 3, tier 2, and tier 1 financial institutions, respectively.\507\ This translates into a market-level savings of $5,600,000 to $7,700,000 per year. The net present value of this savings over five years is $23,000,000 to $31,700,000. --------------------------------------------------------------------------- \506\ These estimates are for financial institutions that meet the threshold for reporting closed-end mortgage loans, but not for reporting of open-end lines of credit or quarterly reporting, and do not include potential cost savings from operational improvements and additional help sources. \507\ These estimates are for financial institutions that meet the threshold for reporting closed-end mortgage loans, but not for reporting of open-end lines of credit or quarterly reporting. --------------------------------------------------------------------------- Current HMDA data points--ongoing costs to covered persons. Specific to the current set of HMDA data points, the final rule increases the number of data fields by 12 on net, and alters the information provided for 19 other fields. The cost impact of these changes on covered persons will vary by data field. For example, some data fields may depend on multiple sub-components or information from multiple platforms. To capture these potential differences, the Bureau estimated different costs depending on whether a data field is aligned with ULDD, MISMO, another regulation, or is a completely new data field. The four denial reason fields are new data fields not aligned with MISMO, ULDD or another regulation; number of units, which along with construction method replaces property type, is aligned with ULDD; the eight additional ethnicity data fields are not aligned with MISMO, ULDD or another regulation; and MSA/MD, which is being excluded, is also aligned with ULDD.\508\ This net increase of 12 data fields increases the costs of transcribing data, transferring data to HMS, researching questions, checking post-submission edits, training, exam assistance, conducting annual edits/checks, and conducting external audits. The Bureau estimates that this component of the final rule will increase operational costs by approximately $460, $3,100, and $8,000 per year for representative tier 3, tier 2, and tier 1 financial institutions, respectively.\509\ --------------------------------------------------------------------------- \508\ Although some institutions are required by their regulator to report denial reasons, Regulation C does not currently require reporting of denial reasons, so the Bureau treated these data fields as new data fields. The cost estimates discussed below are adjusted to reflect that some institutions already report these data fields. \509\ These estimates are for financial institutions that meet the threshold for reporting closed-end mortgage loans, but not for reporting of open-end lines of credit or quarterly reporting, and do not include potential cost savings from operational improvements and additional help sources. --------------------------------------------------------------------------- Number of units will be a new data field that all financial institutions will be required to report, and MSA/MD is an existing data field that will no longer be required. Although the three current denial reasons are considered new data fields, operationally, they will only be new data fields for reporters currently choosing not to report them, or currently not being required by their regulator to report them. In the 2013 HMDA data, approximately 30 percent of HMDA reporters did not provide denial reasons, and approximately 25 percent of all denials did not have data [[Page 66290]] regarding the reason for denial. Further analysis reveals that, compared to other HMDA reporters, HMDA reporters currently providing data regarding denial reasons had larger loan/application registers and reported almost twice as many denials. Therefore, requiring mandatory reporting of denial reasons will only impact about 30 percent of reporters, and these reporters will likely be smaller institutions. The additional denial reason and the eight additional ethnicity data fields are all new data fields all financial institutions will have to report. Taking all of this into consideration, the Bureau estimates the market- level cost of increasing the number of current HMDA data fields by 12 on net in the final rule to be between $8,900,000 and $15,200,000. Using a 7 percent discount rate, the net present value of the cost increase over five years is $36,500,000 to $62,100,000. Considering operational improvements, the final rule will increase operational costs by approximately $400, $2,100, and $6,500 per year for representative tier 3, tier 2, and tier 1 financial institutions, respectively.\510\ This translates into a market-level cost of between $6,700,000 and $10,800,000. Using a 7 percent discount rate, the net present value over five years will be a cost increase of $27,500,000 to $44,100,000. --------------------------------------------------------------------------- \510\ These estimates are for financial institutions that meet the threshold for reporting closed-end mortgage loans, but not for reporting of open-end lines of credit or quarterly reporting. --------------------------------------------------------------------------- The primary cost impact of modifying 19 existing data fields, two of which align with ULDD, will be the occurrence of one-time costs to modify current reporting policies and procedures, update software systems, and conduct training and planning. These cost impacts will generally be addressed in the discussion of one-time costs below. The one exception is the requirement that financial institutions obtain and report an LEI instead of the current reporter's ID. The Bureau estimates that the one-time cost of acquiring an LEI is approximately $200 with an ongoing cost of approximately $100 per year. This translates into an estimated market-level impact of $1,400,000 in one- time costs and an increase of $720,000 in ongoing costs per year. For one-time costs, using a 7 percent discount rate and five-year amortization window, the annualized cost is $351,000. For ongoing costs, using a 7 percent discount rate, the net present value over five years is an increase in costs of approximately $3,000,000. Current HMDA data points--alternatives considered. Apart from the revisions discussed above, the Bureau considered requiring a detailed enumeration of the subordinate lien category. This change to lien status was included in the proposal because the Bureau believed that more detailed enumeration would provide useful information for analysis and would reduce the reporting burden by making the definition of lien status consistent with MISMO. Following numerous commenters that pointed out that very few loans would have third or higher liens and that more granularity would actually increase rather than reduce reporting burden, the Bureau decided to maintain the definition of lien status currently in HMDA. To the extent that changes were adopted for any individual current data point, the costs and benefits of that decision are addressed in the section-by-section analysis of the relevant provision above. New HMDA data points. The final rule requires financial institutions to report 50 additional data fields under HMDA. This number does not include unique loan ID, rate spread, number of units, or construction method, each of which replaces a data field currently reported under HMDA. The Dodd-Frank Act explicitly identified 13 additional data points. Excluding unique loan ID and rate spread, which replace data fields currently reported under HMDA, the remaining 11 Dodd-Frank Act-identified data points translate into 22 new data fields financial institutions will have to report on their loan/application registers. To fill information and data gaps, the Bureau is adopting 13 data points, which translates into an additional 28 new data fields financial institutions will have to report on their loan/application register. For these 50 additional data fields, 19 are aligned with ULDD, 12 are aligned with MISMO, and one is aligned with another regulation. The remaining 18 data fields are not in MISMO or ULDD, or aligned with another regulation.\511\ --------------------------------------------------------------------------- \511\ Some data fields were aligned with multiple sources. For the consideration of costs and benefits, the Bureau assigned each data field to one source. The following hierarchy was used for data fields aligned to multiple sources: (1) ULDD, (2) MISMO, (3) another regulation, and (4) not aligned to another source. --------------------------------------------------------------------------- New HMDA data points--benefits to consumers. The additional data points will have several benefits to consumers. First, the additional fields will improve the usefulness of HMDA data for analyzing mortgage markets by regulators and the public. For example, data points such as non-amortizing features, term of introductory interest rate, prepayment penalty term, and the open-end line of credit indicator are related to certain high-risk lending concerns, and reporting this information will enable a better understanding of the types of products and features consumers are receiving. Recent research has indicated that each of these products and product characteristics have increased likelihoods of default and foreclosure and may have exacerbated the recent housing crisis. In addition to being better able to identify some of the risk factors that played a role in the recent financial crisis, the new HMDA data points on pricing and underwriting will improve current research efforts to understand mortgage markets. All of these enhancements will allow for improved monitoring of trends in mortgage markets and help identify and prevent problems that could potentially harm consumers and society overall. Second, the additional data points will help improve current policy efforts designed to address various market failures. As discussed previously, the mortgage market is characterized by information asymmetry, and this inherent deficiency was made apparent during the financial crisis. In response to the recent financial crisis, the government has pursued a number of policies aimed at regulating the market and protecting consumers. The additional data points will help inform future policy-making efforts by improving consideration of the benefits and costs associated with various choices, resulting in more effective policies. As an example, many recent regulations have limited the types of risky mortgage products that lenders can make to borrowers without fully considering borrowers' ability to repay. New data fields on non-amortizing features, term of introductory interest rate, prepayment penalty term, and debt-to-income ratio can assist future assessment of the effectiveness of such regulations and facilitate adjustments when needed. Third, the additional data points will help determine whether financial institutions are serving the housing needs of their communities and help public officials target public investment to better attract private investment. For example, the data points related to manufactured housing will reveal more information about this segment of the market. Borrowers in manufactured housing are typically more financially vulnerable than borrowers in site-built housing and may deserve closer attention from government agencies and community groups. Similarly, the data points related to multifamily dwellings [[Page 66291]] will reveal more information about this segment of the market, which mostly serves low- to moderate-income renters who live in these financed units. Advocacy groups and government agencies have raised concerns over affordability issues faced by individuals living in multifamily dwellings, who also tend to be more financially vulnerable. Overall, by permitting a better and more comprehensive understanding of these markets, the final rule will improve the usefulness of HMDA data for assessing the supply and demand of credit, and financial institutions' treatment of applicants and borrowers in these communities. Fourth, the Bureau believes that the additional data points will improve current processes used to identify possible discriminatory lending patterns and enforce antidiscrimination statutes. Financial regulators and enforcement agencies use HMDA data in their initial prioritization and screening processes to select institutions for examination and as the base dataset during fair lending reviews. The additional data will allow for improved segmentation during these analyses, so that applications are compared to other applications for similar products. For example, underwriting and pricing policies often differ for open-end lines of credit, closed-end home-equity loans, reverse mortgages, and products with different amortization types. Currently, these products are all combined during prioritization and screening analyses. With additional data fields identifying these products, separate analyses can be conducted for each product, which will more accurately reflect outcomes for consumers. As a second example, pricing often differs across delivery channels, because pricing policies and processing differ, and because intermediaries, such as mortgage brokers, add an additional layer to the complexity of mortgage pricing. The addition of the origination channel data point will permit the separation of originations for pricing analyses, allowing for a better understanding of the drivers of pricing outcomes. Improved segmentation improves the accuracy of fair lending analyses, which improves the usefulness of HMDA to identify potentially disadvantaged consumers. Additionally, the new HMDA data points on pricing will greatly improve the usefulness of HMDA data for assessing pricing outcomes during fair lending analyses. Currently, the rate spread data field is the only quantitative pricing measure included in the HMDA data. This data field includes rate spread data only for higher-priced mortgage loans, which currently comprise less than 5 percent of originated loans in the HMDA data. Thus, the usefulness of this data field is highly limited in today's environment, and for the foreseeable future. In addition, mortgage products and pricing structure are inherently complex. The rate spread data are based on the APR. APR alone, though a useful summary measure that is commonly recognizable to borrowers, fails to capture all of the underlying complexities that go into mortgage pricing. Adding discount points, lender credits, and interest rate will provide a much clearer understanding of the trade-offs between rates and points that are the foundation of mortgage pricing. The total loan costs, lender credits, and origination charge data fields will provide a deeper understanding of fees, which form the third component of mortgage pricing. Furthermore, many of the new HMDA data points capture legitimate factors that financial institutions use in underwriting and pricing that are currently lacking in the HMDA data, which will help regulators and government enforcement agencies to better understand disparities in outcomes. Many, if not all, lenders consider data points such as credit score, CLTV, DTI, and AUS results when either underwriting or pricing mortgage applications. The addition of these types of data points will help users understand patterns in underwriting and pricing outcomes and thus better assess the fair lending risk presented by those outcomes. Finally, the addition of the age data field will allow users to analyze outcomes for different age groups during fair lending analyses. Although consumers are protected against discrimination on the basis of age by ECOA and Regulation B, HMDA data currently lack a direct means of measuring the age of applicants. This limits the ability of government agencies and community groups to monitor and enforce violations of ECOA and Regulation B prohibitions against age discrimination in mortgage markets. Older individuals, in particular, are potentially at a higher risk of age discrimination, as well as unfair, deceptive, or abusive acts or practices. These data are especially important as an increased number of baby boomers enter retirement. The addition of the age data field will allow users to identify potential differential treatment of older Americans for various mortgage products. For example, reverse mortgages are designed to serve senior consumers and are priced based on age factors, providing an illustration of the importance of adding this data field to the HMDA data. Age data might also help inform housing policies designed to assist seniors in maintaining or obtaining home ownership, and building or utilizing home equity for improved social welfare. The new HMDA data fields will reduce the false positive rates that occur when inadequate information causes regulators and enforcement agencies to initially misidentify financial institutions with low fair lending risk as having a high risk of fair lending violations. Better alignment between the degrees of regulatory scrutiny and fair lending risk will increase the likelihood of identifying any instances where consumers are being illegally disadvantaged, thereby ultimately benefitting consumers. New HMDA data points--costs to consumers. The addition of 50 data fields will not impose any direct costs on consumers. However, consumers may bear some indirect costs if financial institutions pass on some or all of the costs imposed on them by the final rule. Following microeconomic principles, the Bureau believes that financial institutions will pass on increased variable costs to future mortgage applicants, but will absorb one-time costs and increased fixed costs if markets are perfectly competitive and financial institutions are profit maximizers. The Bureau estimates that the impact of the additional 50 data fields on variable costs per application is approximately $22 for a representative tier 3 financial institution, $0.62 for a representative tier 2 financial institution, and $0.05 for a representative tier 1 financial institution.\512\ This expense will be amortized over the life of the loan and represents a small increase in the cost of a mortgage loan. Therefore, the Bureau does not anticipate any material adverse effect on credit access in the long or short term if financial institutions pass on these costs to consumers. --------------------------------------------------------------------------- \512\ These estimates are for financial institutions that meet the threshold for reporting closed-end mortgage loans, and not for reporting of open-end lines of credit or quarterly reporting. --------------------------------------------------------------------------- During the Small Business Review Panel process, some small entity representatives noted that they would attempt to pass on all increased compliance costs associated with the final rule, but that this would be difficult in the current market where profit margins for mortgages are tight. In addition, some small entity representatives noted that they would attempt to pass on costs through higher [[Page 66292]] fees on other products offered, leave geographic or product markets, or spend less time on customer service. Many comments to the proposed rule echoed similar sentiments that the proposal would likely increase the cost of credit for consumers. As discussed above, the Bureau believes that any costs passed on to consumers will be amortized over the life of a loan and represent a negligible increase in the cost of a mortgage loan. Therefore, the Bureau does not anticipate any material adverse effect on credit access in the long or short term even if financial institutions pass on these costs to consumers. New HMDA data points--benefits to covered persons. The Bureau believes that the additional data points will improve current processes used to identify possible discriminatory lending patterns, which could reduce the burden of financial institutions subject to fair lending examinations or investigations. Financial regulators and enforcement agencies use HMDA data in their initial prioritization and screening processes to select institutions for examination or investigation, and as the base dataset during fair lending reviews. During prioritization analyses, the additional data points will provide information about the legitimate factors used in underwriting and pricing that are currently lacking in the HMDA data, helping government agencies better understand disparities in outcomes. They will also allow for improved segmentation, so that applications are compared to other applications for similar products. Finally, the additional data points on pricing measures will greatly enhance screening analyses of pricing decisions. All of these improvements will reduce false positives resulting from inadequate information. Examination resources will be used more efficiently, so that lenders at low risk of fair lending violations receive a reduced level of regulatory scrutiny. New HMDA data points--one-time costs to covered persons. The new data points included in the final rule will impose one-time costs on HMDA reporters. Management, operations, legal, and compliance personnel will likely require time to learn the new reporting requirements and assess legal and compliance risks. Financial institutions that use vendors for HMDA compliance will incur one-time costs associated with software installation, troubleshooting, and testing. The Bureau is aware that these activities will take time and that the costs may be sensitive to the time available for them. Financial institutions that maintain their own reporting systems will incur one-time costs to develop, prepare, and implement the necessary modifications to those systems. In all cases, financial institutions will need to update training materials to reflect new requirements and may incur certain one-time costs for providing initial training to current employees. The Bureau expects these one-time costs to be relatively small for less complex financial institutions. These entities use less complex reporting processes, so the tasks involved are more manual than automated and new requirements may involve greater use of established processes. As a result, compliance will likely require straightforward changes in systems and workplace practices and therefore impose relatively low one-time costs. The Bureau estimates the additional reporting requirements will impose on average estimated one-time costs of $3,000 for tier 3 financial institutions, $250,000 for tier 2 financial institutions, and $800,000 for tier 1 financial institutions without considering the expansion of transactional coverage to include expanded reporting of open-end lines of credit, closed-end home-equity loans, and reverse mortgages.\513\ Including the estimated one-time costs to modify processes and systems for these expanded reporting requirements, the Bureau estimates that the total one-time costs will be $3,000 for tier 3 institutions, $375,000 for tier 2 institutions, and $1,200,000 for tier 1 institutions. In total, this yields an overall market impact between $725,900,000 and $1,339,100,000. Using a 7 percent discount rate and a five-year amortization window, the annualized one-time cost is $177,000,000 to $326,600,000. As a frame of reference for these market-level, one-time cost estimates, the total non-interest expenses of current HMDA reporters were approximately $420 billion in 2012. The upper bound estimate of $1,339,100,000 is approximately 0.3 percent of the total annual non-interest expenses.\514\ Because these costs are one-time investments, financial institutions are expected to amortize these costs over a period of years. --------------------------------------------------------------------------- \513\ The Bureau realizes that the impact of one-time costs varies by institution due to many factors, such as size, operational structure, and product complexity, and that this variance exists on a continuum that is impossible to fully capture. As a result, the one-time cost estimates will be high for some financial institutions, and low for others. \514\ The Bureau estimated the total non-interest expense for banks, thrifts and credit unions that reported to HMDA based on Call Report and NCUA Call Report data for depository institutions and credit unions, and NMLS data for nondepository institutions, all matched with 2012 HMDA reporters. --------------------------------------------------------------------------- New HMDA data points--ongoing costs to covered persons. The final rule requires financial institutions to report 50 additional data fields. Adding these additional data fields increases the cost of many operational steps required to report data, including transcribing data, transferring data to HMS, conducting annual edits/checks, and conducting external audits. The Bureau estimates that the impact of the additional 50 data fields on annual operational costs is approximately $2,400 for a representative tier 3 financial institution, $15,800 for a representative tier 2 financial institution, and $38,600 for a representative tier 1 financial institution.\515\ This translates into a market-level cost of $54,600,000 to $92,900,000 per year. Using a 7 percent discount rate, the net present value of this cost over five years is $224,000,000 to $381,000,000. Considering operational improvements, the estimated increase in the operational cost of reporting these 50 additional data fields is approximately $2,100, $10,900, and $31,000 per year for representative tier 3, tier 2, and tier 1 financial institutions, respectively. This translates into a market-level cost of $41,000,000 to $66,100,000 per year. The net present value of this impact over five years will be a cost increase of $168,100,000 to $271,100,000. --------------------------------------------------------------------------- \515\ These estimates are for financial institutions that meet the threshold for reporting closed-end mortgage loans, but not for reporting of open-end lines of credit or quarterly reporting. --------------------------------------------------------------------------- New HMDA data points--alternatives considered. To the extent that changes were adopted for any individual data point not identified by the Dodd-Frank Act, the costs and benefits of that decision are addressed in the section-by-section analysis of the relevant provision above. Assessing the regulation as a whole, however, the Bureau considered removing some or all of the discretionary data points. As explained in greater detail in the section-by-section analysis above, the Bureau believes that the final rule balances the benefits of improved data with the burden of reporting. Removing the discretionary data points would deprive communities, researchers, and public officials of important data beneficial to identifying potentially unlawful discriminatory lending patterns, targeting public investment, and determining whether financial institutions are serving the housing needs of their communities. For example, information regarding origination charges, discount points, [[Page 66293]] interest rate, and lender credits will provide a much clearer understanding of the trade-offs between fees, rates, and points that are the foundation of mortgage pricing and the cost of housing transactions. Eliminating the discretionary data points would also increase false positives and inefficiency in evaluating the lending activity of financial institutions. As explained above, many of the additional data points capture factors that financial institutions use in underwriting and pricing that are currently lacking in the HMDA data, such as CLTV, DTI, and AUS results. On the burden side, the primary driver of HMDA costs is establishing and maintaining systems to collect and report data, not the costs associated with collecting and reporting a particular data field. Therefore, removing discretionary data points would cause a significant loss of data that would not be justified by the relatively small reduction in burden. 6. The Modifications to Disclosure and Reporting Requirements The final rule will make several changes to the disclosure and reporting requirements under Regulation C. The first change concerns the modified loan/application register and the disclosure statement that a financial institution must make available to the public. Regulation C currently requires that a financial institution must make its ``modified'' loan/application register available to the public after removing three fields to protect applicant and borrower privacy: The application or loan ID, the date that the application was received, and the date that action was taken. Regulation C also requires financial institutions to make available to the public their disclosure statements, which are a series of tables describing an institution's HMDA data for the previous calendar year. The final rule requires financial institutions to make their modified loan/application registers and disclosure statements available to the public by making available brief notices referring members of the public seeking these data products to the Bureau's Web site to obtain them. Second, the Bureau is requiring that a financial institution that reported for the preceding calendar year at least 60,000 covered loans and applications, excluding purchased covered loans, submit its HMDA data for the first three quarters of the calendar year on a quarterly basis in addition to submitting its HMDA data for the entire calendar year on an annual basis. Based on 2013 HMDA data, 29 financial institutions reported at least 60,000 covered loans and applications, excluding purchased covered loans, in 2013, which comprised approximately 50 percent of the market. Although this estimate does not include the expansion of reporting of open-end lines of credit, the Bureau has determined that the requirement to report these products under the final rule is unlikely to have a significant impact on the number of financial institutions that would be required to report quarterly. Errors or omissions in the data that such financial institutions report on a quarterly basis will not be considered violations of HMDA or Regulation C if the financial institution makes a good-faith effort to report all required data fully and accurately within sixty calendar days after the end of each calendar quarter and corrects or completes the data prior to submitting its annual loan/ application register. Finally, the final rule will eliminate the option for financial institutions with 25 or fewer entries to submit the loan/application register in paper format. Benefits to consumers. The final rule eliminates the option of paper reporting for financial institutions reporting 25 or fewer records, and provides that financial institutions shall make their disclosure statements available to the public through a notice that clearly conveys that the disclosure statement may be obtained on the Bureau's Web site. These provisions will have little direct benefit to most consumers because they do not significantly change the substance, collection, or release of the information required to be reported. However, the requirement that financial institutions make their modified loan/application registers available to the public by making available a brief notice referring members of the public to the Bureau's Web site will generally benefit some consumers. This provision will increase the availability of modified loan/application registers by providing one easily accessible location where members of the public will be able to access all modified loan/application registers for all financial institutions required to report under the statute. Although this benefit is limited somewhat by the fact that the modified loan/ application register is currently available for download in the agencies' release made available on the FFIEC Web site, the agencies' release is typically not available until almost six months after the modified loan/application register must be made available. Quarterly reporting by large volume financial institutions may have a number of benefits to consumers. Currently, there is significant delay between the time that final action is taken on an application and the time information about the application or loan is reported to regulators pursuant to Regulation C. This time delay ranges from two months if the date of final action occurs during December to 14 months if the date of final action occurs during January of the reporting year. The Bureau believes that timelier data will improve the ability of the regulators to identify current trends in mortgage markets, detect early warning signs of future housing finance crises, and determine, in much closer to ``real time,'' whether financial institutions are fulfilling their obligations to serve the housing needs of communities in which they are located, whether opportunities exist for public investment to attract private investment in communities, and whether there are possible discriminatory lending patterns. Also, timelier identification of risks and troublesome trends in mortgage markets by the Bureau and the appropriate agencies will allow for more effective interventions by public officials. Finally, the Bureau intends to release aggregate quarterly data or analysis to the public more frequently than annually, which would improve the ability of members of the public to use the data in a timely manner. Costs to consumers. The adopted changes requiring financial institutions to make their disclosure statements and modified loan/ application registers available to the public by providing brief notices referring members of the public to the Bureau's Web site, to eliminate the option of paper reporting for financial institutions reporting 25 or fewer records, and to require quarterly reporting by financial institutions that reported at least 60,000 covered loans or applications, excluding purchased covered loans, in the preceding year will impose only minimal direct costs on consumers. Permitting financial institutions to make their disclosure statements and modified loan/application register data available to the public through notices that clearly convey that the disclosure statements and modified loan/ application register data may be obtained on the Bureau's Web site will require consumers to obtain these disclosure statements online. Given the prevalence of internet access and the ease of using the Bureau's Web site, the Bureau believes these adopted changes will impose minimal direct costs on consumers. Any potential costs to consumers of obtaining disclosure statements and modified loan/application register data online are likely no greater than the [[Page 66294]] costs of obtaining disclosure statements and modified loan/application register data from the physical offices of financial institutions, or from a floppy disk or other electronic data storage medium that may be used with a personal computer, as contemplated by HMDA section 304(k)(1)(b). However, consumers may bear some indirect costs of the changes in the final rule if financial institutions pass on some or all of their increased costs to consumers. Following microeconomic principles, the Bureau believes that financial institutions will pass on increased variable costs to future loan applicants but will absorb one-time costs and increased fixed costs if financial institutions are profit maximizers and the market is perfectly competitive. The Bureau defines variable costs as costs that depend on the number of applications received. Based on initial outreach efforts, five of the 18 operational tasks are variable cost tasks: Transcribing data, resolving reportability questions, transferring data to an HMS, geocoding, and researching questions. The Bureau believes that the four changes discussed in this section will have either no, or only a minimal, effect on these variable cost tasks. Quarterly reporting, as well as the requirements that financial institutions make their disclosure statements and modified loan/ application registers available to the public by making available a brief notice referring members of the public to the Bureau's Web site, will not impact any variable-cost operational steps. Hence, these three revisions in the final rule will not lead financial institutions to pass through some of the incremental costs to consumers in a perfectly competitive market with profit-maximizing financial institutions. Eliminating the option of paper reporting for financial institutions may increase transcription costs for financial institutions that currently qualify for this option and report HMDA data in paper form. However, given the closed-end and open-end reporting thresholds, very few, if any, financial institutions would meet the threshold for paper reporting. Given these factors, the Bureau estimates that the impact of this cost is negligible. Benefits to covered persons. The Bureau believes that eliminating the option of paper reporting and requiring quarterly reporting for certain financial institutions will provide little direct benefit to covered persons. However, the requirement that financial institutions make their modified loan/application registers available to the public by providing a brief notice referring members of the public to the Bureau's Web site will benefit covered persons. This provision reduces costs to financial institutions associated with preparing and making available to the public the modified loan/application register and eliminates a financial institution's risk of missing the deadline to make it available. It also eliminates the risks to financial institutions making errors in preparing the modified loan/application register that could result in the unintended disclosure of data. Initial outreach efforts indicated that tier 3 financial institutions rarely receive requests for modified loan/application register data. However, some tier 3 financial institutions indicated that they nevertheless prepare the data in preparation for requests. The Bureau has represented this cost as equivalent to preparing one modified loan/application register dataset each year. The Bureau estimates that representative tier 2 and tier 1 financial institutions receive three and 15 requests for modified loan/application register data each year, respectively. Based on these estimated volumes, the Bureau estimates that this revision in the final rule will reduce ongoing operational costs by approximately $130 per year for a representative tier 3 financial institution, approximately $310 per year for a representative tier 2 financial institution, and approximately $770 per year for a representative tier 1 financial institution. This translates into a market-level reduction in cost of approximately $1,500,000 to $2,000,000 per year. Using a 7 percent discount rate, the net present value of this savings over five years is $6,100,000 to $8,200,000. Similarly, permitting a financial institution to make its disclosure statements available to the public through a notice that clearly conveys that the disclosure statement may be obtained on the Bureau's Web site will free financial institutions from having to download and print their disclosure statements in order to provide them to requesters. Initial outreach efforts indicated that tier 3 financial institutions rarely receive requests for disclosure statements. However, some tier 3 financial institutions indicated that they nevertheless download and print a disclosure statement in preparation for requests. The Bureau has represented this cost as equivalent to receiving one request for a disclosure statement each year. The Bureau estimates that on average tier 2 and tier 1 financial institutions receive three and 15 requests for disclosure statements each year, respectively. Based on these estimated volumes, the Bureau estimates that this change will reduce ongoing operational costs by approximately $15 per year for a representative tier 3 financial institution, approximately $50 per year for a representative tier 2 financial institution, and approximately $250 per year for a representative tier 1 financial institution. This translates into a market-level reduction in cost of approximately $250,000 to $333,000 per year. Using a 7 percent discount rate, the net present value of this savings over five years is $1,015,000 to $1,366,000. One-time costs to covered persons. The Bureau believes that the provisions requiring financial institutions to make their disclosure statements and modified loan/application registers available to the public by providing brief notices referring members of the public to the Bureau's Web site will require a one-time cost to create the notice. However the Bureau believes that the one-time cost to create these notices will be negligible. Similarly, the Bureau believes that the revisions in the final rule to require quarterly reporting by large volume financial institutions, and to eliminate the option of paper reporting, will not impose any significant one-time costs on covered persons. Ongoing costs to covered persons. The Bureau believes that the provisions requiring financial institutions to make their disclosure statements and modified loan/application registers available to the public by providing brief notices referring members of the public to the Bureau's Web site will not increase ongoing costs to covered persons. Eliminating the option of paper reporting for financial institutions reporting 25 or fewer records may increase transcription costs for financial institutions that currently maintain all HMDA data in paper form. However, as discussed above, the Bureau believes that the number of financial institutions that do this is very low, especially given changes to the institutional coverage criteria, planned improvements to the data submission process and the small size of the loan/application register at issue (25 or fewer records). Therefore, the Bureau estimates that the impact of this cost is negligible. Quarterly reporting will increase ongoing costs to covered persons, as costs will increase for annual edits and internal checks, checking post-submission edits, filing post-submission edits, internal audits, and external audits. The Bureau estimates that this change will increase ongoing operational costs by approximately [[Page 66295]] $31,000 per year for a representative tier 1 financial institution.\516\ --------------------------------------------------------------------------- \516\ The Bureau also estimates that this change will increase ongoing operational costs by approximately $800 and $5,000 per year for representative tier 3 and 2 institutions, respectively, were these institutions required to report quarterly. However, since the Bureau believes that all the financial institutions subject to quarterly reporting under the final rule will be tier 1 institutions, the estimates for tier 3 and tier 2 institutions have been excluded. These estimates are for financial institutions that meet the threshold for reporting closed-end mortgage loans, but not for reporting of open-end lines of credit. --------------------------------------------------------------------------- Based on 2013 HMDA data, 29 financial institutions reported at least 60,000 covered loan and applications, combined, excluding purchased covered loans, in 2013, which is substantially larger than the average loan/application register sizes of the representative tier 3 institutions (50 records), tier 2 institutions (1,000 records), and is also above the loan/application register size of the representative tier 1 institutions (50,000) assumed by the Bureau. Therefore, the Bureau believes that it is reasonable to regard all of these institutions as tier 1 HMDA reporters. This yields an estimated market cost of $899,000 (= 29 * $31,000). Using a 7 percent discount rate, the net present value of this impact over five years will be approximately an increase in costs of $3,700,000. G. Potential Specific Impacts of the Final Rule 1. Depository Institutions and Credit Unions With $10 Billion or Less in Total Assets, as Described in Section 1026 As discussed above, the final rule makes certain changes to the institutional and transactional coverage of Regulation C and modifies the disclosure and reporting requirements. The Bureau believes that the benefits of these revisions for depository institutions and credit unions with $10 billion or less in total assets will be similar to the benefits to creditors as a whole, as discussed above. The only potential difference would be the benefits of aligning current and new HMDA data points to industry standards, which will likely create higher benefits for larger institutions. Regarding costs, other than as noted here, the Bureau also believes that the impact of the final rule on the depository institutions and credit unions with $10 billion or less in total assets will be similar to the impact for creditors as a whole. The primary difference in the impact on these institutions is likely to come from differences in the level of complexity of operations, compliance systems, and software of these institutions. The three representative lender types, which the Bureau analyzed when considering the benefits, costs and impacts of the final rule, incorporate differences in complexity and infrastructure across financial institutions, and the effect of these differences on impacts of the final rule. Based on Call Report data for December 2013, 13,454 of 13,565 depository institutions and credit unions had $10 billion or less in total assets. Based on 2013 HMDA data, and the reporting requirement for closed-end mortgage loans in the final rule, approximately 4,800 of these depository institutions and credit unions would be required to report data on closed-end mortgage loans. Six of the estimated 29 institutions that would have been required to report on a quarterly basis in 2014 had the final rule been in effect were depository institutions or credit unions with $10 billion or less in total assets. Given their large loan/application register volumes, all of these institutions are assumed to be tier 1 institutions. Finally, approximately 749 institutions will meet the threshold for open-end lines of credit and be required to report data on these products. The Bureau estimates that 660 of these institutions are depository institutions and credit unions with $10 billion or less in total assets. Under all of these assumptions, the Bureau estimates that the market-level impact of the final rule on operational costs for depository institutions and credit unions with $10 billion or less in total assets will be a cost of between $27,600,000 and $44,500,000. Using a discount rate of 7 percent, the net present value of this cost over five years is between $113,000,000 and $182,500,000. Regarding one-time costs, the Bureau estimates that the market-level impact of the final rule for depository institutions and credit unions with $10 billion or less in total assets is between $637,200,000 and $1,252,300,000. Using a 7 percent discount rate and a five-year amortization window, the annualized one-time cost is between $155,400,000 and $305,400,000. 2. Impact of the Provisions in the Final Rule on Consumers in Rural Areas The Bureau believes that the provisions in the final rule will not impose direct costs to consumers in rural areas. However, as with all consumers, consumers in rural areas may bear some indirect costs of the final rule. This will occur if financial institutions serving rural areas are HMDA reporters and if these institutions pass on some or all of the cost increase to consumers. Recent research suggests that financial institutions that primarily serve rural areas are generally not HMDA reporters.\517\ The Housing Assistance Council (HAC) suggests that the asset and geographic coverage criteria disproportionately exempt small lenders operating in rural communities. For example, HAC uses 2009 Call Report data to show that approximately 700 FDIC-insured lending institutions had assets totaling less than the HMDA institutional coverage threshold and were headquartered in rural communities. These institutions, which would not be HMDA reporters, may represent one of the few sources of credit for many rural areas. Research by economists at the Federal Reserve Board also suggests that HMDA's coverage of rural areas is limited, especially areas further from MSAs.\518\ If a large portion of the rural housing market is serviced by financial institutions that are not HMDA reporters, any indirect impact of the changes on consumers in rural areas will be limited, as the changes directly involve none of those financial institutions. --------------------------------------------------------------------------- \517\ See Keith Wiley, Housing Assistance Council, What Are We Missing? HMDA Asset-Excluded Filers, (2011), available at http://www.ruralhome.org/storage/documents/smallbanklending.pdf; Lance George and Keith Wiley, Housing Assistance Council, Improving HMDA: A Need to Better Understand Rural Mortgage Markets, (2010), available at http://www.ruralhome.org/storage/documents/notehmdasm.pdf. \518\ Robert B. Avery, Kenneth P. Brevoort, and Glenn B. Canner, Opportunities and Issues in Using HMDA Data, 29 J. of Real Estate Research 352 (2007), available at http://pages.jh.edu/jrer/papers/pdf/past/vol29n04/02.351_380.pdf. --------------------------------------------------------------------------- Although some research suggests that HMDA currently does not cover a significant number of financial institutions serving the rural housing market, HMDA data do contain information for some covered loans involving properties in rural areas. These data can be used to estimate the number of HMDA reporters servicing rural areas, and the number of consumers in rural areas that might potentially be affected by the changes to Regulation C. For this analysis, the Bureau uses non-MSA areas as a proxy for rural areas, with the understanding that portions of MSAs and non-MSAs may contain urban and rural territory and populations. In 2013, 5,678 HMDA reporters reported applications, originations, or purchased loans for property located in geographic areas outside of an MSA.\519\ This count [[Page 66296]] provides some sense of the number of financial institutions that could potentially pass on impacts of the final rule to consumers in rural areas.\520\ In total, these 5,678 financial institutions reported 1,989,000 applications, originations, or purchased loans for properties in non-MSA areas. This number provides some sense of the number of consumers in rural areas that could potentially be impacted indirectly by the changes in the final rule. In general, individual financial institutions report small numbers of closed-end mortgage loans from non-MSAs, as approximately 70 percent reported fewer than 100 closed- end mortgage loans from non-MSAs. --------------------------------------------------------------------------- \519\ These counts exclude preapproval requests that were denied or approved but not accepted, because geographic information is typically not available for these transactions. \520\ These counts do not include the estimated 750 or so financial institutions that will be required to report open-end lines of credit, or the estimated 75-450 nondepository institutions that will be required to report due to the coverage threshold being reduced from 100 to 25. In both instances, data required to estimate how many of these institutions serve rural areas is limited. To the extent that some do serve rural areas, the numbers presented will be underestimates. --------------------------------------------------------------------------- Following microeconomic principles, the Bureau believes that financial institutions will pass on increased variable costs to future mortgage applicants but will absorb one-time costs and increased fixed costs if financial institutions are profit maximizers and the market is perfectly competitive.\521\ The Bureau defines variable costs as costs that depend on the number of applications received. Based on initial outreach efforts, the following five operational steps affect variable costs: Transcribing data, resolving reportability questions, transferring data to an HMS, geocoding, and researching questions. The primary impact of the final rule on these operational steps is an increase in time spent per task. Overall, the Bureau estimates that the impact of the final rule on variable costs per application is $23 for a representative tier 3 financial institution, $0.20 for a representative tier 2 financial institution, and $0.10 for a representative tier 1 financial institution.\522\ The 5,678 financial institutions that served rural areas would attempt to pass these variable costs on to all future mortgage customers, including the estimated 2 million consumers from rural areas. Amortized over the life of the loan, this expense would represent a negligible increase in the cost of a mortgage loan. Therefore, the Bureau does not anticipate any material adverse effect on credit access in the long or short term even if these financial institutions pass on these costs to consumers. --------------------------------------------------------------------------- \521\ If markets are not perfectly competitive or financial institutions are not profit maximizers then what financial institutions pass on may differ. For example, they may attempt to pass on one-time costs and increases in fixed costs, or they may not be able to pass on variable costs. \522\ These cost estimates do not incorporate the impact of operational improvements and additional help sources. These estimates are for financial institutions that meet the threshold for reporting closed-end mortgage loans, but not for reporting of open- end lines of credit or quarterly reporting. --------------------------------------------------------------------------- During the Small Business Review Panel process, some small entity representatives noted that they would attempt to pass on all increased compliance costs associated with the final rule, but that this would depend upon the competiveness of the market in which they operate, especially for smaller financial institutions. In addition, some small entity representatives noted that they would attempt to pass on costs through higher fees on other products, exit geographic or product markets, or spend less time on customer service. The similar concern was echoed by some industry comments to the proposal. To the extent that the market is less than perfectly competitive and the lenders are able to pass on a greater amount of these compliance costs, the costs to consumers will be slightly larger than the estimates described above. Nevertheless, the Bureau believes that the potential costs that will be passed on to consumers are small. On the benefit side, the expanded institutional and transactional coverage, and reporting requirements may indirectly benefit consumers in rural areas to the extent that HMDA reporters serve these areas. Specifically, the revisions in the final rule will provide the public and public officials with information to help determine whether financial institutions are serving the housing needs of rural communities, to target public investment to attract private investment in rural communities, and to identify possible discriminatory lending patterns and enforce antidiscrimination statutes. Given the differences between rural and non-rural markets in structure, demand, supply, and competition level, consumers in rural areas may experience benefits and costs from the final rule that are different than those experienced by consumers in general. To the extent that the impacts of the final rule on creditors differ by type of creditor, this may affect the costs and benefits of the final rule on consumers in rural areas. The Bureau solicited feedback regarding the impact of the proposed rule on consumers in rural areas. One national trade association commenter cited a study from several individuals at the Mercatus Center at George Mason University that found compliance burden had increased for over 90 percent of community banks surveyed, and that banks in rural areas were particularly impacted. This survey focused on the overall burden of all recent regulation, and did not focus on the burden specific to HMDA. Therefore, the Bureau was unable to determine how much of the increased cost to attribute to the final HMDA rule and has not revised the estimates contained in this part based on the particular study cited by the commenter. III. Final Regulatory Flexibility Act Analysis The Regulatory Flexibility Act (RFA) generally requires an agency to conduct an initial regulatory flexibility analysis (IRFA) and a final regulatory flexibility analysis (FRFA) of any rule for which notice-and-comment procedures are required by 5 U.S.C. 553.\523\ These analyses must describe the impact of the rule on small entities.\524\ An IRFA or FRFA is not required if the agency certifies that the rule will not have a significant economic impact on a substantial number of small entities.\525\ The Bureau is also subject to certain additional procedures under the RFA involving the convening of a panel to consult with small business representatives prior to proposing a rule for which an IRFA is required.\526\ --------------------------------------------------------------------------- \523\ 5 U.S.C. 601 et. seq. \524\ For purposes of assessing the impacts of the final rule on small entities, ``small entities'' is defined in the RFA to include small businesses, small not-for-profit organizations, and small government jurisdictions. 5 U.S.C. 601(6). A ``small business'' is determined by application of Small Business Administration regulations and reference to the North American Industry Classification System (NAICS) classifications and size standards. 5 U.S.C. 601(3). A ``small organization'' is any ``not-for-profit enterprise which is independently owned and operated and is not dominant in its field.'' 5 U.S.C. 601(4). A ``small governmental jurisdiction'' is the government of a city, county, town, township, village, school district, or special district with a population of less than 50,000. 5 U.S.C. 601(5). \525\ 5 U.S.C. 605(b). \526\ 5 U.S.C. 609. --------------------------------------------------------------------------- In the proposal, the Bureau did not certify that the proposed rule would not have a significant economic impact on a substantial number of small entities within the meaning of the RFA. Accordingly, the Bureau convened and chaired a Small Business Review Panel under the Small Business Regulatory Enforcement Fairness Act (SBREFA) to consider the impact of the proposed rule on small entities that would be subject to that rule and to obtain feedback from representatives of such small entities. The 2014 HMDA Proposal preamble included detailed information on the Small Business Review Panel. The Panel's advice and recommendations [[Page 66297]] are found in the Small Business Review Panel Final Report \527\ and were discussed in the section-by-section analysis of the proposed rule. The 2014 HMDA Proposal also contained an IFRA pursuant to section 603 of the RFA. In this IRFA, the Bureau solicited comment on any costs, recordkeeping requirements, compliance requirements, or changes in operating procedures arising from the application of the proposed rule to small businesses; comment regarding any Federal rules that would duplicate, overlap, or conflict with the proposed rule; and comment on alternative means of compliance for small entities. Comments addressing individual provisions of the final rule are addressed in the section- by-section analysis above. Comments addressing the impact on small entities are discussed below. Many of these comments implicated individual provisions of the final rule or the Bureau's Dodd-Frank Act section 1022 discussion, and are also addressed in those parts. --------------------------------------------------------------------------- \527\ See Final Report of the Small Business Review Panel on the CFPB's Proposals Under Consideration for the Home Mortgage Disclosure Act (HMDA) Rulemaking (April 24, 2014), http://files.consumerfinance.gov/f/201407_cfpb_report_hmda_sbrefa.pdf. --------------------------------------------------------------------------- Based on the comments received, and for the reasons stated below, the Bureau believes the final rule will have a significant economic impact on a substantial number of small entities. Accordingly, the Bureau has prepared the following final regulatory flexibility analysis pursuant to section 604 of the RFA. A. Statement of the Need for, and Objectives of, the Rule The Bureau is publishing the final rule to implement section 1094 of the Dodd-Frank Act, which amended HMDA to improve the utility of the HMDA data.\528\ HMDA was intended to provide the public with information that can be used to help determine whether financial institutions are serving the housing needs of their communities, to assist public officials in distributing public-sector investment so as to attract private investment, and to assist in identifying possible discriminatory lending patterns and enforcing antidiscrimination statutes. Historically, HMDA has been implemented by the Board through Regulation C, 12 CFR part 203. In 2011, the Bureau established a new Regulation C, 12 CFR part 1003, substantially duplicating the Board's Regulation C, making only non-substantive, technical, formatting, and stylistic changes. Congress has periodically modified HMDA, and the Board routinely updated Regulation C, in order to ensure that the data continued to fulfill HMDA's purposes. In 2010, Congress responded to the mortgage crisis by passing the Dodd-Frank Act, which enacted changes to HMDA, as well as directing reforms to the mortgage market and the broader financial system. In addition to transferring rulemaking authority for HMDA from the Board to the Bureau, section 1094 of the Dodd-Frank Act, among other things, directed the Bureau to implement changes requiring the collection and reporting of several new data points, and authorized the Bureau to require financial institutions to collect and report such other information as the Bureau may require. --------------------------------------------------------------------------- \528\ Dodd-Frank Act, Public Law 111-203, section 1094, 124 Stat. 1376, 2097 (2010). --------------------------------------------------------------------------- A full discussion of the reasons for the final rule may be found in parts V and VII, above. Briefly, the rule addresses the market failures caused by the underproduction of public mortgage data and the information asymmetries in credit markets through improved institutional and transactional coverage and additional information about underwriting, pricing, and property characteristics. The final rule will improve the ability of regulators, industry, advocates, researchers, and economists to assess housing needs, public investment, possible discrimination, and market trends. B. Statement of the Significant Issues Raised by the Public Comments in Response to the IRFA, a Statement of the Assessment of the Agency of Such Issues, and a Statement of Any Changes Made as a Result of Such Comments In accordance with section 603(a) of the RFA, the Bureau prepared an IRFA. In the IRFA, the Bureau estimated the possible compliance costs for small entities with respect to each major component of the rule against a pre-statute baseline. The Bureau requested comment on the IRFA. Very few commenters specifically addressed the IRFA. Comments that repeated the same issues raised by the Office of Advocacy of the U.S. Small Business Administration are addressed in part VIII.C, below. Other comments related to small financial institutions are discussed here. As discussed in the section 1022 analysis in part VII above, several commenters addressed the impact of the proposed rule on small financial institutions. Several industry commenters stated that the proposed rule would create a competitive disadvantage for small financial institutions. For example, these commenters noted that larger financial institutions would be able to distribute the cost of compliance across a larger transaction base. Several industry commenters cited reports from Goldman Sachs and Banking Compliance Index figures to support claims that regulatory burdens were disproportionally affecting small financial institutions and preventing low income consumers from accessing certain financial products. Another industry commenter cited the decline in HMDA reporters from 2012 to 2013 as evidence that small financial institutions have left the market. The Bureau presented separate impact estimates for low-, moderate-, and high-complexity institutions, broadly reflecting differences in impact across institutions of different size, and has recognized that on average the smaller institution will incur slightly higher compliance costs per HMDA record due to the final rule than larger institutions. However, the magnitude of such impact on a per application basis is fairly small. Specifically, for low-complexity institutions, which best represent small institutions, the estimated impact on operational costs, after the operational modifications the Bureau is making, is approximately $1,900 per year.\529\ This translates into approximately a $38 increase in per application costs. Based on recent survey estimates of net income from the MBA, this impact represents approximately 1.3 percent ($38/$2,900) of net income per origination for mid/medium sized banks, which the Bureau views as relatively small. Therefore, the Bureau concludes that the final rule will have little impact on any competitive disadvantage faced by small institutions. --------------------------------------------------------------------------- \529\ This estimate applies to financial institutions that meet the threshold for reporting closed-end mortgage loans, but not for reporting of open-end lines of creditor quarterly reporting. --------------------------------------------------------------------------- Other industry commenters believed that the proposal would likely increase the cost of credit for consumers. Several of these commenters cited the cost of systems modifications associated with reporting home- equity lines of credit. A few commenters claimed that certain small financial institutions, such as small credit unions, small farm credit lenders, or small banks, would be faced with difficult choices, such as merging, raising prices, originating fewer loans, or exiting the market. A small number of industry commenters stated that they would double their origination fees as a result of the proposed rule. A national trade association commenter cited, among other things, a study from several individuals at the Mercatus Center at [[Page 66298]] George Mason University and a survey of its members showing that small financial institutions were decreasing their mortgage lending activity in response to increased regulatory burdens. Similarly, other industry commenters pointed to a report from Goldman Sachs showing that higher regulatory costs had priced some low-income consumers out of the credit card and mortgage markets. Following standard economic theory, in a perfectly competitive market where financial institutions are profit maximizers, the affected financial institutions would pass on to consumers the marginal, i.e., variable, cost per application or origination and would absorb the one-time and increased fixed costs of complying with the rule. Overall, the Bureau estimates that the final rule will increase variable costs by $23 per application for representative tier 3 institutions, $0.20 per application for representative tier 2 institutions, and $0.10 per application for representative tier 1 institutions.\530\ These expenses will be amortized over the life of a loan and represent a negligible increase in the cost of a mortgage loan. Therefore, the Bureau does not anticipate any material adverse effect on credit access in the long or short term even if institutions pass on these costs to consumers. --------------------------------------------------------------------------- \530\ These estimates apply to financial institutions that meet the threshold for reporting closed-end mortgage loans, but not for reporting of open-end lines of credit or quarterly reporting. --------------------------------------------------------------------------- Several industry commenters explained that expanding the rule to include commercial-purpose transactions would increase the cost of business credit. These commenters stated that financial institutions would be less willing to take the dwelling of a borrower as collateral, which would decrease the availability of credit. However, as explained above, the Bureau is specifically exempting certain commercial-purpose transactions from the scope of the final rule so that coverage of commercial-purpose transactions is generally maintained at its existing level.\531\ Accordingly, the Bureau expects that the final rule will not have a significant impact on the availability of commercial credit. --------------------------------------------------------------------------- \531\ The revisions to the final rule will require reporting of commercial-purpose lines of credit for the purposes of home purchase, home improvement, or refinancing. Reporting of these loans is not currently required, therefore it is possible that the coverage of commercial-purpose loans will increase slightly, but the Bureau believes that the impact will be minimal. --------------------------------------------------------------------------- Other industry commenters believed that any utilization of the MISMO data standards would burden small entities. These commenters stated that small financial institutions would have to incur training costs to familiarize themselves with MISMO. One national trade association commenter reported that only 22 percent of community banks use MISMO. These commenters believe that MISMO alignment should be optional for small financial institutions. As explained above, the Bureau believes that these commenters have misunderstood the implications of the proposed MISMO alignment. The Bureau did not propose to, and the final rule does not, require any financial institution to use or become familiar with the MISMO data standards. Rather, the rule merely recognizes that many financial institutions are already using the MISMO standard for collecting and transmitting mortgage data, and has utilized similar definitions for certain data points in order to reduce burden. Thus, the rule decreases cost for those institutions that are familiar with MISMO. Financial institutions that are unfamiliar with MISMO may not realize a similar reduction in cost, but they will not experience any increased burden from the utilization of MISMO definitions because the final rule itself and the associated materials contain all of the necessary definitions and instructions for reporting HMDA data. Several industry commenters believed that the Bureau had ignored the comments of the small entity representatives that participated in the Small Business Review Panel or had simply solicited feedback in response to their suggestions. As noted in the IRFA, the small entity representatives made several comments at the SBREFA Panel. Many of these suggestions have been reflected in the final rule. For example, the Bureau heard from small entity representatives that they rarely, if ever, receive requests for their modified loan/application registers, and the Small Business Review Panel recommended that the Bureau consider whether there is a continued need for small institutions to make their modified loan/application registers available. The final rule provides that financial institutions shall make available to the public a notice that clearly conveys that the institution's modified loan/application register may be obtained on the Bureau's Web site. This approach relieves small financial institutions of the obligation to provide the modified loan/application register to the public directly. Additionally, several small entity representatives expressed concern over the operational difficulties of geocoding and the data submission process in general. The Bureau is making operational enhancements and modifications to address these concerns. For example, the Bureau is working to provide implementation support similar to the support provided for the title XIV and TILA-RESPA Integrated Disclosure rules. The Bureau is also improving the geocoding process, creating a web-based HMDA data submission and edit-check system, developing a data-entry tool for small financial institutions that currently use Data Entry Software, and otherwise streamlining the submission and editing process to make it more efficient. All of these enhancements will improve the submission and processing of data, increase clarity, and reduce reporting burden. Finally, small entity representatives requested a two-year look-back period in the loan-volume threshold. The final rule includes a two-year look-back period. Under the final rule, a financial institution that does not meet the loan-volume thresholds established in the final rule and that experiences an unusual and unexpected high origination volume in one year will not be required to begin HMDA reporting unless and until the higher origination volume continues for a second year in a row. In addition to modifying the proposed rule in direct response to suggestions from small entity representatives that participated in the Small Business Review Panel, the Bureau also modified the proposed rule based on responses to the Bureau's requests for feedback that were prompted by the small entity representatives. As one example, the proposed change in transactional coverage to a dwelling-secured basis would have extensively expanded reporting of commercial-purpose loans and lines of credit. In response to comments received about the cost impact of this proposal, the Bureau decided to maintain Regulation C's existing purpose-based coverage test for commercial-purpose transactions, which maintains coverage of commercial-purpose lending generally at existing levels. Similarly, the proposed change in transactional coverage to a dwelling-secured basis would have extensively expanded reporting of consumer-purpose open-end lines of credit. In response to comments received about the cost impact of this proposal, especially about the one-time costs of constructing the infrastructure to report data from a separate business line, the Bureau decided to adopt a separate loan-volume reporting threshold of 100 open-end lines of credit. This threshold will reduce reporting burden for small entities. [[Page 66299]] C. Response to the Chief Counsel for Advocacy of the Small Business Administration and Statement of Any Change Made in the Final Rule as a Result of the Comments The SBA Office of Advocacy (Advocacy) provided a formal comment letter to the Bureau in response to the 2014 HMDA Proposal. Among other things, this letter expressed concern about the following issues: The expanded transactional coverage of the proposal, the analysis of the different loan-volume thresholds suggested by the small entity representatives, the requirement to report the discretionary data points, and the requirement to maintain modified loan/application registers. First, Advocacy expressed concern over the expanded transactional coverage of the proposed rule. The proposed rule would have covered all dwelling-secured closed-end mortgage loans, open-end lines of credit, and reverse mortgages. Advocacy supported the Bureau's decision to eliminate reporting of non-dwelling-secured home improvement loans. However, Advocacy noted that reporting additional transactions was burdensome for small financial institutions and believed that the new transactions might cause certain small financial institutions to become HMDA reporters for the first time. Advocacy urged the Bureau not to adopt the expanded transactional coverage. As described in greater detail in parts V and VII above, the Bureau considered the benefits and costs of the final rule's transactional coverage criteria. With respect to commercial-purpose transactions, the Bureau has decided to withdraw most of the expanded coverage of commercial-purpose loans. The Bureau is now limiting reporting of commercial-purpose loans and lines of credit to those for home purchase, home improvement, or refinancing purposes only. The Bureau is adopting the proposed expansion to consumer-purpose open-end lines of credit and reverse mortgages. Information about these types of transactions serves an important role in fulfilling HMDA's purposes. For example, among other things, data about reverse mortgages will help determine how the housing needs of seniors are being met, while data about open-end lines of credit will help assess housing-related credit being offered in particular communities. Regarding the impact of the new transactions on the loan-volume threshold, the Bureau notes that the 25-loan threshold includes only closed-end mortgage loans. The final rule institutes a separate reporting threshold of 100 open-end lines of credit for institutional coverage. As shown in Table 8 in part VII.F.3, above, compared to the proposal, this separate open-end reporting threshold will achieve a significant reduction in burden by eliminating the number of institutions that would be required to report data concerning their open-end lines of credit, if any, by almost 3,400, most which are likely small financial institutions. The Bureau further estimates that the open-end reporting threshold will require no additional financial institutions to report HMDA data, as compared to the current rule, because it is the Bureau's belief that nondepository institutions commonly are not engaged in dwelling-secured open-end-line-of-credit lending, and the depository institutions and credit unions that will report open-end lines of credit will still be subject to all other reporting requirements and hence can only come from current HMDA reporters.\532\ Therefore, the Bureau believes that the additional types of transactions required by the final rule will not impose a significant burden on small financial institutions or dramatically expand the institutional coverage of the rule. --------------------------------------------------------------------------- \532\ The Bureau estimates under the final rule, about 24 depository institutions and credit unions will report open-end lines of credit but not closed-end mortgage loans. However, even these future open-end-only reporters are not new to HMDA reporting, as they are currently reporting under HMDA but likely will stop reporting closed-end mortgage loans given their closed-end loan volumes fall below the 25-loan closed-end threshold. --------------------------------------------------------------------------- Second, Advocacy believed that the loan-volume threshold was too low. Advocacy also expressed concern over the Bureau's consideration of alternative loan-volume thresholds. Advocacy stated that the 25-loan threshold would exclude approximately 70,000 records from depository institutions and include approximately 30,000 records from nondepository institutions. According to Advocacy, assuming that all excluded institutions were small entities, the proposal would exclude 21 percent of small entities. Finally, Advocacy urged the Bureau to provide a full analysis of the possible loan-volume thresholds suggested by the small entity representatives. Throughout this rulemaking, the Bureau considered several higher loan-volume thresholds. These thresholds were evaluated based on their impact on the goals of the rulemaking, which include simplifying the reporting regime by establishing a uniform loan-volume threshold applicable to both depository and nondepository institutions; eliminating the burden of reporting from low-volume depository institutions while maintaining sufficient data for analysis at the national, local, and institutional levels; and increasing visibility into the home mortgage lending practices of nondepository institutions. As described in parts V and VII.F.3, above, the 25-loan threshold for closed-end mortgage loans appropriately balances multiple competing interests and advances the goals of the rulemaking. The Bureau believes that the threshold reduces burden on small financial institutions while preserving important data about communities and increasing visibility into the lending practices of nondepository institutions. The 25-loan threshold will achieve a significant reduction in burden by eliminating reporting by about 20 percent of depository institutions that are currently reporting. As described in greater detail throughout this discussion, the Bureau estimates that the most significant driver of costs under HMDA is fixed costs associated with the requirement to report, rather than the variable costs associated with any specific aspect of reporting, such as the number or complexity of required data fields or the number of entries. For example, the estimated annual ongoing cost of reporting under the rule is approximately $4,400 for a representative tier 3 financial institution after accounting for operational improvements. Just over $2,300 of this annual ongoing cost is composed of fixed costs. As a comparison, each required data field accounts for approximately $43 of this annual ongoing cost. Thus, the 25-loan threshold for closed-end mortgage loans provides a meaningful reduction in burden. Higher thresholds would further reduce burden but would produce data losses that would undermine the benefits provided by HMDA data. One of the most substantial impacts of any loan-volume threshold is the information that it provides about lending at the community level, including information about vulnerable consumers and the origination activities of smaller lenders. Public officials, community advocates, and researchers rely on HMDA data to analyze access to credit at the neighborhood-level and to target programs to assist underserved communities and consumers. For example, Lawrence, Massachusetts identified a need for homebuyer counseling and education based on HMDA data, which showed a high percentage of high-cost loans compared [[Page 66300]] to surrounding communities.\533\ Similarly, HMDA data helped bring to light discriminatory lending patterns in Chicago neighborhoods, resulting in a large discriminatory lending settlement.\534\ In addition, researchers and consumer advocates analyze HMDA data at the census tract level to identify patterns of discrimination at a national level.\535\ Higher loan-volume thresholds would affect data about more communities and consumers. At a loan-volume threshold set at 100, according to 2013 HMDA data, the number of census tracts that would lose 20 percent of reported data would increase by almost eight times over the number with a threshold set at 25 loans. The number of affected LMI tracts would increase more than six times over the number at the 25-loan level. Tables 5-8 in part VII.F.3 provide additional information about how different reporting thresholds affect the number of financial institutions that would be required to report closed-end mortgage loans, as well as open-end lines of credit. --------------------------------------------------------------------------- \533\ See City of Lawrence, HUD Consolidated Plan 2010-2015, at 68 (2010), available at http://www.cityoflawrence.com/Data/Sites/1/documents/cd/Lawrence_Consolidated_Plan_Final.pdf. Similarly, in 2008 the City of Albuquerque used HMDA data to characterize neighborhoods as ``stable,'' ``prone to gentrification,'' or ``prone to disinvestment'' for purposes of determining the most effective use of housing grants. See City of Albuquerque, Five Year Consolidated Housing Plan and Workforce Housing Plan, at 100 (2008), available at http://www.cabq.gov/family/documents/ConsolidatedWorkforceHousingPlan20082012final.pdf. As another example, Antioch, California, monitors HMDA data, reviews it when selecting financial institutions for contracts and participation in local programs, and supports home purchase programs targeted to households purchasing homes in Census Tracts with low loan origination rates based on HMDA data. See City of Antioch, Fiscal Year 2012-2013 Action Plan, at 29 (2012), available at http://www.ci.antioch.ca.us/CitySvcs/CDBGdocs/Action%20Plan%20FY12-13.pdf. See, e.g., Dara D. Mendez et al., Institutional Racism and Pregnancy Health: Using Home Mortgage Disclosure Act Data to Develop an Index for Mortgage Discrimination at the Community Level, 126 Pub. Health Reports (1974-) Supp. 3, 102-114 (Sept/Oct. 2011) (using HMDA data to analyze discrimination against pregnant women in redlined neighborhoods), available at http://www.publichealthreports.org/issueopen.cfm?articleID=2732. \534\ See, e.g., Yana Kunichoff, Lisa Madigan Credits Reporter with Initiating Largest Discriminatory Lending Settlements in U.S. History (June 14, 2013), http://www.chicagonow.com/chicago-muckrakers/2013/06/lisa-madigan-credits-reporter-with-initiating-largest-discriminatory-lending-settlements-in-u-s-history/ (``During our ongoing litigation . . . the Chicago Reporter study looking at the HMDA data for the City of Chicago came out . . . It was such a startling statistic that I said . . . we have to investigate, we have to find out if this is true . . . We did an analysis of that data that substantiated what the Reporter had already found . . . [W]e ultimately resolved those two lawsuits. They are the largest fair-lending settlements in our nation's history.'') \535\ See, e.g., California Reinvestment Coalition, et al, Paying More for the American Dream VI: Racial Disparities in FHA/VA Lending, at http://www.woodstockinst.org/research/paying-more-american-dream-vi-racial-disparities-fhava-lending. Likewise, researchers have analyzed GSE purchases in census tracts designated as underserved by HUD using HMDA data. James E. Pearce, Fannie Mae and Freddie Mac Mortgage Purchases in Low-Income and High-Minority Neighborhoods: 1994-96, Cityscape: A Journal of Policy Development and Research (2001), available at http://www.huduser.org/periodicals/cityscpe/vol5num3/pearce.pdf. --------------------------------------------------------------------------- Additionally, the Bureau believes that it is important to increase visibility into nondepository institutions' practices due to the lack of adequate data regarding their lending activity. Uniform loan-volume thresholds of fewer than 100 loans annually will expand nondepository institution coverage, because the current test requires reporting by all nondepository institutions that meet the other applicable criteria and originate 100 loans annually.\536\ Therefore, any threshold set at 100 loans would not provide any enhanced insight into nondepository institution lending and a threshold above 100 loans would actually decrease visibility into nondepository institutions' practices and hamper the ability of HMDA users to monitor risks posed to consumers by those institutions. The 25-loan volume threshold, however, achieves a significant expansion of nondepository institution coverage, with about a 40 percent increase in the number of reporting institutions. --------------------------------------------------------------------------- \536\ In addition, nondepository institutions that originate fewer than 100 applicable loans annually are required to report if they have assets of at least $10 million and meet the other criteria. See 12 CFR 1003.2 (definition of financial institution). --------------------------------------------------------------------------- Third, Advocacy stated that most small entities were concerned about the additional proposed data points that were not required by the Dodd-Frank Act. Advocacy believed that complying with the discretionary reporting requirements would impose additional expenses on small entities and might subject them to penalties for reporting errors. Therefore, Advocacy recommended that the Bureau exempt small entities from the reporting requirements regarding data points not mandated by the Dodd-Frank Act. The Bureau considered exempting smaller financial institutions from the requirement to report some or all of the discretionary data points. As described above, however, because under a tiered reporting regime smaller financial institutions would not report all or some of the HMDA data points, tiered reporting would prevent communities and users of HMDA data from learning important information about the lending and underwriting practices of smaller financial institutions, which may differ from those of larger institutions. Second, as discussed above, the primary driver of HMDA costs is establishing and maintaining systems to collect and report data, not the costs associated with collecting and reporting a particular data field. Therefore, tiered reporting would reduce the costs of low-volume depository institutions somewhat, but not significantly. Finally, Advocacy argued that requiring small entities to maintain modified loan/application registers was unduly burdensome because these institutions reported rarely being asked to provide such information to the public. Advocacy recommended removing small entities from this requirement. The Bureau generally agrees with these recommendations. As explained above, the final rule provides that financial institutions shall make available to the public a notice that clearly conveys that the institution's modified loan/application register may be obtained on the Bureau's Web site. This approach relieves all financial institutions, including small entities, of the obligation to provide the modified loan/application register to the public directly. The Bureau is also finalizing its proposal to provide that financial institutions shall make available to the public a notice that clearly conveys that the institution's disclosure statements may be obtained on the Bureau's Web site. This approach relieves all financial institutions, including small entities, of such burdens. D. Description of and Estimate of the Number of Small Entities to Which the Rule Will Apply or an Explanation of Why No Such Estimate Is Available As discussed in the proposal and Small Business Review Panel Report, for purposes of assessing the impacts of the final rule on small entities, ``small entities'' is defined in the RFA to include small businesses, small not-for-profit organizations, and small government jurisdictions.\537\ A ``small business'' is determined by application of Small Business Administration regulations and reference to the North American Industry Classification System (NAICS) classifications and size standards.\538\ A ``small organization'' is any ``not-for-profit enterprise which is independently owned and operated and is not dominant in its field.'' \539\ A ``small governmental jurisdiction'' is the government of a city, county, town, township, village, school district, or [[Page 66301]] special district with a population of less than 50,000.\540\ --------------------------------------------------------------------------- \537\ 5 U.S.C. 601(6). \538\ 5 U.S.C. 601(3). \539\ 5 U.S.C. 601(4). \540\ 5 U.S.C. 601(5). --------------------------------------------------------------------------- The following table provides the Bureau's estimate of the number and types of entities that may be affected by the final rule under consideration: [GRAPHIC] [TIFF OMITTED] TR28OC15.008 E. Projected Reporting, Recordkeeping, and Other Compliance Requirements of the Rule, Including an Estimate of the Classes of Small Entities Which Will Be Subject to the Requirement and the Type of Professional Skills Necessary for the Preparation of the Report 1. Reporting Requirements HMDA requires financial institutions to report certain information related to covered loans to the Bureau or to the appropriate Federal agency.\541\ Under Regulation C, all reportable transactions must be recorded on a loan/application register within 30 calendar days \542\ after the end of the calendar quarter in which final action is taken. Currently, financial institutions must disclose to the public upon request a modified version of the loan/application register submitted to regulators.\543\ Financial institutions must also make their disclosure statements, which are prepared by the FFIEC from data submitted by the institutions, available to the public upon request.\544\ --------------------------------------------------------------------------- \541\ 12 U.S.C. 2803(h)(1). \542\ 12 CFR 1003.4(a). \543\ 12 CFR 1003.5(c). \544\ 12 CFR 1003.5(b). --------------------------------------------------------------------------- The final rule modifies current reporting requirements and imposes new reporting requirements by requiring financial institutions to report additional information required by the Dodd-Frank Act, as well as certain information determined by the Bureau to be necessary and proper to effectuate HMDA's purposes. The rule also modifies the scope of the institutional and transactional coverage thresholds. In addition, under the final rule, financial institutions will make available to the public notices that clearly convey that the institution's disclosure statement and modified loan/application register may be obtained on the Bureau's Web site. Finally, financial institutions that reported at least 60,000 covered loans and applications, combined, excluding purchased loans, in the preceding calendar year will be required to report HMDA data on a quarterly basis to the appropriate Federal agency. These data will only be considered preliminary submissions, and the final rule provides a safe harbor that protects, in certain circumstances, a financial institution from being cited for violations of HMDA or Regulation C for errors and omissions in its quarterly submissions. The section-by-section analysis of the final rule in part V, above, discusses all of the additional required data points and the scope of the final rule in greater detail. 2. Recordkeeping Requirements HMDA currently requires financial institutions to compile and maintain information related to transactions involving covered loans. HMDA section 304(c) requires that information required to be compiled and made [[Page 66302]] available under HMDA section 304, other than loan/application register information required under section 304(j), must be maintained and made available for a period of five years. HMDA section 304(j)(6) requires that loan/application register information for any year shall be maintained and made available, upon request, for three years. Regulation C requires that all reportable transactions be recorded on a loan/application register within thirty calendar days after the end of the calendar quarter in which final action is taken.\545\ Regulation C further specifies that a financial institution shall retain a copy of its submitted loan/application register for its records for at least three years.\546\ --------------------------------------------------------------------------- \545\ 12 CFR 1003.4(a). \546\ 12 CFR 1003.5(a). --------------------------------------------------------------------------- The final rule will not modify the recordkeeping periods for financial institutions. The rule might, however, indirectly require additional recordkeeping in that it will require financial institutions to maintain additional information as a result of the expanded reporting requirements described above. However, the final rule reduces the amount of recordkeeping in other ways. Specifically, although the final rule does not eliminate the requirement that financial institutions retain a copy of their loan/application registers, the final rule does provide that financial institutions shall retain the notices concerning their disclosure statements and modified loan/ application registers, not the disclosure statements or modified loan/ application registers themselves, which may lessen the recordkeeping burden. Benefits to small entities. HMDA is a data reporting statute, so all provisions of the final rule affect reporting requirements. Overall, the final rule has several potential benefits for small entities. A summary of these benefits is provided here, and more detailed discussions of these benefits are provided in the section 1022 discussion in part VII, above. First, the revision to the institutional coverage criteria, which imposes a 25-loan threshold for closed-end mortgage loans, will benefit depository institutions that are not significantly involved in originating dwelling-secured closed-end mortgage loans. The Bureau expects that most of these depository institutions are small entities. These depository institutions will no longer have to report closed-end mortgage loans under HMDA. The Bureau also estimates that most of the depository institutions with closed-end mortgage loan originations falling below the threshold will originate fewer than 100 open-end lines of credit, and thus not be required to report such transactions under HMDA. Therefore, they will no longer have to incur one-time costs, or any current or increased operational costs, imposed by the final rule. Second, the Bureau adopted revisions to transactional coverage criteria that benefit small entities. As one example, the final rule eliminates reporting of non-dwelling-secured home improvement loans. This change reduces reporting burden to small entities to the extent that these entities offer such loans. As a second example, the overall change in transactional coverage to a dwelling-secured basis in the proposed rule extensively expanded reporting of commercial-purpose loans and lines. In response to comments received about the cost impact of this proposal, some of which came from small entities, the Bureau decided to retain Regulation C's existing purpose-based coverage test for commercial-purpose transactions, which maintains coverage of commercial-purpose lending generally at existing levels. Third, the expanded transactional coverage provisions, combined with the additional data points being adopted, will improve the prioritization process that regulators and enforcement agencies use to identify institutions with higher fair lending risk. During prioritization analyses, the additional transactions and data points will allow for improved segmentation, so that applications are compared to other applications for similar products. In addition, the data points will add legitimate factors used in underwriting and pricing that are currently lacking in the HMDA data, helping regulators and government enforcement agencies better understand disparities in outcomes. These improvements will reduce false positives that occur when inadequate information causes lenders with low fair lending risk to be initially misidentified as high-risk. This reduction in false positives will improve allocation of examination resources so that lenders with low fair lending risk receive a reduced level of regulatory scrutiny. For small entities currently receiving regulatory oversight, this could greatly reduce the burden from fair lending examinations and enforcement actions. Fourth, utilizing industry data standards may provide a benefit to some small entities, especially those originating and selling loans to the GSEs. The Bureau believes that promoting consistent data standards for both industry and regulatory use has benefits for market efficiency, market understanding, and market oversight. The efficiencies achieved by aligning HMDA data with widely used industry data standards should grow over time. Specific to small entities, outreach efforts have determined that aligning HMDA with industry data standards will reduce costs for training and researching questions. Fifth, and finally, the additional fields will improve the usefulness of HMDA data for analyzing mortgage markets by the regulators and the public. For instance, data points such as non- amortizing features, introductory interest rate, and prepayment penalty term that are commonly related to higher risk lending will provide a better understanding of the types of products and features consumers are receiving. This will allow for improved monitoring of trends in mortgage markets and help identify problems that could potentially harm consumers and society overall. Lowering the likelihood of future financial crises benefits all financial institutions, including small entities. Costs to small entities. Overall, the final rule has several potential costs for small entities. A summary of these costs is provided here, and more detailed discussions of these costs are provided in the section 1022 analysis in part VII, above. First, the adopted revision to the coverage criteria raises the closed-end mortgage loan reporting threshold for depository institutions from one to 25 loans and lowers the reporting threshold for nondepository institutions from 100 to 25 loans. Based on 2012 HMDA and NMLSR data, the Bureau estimates that an additional 75-450 nondepository institutions will be required to report as a result of this revision. The Bureau expects most of the affected nondepository institutions to be small entities. The additional nondepository institutions that will now be required to report under HMDA will incur one-time start-up costs to develop the necessary reporting infrastructure, as well as the ongoing operational costs to report. Second, for financial institutions subject to the final rule, the adopted revisions to transactional coverage will require reporting of open-end lines of credit, and require reporting of all closed-end home- equity loans and reverse mortgages. To the extent that small entities offer these products, these additional reporting requirements will increase operational costs as costs increase, for example, to transcribe data, resolve reportability questions, transfer data to HMS, and research questions. Third, the final rule adds additional data points identified by the Dodd-Frank Act and that the Bureau believes are necessary to close information gaps. As part of this final rule, the Bureau is [[Page 66303]] aligning all current and final data points to industry data standards to the extent practicable. The additional data points will increase ongoing operational costs, and impose one-time costs as small entities modify reporting infrastructure to incorporate additional fields. The transition to industry data standards will offset this cost slightly through reduced costs of researching questions and training. 3. Estimate of the Classes of Small Entities That Will Be Subject to the Requirement and the Type of Professional Skills Necessary for the Preparation of the Report or Record Section 603(a)(5) of the RFA requires an estimate of the classes of small entities that will be subject to the requirement. The classes of small entities that will be subject to the reporting, recordkeeping, and compliance requirements of the final rule are the same classes of small entities that are identified in part VIII.D, above. Type of professional skills required. Section 604(a)(5) of the RFA also requires an estimate of the type of professional skills necessary for the preparation of the reports or records required by the rule. The recordkeeping and compliance requirements of the final rule that will affect small entities are summarized above. Based on outreach with financial institutions, vendors, and governmental agency representatives, the Bureau classified the operational activities that financial institutions currently use for HMDA data collection and reporting into 18 operational ``tasks'' which can be further grouped into four ``primary tasks.'' These are: 1. Data collection: Transcribing data, resolving reportability questions, and transferring data to an HMS. 2. Reporting and resubmission: Geocoding, standard annual edit and internal checks, researching questions, resolving question responses, checking post-submission edits, filing post-submission documents, creating modified loan/application register, distributing modified loan/application register, distributing disclosure statement, and using vendor HMS software. 3. Compliance and internal audits: Training, internal audits, and external audits. 4. HMDA-related exams: Examination preparation and examination assistance. All of these tasks are related to the preparation of reports or records and most of them are performed by compliance personnel in the compliance department of financial institutions. For some financial institutions, however, the data intake and transcription stage could involve loan officers or processors whose primary function is to obtain or process loan applications. For example, the loan officers would take in government monitoring information from the applicants and input that information into the reporting system. However, the Bureau believes that such roles generally do not require any additional professional skills related to recordkeeping or other compliance requirements of this final rule that are not otherwise required during the ordinary course of business for small entities. The type of professional skills required for compliance varies depending on the particular task involved. For example, data transcription requires data entry skills. Transferring data to an HMS and using vendor HMS software requires knowledge of computer systems and the ability to use them. Researching and resolving reportability questions requires a more complex understanding of the regulatory requirements and the details of the relevant line of business. Geocoding requires skills in using geocoding software, web systems, or, in cases where geocoding is difficult, knowledge of the local area in which the property is located. Standard annual editing, internal checks, and post-submission editing require knowledge of the relevant data systems, data formats, and HMDA regulatory requirements in addition to skills in quality control and assurance. Filing post- submission documents, creating modified loan/application registers, and distributing modified loan/application registers and disclosure statements require skills in information creation, dissemination, and communication. Training, internal audits, and external audits require communications skills, teaching skills, and regulatory knowledge. HMDA- related examination preparation and examination assistance involve knowledge of regulatory requirements, the relevant line of business, and the relevant data systems. Tables 2-4 in part VII.F.2 provide detailed estimates of the costs of conducting each of these operational tasks. The Standard Occupational Classification (SOC) code has compliance officers listed under code 13-1041. The Bureau believes that most of the skills required for preparation of the reports or records related to this final rule are the skills required for job functions performed in this occupation. However, the Bureau recognizes that under this general occupational code there is a high level of heterogeneity in the type of skills required as well as the corresponding labor costs incurred by the financial institutions performing these functions. During the Small Business Review Panel process, some small entity representatives noted that, due to the small size of their institutions, they do not have separate compliance departments exclusively dedicated to HMDA compliance. Their HMDA compliance personnel are often engaged in other corporate compliance functions. To the extent that the compliance personnel of a small entity are divided between HMDA compliance and other functions, the skills required for those personnel may differ from the skills required for fully-dedicated HMDA compliance personnel. For instance, some small entity representatives noted that high-level corporate officers such as CEOs and senior vice presidents could be directly involved in some HMDA tasks. The Bureau acknowledges the possibility that certain aspects of the final rule may require some small entities to hire additional compliance staff. The Bureau has no evidence that such additional staff will possess a qualitatively different set of professional skills than small entity staff employed currently for HMDA purposes. It is possible, however, that compliance with the final rule may emphasize certain skills. For example, additional data points may increase demand for skills involved in researching questions, standard annual editing, and post-submission editing. On the other hand, the Bureau is making operational enhancements and modifications to alleviate some of the compliance burden. For example, the Bureau is working to provide implementation support similar to the support provided for the title XIV and TILA-RESPA Integrated Disclosure rules. The Bureau is also improving the geocoding process, creating a web-based HMDA data submission and edit-check system, developing a data-entry tool for small financial institutions that currently use Data Entry Software, and otherwise streamlining the submission and editing process to make it more efficient. Such enhancements may also change the relative composition of HMDA compliance personnel and the skills involved in recording and reporting data. Nevertheless, the Bureau believes that compliance will still involve the general set of skills identified above. The recordkeeping and reporting requirements associated with the final rule will also involve skills for [[Page 66304]] information technology system development, integration, and maintenance. Financial institutions often use an HMS for HMDA purposes. An HMS could be developed by the institution internally or purchased from a third-party vendor. Under the final rule, the Bureau anticipates that most of these systems will need substantial updates to comply with the new requirements. It is possible that other systems used by financial institutions, such as loan origination systems, might also need modification to be compatible with the updated HMS. The professional skills required for this one-time updating will be related to software development, testing, system engineering, information technology project management, budgeting, and operations. Based on feedback from the small entity representatives, many small business HMDA reporters rely on FFIEC DES tools and do not use a dedicated HMS. The Bureau is working to create a web-based HMDA data submission and edit-check system and develop a data-entry tool for small financial institutions that currently use DES that will allow financial institutions to use the software from multiple terminals in different branches and might reduce the required information technology implementation cost for small financial institutions. F. Description of the Steps the Agency Has Taken To Minimize the Significant Economic Impact on Small Entities The Bureau understands that the new provisions will impose a cost on small entities, and has attempted to mitigate the burden consistent with statutory objectives. The Bureau has adopted a number of modifications to particular provisions designed to reduce burden, which are described in the section-by-section analysis and the section 1022 analysis in parts V and VII, above. Several of the more significant burden-reducing steps reflected in the final rule are also described here. First, by raising the loan-volume threshold applicable to closed- end mortgage loans to 25 loans for depository institutions and adopting a threshold of 100 open-end lines of credit, the Bureau has provided substantial relief to small entities falling below these thresholds. As described in greater detail throughout this discussion, the Bureau estimates that the most significant driver of costs under HMDA is fixed costs associated with the requirement to report, rather than the variable costs associated with any specific aspect of reporting, such as the number or complexity of required data fields or the number of entries. For example, the estimated annual ongoing cost of reporting under the rule is approximately $4,400 for a representative tier 3 financial institution. Just over $2,300 of this annual ongoing cost is composed of fixed costs. As a comparison, each required data field accounts for approximately $43 of this annual ongoing cost. Thus, the closed-end reporting threshold provides a meaningful reduction in burden. Second, the Bureau is providing that financial institutions shall make available to the public notices that clearly convey that the institutions' disclosure statements and modified loan/application registers may be obtained on the Bureau's Web site. This approach relieves all financial institutions, including small entities, of the obligation to provide the disclosure statement and modified loan/ application register to the public directly. It also eliminates the risks to financial institutions from missing the publication deadline and from errors in preparing the modified loan/application register that could result in the unintended disclosure of data. The Bureau believes that these aspects of the final rule will be beneficial to small entities. Third, the Bureau adopted revisions to transactional coverage criteria that benefit small entities. As one example, the final rule eliminates reporting of non-dwelling-secured home improvement loans. This change reduces reporting burden to small entities to the extent that these entities offer such loans. As a second example, the overall change of transactional coverage to a dwelling-secured basis in the proposed rule would have extensively expanded reporting of commercial- purpose loans and lines of credit. In response to comments received about the cost impact of this proposal, some of which came from small entities, the Bureau decided to maintain Regulation C's existing purpose-based coverage test for commercial-purpose transactions, which maintains coverage of commercial-purpose transactions generally at existing levels. Fourth, and finally, the Bureau is making operational enhancements and modifications to improve the data submission process. For example, the Bureau is working to provide implementation support similar to the support provided for title XIV and TILA-RESPA Integrated Disclosure rules. The Bureau is also improving the geocoding process, creating a web-based HMDA data submission and edit-check system, developing a data-entry tool for small financial institutions that currently use Data Entry Software, and otherwise streamlining the submission and editing process to make it more efficient. All of these enhancements will improve the submission and processing of data, increase clarity, and reduce reporting burden. The section-by-section analysis, section 1022 analysis, and response to the comments from the Chief Counsel for Advocacy of the Small Business Administration, above, discuss the steps that the Bureau has considered and rejected, including adopting a higher loan-volume threshold and exempting small entities from the discretionary reporting requirements or from the reporting requirements altogether. G. Description of the Steps the Agency Has Taken To Minimize Any Additional Cost of Credit for Small Entities Section 603(d) of the RFA requires the Bureau to consult with small entities regarding the potential impact of the proposed rule on the cost of credit for small entities and related matters.\547\ To satisfy these statutory requirements, the Bureau provided notification to the Chief Counsel for Advocacy of the SBA in December 2013 that the Bureau would collect the advice and recommendations of the same small entity representatives identified in consultation with the Chief Counsel for Advocacy of the SBA through the Small Business Review Panel outreach concerning any projected impact of the proposed rule on the cost of credit for small entities, as well as any significant alternatives to the proposed rule which accomplish the stated objectives of applicable statutes and which minimize any increase in the cost of credit for small entities.\548\ The Bureau sought to collect the advice and recommendations of the small entity representatives during the Panel Outreach Meeting regarding these issues because, as small financial service providers, the small entity representatives could provide valuable input on any such impact related to the proposed rule.\549\ --------------------------------------------------------------------------- \547\ 5 U.S.C. 603(d). \548\ See 5 U.S.C. 603(d)(2). The Bureau provided this notification as part of the notification and other information provided to the Chief Counsel for Advocacy of the SBA with respect to the Small Business Review Panel outreach pursuant to RFA section 609(b)(1). \549\ See 5 U.S.C. 603(d)(2)(B). --------------------------------------------------------------------------- Following the Small Business Review Panel and as stated in the proposal, the Bureau believed that the rule would have a minimal impact on the cost of business credit. The small entity representatives had few comments on [[Page 66305]] the impact on the cost of business credit, but a few representatives noted that they would likely have to pass additional costs on to business customers. The Bureau noted that the proposed rule would cover certain dwelling-secured loans used for business purposes. As explained above, the final rule does not adopt the proposed expansion of reporting for commercial transactions. The final rule generally requires reporting of consumer-purpose mortgage loans, and exempts loans for a business or commercial purpose unless the loan is a home improvement loan, a home purchase loan, or a refinancing. Maintaining coverage of commercial loans at its current level will minimize the impact of the cost of credit for small entities. The Bureau expects any such increase to be minimal. IV. Paperwork Reduction Act Under the Paperwork Reduction Act of 1995 (PRA),\550\ Federal agencies are generally required to seek approval from the Office of Management and Budget (OMB) for information collection requirements prior to implementation. Further, the Bureau may not conduct or sponsor a collection of information unless OMB approves the collection under the PRA and displays a currently valid OMB control number. Notwithstanding any other provision of law, no person is required to comply with, or is subject to penalty for failure to comply with, a collection of information if the collection instrument does not display a currently valid OMB control number. The information collection requirements contained in Regulation C are currently approved by OMB under OMB control number 3170-0008. --------------------------------------------------------------------------- \550\ 44 U.S.C. 3501 et seq. --------------------------------------------------------------------------- On August 29, 2014, notice of the proposed rule was published in the Federal Register. The Bureau invited comment on: (1) Whether the proposed collection of information is necessary for the proper performance of the Bureau's functions, including whether the information has practical utility; (2) the accuracy of the Bureau's estimate of the burden of the proposed information collection; (3) ways to enhance the quality, utility, and clarity of the information to be collected; and (4) ways to minimize the burden of information collection on respondents, including through the use of automated collection techniques or other forms of information technology. The comment period for the proposal expired on October 29, 2014. The Bureau received almost no comments specifically addressing the PRA notice. One industry commenter noted that the proposal's total estimated burden of 4,700,000 hours per year, if divided evenly among all respondents, was 752 hours, or the equivalent to a full-time employee working 19 weeks. The commenter was concerned with the amount of burden represented by this figure. As the commenter acknowledged, 4,700,000 hours represented the total estimated burden hours imposed by the entire rule, not just the amended provisions, for all persons associated with all HMDA reporters. For any individual financial institution, the estimated burden hours may be far less than the 752- hour estimate derived by the commenter. For example, the Bureau estimates that the total annual burden of all reporting, recordkeeping, and third-party disclosure requirements for a tier 3 financial institution is approximately 134 hours per year. As described below, the final rule amends the information collection requirements contained in Regulation C.\551\ The information collection requirements currently contained in Regulation C remain in effect and are approved by OMB under OMB control number 3170-0008. This final rule contains information collection requirements that have not been approved by the OMB and, therefore, are not effective until OMB approval is obtained. The revised information collection requirements are contained in Sec. Sec. 1003.4 and 1003.5 of the final rule. The Bureau will publish a separate notice in the Federal Register announcing OMB's action on these submissions, which will include the OMB control number and expiration date. --------------------------------------------------------------------------- \551\ 12 CFR part 1003. --------------------------------------------------------------------------- The title of this information collection is Home Mortgage Disclosure (Regulation C). The frequency of response is annually, quarterly, and on-occasion. The Bureau's regulation will require financial institutions that meet certain thresholds to maintain data about originations and purchases of mortgage loans, as well as mortgage loan applications that do not result in originations, to update the information quarterly, and to report the information annually or quarterly. Financial institutions must also make certain information available to the public upon request. The information collection requirements in this final rule will be mandatory.\552\ Certain data fields will be removed or modified before they are made available to the public, as required by the statute and regulation. These removals or modifications will be determined through the Bureau's assessment under its balancing test of the benefits and risks created by the disclosure of loan-level HMDA data. The non- removed and unmodified data will be made publicly available and are not considered confidential. Data not made publicly available are considered confidential under the Bureau's confidentiality regulations, 12 CFR part 1070 et seq., and the Freedom of Information Act.\553\ The likely respondents will be financial institutions--specifically banks, savings associations, or credit unions (depository institutions), and for-profit mortgage-lending institutions (nondepository institutions)-- that meet the tests for coverage under Regulation C. These respondents will be required under the rule to maintain, disclose to the public, and report to Federal agencies, information regarding covered loans and applications for covered loans. --------------------------------------------------------------------------- \552\ See 12 U.S.C. 2801 et seq. \553\ 5 U.S.C. 552(b)(6). --------------------------------------------------------------------------- For the purposes of this PRA analysis, the Bureau estimates that, under the final rule, approximately 1,400 depository institutions that currently report HMDA data will no longer be required to report, and that approximately 75-450 nondepository institutions that currently do not report HMDA data will now be required to report. In 2013, approximately 7,200 financial institutions reported data under HMDA. The adopted coverage changes will reduce the number of reporters by an estimated 950 reporters for an estimated total of approximately 6,250. Under the final rule, the Bureau generally will account for the paperwork burden for all respondents under Regulation C. Using the Bureau's burden estimation methodology, which projects the estimated burden on several types of representative respondents to the entire market, the Bureau believes the total estimated industry burden for the approximately 6,250 respondents \554\ subject to the rule will be approximately 8,300,000 hours per year.\555\ The Bureau [[Page 66306]] expects that the amount of time required to implement each revision of the final rule for a given institution may vary based on the size, complexity, and practices of the respondent. --------------------------------------------------------------------------- \554\ The count of 6,250 is constructed as the number of HMDA reporters in 2013 (7,200) less the estimated 1,400 depository institutions that will no longer have to report under the adopted coverage rules plus the additional 75-450 estimated nondepository institutions that will have to begin reporting under the adopted coverage rules. \555\ The Bureau estimates that, for all HMDA reporters, the burden hours will be approximately 6,851,000 to 9,779,000 hours per year. 8,300,000 is approximately the mid-point of this estimated range. These burden hour estimates include reporting of closed-end mortgage loans, open-end lines of credit, and quarterly reporting. --------------------------------------------------------------------------- In 2013, a total of 145 financial institutions reported HMDA data to the Bureau. Currently, only depository institutions with over $10 billion in assets and their affiliates report their HMDA data to the Bureau. Using 2013 loan/application register sizes as a proxy to assign these 145 financial institutions into tiers yields 84, 39, and 22 tier 1, 2, and 3 financial institutions, respectively.\556\ The Bureau estimates that the current time burden for the Bureau reporters is approximately 690,000 hours per year. Eighteen of these 145 institutions reported over 60,000 HMDA loan/application register records and will therefore be required to report data quarterly. An estimated 74 of these 145 institutions would exceed the open-end reporting threshold of 100 open-end lines of credit. Including the modifications to the information collection requirements contained in the final rule, and the operations modernization measures, the Bureau estimates that the burden for annual and quarterly Bureau reporters will be 1,089,000 and 300,000 hours per year, respectively, for a total estimated burden hours of 1,389,000 per year. This represents an increase of approximately 699,000 burden hours over the estimated burden under the current rule. --------------------------------------------------------------------------- \556\ The Bureau's estimation methodology is fully described in the section 1022 analysis in part VII, above. --------------------------------------------------------------------------- A. Information Collection Requirements 557 --------------------------------------------------------------------------- \557\ A detailed analysis of the burdens and costs described in this part can be found in the Paperwork Reduction Act Supporting Statement that corresponds to this final rule. The Supporting Statement is available at www.reginfo.gov. --------------------------------------------------------------------------- The Bureau believes that the following aspects of the final rule are information collection requirements under the PRA: (1) The requirement that financial institutions maintain copies of their submitted annual loan/application register information for three years and record information regarding reportable transactions for the first three quarters of the calendar year on a quarterly basis; (2) the requirement that financial institutions report HMDA data annually--and, in the case of financial institutions that reported for the preceding calendar year at least 60,000 covered loans and applications, combined, excluding purchased covered loans, for the first three quarters of the calendar year on a quarterly basis--to the appropriate Federal agency; and (3) the requirement that financial institutions provide notices that clearly convey that disclosure statements and modified loan/ application registers may be obtained on the Bureau's Web site and maintain notices of availability of modified loan/application registers for three years and notices of availability of disclosure statements for five years. 1. Recordkeeping Requirements Financial institutions are required to maintain a copy of both the submitted annual loan/application register and a notice of its availability for three years. However, financial institutions no longer have to maintain the modified loan/application register. Similarly, financial institutions are required to maintain the notice of availability of their disclosure statements for five years, but no longer have to maintain the disclosure statements themselves. Therefore, the final rule includes changes that both increase and decrease the documentation or non-data-specific information that financial institutions will have to maintain. The Bureau believes that the net impact of these changes on recordkeeping requirements is minimal. In addition to recordkeeping requirements related to the loan/ application register and disclosure statements, the rule increases the number of data fields, and possibly the number of records, that financial institutions are required to gather and report. The Bureau estimates that the current time burden of reporting for Bureau reporters is approximately 296,000 hours per year. The Bureau estimates that, with the final amendments and the operations modernization, the time burden for annual and quarterly Bureau reporters will be approximately 417,000 and 112,000 hours per year, respectively, for a total estimate of approximately 529,000 burden hours per year. This represents an increase of approximately 233,000 burden hours over the estimated burden under the current rule. 2. Reporting Requirements HMDA is a data reporting statute, so most provisions of the rule affect reporting requirements, as described above. Specifically, financial institutions are required annually to report HMDA data to the Bureau or to the appropriate Federal agency,\558\ and all reportable transactions must be recorded on a loan/application register within 30 calendar days\559\ after the end of the calendar quarter in which final action is taken. Additionally, financial institutions that reported for the preceding calendar year at least 60,000 covered loans and applications, combined, excluding purchased covered loans, will be required to report HMDA data for the first three quarters of the calendar year on a quarterly basis to the Bureau or the appropriate Federal agency. --------------------------------------------------------------------------- \558\ 12 U.S.C. 2803(h)(1). \559\ 12 CFR 1003.4(a). --------------------------------------------------------------------------- The Bureau estimates that the current time burden of reporting for Bureau reporters is approximately 391,000 hours per year. The Bureau estimates that, with the final amendments and the operations modernization, the time burden for annual and quarterly Bureau reporters will be approximately 671,000 and 188,000 hours per year, respectively, for a total estimate of approximately 859,000 burden hours per year. This represents an increase of approximately 468,000 burden hours over the estimated burden under the current rule. 3. Disclosure Requirements The final rule modifies Regulation C's requirements for financial institutions to disclose information to the public. Under the final rule, a financial institution will no longer be required to make available to the public the modified loan/application register itself but must instead make available a notice informing the public that the institution's modified loan/application register may be obtained on the Bureau's Web site. Additionally, the final rule will require financial institutions to make available to the public their disclosure statements by making available a notice that clearly conveys that the disclosure statement may be obtained on the Bureau's Web site and that includes the Bureau's Web site address. The Bureau estimates that the current time burden of disclosure for Bureau reporters is approximately 2,700 hours per year. The Bureau estimates that, with the final amendments and the operations modernization, the time burden for annual and quarterly Bureau reporters will be approximately 360 and 100 hours per year, respectively, for a total estimate of approximately 460 burden hours per year. This represents a decrease of approximately 2,240 burden hours from the estimated burden under the current rule. Burden hours have fallen here because financial institutions will no longer have to make their modified loan/application register [[Page 66307]] or disclosure statements available to the public. 4. One-Time Costs Associated With the Adopted Information Collections Financial institutions' management, legal, and compliance personnel will likely take time to learn new reporting requirements and assess legal and compliance risks. Financial institutions that use vendors for HMDA compliance will incur one-time costs associated with software installation, troubleshooting, and testing. The Bureau is aware that these activities will require time and that the costs may be sensitive to the time available for them. Financial institutions that maintain their own reporting systems will incur one-time costs to develop, prepare, and implement necessary modifications to those systems. In all cases, financial institutions will need to update training materials to reflect new requirements and activities and may incur certain one-time costs for providing initial training to current employees. For current HMDA reporters, the Bureau estimates that the final rule will impose average one-time costs of $3,000 for tier 3 financial institutions, $250,000 for tier 2 financial institutions, and $800,000 for tier 1 financial institutions without considering the expansion of transactional coverage to include additional open-end lines of credit and reverse mortgages.\560\ Including the estimated one-time costs to modify processes and systems for home-equity products, the Bureau estimates that the total one-time costs will be $3,000 for tier 3 institutions, $375,000 for tier 2 institutions, and $1,200,000 for tier 1 institutions. This yields an overall estimated market impact of between $725,900,000 and $1,339,000,000. Using a 7 percent discount rate and a five-year amortization window, the annualized one-time, additional cost is $177,000,000 to $326,600,000. --------------------------------------------------------------------------- \560\ The Bureau realizes that the impact to one-time costs varies by institution due to many factors, such as size, operational structure, and product complexity, and that this variance exists on a continuum that is impossible to fully capture. As a result, the one-time cost estimates will be high for some financial institutions and low for others. --------------------------------------------------------------------------- The revisions to the institutional coverage criteria will require an estimated 75-450 nondepository institutions that are currently not reporting under HMDA to begin reporting. These nondepository institutions will incur start-up costs to develop policies and procedures, infrastructure, and training. Based on outreach discussions with financial institutions prior to the proposal, the Bureau believes that these start-up costs will be approximately $25,000 for tier 3 financial institutions. Although origination volumes for these 75-450 nondepository institutions are slightly higher, the Bureau still expects most of these nondepository institutions to be tier 3 financial institutions. Under this assumption, the estimated overall market cost will be $11,300,000 (= 450 * $25,000). B. Summary of Burden Hours The tables below summarize the estimated annual burdens under Regulation C associated with the information collections described above for Bureau reporters and all HMDA reporters, respectively. The tables combine all three aspects of information collection: Reporting, recordkeeping, and disclosure requirements. The Paperwork Reduction Act Supporting Statement that corresponds with this final rule provides more information as to how these estimates were derived and further detail regarding the burden hours associated with each information collection. The first table presents burden hour estimates for financial institutions that report HMDA data to the Bureau, and the second table provides information for all HMDA reporters. [GRAPHIC] [TIFF OMITTED] TR28OC15.009 [[Page 66308]] [GRAPHIC] [TIFF OMITTED] TR28OC15.010 List of Subjects in 12 CFR Part 1003 Banks, Banking, Credit unions, Mortgages, National banks, Savings associations, Reporting and recordkeeping requirements. Authority and Issuance For the reasons set forth above, the Bureau amends Regulation C, 12 CFR part 1003, as set forth below: PART 1003--HOME MORTGAGE DISCLOSURE (REGULATION C) 0 1. The authority citation for part 1003 continues to read as follows: Authority: 12 U.S.C. 2803, 2804, 2805, 5512, 5581. 0 2. Effective January 1, 2018, Sec. 1003.1 is amended by revising paragraph (c) to read as follows: Sec. 1003.1 Authority, purpose, and scope. * * * * * (c) Scope. This part applies to financial institutions as defined in Sec. 1003.2(g). This part requires a financial institution to submit data to the appropriate Federal agency for the financial institution as defined in Sec. 1003.5(a)(4), and to disclose certain data to the public, about covered loans for which the financial institution receives applications, or that it originates or purchases, and that are secured by a dwelling located in a State of the United States of America, the District of Columbia, or the Commonwealth of Puerto Rico. 0 3. Effective January 1, 2017, Sec. 1003.2 is amended by revising paragraph (1)(iii) and adding paragraph (1)(v) to the definition of ``financial institution'' to read as follows: Sec. 1003.2 Definitions. * * * * * Financial institution means: (1) * * * (iii) In the preceding calendar year, originated at least one home purchase loan (excluding temporary financing such as a construction loan) or refinancing of a home purchase loan, secured by a first lien on a one- to four-family dwelling; * * * * * (v) In each of the two preceding calendar years, originated at least 25 home purchase loans, including refinancings of home purchase loans, that are not excluded from this part pursuant to Sec. 1003.4(d); and * * * * * 0 4. Effective January 1, 2018, Sec. 1003.2 is revised to read as follows: Sec. 1003.2 Definitions. In this part: (a) Act means the Home Mortgage Disclosure Act (HMDA) (12 U.S.C. 2801 et seq.), as amended. (b) Application--(1) In general. Application means an oral or written request for a covered loan that is made in accordance with procedures used by a financial institution for the type of credit requested. (2) Preapproval programs. A request for preapproval for a home purchase loan, other than a home purchase loan that will be an open-end line of credit, a reverse mortgage, or secured by a multifamily dwelling, is an application under this section if the request is reviewed under a program in which the financial institution, after a comprehensive analysis of the creditworthiness of the applicant, issues a written commitment to the applicant valid for a designated period of time to extend a home purchase loan up to a specified amount. The written [[Page 66309]] commitment may not be subject to conditions other than: (i) Conditions that require the identification of a suitable property; (ii) Conditions that require that no material change has occurred in the applicant's financial condition or creditworthiness prior to closing; and (iii) Limited conditions that are not related to the financial condition or creditworthiness of the applicant that the financial institution ordinarily attaches to a traditional home mortgage application. (c) Branch office means: (1) Any office of a bank, savings association, or credit union that is considered a branch by the Federal or State supervisory agency applicable to that institution, excluding automated teller machines and other free-standing electronic terminals; and (2) Any office of a for-profit mortgage-lending institution (other than a bank, savings association, or credit union) that takes applications from the public for covered loans. A for-profit mortgage- lending institution (other than a bank, savings association, or credit union) is also deemed to have a branch office in an MSA or in an MD, if, in the preceding calendar year, it received applications for, originated, or purchased five or more covered loans related to property located in that MSA or MD, respectively. (d) Closed-end mortgage loan means an extension of credit that is secured by a lien on a dwelling and that is not an open-end line of credit under paragraph (o) of this section. (e) Covered loan means a closed-end mortgage loan or an open-end line of credit that is not an excluded transaction under Sec. 1003.3(c). (f) Dwelling means a residential structure, whether or not attached to real property. The term includes but is not limited to a detached home, an individual condominium or cooperative unit, a manufactured home or other factory-built home, or a multifamily residential structure or community. (g) Financial institution means a depository financial institution or a nondepository financial institution, where: (1) Depository financial institution means a bank, savings association, or credit union that: (i) On the preceding December 31 had assets in excess of the asset threshold established and published annually by the Bureau for coverage by the Act, based on the year-to-year change in the average of the Consumer Price Index for Urban Wage Earners and Clerical Workers, not seasonally adjusted, for each twelve month period ending in November, with rounding to the nearest million; (ii) On the preceding December 31, had a home or branch office in an MSA; (iii) In the preceding calendar year, originated at least one home purchase loan or refinancing of a home purchase loan, secured by a first lien on a one- to four-unit dwelling; (iv) Meets one or more of the following two criteria: (A) The institution is federally insured or regulated; or (B) Any loan referred to in paragraph (g)(1)(iii) of this section was insured, guaranteed, or supplemented by a Federal agency, or was intended by the institution for sale to the Federal National Mortgage Association or the Federal Home Loan Mortgage Corporation; and (v) Meets at least one of the following criteria: (A) In each of the two preceding calendar years, originated at least 25 closed-end mortgage loans that are not excluded from this part pursuant to Sec. 1003.3(c)(1) through (10); or (B) In each of the two preceding calendar years, originated at least 100 open-end lines of credit that are not excluded from this part pursuant to Sec. 1003.3(c)(1) through (10); and (2) Nondepository financial institution means a for-profit mortgage-lending institution (other than a bank, savings association, or credit union) that: (i) On the preceding December 31, had a home or branch office in an MSA; and (ii) Meets at least one of the following criteria: (A) In each of the two preceding calendar years, originated at least 25 closed-end mortgage loans that are not excluded from this part pursuant to Sec. 1003.3(c)(1) through (10); or (B) In each of the two preceding calendar years, originated at least 100 open-end lines of credit that are not excluded from this part pursuant to Sec. 1003.3(c)(1) through (10). (h) [Reserved] (i) Home improvement loan means a closed-end mortgage loan or an open-end line of credit that is for the purpose, in whole or in part, of repairing, rehabilitating, remodeling, or improving a dwelling or the real property on which the dwelling is located. (j) Home purchase loan means a closed-end mortgage loan or an open- end line of credit that is for the purpose, in whole or in part, of purchasing a dwelling. (k) Loan/Application Register means both the record of information required to be collected pursuant to Sec. 1003.4 and the record submitted annually or quarterly, as applicable, pursuant to Sec. 1003.5(a). (l) Manufactured home means any residential structure as defined under regulations of the U.S. Department of Housing and Urban Development establishing manufactured home construction and safety standards (24 CFR 3280.2). For purposes of Sec. 1003.4(a)(5), the term also includes a multifamily dwelling that is a manufactured home community. (m) Metropolitan Statistical Area (MSA) and Metropolitan Division (MD). (1) Metropolitan Statistical Area or MSA means a Metropolitan Statistical Area as defined by the U.S. Office of Management and Budget. (2) Metropolitan Division (MD) means a Metropolitan Division of an MSA, as defined by the U.S. Office of Management and Budget. (n) Multifamily dwelling means a dwelling, regardless of construction method, that contains five or more individual dwelling units. (o) Open-end line of credit means an extension of credit that: (1) Is secured by a lien on a dwelling; and (2) Is an open-end credit plan as defined in Regulation Z, 12 CFR 1026.2(a)(20), but without regard to whether the credit is consumer credit, as defined in Sec. 1026.2(a)(12), is extended by a creditor, as defined in Sec. 1026.2(a)(17), or is extended to a consumer, as defined in Sec. 1026.2(a)(11). (p) Refinancing means a closed-end mortgage loan or an open-end line of credit in which a new, dwelling-secured debt obligation satisfies and replaces an existing, dwelling-secured debt obligation by the same borrower. (q) Reverse mortgage means a closed-end mortgage loan or an open- end line of credit that is a reverse mortgage transaction as defined in Regulation Z, 12 CFR 1026.33(a), but without regard to whether the security interest is created in a principal dwelling. 0 5. Effective January 1, 2018, Sec. 1003.3 is amended by revising the heading and adding paragraph (c) to read as follows: Sec. 1003.3 Exempt institutions and excluded transactions. * * * * * (c) Excluded transactions. The requirements of this part do not apply to: (1) A closed-end mortgage loan or open-end line of credit originated or purchased by a financial institution acting in a fiduciary capacity; (2) A closed-end mortgage loan or open-end line of credit secured by a lien on unimproved land; [[Page 66310]] (3) Temporary financing; (4) The purchase of an interest in a pool of closed-end mortgage loans or open-end lines of credit; (5) The purchase solely of the right to service closed-end mortgage loans or open-end lines of credit; (6) The purchase of closed-end mortgage loans or open-end lines of credit as part of a merger or acquisition, or as part of the acquisition of all of the assets and liabilities of a branch office as defined in Sec. 1003.2(c); (7) A closed-end mortgage loan or open-end line of credit, or an application for a closed-end mortgage loan or open-end line of credit, for which the total dollar amount is less than $500; (8) The purchase of a partial interest in a closed-end mortgage loan or open-end line of credit; (9) A closed-end mortgage loan or open-end line of credit used primarily for agricultural purposes; (10) A closed-end mortgage loan or open-end line of credit that is or will be made primarily for a business or commercial purpose, unless the closed-end mortgage loan or open-end line of credit is a home improvement loan under Sec. 1003.2(i), a home purchase loan under Sec. 1003.2(j), or a refinancing under Sec. 1003.2(p); (11) A closed-end mortgage loan, if the financial institution originated fewer than 25 closed-end mortgage loans in each of the two preceding calendar years; or (12) An open-end line of credit, if the financial institution originated fewer than 100 open-end lines of credit in each of the two preceding calendar years. 0 6. Effective January 1, 2018, Sec. 1003.4 is revised to read as follows: Sec. 1003.4 Compilation of reportable data. (a) Data format and itemization. A financial institution shall collect data regarding applications for covered loans that it receives, covered loans that it originates, and covered loans that it purchases for each calendar year. A financial institution shall collect data regarding requests under a preapproval program, as defined in Sec. 1003.2(b)(2), only if the preapproval request is denied, is approved by the financial institution but not accepted by the applicant, or results in the origination of a home purchase loan. The data collected shall include the following items: (1)(i) A universal loan identifier (ULI) for the covered loan or application that can be used to identify and retrieve the covered loan or application file. Except for a purchased covered loan or application described in paragraphs (a)(1)(i)(D) and (E) of this section, the financial institution shall assign and report a ULI that: (A) Begins with the financial institution's Legal Entity Identifier (LEI) that is issued by: (1) A utility endorsed by the LEI Regulatory Oversight Committee; or (2) A utility endorsed or otherwise governed by the Global LEI Foundation (GLEIF) (or any successor of the GLEIF) after the GLEIF assumes operational governance of the global LEI system. (B) Follows the LEI with up to 23 additional characters to identify the covered loan or application, which: (1) May be letters, numerals, or a combination of letters and numerals; (2) Must be unique within the financial institution; and (3) Must not include any information that could be used to directly identify the applicant or borrower; and (C) Ends with a two-character check digit, as prescribed in appendix C to this part. (D) For a purchased covered loan that any financial institution has previously assigned or reported with a ULI under this part, the financial institution that purchases the covered loan must use the ULI that was assigned or previously reported for the covered loan. (E) For an application that was previously reported with a ULI under this part and that results in an origination during the same calendar year that is reported in a subsequent reporting period pursuant to Sec. 1003.5(a)(1)(ii), the financial institution may report the same ULI for the origination that was previously reported for the application. (ii) Except for purchased covered loans, the date the application was received or the date shown on the application form. (2) Whether the covered loan is, or in the case of an application would have been, insured by the Federal Housing Administration, guaranteed by the Veterans Administration, or guaranteed by the Rural Housing Service or the Farm Service Agency. (3) Whether the covered loan is, or the application is for, a home purchase loan, a home improvement loan, a refinancing, a cash-out refinancing, or for a purpose other than home purchase, home improvement, refinancing, or cash-out refinancing. (4) Whether the application or covered loan involved a request for a preapproval of a home purchase loan under a preapproval program. (5) Whether the construction method for the dwelling related to the property identified in paragraph (a)(9) of this section is site-built or a manufactured home. (6) Whether the property identified in paragraph (a)(9) of this section is or will be used by the applicant or borrower as a principal residence, as a second residence, or as an investment property. (7) The amount of the covered loan or the amount applied for, as applicable. (i) For a closed-end mortgage loan, other than a purchased loan, an assumption, or a reverse mortgage, the amount to be repaid as disclosed on the legal obligation. For a purchased closed-end mortgage loan or an assumption of a closed-end mortgage loan, the unpaid principal balance at the time of purchase or assumption. (ii) For an open-end line of credit, other than a reverse mortgage open-end line of credit, the amount of credit available to the borrower under the terms of the plan. (iii) For a reverse mortgage, the initial principal limit, as determined pursuant to section 255 of the National Housing Act (12 U.S.C. 1715z-20) and implementing regulations and mortgagee letters issued by the U.S. Department of Housing and Urban Development. (8) The following information about the financial institution's action: (i) The action taken by the financial institution, recorded as one of the following: (A) Whether a covered loan was originated or purchased; (B) Whether an application for a covered loan that did not result in the origination of a covered loan was approved but not accepted, denied, withdrawn by the applicant, or closed for incompleteness; and (C) Whether a preapproval request that did not result in the origination of a home purchase loan was denied or approved but not accepted. (ii) The date of the action taken by the financial institution. (9) The following information about the location of the property securing the covered loan or, in the case of an application, proposed to secure the covered loan: (i) The property address; and (ii) If the property is located in an MSA or MD in which the financial institution has a home or branch office, or if the institution is subject to paragraph (e) of this section, the location of the property by: (A) State; (B) County; and (C) Census tract if the property is located in a county with a population of more than 30,000 according to the most recent decennial census conducted by the U.S. Census Bureau. [[Page 66311]] (10) The following information about the applicant or borrower: (i) Ethnicity, race, and sex, and whether this information was collected on the basis of visual observation or surname; (ii) Age; and (iii) Except for covered loans or applications for which the credit decision did not consider or would not have considered income, the gross annual income relied on in making the credit decision or, if a credit decision was not made, the gross annual income relied on in processing the application. (11) The type of entity purchasing a covered loan that the financial institution originates or purchases and then sells within the same calendar year. (12)(i) For covered loans subject to Regulation Z, 12 CFR part 1026, other than assumptions, purchased covered loans, and reverse mortgages, the difference between the covered loan's annual percentage rate and the average prime offer rate for a comparable transaction as of the date the interest rate is set. (ii) ``Average prime offer rate'' means an annual percentage rate that is derived from average interest rates, points, and other loan pricing terms currently offered to consumers by a representative sample of creditors for mortgage loans that have low-risk pricing characteristics. The Bureau publishes average prime offer rates for a broad range of types of transactions in tables updated at least weekly, as well as the methodology the Bureau uses to derive these rates. (13) For covered loans subject to the Home Ownership and Equity Protection Act of 1994, as implemented in Regulation Z, 12 CFR 1026.32, whether the covered loan is a high-cost mortgage under Regulation Z, 12 CFR 1026.32(a). (14) The lien status (first or subordinate lien) of the property identified under paragraph (a)(9) of this section. (15)(i) Except for purchased covered loans, the credit score or scores relied on in making the credit decision and the name and version of the scoring model used to generate each credit score. (ii) For purposes of this paragraph (a)(15), ``credit score'' has the meaning set forth in 15 U.S.C. 1681g(f)(2)(A). (16) The principal reason or reasons the financial institution denied the application, if applicable. (17) For covered loans subject to Regulation Z, 12 CFR 1026.43(c), the following information: (i) If a disclosure is provided for the covered loan pursuant to Regulation Z, 12 CFR 1026.19(f), the amount of total loan costs, as disclosed pursuant to Regulation Z, 12 CFR 1026.38(f)(4); or (ii) If the covered loan is not subject to the disclosure requirements in Regulation Z, 12 CFR 1026.19(f), and is not a purchased covered loan, the total points and fees charged in connection with the covered loan, expressed in dollars and calculated pursuant to Regulation Z, 12 CFR 1026.32(b)(1). (18) For covered loans subject to the disclosure requirements in Regulation Z, 12 CFR 1026.19(f), the total of all itemized amounts that are designated borrower-paid at or before closing, as disclosed pursuant to Regulation Z, 12 CFR 1026.38(f)(1). (19) For covered loans subject to the disclosure requirements in Regulation Z, 12 CFR 1026.19(f), the points paid to the creditor to reduce the interest rate, expressed in dollars, as described in Regulation Z, 12 CFR 1026.37(f)(1)(i), and disclosed pursuant to Regulation Z, 12 CFR 1026.38(f)(1). (20) For covered loans subject to the disclosure requirements in Regulation Z, 12 CFR 1026.19(f), the amount of lender credits, as disclosed pursuant to Regulation Z, 12 CFR 1026.38(h)(3). (21) The interest rate applicable to the approved application, or to the covered loan at closing or account opening. (22) For covered loans or applications subject to Regulation Z, 12 CFR part 1026, other than reverse mortgages or purchased covered loans, the term in months of any prepayment penalty, as defined in Regulation Z, 12 CFR 1026.32(b)(6)(i) or (ii), as applicable. (23) Except for purchased covered loans, the ratio of the applicant's or borrower's total monthly debt to the total monthly income relied on in making the credit decision. (24) Except for purchased covered loans, the ratio of the total amount of debt secured by the property to the value of the property relied on in making the credit decision. (25) The scheduled number of months after which the legal obligation will mature or terminate or would have matured or terminated. (26) The number of months, or proposed number of months in the case of an application, until the first date the interest rate may change after closing or account opening. (27) Whether the contractual terms include or would have included any of the following: (i) A balloon payment as defined in Regulation Z, 12 CFR 1026.18(s)(5)(i); (ii) Interest-only payments as defined in Regulation Z, 12 CFR 1026.18(s)(7)(iv); (iii) A contractual term that would cause the covered loan to be a negative amortization loan as defined in Regulation Z, 12 CFR 1026.18(s)(7)(v); or (iv) Any other contractual term that would allow for payments other than fully amortizing payments, as defined in Regulation Z, 12 CFR 1026.43(b)(2), during the loan term, other than the contractual terms described in this paragraph (a)(27)(i), (ii), and (iii). (28) The value of the property securing the covered loan or, in the case of an application, proposed to secure the covered loan relied on in making the credit decision. (29) If the dwelling related to the property identified in paragraph (a)(9) of this section is a manufactured home and not a multifamily dwelling, whether the covered loan is, or in the case of an application would have been, secured by a manufactured home and land, or by a manufactured home and not land. (30) If the dwelling related to the property identified in paragraph (a)(9) of this section is a manufactured home and not a multifamily dwelling, whether the applicant or borrower: (i) Owns the land on which it is or will be located or, in the case of an application, did or would have owned the land on which it would have been located, through a direct or indirect ownership interest; or (ii) Leases or, in the case of an application, leases or would have leased the land through a paid or unpaid leasehold. (31) The number of individual dwelling units related to the property securing the covered loan or, in the case of an application, proposed to secure the covered loan. (32) If the property securing the covered loan or, in the case of an application, proposed to secure the covered loan includes a multifamily dwelling, the number of individual dwelling units related to the property that are income-restricted pursuant to Federal, State, or local affordable housing programs. (33) Except for purchased covered loans, the following information about the application channel of the covered loan or application: (i) Whether the applicant or borrower submitted the application for the covered loan directly to the financial institution; and (ii) Whether the obligation arising from the covered loan was, or in the case of an application, would have been initially payable to the financial institution. (34) For a covered loan or application, the unique identifier assigned by the Nationwide Mortgage Licensing System and Registry for the mortgage loan [[Page 66312]] originator, as defined in Regulation G, 12 CFR 1007.102, or Regulation H, 12 CFR 1008.23, as applicable. (35)(i) Except for purchased covered loans, the name of the automated underwriting system used by the financial institution to evaluate the application and the result generated by that automated underwriting system. (ii) For purposes of this paragraph (a)(35), an ``automated underwriting system'' means an electronic tool developed by a securitizer, Federal government insurer, or Federal government guarantor that provides a result regarding the credit risk of the applicant and whether the covered loan is eligible to be originated, purchased, insured, or guaranteed by that securitizer, Federal government insurer, or Federal government guarantor. (36) Whether the covered loan is, or the application is for, a reverse mortgage. (37) Whether the covered loan is, or the application is for, an open-end line of credit. (38) Whether the covered loan is, or the application is for a covered loan that will be, made primarily for a business or commercial purpose. (b) Collection of data on ethnicity, race, sex, age, and income. (1) A financial institution shall collect data about the ethnicity, race, and sex of the applicant or borrower as prescribed in appendix B to this part. (2) Ethnicity, race, sex, age, and income data may but need not be collected for covered loans purchased by a financial institution. (c)-(d) [Reserved] (e) Data reporting for banks and savings associations that are required to report data on small business, small farm, and community development lending under CRA. Banks and savings associations that are required to report data on small business, small farm, and community development lending under regulations that implement the Community Reinvestment Act of 1977 (12 U.S.C. 2901 et seq.) shall also collect the information required by paragraph 4(a)(9) of this section for property located outside MSAs and MDs in which the institution has a home or branch office, or outside any MSA. (f) Quarterly recording of data. A financial institution shall record the data collected pursuant to this section on a loan/ application register within 30 calendar days after the end of the calendar quarter in which final action is taken (such as origination or purchase of a covered loan, sale of a covered loan in the same calendar year it is originated or purchased, or denial or withdrawal of an application). 0 7. Effective January 1, 2018, Sec. 1003.5 is amended by revising paragraphs (b) through (f) to read as follows: Sec. 1003.5 Disclosure and reporting. * * * * * (b) Disclosure statement. (1) The Federal Financial Institutions Examination Council (FFIEC) will make available a disclosure statement based on the data each financial institution submits for the preceding calendar year pursuant to paragraph (a) of this section. (2) No later than three business days after receiving notice from the FFIEC that a financial institution's disclosure statement is available, the financial institution shall make available to the public upon request at its home office, and each branch office physically located in each MSA and each MD, a written notice that clearly conveys that the institution's disclosure statement may be obtained on the Bureau's Web site at www.consumerfinance.gov/hmda. (c) Modified loan/application register. (1) A financial institution shall make available to the public upon request at its home office, and each branch office physically located in each MSA and each MD, a written notice that clearly conveys that the institution's loan/ application register, as modified by the Bureau to protect applicant and borrower privacy, may be obtained on the Bureau's Web site at www.consumerfinance.gov/hmda. (2) A financial institution shall make available the notice required by paragraph (c)(1) of this section following the calendar year for which the data are collected. (d) Availability of written notices. (1) A financial institution shall make the notice required by paragraph (c) of this section available to the public for a period of three years and the notice required by paragraph (b)(2) of this section available to the public for a period of five years. An institution shall make these notices available during the hours the office is normally open to the public for business. (2) A financial institution may make available to the public, at its discretion and in addition to the written notices required by paragraphs (b)(2) or (c)(1) of this section, as applicable, its disclosure statement or its loan/application register, as modified by the Bureau to protect applicant and borrower privacy. A financial institution may impose a reasonable fee for any cost incurred in providing or reproducing these data. (e) Posted notice of availability of data. A financial institution shall post a general notice about the availability of its HMDA data in the lobby of its home office and of each branch office physically located in each MSA and each MD. This notice must clearly convey that the institution's HMDA data is available on the Bureau's Web site at www.consumerfinance.gov/hmda. (f) Aggregated data. Using data submitted by financial institutions pursuant to paragraph (a) of this section, the FFIEC will make available aggregate data for each MSA and MD, showing lending patterns by property location, age of housing stock, and income level, sex, ethnicity, and race. 0 8. Effective January 1, 2019, Sec. 1003.5 is revised to read as follows: Sec. 1003.5 Disclosure and reporting. (a) Reporting to agency. (1)(i) Annual reporting. By March 1 following the calendar year for which data are collected and recorded as required by Sec. 1003.4, a financial institution shall submit its annual loan/application register in electronic format to the appropriate Federal agency at the address identified by such agency. An authorized representative of the financial institution with knowledge of the data submitted shall certify to the accuracy and completeness of data submitted pursuant to this paragraph (a)(1)(i). The financial institution shall retain a copy of its annual loan/application register submitted pursuant to this paragraph (a)(1)(i) for its records for at least three years. (ii) [Reserved] (iii) When the last day for submission of data prescribed under this paragraph (a)(1) falls on a Saturday or Sunday, a submission shall be considered timely if it is submitted on the next succeeding Monday. (2) A financial institution that is a subsidiary of a bank or savings association shall complete a separate loan/application register. The subsidiary shall submit the loan/application register, directly or through its parent, to the appropriate Federal agency for the subsidiary's parent at the address identified by the agency. (3) A financial institution shall provide with its submission: (i) Its name; (ii) The calendar year the data submission covers pursuant to paragraph (a)(1)(i) of this section or calendar quarter and year the data submission covers pursuant to paragraph (a)(1)(ii) of this section; (iii) The name and contact information of a person who may be contacted with questions about the institution's submission; (iv) Its appropriate Federal agency; (v) The total number of entries contained in the submission; [[Page 66313]] (vi) Its Federal Taxpayer Identification number; and (vii) Its Legal Entity Identifier (LEI) as described in Sec. 1003.4(a)(1)(i)(A). (4) For purposes of paragraph (a) of this section, ``appropriate Federal agency'' means the appropriate agency for the financial institution as determined pursuant to section 304(h)(2) of the Home Mortgage Disclosure Act (12 U.S.C. 2803(h)(2)) or, with respect to a financial institution subject to the Bureau's supervisory authority under section 1025(a) of the Consumer Financial Protection Act of 2010 (12 U.S.C. 5515(a)), the Bureau. (5) Procedures for the submission of data pursuant to paragraph (a) of this section are available at www.consumerfinance.gov/hmda. (b) Disclosure statement. (1) The Federal Financial Institutions Examination Council (FFIEC) will make available a disclosure statement based on the data each financial institution submits for the preceding calendar year pursuant to paragraph (a)(1)(i) of this section. (2) No later than three business days after receiving notice from the FFIEC that a financial institution's disclosure statement is available, the financial institution shall make available to the public upon request at its home office, and each branch office physically located in each MSA and each MD, a written notice that clearly conveys that the institution's disclosure statement may be obtained on the Bureau's Web site at www.consumerfinance.gov/hmda. (c) Modified loan/application register. (1) A financial institution shall make available to the public upon request at its home office, and each branch office physically located in each MSA and each MD, a written notice that clearly conveys that the institution's loan/ application register, as modified by the Bureau to protect applicant and borrower privacy, may be obtained on the Bureau's Web site at www.consumerfinance.gov/hmda. (2) A financial institution shall make available the notice required by paragraph (c)(1) of this section following the calendar year for which the data are collected. (d) Availability of written notices. (1) A financial institution shall make the notice required by paragraph (c) of this section available to the public for a period of three years and the notice required by paragraph (b)(2) of this section available to the public for a period of five years. An institution shall make these notices available during the hours the office is normally open to the public for business. (2) A financial institution may make available to the public, at its discretion and in addition to the written notices required by paragraphs (b)(2) or (c)(1) of this section, as applicable, its disclosure statement or its loan/application register, as modified by the Bureau to protect applicant and borrower privacy. A financial institution may impose a reasonable fee for any cost incurred in providing or reproducing these data. (e) Posted notice of availability of data. A financial institution shall post a general notice about the availability of its HMDA data in the lobby of its home office and of each branch office physically located in each MSA and each MD. This notice must clearly convey that the institution's HMDA data is available on the Bureau's Web site at www.consumerfinance.gov/hmda. (f) Aggregated data. Using data submitted by financial institutions pursuant to paragraph (a)(1)(i) of this section, the FFIEC will make available aggregate data for each MSA and MD, showing lending patterns by property location, age of housing stock, and income level, sex, ethnicity, and race. 0 9. Effective January 1, 2020, Sec. 1003.5 is amended by adding paragraph (a)(1)(ii) to read as follows: Sec. 1003.5 Disclosure and reporting. (a) * * * (1) * * * (ii) Quarterly reporting. Within 60 calendar days after the end of each calendar quarter except the fourth quarter, a financial institution that reported for the preceding calendar year at least 60,000 covered loans and applications, combined, excluding purchased covered loans, shall submit to the appropriate Federal agency its loan/ application register containing all data required to be recorded for that quarter pursuant to Sec. 1003.4(f). The financial institution shall submit its quarterly loan/application register pursuant to this paragraph (a)(1)(ii) in electronic format at the address identified by the appropriate Federal agency for the institution. * * * * * 0 10. Effective January 1, 2019, Sec. 1003.6 is revised to read as follows: Sec. 1003.6 Enforcement. (a) Administrative enforcement. A violation of the Act or this part is subject to administrative sanctions as provided in section 305 of the Act (12 U.S.C. 2804), including the imposition of civil money penalties, where applicable. Compliance is enforced by the agencies listed in section 305 of the Act. (b) Bona fide errors. (1) An error in compiling or recording data for a covered loan or application is not a violation of the Act or this part if the error was unintentional and occurred despite the maintenance of procedures reasonably adapted to avoid such an error. (2) An incorrect entry for a census tract number is deemed a bona fide error, and is not a violation of the Act or this part, provided that the financial institution maintains procedures reasonably adapted to avoid such an error. (c) Quarterly recording and reporting. (1) If a financial institution makes a good-faith effort to record all data required to be recorded pursuant to Sec. 1003.4(f) fully and accurately within 30 calendar days after the end of each calendar quarter, and some data are nevertheless inaccurate or incomplete, the inaccuracy or omission is not a violation of the Act or this part provided that the institution corrects or completes the data prior to submitting its annual loan/ application register pursuant to Sec. 1003.5(a)(1)(i). (2) If a financial institution required to comply with Sec. 1003.5(a)(1)(ii) makes a good-faith effort to report all data required to be reported pursuant to Sec. 1003.5(a)(1)(ii) fully and accurately within 60 calendar days after the end of each calendar quarter, and some data are nevertheless inaccurate or incomplete, the inaccuracy or omission is not a violation of the Act or this part provided that the institution corrects or completes the data prior to submitting its annual loan/application register pursuant to Sec. 1003.5(a)(1)(i). 0 11. Effective January 1, 2018, in Appendix A to Part 1003: 0 a. New subheading Transition Requirements for Data Collected in 2017 and Submitted in 2018 and paragraph 1 under that subheading are added immediately after the ``Paperwork Reduction Act Notice'' paragraph. 0 b. Paragraphs II.A and B are revised, and paragraph II.C is added. The additions and revisions read as follows: Appendix A to Part 1003--Form and Instructions for Completion of HMDA Loan/Application Register Paperwork Reduction Act Notice * * * * * Transition Requirements for Data Collected in 2017 and Submitted in 2018 1. The instructions for completion of the loan/application register in part I of this appendix applies to data collected during the 2017 calendar year and reported in 2018. Part I of this appendix does not apply to data [[Page 66314]] collected pursuant to the amendments to Regulation C effective January 1, 2018. * * * * * II. Appropriate Federal Agencies for HMDA Reporting A. A financial institution shall submit its loan/application register in electronic format to the appropriate Federal agency at the address identified by such agency. The appropriate Federal agency for a financial institution is determined pursuant to section 304(h)(2) of the Home Mortgage Disclosure Act (12 U.S.C. 2803(h)(2)) or, with respect to a financial institution subject to the Bureau's supervisory authority under section 1025(a) of the Consumer Financial Protection Act of 2010 (12 U.S.C. 5515(a)), is the Bureau. B. Procedures for the submission of the loan/application register are available at www.consumerfinance.gov/hmda. C. An authorized representative of the financial institution with knowledge of the data submitted shall certify to the accuracy and completeness of the data submitted. * * * * * Appendix A to Part 1003--[Removed and Reserved] 0 12. Effective January 1, 2019, Appendix A to Part 1003 is removed and reserved. 0 13. Effective January 1, 2018, Appendix B to Part 1003 is revised to read as follows: Appendix B to Part 1003--Form and Instructions for Data Collection on Ethnicity, Race, and Sex You may list questions regarding the ethnicity, race, and sex of the applicant on your loan application form, or on a separate form that refers to the application. (See the sample data collection form below for model language.) 1. You must ask the applicant for this information (but you cannot require the applicant to provide it) whether the application is taken in person, by mail or telephone, or on the internet. For applications taken by telephone, you must state the information in the collection form orally, except for that information which pertains uniquely to applications taken in writing, for example, the italicized language in the sample data collection form. 2. Inform the applicant that Federal law requires this information to be collected in order to protect consumers and to monitor compliance with Federal statutes that prohibit discrimination against applicants on these bases. Inform the applicant that if the information is not provided where the application is taken in person, you are required to note the information on the basis of visual observation or surname. 3. If you accept an application through electronic media with a video component, you must treat the application as taken in person. If you accept an application through electronic media without a video component (for example, facsimile), you must treat the application as accepted by mail. 4. For purposes of Sec. 1003.4(a)(10)(i), if a covered loan or application includes a guarantor, you do not report the guarantor's ethnicity, race, and sex. 5. If there are no co-applicants, you must report that there is no co-applicant. If there is more than one co-applicant, you must provide the ethnicity, race, and sex only for the first co-applicant listed on the collection form. A co-applicant may provide an absent co-applicant's ethnicity, race, and sex on behalf of the absent co- applicant. If the information is not provided for an absent co- applicant, you must report ``information not provided by applicant in mail, internet, or telephone application'' for the absent co- applicant. 6. When you purchase a covered loan and you choose not to report the applicant's or co-applicant's ethnicity, race, and sex, you must report that the requirement is not applicable. 7. You must report that the requirement to report the applicant's or co-applicant's ethnicity, race, and sex is not applicable when the applicant or co-applicant is not a natural person (for example, a corporation, partnership, or trust). For example, for a transaction involving a trust, you must report that the requirement to report the applicant's ethnicity, race, and sex is not applicable if the trust is the applicant. On the other hand, if the applicant is a natural person, and is the beneficiary of a trust, you must report the applicant's ethnicity, race, and sex. 8. You must report the ethnicity, race, and sex of an applicant as provided by the applicant. For example, if an applicant selects the ``Mexican'' box the institution reports ``Mexican'' for the ethnicity of the applicant. If an applicant selects the ``Asian'' box the institution reports ``Asian'' for the race of the applicant. Only an applicant may self-identify as being of a particular Hispanic or Latino subcategory (Mexican, Puerto Rican, Cuban, Other Hispanic or Latino) or of a particular Asian subcategory (Asian Indian, Chinese, Filipino, Japanese, Korean, Vietnamese, Other Asian) or of a particular Native Hawaiian or Other Pacific Islander subcategory (Native Hawaiian, Guamanian or Chamorro, Samoan, Other Pacific Islander) or of a particular American Indian or Alaska Native enrolled or principal tribe. 9. You must offer the applicant the option of selecting more than one ethnicity or race. If an applicant selects more than one ethnicity or race, you must report each selected designation, subject to the limits described below. i. Ethnicity--Aggregate categories and subcategories. There are two aggregate ethnicity categories: Hispanic or Latino; and Not Hispanic or Latino. If an applicant selects Hispanic or Latino, the applicant may also select up to four ethnicity subcategories: Mexican; Puerto Rican; Cuban; and Other Hispanic or Latino. You must report each aggregate ethnicity category and each ethnicity subcategory selected by the applicant. ii. Ethnicity--Other subcategories. If an applicant selects the Other Hispanic or Latino ethnicity subcategory, the applicant may also provide a particular Hispanic or Latino ethnicity not listed in the standard subcategories. In such a case, you must report both the selection of Other Hispanic or Latino and the additional information provided by the applicant. iii. Race--Aggregate categories and subcategories. There are five aggregate race categories: American Indian or Alaska Native; Asian; Black or African American; Native Hawaiian or Other Pacific Islander; and White. The Asian and the Native Hawaiian or Other Pacific Islander aggregate categories have seven and four subcategories, respectively. The Asian race subcategories are: Asian Indian; Chinese, Filipino; Japanese; Korean; Vietnamese; and Other Asian. The Native Hawaiian or Other Pacific Islander race subcategories are: Native Hawaiian; Guamanian or Chamorro; Samoan; and Other Pacific Islander. You must report every aggregate race category selected by the applicant. If the applicant also selects one or more race subcategories, you must report each race subcategory selected by the applicant, except that you must not report more than a total of five aggregate race categories and race subcategories combined. For example, if the applicant selects all five aggregate race categories and also selects some race subcategories, you report only the five aggregate race categories. On the other hand, if the applicant selects the White, Asian, and Native Hawaiian or Other Pacific Islander aggregate race categories, and the applicant also selects the Korean, Vietnamese, and Samoan race subcategories, you must report White, Asian, Native Hawaiian or Other Pacific Islander, and any two, at your option, of the three race subcategories selected by the applicant. In this example, you must report White, Asian, and Native Hawaiian or Other Pacific Islander, and in addition you must report (at your option) either Korean and Vietnamese, Korean and Samoan, or Vietnamese and Samoan. To determine how to report an Other race subcategory for purposes of the five-race maximum, see paragraph 9.iv below. iv. Race--Other subcategories. If an applicant selects the Other Asian race subcategory or the Other Pacific Islander race subcategory, the applicant may also provide a particular Other Asian or Other Pacific Islander race not listed in the standard subcategories. In either such case, you must report both the selection of Other Asian or Other Pacific Islander, as applicable, and the additional information provided by the applicant, subject to the five-race maximum. In all such cases where the applicant has selected an Other race subcategory and also provided additional information, for purposes of the maximum of five reportable race categories and race subcategories combined set forth above, the Other race subcategory and additional information provided by the applicant together constitute only one selection. Thus, using the same facts in the example offered in paragraph 9.iii above, if the applicant also selected Other Asian and entered ``Thai'' in the space provided, Other Asian and Thai are considered one selection. You must report any two (at your option) of the four race subcategories selected by the applicant, Korean, Vietnamese, Other Asian-Thai, and [[Page 66315]] Samoan, in addition to the three aggregate race categories selected by the applicant. 10. If the applicant chooses not to provide the information for an application taken in person, note this fact on the collection form and then collect the applicant's ethnicity, race, and sex on the basis of visual observation or surname. You must report whether the applicant's ethnicity, race, and sex was collected on the basis of visual observation or surname. When you collect an applicant's ethnicity, race, and sex on the basis of visual observation or surname, you must select from the following aggregate categories: Ethnicity (Hispanic or Latino; not Hispanic or Latino); race (American Indian or Alaska Native; Asian; Black or African American; Native Hawaiian or Other Pacific Islander; White); sex (male; female). 11. If the applicant declines to answer these questions by checking the ``I do not wish to provide this information'' box on an application that is taken by mail or on the internet, or declines to provide this information by stating orally that he or she does not wish to provide this information on an application that is taken by telephone, you must report ``information not provided by applicant in mail, internet, or telephone application.'' 12. If the applicant begins an application by mail, internet, or telephone, and does not provide the requested information on the application but does not check or select the ``I do not wish to provide this information'' box on the application, and the applicant meets in person with you to complete the application, you must request the applicant's ethnicity, race, and sex. If the applicant does not provide the requested information during the in-person meeting, you must collect the information on the basis of visual observation or surname. If the meeting occurs after the application process is complete, for example, at closing or account opening, you are not required to obtain the applicant's ethnicity, race, and sex. 13. When an applicant provides the requested information for some but not all fields, you report the information that was provided by the applicant, whether partial or complete. If an applicant provides partial or complete information on ethnicity, race, and sex and also checks the ``I do not wish to provide this information'' box on an application that is taken by mail or on the internet, or makes that selection when applying by telephone, you must report the information on ethnicity, race, and sex that was provided by the applicant. [[Page 66316]] [GRAPHIC] [TIFF OMITTED] TR28OC15.011 0 14. Effective January 1, 2018, Appendix C to Part 1003 is added to read as follows: Appendix C to Part 1003--Procedures for Generating a Check Digit and Validating a ULI The check digit for the Universal Loan Identifier (ULI) pursuant to Sec. 1003.4(a)(1)(i)(C) is calculated using the ISO/IEC 7064, MOD 97-10 as it appears on the International Standard ISO/IEC 7064:2003, which is published by the International Organization for Standardization (ISO). (copyright)ISO. This material is reproduced from ISO/IEC 7064:2003 with permission of the American National Standards Institute (ANSI) on behalf of ISO. All rights reserved. [[Page 66317]] Generating A Check Digit Step 1: Starting with the leftmost character in the string that consists of the combination of the Legal Entity Identifier (LEI) pursuant to Sec. 1003.4(a)(1)(i)(A) and the additional characters identifying the covered loan or application pursuant to Sec. 1003.4(a)(1)(i)(B), replace each alphabetic character with numbers in accordance with Table I below to obtain all numeric values in the string. Table I--Alphabetic To Numeric Conversion Table The alphabetic characters are not case-sensitive and each letter, whether it is capitalized or in lower-case, is equal to the same value as each letter illustrates in the conversion table. For example, A and a are each equal to 10. ------------------------------------------------------------------------ ------------------------------------------------------------------------ A = 10 H = 17 O = 24 V = 31 B = 11 I = 18 P = 25 W = 32 C = 12 J = 19 Q = 26 X = 33 D = 13 K = 20 R = 27 Y = 34 E = 14 L = 21 S = 28 Z = 35 F = 15 M = 22 T = 29 G = 16 N = 23 U = 30 ------------------------------------------------------------------------ Step 2: After converting the combined string of characters to all numeric values, append two zeros to the rightmost positions. Step 3: Apply the mathematical function mod=(n,97) where n= the number obtained in step 2 above and 97 is the divisor. Alternatively, to calculate without using the modulus operator, divide the numbers in step 2 above by 97. Truncate the remainder to three digits and multiply it by .97. Round the result to the nearest whole number. Step 4: Subtract the result in step 3 from 98. If the result is one digit, add a leading 0 to make it two digits. Step 5: The two digits in the result from step 4 is the check digit. Append the resulting check digit to the rightmost position in the combined string of characters described in step 1 above to generate the ULI. Example For example, assume the LEI for a financial institution is 10Bx939c5543TqA1144M and the financial institution assigned the following string of characters to identify the covered loan: 999143X. The combined string of characters is 10Bx939c5543TqA1144M999143X. Step 1: Starting with the leftmost character in the combined string of characters, replace each alphabetic character with numbers in accordance with Table I above to obtain all numeric values in the string. The result is 10113393912554329261011442299914333. Step 2: Append two zeros to the rightmost positions in the combined string. The result is 1011339391255432926101144229991433300. Step 3: Apply the mathematical function mod=(n,97) where n= the number obtained in step 2 above and 97 is the divisor. The result is 60. Alternatively, to calculate without using the modulus operator, divide the numbers in step 2 above by 97. The result is 1042617929129312294946332267952920.618556701030928. Truncate the remainder to three digits, which is .618, and multiply it by .97. The result is 59.946. Round this result to the nearest whole number, which is 60. Step 4: Subtract the result in step 3 from 98. The result is 38. Step 5: The two digits in the result from step 4 is the check digit. Append the check digit to the rightmost positions in the combined string of characters that consists of the LEI and the string of characters assigned by the financial institution to identify the covered loan to obtain the ULI. In this example, the ULI would be 10Bx939c5543TqA1144M999143X38. Validating A ULI To determine whether the ULI contains a transcription error using the check digit calculation, the procedures are described below. Step 1: Starting with the leftmost character in the ULI, replace each alphabetic character with numbers in accordance with Table I above to obtain all numeric values in the string. Step 2: Apply the mathematical function mod=(n,97) where n=the number obtained in step 1 above and 97 is the divisor. Step 3: If the result is 1, the ULI does not contain transcription errors. Example For example, the ULI assigned to a covered loan is 10Bx939c5543TqA1144M999143X38. Step 1: Starting with the leftmost character in the ULI, replace each alphabetic character with numbers in accordance with Table I above to obtain all numeric values in the string. The result is 1011339391255432926101144229991433338. Step 2: Apply the mathematical function mod=(n,97) where n is the number obtained in step 1 above and 97 is the divisor. Step 3: The result is 1. The ULI does not contain transcription errors. 0 15. Effective January 1, 2018, Supplement I to Part 1003 is revised to read as follows: Supplement I to Part 1003--Official Interpretations Introduction 1. Status. The commentary in this supplement is the vehicle by which the Bureau of Consumer Financial Protection issues formal interpretations of Regulation C (12 CFR part 1003). Section 1003.2--Definitions 2(b) Application 1. Consistency with Regulation B. Bureau interpretations that appear in the official commentary to Regulation B (Equal Credit Opportunity Act, 12 CFR part 1002, Supplement I) are generally applicable to the definition of application under Regulation C. However, under Regulation C the definition of an application does not include prequalification requests. 2. Prequalification. A prequalification request is a request by a prospective loan applicant (other than a request for preapproval) for a preliminary determination on whether the prospective loan applicant would likely qualify for credit under an institution's standards, or for a determination on the amount of credit for which the prospective applicant would likely qualify. Some institutions evaluate prequalification requests through a procedure that is separate from the institution's normal loan application process; others use the same process. In either case, Regulation C does not require an institution to report prequalification requests on the loan/application register, even though these requests may constitute applications under Regulation B for purposes of adverse action notices. 3. Requests for preapproval. To be a preapproval program as defined in Sec. 1003.2(b)(2), the written commitment issued under the program must result from a comprehensive review of the creditworthiness of the applicant, including such verification of income, resources, and other matters as is typically done by the institution as part of its normal credit evaluation program. In addition to conditions involving the identification of a suitable property and verification that no material change has occurred in the applicant's financial condition or creditworthiness, the written commitment may be subject only to other conditions (unrelated to the financial condition or creditworthiness of the applicant) that the lender ordinarily attaches to a traditional home mortgage application approval. These conditions are limited to conditions such as requiring an acceptable title insurance binder or a certificate indicating clear termite inspection, and, in the case where the applicant plans to use the proceeds from the sale of the applicant's present home to purchase a new home, a settlement statement showing adequate proceeds from the sale of the present home. Regardless of its name, a program that satisfies the definition of a preapproval program in Sec. 1003.2(b)(2) is a preapproval program for purposes of Regulation C. Conversely, a program that a financial institution describes as a ``preapproval program'' that does not satisfy the requirements of Sec. 1003.2(b)(2) is not a preapproval program for purposes of Regulation C. If a financial institution does not regularly use the procedures specified in Sec. 1003.2(b)(2), but instead considers requests for preapprovals on an ad hoc basis, the financial institution need not treat ad hoc requests as part of a preapproval program for purposes of Regulation C. A financial institution should, however, be generally consistent in following uniform procedures for considering such ad hoc requests. 2(c) Branch Office Paragraph 2(c)(1) 1. Credit unions. For purposes of Regulation C, a ``branch'' of a credit union is any office where member accounts are established or loans are made, whether or not the office has been approved as a branch by a Federal or State agency. (See 12 U.S.C. 1752.) 2. Bank, savings association, or credit unions. A branch office of a bank, savings [[Page 66318]] association, or credit union does not include a loan-production office if the loan-production office is not considered a branch by the Federal or State supervisory authority applicable to that institution. A branch office also does not include the office of an affiliate or of a third party, such as a third-party broker. Paragraph 2(c)(2) 1. General. A branch office of a for-profit mortgage lending institution, other than a bank savings association or credit union, does not include the office of an affiliate or of a third party, such as a third-party broker. 2(d) Closed-end Mortgage Loan 1. Dwelling-secured. Section 1003.2(d) defines a closed-end mortgage loan as an extension of credit that is secured by a lien on a dwelling and that is not an open-end line of credit under Sec. 1003.2(o). Thus, for example, a loan to purchase a dwelling and secured only by a personal guarantee is not a closed-end mortgage loan because it is not dwelling-secured. 2. Extension of credit. Under Sec. 1003.2(d), a dwelling- secured loan is not a closed-end mortgage loan unless it involves an extension of credit. Thus, some transactions completed pursuant to installment sales contracts, such as some land contracts, are not closed-end mortgage loans because no credit is extended. For example, if a land contract provides that, upon default, the contract terminates, all previous payments will be treated as rent, and the borrower is under no obligation to make further payments, the transaction is not a closed-end mortgage loan. In general, extension of credit under Sec. 1003.2(d) refers to the granting of credit only pursuant to a new debt obligation. Thus, except as described in comments 2(d)-2.i and .ii, if a transaction modifies, renews, extends, or amends the terms of an existing debt obligation, but the existing debt obligation is not satisfied and replaced, the transaction is not a closed-end mortgage loan under Sec. 1003.2(d) because there has been no new extension of credit. The phrase extension of credit thus is defined differently under Regulation C than under Regulation B, 12 CFR part 1002. i. Assumptions. For purposes of Regulation C, an assumption is a transaction in which an institution enters into a written agreement accepting a new borrower in place of an existing borrower as the obligor on an existing debt obligation. For purposes of Regulation C, assumptions include successor-in-interest transactions, in which an individual succeeds the prior owner as the property owner and then assumes the existing debt secured by the property. Under Sec. 1003.2(d), assumptions are extensions of credit even if the new borrower merely assumes the existing debt obligation and no new debt obligation is created. See also comment 2(j)-5. ii. New York State consolidation, extension, and modification agreements. A transaction completed pursuant to a New York State consolidation, extension, and modification agreement and classified as a supplemental mortgage under New York Tax Law section 255, such that the borrower owes reduced or no mortgage recording taxes, is an extension of credit under Sec. 1003.2(d). Comments 2(i)-1, 2(j)-5, and 2(p)-2 clarify whether such transactions are home improvement loans, home purchase loans, or refinancings, respectively. 2(f) Dwelling 1. General. The definition of a dwelling is not limited to the principal or other residence of the applicant or borrower, and thus includes vacation or second homes and investment properties. 2. Multifamily residential structures and communities. A dwelling also includes a multifamily residential structure or community such as an apartment, condominium, cooperative building or complex, or a manufactured home community. A loan related to a manufactured home community is secured by a dwelling for purposes of Sec. 1003.2(f) even if it is not secured by any individual manufactured homes, but only by the land that constitutes the manufactured home community including sites for manufactured homes. However, a loan related to a multifamily residential structure or community that is not a manufactured home community is not secured by a dwelling for purposes of Sec. 1003.2(f) if it is not secured by any individual dwelling units and is, for example, instead secured only by property that only includes common areas, or is secured only by an assignment of rents or dues. 3. Exclusions. Recreational vehicles, including boats, campers, travel trailers, and park model recreational vehicles, are not considered dwellings for purposes of Sec. 1003.2(f), regardless of whether they are used as residences. Houseboats, floating homes, and mobile homes constructed before June 15, 1976, are also excluded, regardless of whether they are used as residences. Also excluded are transitory residences such as hotels, hospitals, college dormitories, and recreational vehicle parks, and structures originally designed as dwellings but used exclusively for commercial purposes, such as homes converted to daycare facilities or professional offices. 4. Mixed-use properties. A property used for both residential and commercial purposes, such as a building containing apartment units and retail space, is a dwelling if the property's primary use is residential. An institution may use any reasonable standard to determine the primary use of the property, such as by square footage or by the income generated. An institution may select the standard to apply on a case-by-case basis. 5. Properties with service and medical components. For purposes of Sec. 1003.2(f), a property used for both long-term housing and to provide related services, such as assisted living for senior citizens or supportive housing for persons with disabilities, is a dwelling and does not have a non-residential purpose merely because the property is used for both housing and to provide services. However, transitory residences that are used to provide such services are not dwellings. See comment 2(f)-3. Properties that are used to provide medical care, such as skilled nursing, rehabilitation, or long-term medical care, also are not dwellings. See comment 2(f)-3. If a property that is used for both long-term housing and to provide related services also is used to provide medical care, the property is a dwelling if its primary use is residential. An institution may use any reasonable standard to determine the property's primary use, such as by square footage, income generated, or number of beds or units allocated for each use. An institution may select the standard to apply on a case-by-case basis. 2(g) Financial Institution 1. Preceding calendar year and preceding December 31. The definition of financial institution refers both to the preceding calendar year and the preceding December 31. These terms refer to the calendar year and the December 31 preceding the current calendar year. For example, in 2019, the preceding calendar year is 2018 and the preceding December 31 is December 31, 2018. Accordingly, in 2019, Financial Institution A satisfies the asset-size threshold described in Sec. 1003.2(g)(1)(i) if its assets exceeded the threshold specified in comment 2(g)-2 on December 31, 2018. Likewise, in 2020, Financial Institution A does not meet the loan- volume test described in Sec. 1003.2(g)(1)(v)(A) if it originated fewer than 25 closed-end mortgage loans during either 2018 or 2019. 2. [Reserved] 3. Merger or acquisition--coverage of surviving or newly formed institution. After a merger or acquisition, the surviving or newly formed institution is a financial institution under Sec. 1003.2(g) if it, considering the combined assets, location, and lending activity of the surviving or newly formed institution and the merged or acquired institutions or acquired branches, satisfies the criteria included in Sec. 1003.2(g). For example, A and B merge. The surviving or newly formed institution meets the loan threshold described in Sec. 1003.2(g)(1)(v)(B) if the surviving or newly formed institution, A, and B originated a combined total of at least 100 open-end lines of credit in each of the two preceding calendar years. Likewise, the surviving or newly formed institution meets the asset-size threshold in Sec. 1003.2(g)(1)(i) if its assets and the combined assets of A and B on December 31 of the preceding calendar year exceeded the threshold described in Sec. 1003.2(g)(1)(i). Comment 2(g)-4 discusses a financial institution's responsibilities during the calendar year of a merger. 4. Merger or acquisition--coverage for calendar year of merger or acquisition. The scenarios described below illustrate a financial institution's responsibilities for the calendar year of a merger or acquisition. For purposes of these illustrations, a ``covered institution'' means a financial institution, as defined in Sec. 1003.2(g), that is not exempt from reporting under Sec. 1003.3(a), and ``an institution that is not covered'' means either an institution that is not a financial institution, as defined in Sec. 1003.2(g), or an institution that is exempt from reporting under Sec. 1003.3(a). i. Two institutions that are not covered merge. The surviving or newly formed institution meets all of the requirements necessary to be a covered institution. No data collection is required for the calendar year of the merger (even though the merger creates [[Page 66319]] an institution that meets all of the requirements necessary to be a covered institution). When a branch office of an institution that is not covered is acquired by another institution that is not covered, and the acquisition results in a covered institution, no data collection is required for the calendar year of the acquisition. ii. A covered institution and an institution that is not covered merge. The covered institution is the surviving institution, or a new covered institution is formed. For the calendar year of the merger, data collection is required for covered loans and applications handled in the offices of the merged institution that was previously covered and is optional for covered loans and applications handled in offices of the merged institution that was previously not covered. When a covered institution acquires a branch office of an institution that is not covered, data collection is optional for covered loans and applications handled by the acquired branch office for the calendar year of the acquisition. iii. A covered institution and an institution that is not covered merge. The institution that is not covered is the surviving institution, or a new institution that is not covered is formed. For the calendar year of the merger, data collection is required for covered loans and applications handled in offices of the previously covered institution that took place prior to the merger. After the merger date, data collection is optional for covered loans and applications handled in the offices of the institution that was previously covered. When an institution remains not covered after acquiring a branch office of a covered institution, data collection is required for transactions of the acquired branch office that take place prior to the acquisition. Data collection by the acquired branch office is optional for transactions taking place in the remainder of the calendar year after the acquisition. iv. Two covered institutions merge. The surviving or newly formed institution is a covered institution. Data collection is required for the entire calendar year of the merger. The surviving or newly formed institution files either a consolidated submission or separate submissions for that calendar year. When a covered institution acquires a branch office of a covered institution, data collection is required for the entire calendar year of the merger. Data for the acquired branch office may be submitted by either institution. 5. Originations. Whether an institution is a financial institution depends in part on whether the institution originated at least 25 closed-end mortgage loans in each of the two preceding calendar years or at least 100 open-end lines of credit in each of the two preceding calendar years. Comments 4(a)-2 through -4 discuss whether activities with respect to a particular closed-end mortgage loan or open-end line of credit constitute an origination for purposes of Sec. 1003.2(g). 6. Branches of foreign banks--treated as banks. A Federal branch or a State-licensed or insured branch of a foreign bank that meets the definition of a ``bank'' under section 3(a)(1) of the Federal Deposit Insurance Act (12 U.S.C. 1813(a)) is a bank for the purposes of Sec. 1003.2(g). 7. Branches and offices of foreign banks and other entities-- treated as nondepository financial institutions. A Federal agency, State-licensed agency, State-licensed uninsured branch of a foreign bank, commercial lending company owned or controlled by a foreign bank, or entity operating under section 25 or 25A of the Federal Reserve Act, 12 U.S.C. 601 and 611 (Edge Act and agreement corporations) may not meet the definition of ``bank'' under the Federal Deposit Insurance Act and may thereby fail to satisfy the definition of a depository financial institution under Sec. 1003.2(g)(1). An entity is nonetheless a financial institution if it meets the definition of nondepository financial institution under Sec. 1003.2(g)(2). 2(i) Home Improvement Loan 1. General. Section 1003.2(i) defines a home improvement loan as a closed-end mortgage loan or an open-end line of credit that is for the purpose, in whole or in part, of repairing, rehabilitating, remodeling, or improving a dwelling or the real property on which the dwelling is located. For example, a closed-end mortgage loan obtained to repair a dwelling by replacing a roof is a home improvement loan under Sec. 1003.2(i). A loan or line of credit is a home improvement loan even if only a part of the purpose is for repairing, rehabilitating, remodeling, or improving a dwelling. For example, an open-end line of credit obtained in part to remodel a kitchen and in part to pay college tuition is a home improvement loan under Sec. 1003.2(i). Similarly, for example, a loan that is completed pursuant to a New York State consolidation, extension, and modification agreement and that is classified as a supplemental mortgage under New York Tax Law section 255, such that the borrower owes reduced or no mortgage recording taxes, is a home improvement loan if any of the loan's funds are for home improvement purposes. See also comment 2(d)-2.ii. 2. Improvements to real property. Home improvements include improvements both to a dwelling and to the real property on which the dwelling is located (for example, installation of a swimming pool, construction of a garage, or landscaping). 3. Commercial and other loans. A home improvement loan may include a closed-end mortgage loan or an open-end line of credit originated outside an institution's residential mortgage lending division, such as a loan or line of credit to improve an apartment building originated in the commercial loan department. 4. Mixed-use property. A closed-end mortgage loan or an open-end line of credit to improve a dwelling used for residential and commercial purposes (for example, a building containing apartment units and retail space), or the real property on which such a dwelling is located, is a home improvement loan if the loan's proceeds are used either to improve the entire property (for example, to replace the heating system), or if the proceeds are used primarily to improve the residential portion of the property. An institution may use any reasonable standard to determine the primary use of the loan proceeds. An institution may select the standard to apply on a case-by-case basis. 5. Multiple-purpose loans. A closed-end mortgage loan or an open-end line of credit may be used for multiple purposes. For example, a closed-end mortgage loan that is a home improvement loan under Sec. 1003.2(i) may also be a refinancing under Sec. 1003.2(p) if the transaction is a cash-out refinancing and the funds will be used to improve a home. Such a transaction is a multiple- purpose loan. Comment 4(a)(3)-3 provides details about how to report multiple-purpose covered loans. 6. Statement of borrower. In determining whether a closed-end mortgage loan or an open-end line of credit, or an application for a closed-end mortgage loan or an open-end line of credit, is for home improvement purposes, an institution may rely on the applicant's or borrower's stated purpose(s) for the loan or line of credit at the time the application is received or the credit decision is made. An institution need not confirm that the borrower actually uses any of the funds for the stated purpose(s). 2(j) Home Purchase Loan 1. Multiple properties. A home purchase loan includes a closed- end mortgage loan or an open-end line of credit secured by one dwelling and used to purchase another dwelling. For example, if a person obtains a home-equity loan or a reverse mortgage secured by dwelling A to purchase dwelling B, the home-equity loan or the reverse mortgage is a home purchase loan under Sec. 1003.2(j). 2. Commercial and other loans. A home purchase loan may include a closed-end mortgage loan or an open-end line of credit originated outside an institution's residential mortgage lending division, such as a loan or line of credit to purchase an apartment building originated in the commercial loan department. 3. Construction and permanent financing. A home purchase loan includes both a combined construction/permanent loan and the permanent financing that replaces a construction-only loan. A home purchase loan does not include a construction-only loan that is designed to be replaced by permanent financing at a later time, which is excluded from Regulation C as temporary financing under Sec. 1003.3(c)(3). Comment 3(c)(3)-1 provides additional details about transactions that are excluded as temporary financing. 4. Second mortgages that finance the downpayments on first mortgages. If an institution making a first mortgage loan to a home purchaser also makes a second mortgage loan or line of credit to the same purchaser to finance part or all of the home purchaser's downpayment, both the first mortgage loan and the second mortgage loan or line of credit are home purchase loans. 5. Assumptions. Under Sec. 1003.2(j), an assumption is a home purchase loan when an institution enters into a written agreement accepting a new borrower as the obligor on an existing obligation to finance the new borrower's purchase of the dwelling securing the existing obligation, if the resulting obligation is a closed-end mortgage loan or an open-end line of credit. A transaction in [[Page 66320]] which borrower B finances the purchase of borrower A's dwelling by assuming borrower A's existing debt obligation and that is completed pursuant to a New York State consolidation, extension, and modification agreement and is classified as a supplemental mortgage under New York Tax Law section 255, such that the borrower owes reduced or no mortgage recording taxes, is an assumption and a home purchase loan. See comment 2(d)-2.ii. On the other hand, a transaction in which borrower B, a successor-in-interest, assumes borrower A's existing debt obligation only after acquiring title to borrower A's dwelling is not a home purchase loan because borrower B did not assume the debt obligation for the purpose of purchasing a dwelling. See Sec. 1003.4(a)(3) and comment 4(a)(3)-4 for guidance about how to report covered loans that are not home improvement loans, home purchase loans, or refinancings. 6. Multiple-purpose loans. A closed-end mortgage loan or an open-end line of credit may be used for multiple purposes. For example, a closed-end mortgage loan that is a home purchase loan under Sec. 1003.2(j) may also be a home improvement loan under Sec. 1003.2(i) and a refinancing under Sec. 1003.2(p) if the transaction is a cash-out refinancing and the funds will be used to purchase and improve a dwelling. Such a transaction is a multiple- purpose loan. Comment 4(a)(3)-3 provides details about how to report multiple-purpose covered loans. 2(l) Manufactured Home 1. Definition of a manufactured home. The definition in Sec. 1003.2(l) refers to the Federal building code for manufactured housing established by the U.S. Department of Housing and Urban Development (HUD) (24 CFR part 3280.2). Modular or other factory- built homes that do not meet the HUD code standards are not manufactured homes for purposes of Sec. 1003.2(l). Recreational vehicles are excluded from the HUD code standards pursuant to 24 CFR 3282.8(g) and are also excluded from the definition of dwelling for purposes of Sec. 1003.2(f). See comment 2(f)-3. 2. Identification. A manufactured home will generally bear a data plate affixed in a permanent manner near the main electrical panel or other readily accessible and visible location noting its compliance with the Federal Manufactured Home Construction and Safety Standards in force at the time of manufacture and providing other information about its manufacture pursuant to 24 CFR 3280.5. A manufactured home will generally also bear a HUD Certification Label pursuant to 24 CFR 3280.11. 2(m) Metropolitan Statistical Area (MD) or Metropolitan Division (MD). 1. Use of terms ``Metropolitan Statistical Area (MSA)'' and ``Metropolitan Division (MD).'' The U.S. Office of Management and Budget (OMB) defines Metropolitan Statistical Areas (MSAs) and Metropolitan Divisions (MDs) to provide nationally consistent definitions for collecting, tabulating, and publishing Federal statistics for a set of geographic areas. For all purposes under Regulation C, if an MSA is divided by OMB into MDs, the appropriate geographic unit to be used is the MD; if an MSA is not so divided by OMB into MDs, the appropriate geographic unit to be used is the MSA. 2(n) Multifamily Dwelling 1. Multifamily residential structures. The definition of dwelling in Sec. 1003.2(f) includes multifamily residential structures and the corresponding commentary provides guidance on when such residential structures are included in that definition. See comments 2(f)-2 through -5. 2. Special reporting requirements for multifamily dwellings. The definition of multifamily dwelling in Sec. 1003.2(n) includes a dwelling, regardless of construction method, that contains five or more individual dwelling units. Covered loans secured by a multifamily dwelling are subject to additional reporting requirements under Sec. 1003.4(a)(32), but are not subject to reporting requirements under Sec. 1003.4(a)(4), (10)(iii), (23), (29), or (30). 2(o) Open-End Line of Credit 1. General. Section 1003.2(o) defines an open-end line of credit as an extension of credit that is secured by a lien on a dwelling and that is an open-end credit plan as defined in Regulation Z, 12 CFR 1026.2(a)(20), but without regard to whether the credit is consumer credit, as defined in Sec. 1026.2(a)(12), is extended by a creditor, as defined in Sec. 1026.2(a)(17), or is extended to a consumer, as defined in Sec. 1026.2(a)(11). Aside from these distinctions, institutions may rely on 12 CFR 1026.2(a)(20) and its related commentary in determining whether a transaction is an open- end line of credit under Sec. 1003.2(o). For example, assume a business-purpose transaction that is exempt from Regulation Z pursuant to Sec. 1026.3(a)(1) but that otherwise is open-end credit under Regulation Z Sec. 1026.2(a)(20). The business-purpose transaction is an open-end line of credit under Regulation C, provided the other requirements of Sec. 1003.2(o) are met. Similarly, assume a transaction in which the person extending open- end credit is a financial institution under Sec. 1003.2(g) but is not a creditor under Regulation Z, Sec. 1026.2(a)(17). In this example, the transaction is an open-end line of credit under Regulation C, provided the other requirements of Sec. 1003.2(o) are met. 2. Extension of credit. Extension of credit has the same meaning under Sec. 1003.2(o) as under Sec. 1003.2(d) and comment 2(d)-2. Thus, for example, a renewal of an open-end line of credit is not an extension of credit under Sec. 1003.2(o) and is not covered by Regulation C unless the existing debt obligation is satisfied and replaced. Likewise, under Sec. 1003.2(o), each draw on an open-end line of credit is not an extension of credit. 2(p) Refinancing 1. General. Section 1003.2(p) defines a refinancing as a closed- end mortgage loan or an open-end line of credit in which a new, dwelling-secured debt obligation satisfies and replaces an existing, dwelling-secured debt obligation by the same borrower. Except as described in comment 2(p)-2, whether a refinancing has occurred is determined by reference to whether, based on the parties' contract and applicable law, the original debt obligation has been satisfied or replaced by a new debt obligation. Whether the original lien is satisfied is irrelevant. For example: i. A new closed-end mortgage loan that satisfies and replaces one or more existing closed-end mortgage loans is a refinancing under Sec. 1003.2(p). ii. A new open-end line of credit that satisfies and replaces an existing closed-end mortgage loan is a refinancing under Sec. 1003.2(p). iii. Except as described in comment 2(p)-2, a new debt obligation that renews or modifies the terms of, but that does not satisfy and replace, an existing debt obligation, is not a refinancing under Sec. 1003.2(p). 2. New York State consolidation, extension, and modification agreements. Where a transaction is completed pursuant to a New York State consolidation, extension, and modification agreement and is classified as a supplemental mortgage under New York Tax Law sectionSec. 255, such that the borrower owes reduced or no mortgage recording taxes, and where, but for the agreement, the transaction would have met the definition of a refinancing under Sec. 1003.2(p), the transaction is considered a refinancing under Sec. 1003.2(p). See also comment 2(d)-2.ii. 3. Existing debt obligation. A closed-end mortgage loan or an open-end line of credit that satisfies and replaces one or more existing debt obligations is not a refinancing under Sec. 1003.2(p) unless the existing debt obligation (or obligations) also was secured by a dwelling. For example, assume that a borrower has an existing $30,000 closed-end mortgage loan and obtains a new $50,000 closed-end mortgage loan that satisfies and replaces the existing $30,000 loan. The new $50,000 loan is a refinancing under Sec. 1003.2(p). However, if the borrower obtains a new $50,000 closed-end mortgage loan that satisfies and replaces an existing $30,000 loan secured only by a personal guarantee, the new $50,000 loan is not a refinancing under Sec. 1003.2(p). See Sec. 1003.4(a)(3) and related commentary for guidance about how to report the loan purpose of such transactions, if they are not otherwise excluded under Sec. 1003.3(c). 4. Same borrower. Section 1003.2(p) provides that, even if all of the other requirements of Sec. 1003.2(p) are met, a closed-end mortgage loan or an open-end line of credit is not a refinancing unless the same borrower undertakes both the existing and the new obligation(s). Under Sec. 1003.2(p), the ``same borrower'' undertakes both the existing and the new obligation(s) even if only one borrower is the same on both obligations. For example, assume that an existing closed-end mortgage loan (obligation X) is satisfied and replaced by a new closed-end mortgage loan (obligation Y). If borrowers A and B both are obligated on obligation X, and only borrower B is obligated on obligation Y, then obligation Y is a refinancing under Sec. 1003.2(p), assuming the other requirements of Sec. 1003.2(p) are met, because borrower B is obligated on both transactions. On the other hand, if only borrower A is obligated on obligation X, and only borrower B is obligated on obligation Y, then obligation Y is not a refinancing under [[Page 66321]] Sec. 1003.2(p). For example, assume that two spouses are divorcing. If both spouses are obligated on obligation X, but only one spouse is obligated on obligation Y, then obligation Y is a refinancing under Sec. 1003.2(p), assuming the other requirements of Sec. 1003.2(p) are met. On the other hand, if only spouse A is obligated on obligation X, and only spouse B is obligated on obligation Y, then obligation Y is not a refinancing under Sec. 1003.2(p). See Sec. 1003.4(a)(3) and related commentary for guidance about how to report the loan purpose of such transactions, if they are not otherwise excluded under Sec. 1003.3(c). 5. Two or more debt obligations. Section 1003.2(p) provides that, to be a refinancing, a new debt obligation must satisfy and replace an existing debt obligation. Where two or more new obligations replace an existing obligation, each new obligation is a refinancing if, taken together, the new obligations satisfy the existing obligation. Similarly, where one new obligation replaces two or more existing obligations, the new obligation is a refinancing if it satisfies each of the existing obligations. 6. Multiple-purpose loans. A closed-end mortgage loan or an open-end line of credit may be used for multiple purposes. For example, a closed-end mortgage loan that is a refinancing under Sec. 1003.2(p) may also be a home improvement loan under Sec. 1003.2(i) and be used for other purposes if the refinancing is a cash-out refinancing and the funds will be used both for home improvement and to pay college tuition. Such a transaction is a multiple-purpose loan. Comment 4(a)(3)-3 provides details about how to report multiple-purpose covered loans. Section 1003.3--Exempt Institutions and Excluded Transactions 3(c) Excluded Transactions Paragraph 3(c)(1) 1. Financial institution acting in a fiduciary capacity. Section 1003.3(c)(1) provides that a closed-end mortgage loan or an open-end line of credit originated or purchased by a financial institution acting in a fiduciary capacity is an excluded transaction. A financial institution acts in a fiduciary capacity if, for example, the financial institution acts as a trustee. Paragraph 3(c)(2) 1. Loan or line of credit secured by a lien on unimproved land. Section 1003.3(c)(2) provides that a closed-end mortgage loan or an open-end line of credit secured by a lien on unimproved land is an excluded transaction. A loan or line of credit is secured by a lien on unimproved land if the loan or line of credit is secured by vacant or unimproved property, unless the institution knows, based on information that it receives from the applicant or borrower at the time the application is received or the credit decision is made, that the proceeds of that loan or credit line will be used within two years after closing or account opening to construct a dwelling on, or to purchase a dwelling to be placed on, the land. A loan or line of credit that is not excludable under Sec. 1003.3(c)(2) nevertheless may be excluded, for example, as temporary financing under Sec. 1003.3(c)(3). Paragraph 3(c)(3) 1. Temporary financing. Section 1003.3(c)(3) provides that closed-end mortgage loans or open-end lines of credit obtained for temporary financing are excluded transactions. A loan or line of credit is considered temporary financing and excluded under Sec. 1003.3(c)(3) if the loan or line of credit is designed to be replaced by permanent financing at a later time. For example: i. Lender A extends credit in the form of a bridge or swing loan to finance a borrower's down payment on a home purchase. The borrower pays off the bridge or swing loan with funds from the sale of his or her existing home and obtains permanent financing for his or her new home from Lender A. The bridge or swing loan is excluded as temporary financing under Sec. 1003.3(c)(3). ii. Lender A extends credit to finance construction of a dwelling. A new extension of credit for permanent financing for the dwelling will be obtained, either from Lender A or from another lender, and either through a refinancing of the initial construction loan or a separate loan. The initial construction loan is excluded as temporary financing under Sec. 1003.3(c)(3). iii. Assume the same scenario as in comment 3(c)(3)-1.ii, except that the initial construction loan is, or may be, renewed one or more times before the permanent financing is made. The initial construction loan, including any renewal thereof, is excluded as temporary financing under Sec. 1003.3(c)(3). iv. Lender A extends credit to finance construction of a dwelling. The loan automatically will convert to permanent financing with Lender A once the construction phase is complete. Under Sec. 1003.3(c)(3), the loan is not designed to be replaced by permanent financing and therefore the temporary financing exclusion does not apply. See also comment 2(j)-3. v. Lender A originates a loan with a nine-month term to enable an investor to purchase a home, renovate it, and re-sell it before the term expires. Under Sec. 1003.3(c)(3), the loan is not designed to be replaced by permanent financing and therefore the temporary financing exclusion does not apply. Such a transaction is not temporary financing under Sec. 1003.3(c)(3) merely because its term is short. Paragraph 3(c)(4) 1. Purchase of an interest in a pool of loans. Section 1003.3(c)(4) provides that the purchase of an interest in a pool of closed-end mortgage loans or open-end lines of credit is an excluded transaction. The purchase of an interest in a pool of loans or lines of credit includes, for example, mortgage-participation certificates, mortgage-backed securities, or real estate mortgage investment conduits. Paragraph 3(c)(6) 1. Mergers and acquisitions. Section 1003.3(c)(6) provides that the purchase of closed-end mortgage loans or open-end lines of credit as part of a merger or acquisition, or as part of the acquisition of all of the assets and liabilities of a branch office, are excluded transactions. If a financial institution acquires loans or lines of credit in bulk from another institution (for example, from the receiver for a failed institution), but no merger or acquisition of an institution, or acquisition of a branch office, is involved and no other exclusion applies, the acquired loans or lines of credit are covered loans and are reported as described in comment 4(a)-1.iii. Paragraph 3(c)(8) 1. Partial interest. Section 1003.3(c)(8) provides that the purchase of a partial interest in a closed-end mortgage loan or an open-end line of credit is an excluded transaction. If an institution acquires only a partial interest in a loan or line of credit, the institution does not report the transaction even if the institution participated in the underwriting and origination of the loan or line of credit. If an institution acquires a 100 percent interest in a loan or line of credit, the transaction is not excluded under Sec. 1003.3(c)(8). Paragraph 3(c)(9) 1. Loan or line of credit used primarily for agricultural purposes. Section 1003.3(c)(9) provides that an institution does not report a closed-end mortgage loan or an open-end line of credit used primarily for agricultural purposes. A loan or line of credit is used primarily for agricultural purposes if its funds will be used primarily for agricultural purposes, or if the loan or line of credit is secured by a dwelling that is located on real property that is used primarily for agricultural purposes (e.g., a farm). An institution may refer to comment 3(a)-8 in the official interpretations of Regulation Z, 12 CFR part 1026, supplement I, for guidance on what is an agricultural purpose. An institution may use any reasonable standard to determine the primary use of the property. An institution may select the standard to apply on a case- by-case basis. Paragraph 3(c)(10) 1. General. Section 1003.3(c)(10) provides a special rule for reporting a closed-end mortgage loan or an open-end line of credit that is or will be made primarily for a business or commercial purpose. If an institution determines that a closed-end mortgage loan or an open-end line of credit primarily is for a business or commercial purpose, then the loan or line of credit is a covered loan only if it is a home improvement loan under Sec. 1003.2(i), a home purchase loan under Sec. 1003.2(j), or a refinancing under Sec. 1003.2(p) and no other exclusion applies. Section 1003.3(c)(10) does not categorically exclude all business- or commercial-purpose loans and lines of credit from coverage. 2. Primary purpose. An institution must determine in each case if a closed-end mortgage loan or an open-end line of credit primarily is for a business or commercial purpose. If a closed-end mortgage loan or an open-end line of credit is deemed to be primarily for a business, commercial, or organizational purpose under Regulation Z, 12 CFR 1026.3(a) and its related commentary, then the loan or line of credit also is deemed [[Page 66322]] to be primarily for a business or commercial purpose under Sec. 1003.3(c)(10). 3. Examples--covered business- or commercial-purpose transactions. The following are examples of closed-end mortgage loans and open-end lines of credit that are not excluded from reporting under Sec. 1003.3(c)(10), because they primarily are for a business or commercial purpose, but they also meet the definition of a home improvement loan under Sec. 1003.2(i), a home purchase loan under Sec. 1003.2(j), or a refinancing under Sec. 1003.2(p): i. A closed-end mortgage loan or an open-end line of credit to purchase or to improve a multifamily dwelling or a single-family investment property, or a refinancing of a closed-end mortgage loan or an open-end line of credit secured by a multifamily dwelling or a single-family investment property; ii. A closed-end mortgage loan or an open-end line of credit to improve an office, for example a doctor's office, that is located in a dwelling; and iii. A closed-end mortgage loan or an open-end line of credit to a corporation, if the funds from the loan or line of credit will be used to purchase or to improve a dwelling, or if the transaction is a refinancing. 4. Examples--excluded business- or commercial-purpose transactions. The following are examples of closed-end mortgage loans and open-end lines of credit that are not covered loans because they primarily are for a business or commercial purpose, but they do not meet the definition of a home improvement loan under Sec. 1003.2(i), a home purchase loan under Sec. 1003.2(j), or a refinancing under Sec. 1003.2(p): i. A closed-end mortgage loan or an open-end line of credit whose funds will be used primarily to improve or expand a business, for example to renovate a family restaurant that is not located in a dwelling, or to purchase a warehouse, business equipment, or inventory; ii. A closed-end mortgage loan or an open-end line of credit to a corporation whose funds will be used primarily for business purposes, such as to purchase inventory; and iii. A closed-end mortgage loan or an open-end line of credit whose funds will be used primarily for business or commercial purposes other than home purchase, home improvement, or refinancing, even if the loan or line of credit is cross-collateralized by a covered loan. Paragraph 3(c)(11) 1. General. Section 1003.3(c)(11) provides that a closed-end mortgage loan is an excluded transaction if a financial institution originated fewer than 25 closed-end mortgage loans in each of the two preceding calendar years. For example, assume that a bank is a financial institution in 2022 under Sec. 1003.2(g) because it originated 200 open-end lines of credit in 2020, 250 open-end lines of credit in 2021, and met all of the other requirements under Sec. 1003.2(g)(1). Also assume that the bank originated 10 and 20 closed- end mortgage loans in 2020 and 2021, respectively. The open-end lines of credit that the bank originated, or for which it received applications, during 2022 are covered loans and must be reported, unless they otherwise are excluded transactions under Sec. 1003.3(c). However, the closed-end mortgage loans that the bank originated, or for which it received applications, during 2022 are excluded transactions under Sec. 1003.3(c)(11) and need not be reported. See comments 4(a)-2 through -4 for guidance about the activities that constitute an origination. Paragraph 3(c)(12) 1. General. Section 1003.3(c)(12) provides that an open-end line of credit is an excluded transaction if a financial institution originated fewer than 100 open-end lines of credit in each of the two preceding calendar years. For example, assume that a bank is a financial institution in 2022 under Sec. 1003.2(g) because it originated 50 closed-end mortgage loans in 2020, 75 closed-end mortgage loans in 2021, and met all of the other requirements under Sec. 1003.2(g)(1). Also assume that the bank originated 75 and 85 open-end lines of credit in 2020 and 2021, respectively. The closed- end mortgage loans that the bank originated, or for which it received applications, during 2022 are covered loans and must be reported, unless they otherwise are excluded transactions under Sec. 1003.3(c). However, the open-end lines of credit that the bank originated, or for which it received applications, during 2022 are excluded transactions under Sec. 1003.3(c)(12) and need not be reported. See comments 4(a)-2 through -4 for guidance about the activities that constitute an origination. Section 1003.4--Compilation of Reportable Data 4(a) Data Format and Itemization 1. General. Section 1003.4(a) describes a financial institution's obligation to collect data on applications it received, on covered loans that it originated, and on covered loans that it purchased during the calendar year covered by the loan/ application register. i. A financial institution reports these data even if the covered loans were subsequently sold by the institution. ii. A financial institution reports data for applications that did not result in an origination but on which actions were taken-for example, an application that the institution denied, that it approved but that was not accepted, that it closed for incompleteness, or that the applicant withdrew during the calendar year covered by the loan/application register. A financial institution is required to report data regarding requests under a preapproval program (as defined in Sec. 1003.2(b)(2)) only if the preapproval request is denied, results in the origination of a home purchase loan, or was approved but not accepted. iii. If a financial institution acquires covered loans in bulk from another institution (for example, from the receiver for a failed institution), but no merger or acquisition of an institution, or acquisition of a branch office, is involved, the acquiring financial institution reports the covered loans as purchased loans. iv. A financial institution reports the data for an application on the loan/application register for the calendar year during which the application was acted upon even if the institution received the application in a previous calendar year. 2. Originations and applications involving more than one institution. Section 1003.4(a) requires a financial institution to collect certain information regarding applications for covered loans that it receives and regarding covered loans that it originates. The following provides guidance on how to report originations and applications involving more than one institution. The discussion below assumes that all of the parties are financial institutions as defined by Sec. 1003.2(g). The same principles apply if any of the parties is not a financial institution. Comment 4(a)-3 provides examples of transactions involving more than one institution, and comment 4(a)-4 discusses how to report actions taken by agents. i. Only one financial institution reports each originated covered loan as an origination. If more than one institution was involved in the origination of a covered loan, the financial institution that made the credit decision approving the application before closing or account opening reports the loan as an origination. It is not relevant whether the loan closed or, in the case of an application, would have closed in the institution's name. If more than one institution approved an application prior to closing or account opening and one of those institutions purchased the loan after closing, the institution that purchased the loan after closing reports the loan as an origination. If a financial institution reports a transaction as an origination, it reports all of the information required for originations, even if the covered loan was not initially payable to the financial institution that is reporting the covered loan as an origination. ii. In the case of an application for a covered loan that did not result in an origination, a financial institution reports the action it took on that application if it made a credit decision on the application or was reviewing the application when the application was withdrawn or closed for incompleteness. It is not relevant whether the financial institution received the application from the applicant or from another institution, such as a broker, or whether another financial institution also reviewed and reported an action taken on the same application. 3. Examples--originations and applications involving more than one institution. The following scenarios illustrate how an institution reports a particular application or covered loan. The illustrations assume that all of the parties are financial institutions as defined by Sec. 1003.2(g). However, the same principles apply if any of the parties is not a financial institution. i. Financial Institution A received an application for a covered loan from an applicant and forwarded that application to Financial Institution B. Financial Institution B reviewed the application and approved the loan prior to closing. The loan closed in Financial Institution A's name. Financial Institution B purchased the loan from Financial Institution A after closing. Financial Institution B was not acting as [[Page 66323]] Financial Institution A's agent. Since Financial Institution B made the credit decision prior to closing, Financial Institution B reports the transaction as an origination, not as a purchase. Financial Institution A does not report the transaction. ii. Financial Institution A received an application for a covered loan from an applicant and forwarded that application to Financial Institution B. Financial Institution B reviewed the application before the loan would have closed, but the application did not result in an origination because Financial Institution B denied the application. Financial Institution B was not acting as Financial Institution A's agent. Since Financial Institution B made the credit decision, Financial Institution B reports the application as a denial. Financial Institution A does not report the application. If, under the same facts, the application was withdrawn before Financial Institution B made a credit decision, Financial Institution B would report the application as withdrawn and Financial Institution A would not report the application. iii. Financial Institution A received an application for a covered loan from an applicant and approved the application before closing the loan in its name. Financial Institution A was not acting as Financial Institution B's agent. Financial Institution B purchased the covered loan from Financial Institution A. Financial Institution B did not review the application before closing. Financial Institution A reports the loan as an origination. Financial Institution B reports the loan as a purchase. iv. Financial Institution A received an application for a covered loan from an applicant. If approved, the loan would have closed in Financial Institution B's name. Financial Institution A denied the application without sending it to Financial Institution B for approval. Financial Institution A was not acting as Financial Institution B's agent. Since Financial Institution A made the credit decision before the loan would have closed, Financial Institution A reports the application. Financial Institution B does not report the application. v. Financial Institution A reviewed an application and made the credit decision to approve a covered loan using the underwriting criteria provided by a third party (e.g., another financial institution, Fannie Mae, or Freddie Mac). The third party did not review the application and did not make a credit decision prior to closing. Financial Institution A was not acting as the third party's agent. Financial Institution A reports the application or origination. If the third party purchased the loan and is subject to Regulation C, the third party reports the loan as a purchase whether or not the third party reviewed the loan after closing. Assume the same facts, except that Financial Institution A approved the application, and the applicant chose not to accept the loan from Financial Institution A. Financial Institution A reports the application as approved but not accepted and the third party, assuming the third party is subject to Regulation C, does not report the application. vi. Financial Institution A reviewed and made the credit decision on an application based on the criteria of a third-party insurer or guarantor (for example, a government or private insurer or guarantor). Financial Institution A reports the action taken on the application. vii. Financial Institution A received an application for a covered loan and forwarded it to Financial Institutions B and C. Financial Institution A made a credit decision, acting as Financial Institution D's agent, and approved the application. The applicant did not accept the loan from Financial Institution D. Financial Institution D reports the application as approved but not accepted. Financial Institution A does not report the application. Financial Institution B made a credit decision, approving the application, the applicant accepted the offer of credit from Financial Institution B, and credit was extended. Financial Institution B reports the origination. Financial Institution C made a credit decision and denied the application. Financial Institution C reports the application as denied. 4. Agents. If a financial institution made the credit decision on a covered loan or application through the actions of an agent, the institution reports the application or origination. State law determines whether one party is the agent of another. For example, acting as Financial Institution A's agent, Financial Institution B approved an application prior to closing and a covered loan was originated. Financial Institution A reports the loan as an origination. 5. Purchased loans. i. A financial institution is required to collect data regarding covered loans it purchases. For purposes of Sec. 1003.4(a), a purchase includes a repurchase of a covered loan, regardless of whether the institution chose to repurchase the covered loan or was required to repurchase the covered loan because of a contractual obligation and regardless of whether the repurchase occurs within the same calendar year that the covered loan was originated or in a different calendar year. For example, assume that Financial Institution A originates or purchases a covered loan and then sells it to Financial Institution B, who later requires Financial Institution A to repurchase the covered loan pursuant to the relevant contractual obligations. Financial Institution B reports the purchase from Financial Institution A, assuming it is a financial institution as defined under Sec. 1003.2(g). Financial Institution A reports the repurchase from Financial Institution B as a purchase. ii. In contrast, for purposes of Sec. 1003.4(a), a purchase does not include a temporary transfer of a covered loan to an interim funder or warehouse creditor as part of an interim funding agreement under which the originating financial institution is obligated to repurchase the covered loan for sale to a subsequent investor. Such agreements, often referred to as ``repurchase agreements,'' are sometimes employed as functional equivalents of warehouse lines of credit. Under these agreements, the interim funder or warehouse creditor acquires legal title to the covered loan, subject to an obligation of the originating institution to repurchase at a future date, rather than taking a security interest in the covered loan as under the terms of a more conventional warehouse line of credit. To illustrate, assume Financial Institution A has an interim funding agreement with Financial Institution B to enable Financial Institution B to originate loans. Assume further that Financial Institution B originates a covered loan and that, pursuant to this agreement, Financial Institution A takes a temporary transfer of the covered loan until Financial Institution B arranges for the sale of the covered loan to a subsequent investor and that Financial Institution B repurchases the covered loan to enable it to complete the sale to the subsequent investor (alternatively, Financial Institution A may transfer the covered loan directly to the subsequent investor at Financial Institution B's direction, pursuant to the interim funding agreement). The subsequent investor could be, for example, a financial institution or other entity that intends to hold the loan in portfolio, a GSE or other securitizer, or a financial institution or other entity that intends to package and sell multiple loans to a GSE or other securitizer. In this example, the temporary transfer of the covered loan from Financial Institution B to Financial Institution A is not a purchase, and any subsequent transfer back to Financial Institution B for delivery to the subsequent investor is not a purchase, for purposes of Sec. 1003.4(a). Financial Institution B reports the origination of the covered loan as well as its sale to the subsequent investor. If the subsequent investor is a financial institution under Sec. 1003.2(g), it reports a purchase of the covered loan pursuant to Sec. 1003.4(a), regardless of whether it acquired the covered loan from Financial Institution B or directly from Financial Institution A. Paragraph 4(a)(1)(i) 1. ULI--uniqueness. Section 1003.4(a)(1)(i)(B)(2) requires a financial institution that assigns a universal loan identifier (ULI) to each covered loan or application (except as provided in Sec. 1003.4(a)(1)(i)(D) and (E)) to ensure that the character sequence it assigns is unique within the institution and used only for the covered loan or application. A financial institution should assign only one ULI to any particular covered loan or application, and each ULI should correspond to a single application and ensuing loan in the case that the application is approved and a loan is originated. A financial institution may use a ULI that was reported previously to refer only to the same loan or application for which the ULI was used previously or a loan that ensues from an application for which the ULI was used previously. A financial institution may not report an application for a covered loan in 2030 using the same ULI that was reported for a covered loan that was originated in 2020. Similarly, refinancings or applications for refinancing should be assigned a different ULI than the loan that is being refinanced. A financial institution with multiple branches must ensure that its branches do not use the same ULI to refer to multiple covered loans or applications. 2. ULI--privacy. Section 1003.4(a)(1)(i)(B)(3) prohibits a financial institution from including information that could be used to directly identify the applicant or borrower in the identifier that it [[Page 66324]] assigns for the application or covered loan of the applicant or borrower. Information that could be used to directly identify the applicant or borrower includes, but is not limited to, the applicant's or borrower's name, date of birth, Social Security number, official government-issued driver's license or identification number, alien registration number, government passport number, or employer or taxpayer identification number. 3. ULI--purchased covered loan. If a financial institution has previously reported a covered loan with a ULI under this part, a financial institution that purchases that covered loan must use the ULI that was previously reported under this part. For example, if a loan origination was previously reported under this part with a ULI, the financial institution that purchases the covered loan would report the purchase of the covered loan using the same ULI. A financial institution that purchases a covered loan must use the ULI that was assigned by the financial institution that originated the covered loan. For example, if a financial institution that submits an annual loan/application register pursuant to Sec. 1003.5(a)(1)(i) originates a covered loan that is purchased by a financial institution that submits a quarterly loan/application register pursuant to Sec. 1003.5(a)(1)(ii), the financial institution that purchased the covered loan must use the ULI that was assigned by the financial institution that originated the covered loan. A financial institution that purchases a covered loan assigns a ULI and records and submits it in its loan/application register pursuant to Sec. 1003.5(a)(1)(i) or (ii), whichever is applicable, if the covered loan was not assigned a ULI by the financial institution that originated the loan because, for example, the loan was originated prior to January 1, 2018. 4. ULI--reinstated or reconsidered application. A financial institution may, at its option, use a ULI previously reported under this part if, during the same calendar year, an applicant asks the institution to reinstate a counteroffer that the applicant previously did not accept or asks the financial institution to reconsider an application that was previously denied, withdrawn, or closed for incompleteness. For example, if a financial institution reports a denied application in its second-quarter 2020 data submission, pursuant to Sec. 1003.5(a)(1)(ii), but then reconsiders the application, which results in an origination in the third quarter of 2020, the financial institution may report the origination in its third-quarter 2020 data submission using the same ULI that was reported for the denied application in its second- quarter 2020 data submission, so long as the financial institution treats the transaction as a continuation of the application. However, a financial institution may not use a ULI previously reported if it reinstates or reconsiders an application that occurred and was reported in a prior calendar year. For example, if a financial institution reports a denied application in its fourth- quarter 2020 data submission, pursuant to Sec. 1003.5(a)(1)(ii), but then reconsiders the application resulting in an origination in the first quarter of 2021, the financial institution reports a denied application under the original ULI in its fourth-quarter 2020 data submission and an approved application with a different ULI in its first-quarter 2021 data submission, pursuant to Sec. 1003.5(a)(1)(ii). 5. ULI--check digit. Section 1003.(4)(a)(1)(i)(C) requires that the two right-most characters in the ULI represent the check digit. Appendix C prescribes the requirements for generating a check digit and validating a ULI. Paragraph 4(a)(1)(ii) 1. Application date--consistency. Section 1003.4(a)(1)(ii) requires that, in reporting the date of application, a financial institution report the date it received the application, as defined under Sec. 1003.2(b), or the date shown on the application form. Although a financial institution need not choose the same approach for its entire HMDA submission, it should be generally consistent (such as by routinely using one approach within a particular division of the institution or for a category of loans). If the financial institution chooses to report the date shown on the application form and the institution retains multiple versions of the application form, the institution reports the date shown on the first application form satisfying the application definition provided under Sec. 1003.2(b). 2. Application date--indirect application. For an application that was not submitted directly to the financial institution, the institution may report the date the application was received by the party that initially received the application, the date the application was received by the institution, or the date shown on the application form. Although an institution need not choose the same approach for its entire HMDA submission, it should be generally consistent (such as by routinely using one approach within a particular division of the institution or for a category of loans). 3. Application date--reinstated application. If, within the same calendar year, an applicant asks a financial institution to reinstate a counteroffer that the applicant previously did not accept (or asks the institution to reconsider an application that was denied, withdrawn, or closed for incompleteness), the institution may treat that request as the continuation of the earlier transaction using the same ULI or as a new transaction with a new ULI. If the institution treats the request for reinstatement or reconsideration as a new transaction, it reports the date of the request as the application date. If the institution does not treat the request for reinstatement or reconsideration as a new transaction, it reports the original application date. Paragraph 4(a)(2) 1. Loan type--general. If a covered loan is not, or in the case of an application would not have been, insured by the Federal Housing Administration, guaranteed by the Veterans Administration, or guaranteed by the Rural Housing Service or the Farm Service Agency, an institution complies with Sec. 1003.4(a)(2) by reporting the covered loan as not insured or guaranteed by the Federal Housing Administration, Veterans Administration, Rural Housing Service, or Farm Service Agency. Paragraph 4(a)(3) 1. Purpose--statement of applicant. A financial institution may rely on the oral or written statement of an applicant regarding the proposed use of covered loan proceeds. For example, a lender could use a check-box or a purpose line on a loan application to determine whether the applicant intends to use covered loan proceeds for home improvement purposes. If an applicant provides no statement as to the proposed use of covered loan proceeds and the covered loan is not a home purchase loan, cash-out refinancing, or refinancing, a financial institution reports the covered loan as for a purpose other than home purchase, home improvement, refinancing, or cash-out refinancing for purposes of Sec. 1003.4(a)(3). 2. Purpose--refinancing and cash-out refinancing. Section 1003.4(a)(3) requires a financial institution to report whether a covered loan is, or an application is for, a refinancing or a cash- out refinancing. A financial institution reports a covered loan or an application as a cash-out refinancing if it is a refinancing as defined by Sec. 1003.2(p) and the institution considered it to be a cash-out refinancing in processing the application or setting the terms (such as the interest rate or origination charges) under its guidelines or an investor's guidelines. For example: i. Assume a financial institution considers an application for a loan product to be a cash-out refinancing under an investor's guidelines because of the amount of cash received by the borrower at closing or account opening. Assume also that under the investor's guidelines, the applicant qualifies for the loan product and the financial institution approves the application, originates the covered loan, and sets the terms of the covered loan consistent with the loan product. In this example, the financial institution would report the covered loan as a cash-out refinancing for purposes of Sec. 1003.4(a)(3). ii. Assume a financial institution does not consider an application for a covered loan to be a cash-out refinancing under its own guidelines because the amount of cash received by the borrower does not exceed a certain threshold. Assume also that the institution approves the application, originates the covered loan, and sets the terms of the covered loan consistent with its own guidelines applicable to refinancings other than cash-out refinancings. In this example, the financial institution would report the covered loan as a refinancing for purposes of Sec. 1003.4(a)(3). iii. Assume a financial institution does not distinguish between a cash-out refinancing and a refinancing under its own guidelines, and sets the terms of all refinancings without regard to the amount of cash received by the borrower at closing or account opening, and does not offer loan products under investor guidelines. In this example, the financial institution reports all covered loans and applications for covered loans that are defined by Sec. 1003.2(p) as refinancings for purposes of Sec. 1003.4(a)(3). 3. Purpose--multiple-purpose loan. Section 1003.4(a)(3) requires a financial institution to report the purpose of a covered loan or application. If a covered loan is a [[Page 66325]] home purchase loan as well as a home improvement loan, a refinancing, or a cash-out refinancing, an institution complies with Sec. 1003.4(a)(3) by reporting the loan as a home purchase loan. If a covered loan is a home improvement loan as well as a refinancing or cash-out refinancing, but the covered loan is not a home purchase loan, an institution complies with Sec. 1003.4(a)(3) by reporting the covered loan as a refinancing or a cash-out refinancing, as appropriate. If a covered loan is a refinancing or cash-out refinancing as well as for another purpose, such as for the purpose of paying educational expenses, but the covered loan is not a home purchase loan, an institution complies with Sec. 1003.4(a)(3) by reporting the covered loan as a refinancing or a cash-out refinancing, as appropriate. See comment 4(a)(3)-2. If a covered loan is a home improvement loan as well as for another purpose, but the covered loan is not a home purchase loan, a refinancing, or cash-out refinancing, an institution complies with Sec. 1003.4(a)(3) by reporting the covered loan as a home improvement loan. See comment 2(i)-1. 4. Purpose--other. If a covered loan is not, or an application is not for, a home purchase loan, a home improvement loan, a refinancing, or a cash-out refinancing, a financial institution complies with Sec. 1003.4(a)(3) by reporting the covered loan or application as for a purpose other than home purchase, home improvement, refinancing, or cash-out refinancing. For example, if a covered loan is for the purpose of paying educational expenses, the financial institution complies with Sec. 1003.4(a)(3) by reporting the covered loan as for a purpose other than home purchase, home improvement, refinancing, or cash-out refinancing. Section 1003.4(a)(3) also requires an institution to report a covered loan or application as for a purpose other than home purchase, home improvement, refinancing, or cash-out refinancing if it is a refinancing but, under the terms of the agreement, the financial institution was unconditionally obligated to refinance the obligation subject to conditions within the borrower's control. 5. Purpose--business or commercial purpose loans. If a covered loan primarily is for a business or commercial purpose as described in Sec. 1003.3(c)(10) and comment 3(c)(10)-2 and is a home purchase loan, home improvement loan, or a refinancing, Sec. 1003.4(a)(3) requires the financial institution to report the applicable loan purpose. If a loan primarily is for a business or commercial purpose but is not a home purchase loan, home improvement loan, or a refinancing, the loan is an excluded transaction under Sec. 1003.3(c)(10). Paragraph 4(a)(4) 1. Request under a preapproval program. Section 1003.4(a)(4) requires a financial institution to report whether an application or covered loan involved a request for a preapproval of a home purchase loan under a preapproval program as defined by Sec. 1003.2(b)(2). If an application or covered loan did not involve a request for a preapproval of a home purchase loan under a preapproval program as defined by Sec. 1003.2(b)(2), a financial institution complies with Sec. 1003.4(a)(4) by reporting that the application or covered loan did not involve such a request, regardless of whether the institution has such a program and the applicant did not apply through that program or the institution does not have a preapproval program as defined by Sec. 1003.2(b)(2). 2. Scope of requirement. A financial institution reports that the application or covered loan did not involve a preapproval request for a purchased covered loan; an application or covered loan for any purpose other than a home purchase loan; an application for a home purchase loan or a covered loan that is a home purchase loan secured by a multifamily dwelling; an application or covered loan that is an open-end line of credit or a reverse mortgage; or an application that is denied, withdrawn by the applicant, or closed for incompleteness. Paragraph 4(a)(5) 1. Modular homes and prefabricated components. Covered loans or applications related to modular homes should be reported with a construction method of site-built, regardless of whether they are on-frame or off-frame modular homes. Modular homes comply with local or other recognized buildings codes rather than standards established by the National Manufactured Housing Construction and Safety Standards Act, 42 U.S.C. 5401 et seq. Modular homes are not required to have HUD Certification Labels under 24 CFR 3280.11 or data plates under 24 CFR 3280.5. Modular homes may have a certification from a State licensing agency that documents compliance with State or other applicable building codes. On-frame modular homes are constructed on permanent metal chassis similar to those used in manufactured homes. The chassis are not removed on site and are secured to the foundation. Off-frame modular homes typically have floor construction similar to the construction of other site-built homes, and the construction typically includes wooden floor joists and does not include permanent metal chassis. Dwellings built using prefabricated components assembled at the dwelling's permanent site should also be reported with a construction method of site-built. 2. Multifamily dwelling. For a covered loan or an application for a covered loan related to a multifamily dwelling, the financial institution should report the construction method as site-built unless the multifamily dwelling is a manufactured home community, in which case the financial institution should report the construction method as manufactured home. 3. Multiple properties. See comment 4(a)(9)-2 regarding transactions involving multiple properties with more than one property taken as security. Paragraph 4(a)(6) 1. Multiple properties. See comment 4(a)(9)-2 regarding transactions involving multiple properties with more than one property taken as security. 2. Principal residence. Section 1003.4(a)(6) requires a financial institution to identify whether the property to which the covered loan or application relates is or will be used as a residence that the applicant or borrower physically occupies and uses, or will occupy and use, as his or her principal residence. For purposes of Sec. 1003.4(a)(6), an applicant or borrower can have only one principal residence at a time. Thus, a vacation or other second home would not be a principal residence. However, if an applicant or borrower buys or builds a new dwelling that will become the applicant's or borrower's principal residence within a year or upon the completion of construction, the new dwelling is considered the principal residence for purposes of applying this definition to a particular transaction. 3. Second residences. Section 1003.4(a)(6) requires a financial institution to identify whether the property to which the loan or application relates is or will be used as a second residence. For purposes of Sec. 1003.4(a)(6), a property is a second residence of an applicant or borrower if the property is or will be occupied by the applicant or borrower for a portion of the year and is not the applicant's or borrower's principal residence. For example, if a person purchases a property, occupies the property for a portion of the year, and rents the property for the remainder of the year, the property is a second residence for purposes of Sec. 1003.4(a)(6). Similarly, if a couple occupies a property near their place of employment on weekdays, but the couple returns to their principal residence on weekends, the property near the couple's place of employment is a second residence for purposes of Sec. 1003.4(a)(6). 4. Investment properties. Section 1003.4(a)(6) requires a financial institution to identify whether the property to which the covered loan or application relates is or will be used as an investment property. For purposes of Sec. 1003.4(a)(6), a property is an investment property if the borrower does not, or the applicant will not, occupy the property. For example, if a person purchases a property, does not occupy the property, and generates income by renting the property, the property is an investment property for purposes of Sec. 1003.4(a)(6). Similarly, if a person purchases a property, does not occupy the property, and does not generate income by renting the property, but intends to generate income by selling the property, the property is an investment property for purposes of Sec. 1003.4(a)(6). Section 1003.4(a)(6) requires a financial institution to identify a property as an investment property if the borrower or applicant does not or will not occupy the property, even if the borrower or applicant does not consider the property as owned for investment purposes. For example, if a corporation purchases a property that is a dwelling under Sec. 1003.2(f), that it does not occupy, but that is for the long-term residential use of its employees, the property is an investment property for purposes of Sec. 1003.4(a)(6), even if the corporation considers the property as owned for business purposes rather than investment purposes, does not generate income by renting the property, and does not intend to generate income by selling the property at some point in time. If the property is for transitory use by employees, the property would not be considered a dwelling under Sec. 1003.2(f). See comment 2(f)-3. [[Page 66326]] 5. Purchased covered loans. For purchased covered loans, a financial institution may report principal residence unless the loan documents or application indicate that the property will not be occupied as a principal residence. Paragraph 4(a)(7) 1. Covered loan amount--counteroffer. If an applicant accepts a counteroffer for an amount different from the amount for which the applicant applied, the financial institution reports the covered loan amount granted. If an applicant does not accept a counteroffer or fails to respond, the institution reports the amount initially requested. 2. Covered loan amount--application approved but not accepted or preapproval request approved but not accepted. A financial institution reports the covered loan amount that was approved. 3. Covered loan amount--preapproval request denied, application denied, closed for incompleteness or withdrawn. For a preapproval request that was denied, and for an application that was denied, closed for incompleteness, or withdrawn, a financial institution reports the amount for which the applicant applied. 4. Covered loan amount--multiple-purpose loan. A financial institution reports the entire amount of the covered loan, even if only a part of the proceeds is intended for home purchase, home improvement, or refinancing. 5. Covered loan amount--closed-end mortgage loan. For a closed- end mortgage loan, other than a purchased loan, an assumption, or a reverse mortgage, a financial institution reports the amount to be repaid as disclosed on the legal obligation. For a purchased closed- end mortgage loan or an assumption of a closed-end mortgage loan, a financial institution reports the unpaid principal balance at the time of purchase or assumption. 6. Covered loan amount--open-end line of credit. For an open-end line of credit, a financial institution reports the entire amount of credit available to the borrower under the terms of the open-end plan, including a purchased open-end line of credit and an assumption of an open-end line of credit, but not for a reverse mortgage open-end line of credit. 7. Covered loan amount--refinancing. For a refinancing, a financial institution reports the amount of credit extended under the terms of the new debt obligation. 8. Covered loan amount--home improvement loan. A financial institution reports the entire amount of a home improvement loan, even if only a part of the proceeds is intended for home improvement. 9. Covered loan amount--non-federally insured reverse mortgage. A financial institution reports the initial principal limit of a non-federally insured reverse mortgage as set forth in Sec. 1003.4(a)(7)(iii). Paragraph 4(a)(8)(i) 1. Action taken--covered loan originated. A financial institution reports that the covered loan was originated if the financial institution made a credit decision approving the application before closing or account opening and that credit decision results in an extension of credit. The same is true for an application that began as a request for a preapproval that subsequently results in a covered loan being originated. See comments 4(a)-2 through -4 for guidance on transactions in which more than one institution is involved. 2. Action taken--covered loan purchased. A financial institution reports that the covered loan was purchased if the covered loan was purchased by the financial institution after closing or account opening and the financial institution did not make a credit decision on the application prior to closing or account opening, or if the financial institution did make a credit decision on the application prior to closing or account opening, but is repurchasing the loan from another entity that the loan was sold to. See comment 4(a)-5. See comments 4(a)-2 through -4 for guidance on transactions in which more than one financial institution is involved. 3. Action taken--application approved but not accepted. A financial institution reports application approved but not accepted if the financial institution made a credit decision approving the application before closing or account opening, subject solely to outstanding conditions that are customary commitment or closing conditions, but the applicant or the party that initially received the application fails to respond to the financial institution's approval within the specified time, or the closed-end mortgage loan was not otherwise consummated or the account was not otherwise opened. See comment 4(a)(8)(i)-13. 4. Action taken--application denied. A financial institution reports that the application was denied if it made a credit decision denying the application before an applicant withdraws the application or the file is closed for incompleteness. See comments 4(a)-2 through -4 for guidance on transactions in which more than one institution is involved. 5. Action taken--application withdrawn. A financial institution reports that the application was withdrawn when the application is expressly withdrawn by the applicant before the financial institution makes a credit decision denying the application, before the financial institution makes a credit decision approving the application, or before the file is closed for incompleteness. A financial institution also reports application withdrawn if the financial institution provides a conditional approval specifying underwriting or creditworthiness conditions, pursuant to comment 4(a)(8)(i)-13, and the application is expressly withdrawn by the applicant before the applicant satisfies all specified underwriting or creditworthiness conditions. A preapproval request that is withdrawn is not reportable under HMDA. See Sec. 1003.4(a). 6. Action taken--file closed for incompleteness. A financial institution reports that the file was closed for incompleteness if the financial institution sent a written notice of incompleteness under Regulation B, 12 CFR 1002.9(c)(2), and the applicant did not respond to the request for additional information within the period of time specified in the notice before the applicant satisfies all underwriting or creditworthiness conditions. See comment 4(a)(8)(i)- 13. If a financial institution then provides a notification of adverse action on the basis of incompleteness under Regulation B, 12 CFR 1002.9(c)(i), the financial institution may report the action taken as either file closed for incompleteness or application denied. A preapproval request that is closed for incompleteness is not reportable under HMDA. See Sec. 1003.4(a). 7. Action taken--preapproval request denied. A financial institution reports that the preapproval request was denied if the application was a request for a preapproval under a preapproval program as defined in Sec. 1003.2(b)(2) and the institution made a credit decision denying the preapproval request. 8. Action taken--preapproval request approved but not accepted. A financial institution reports that the preapproval request was approved but not accepted if the application was a request for a preapproval under a preapproval program as defined in Sec. 1003.2(b)(2) and the institution made a credit decision approving the preapproval request but the application did not result in a covered loan originated by the financial institution. 9. Action taken--counteroffers. If a financial institution makes a counteroffer to lend on terms different from the applicant's initial request (for example, for a shorter loan maturity, with a different interest rate, or in a different amount) and the applicant does not accept the counteroffer or fails to respond, the institution reports the action taken as a denial on the original terms requested by the applicant. If the applicant accepts, the financial institution reports the action taken as covered loan originated. 10. Action taken--rescinded transactions. If a borrower rescinds a transaction after closing and before a financial institution is required to submit its loan/application register containing the information for the transaction under Sec. 1003.5(a), the institution reports the transaction as an application that was approved but not accepted. 11. Action taken--purchased covered loans. An institution reports the covered loans that it purchased during the calendar year. An institution does not report the covered loans that it declined to purchase, unless, as discussed in comments 4(a)-2 through -4, the institution reviewed the application prior to closing, in which case it reports the application or covered loan according to comments 4(a)-2 through -4. 12. Action taken--repurchased covered loans. See comment 4(a)-5 regarding reporting requirements when a covered loan is repurchased by the originating financial institution. 13. Action taken--conditional approvals. If an institution issues an approval other than a commitment pursuant to a preapproval program as defined under Sec. 1003.2(b)(2), and that approval is subject to the applicant meeting certain conditions, the institution reports the action taken as provided below dependent on whether the conditions are solely customary commitment or closing conditions or if the conditions include any underwriting or creditworthiness conditions. [[Page 66327]] i. Action taken examples. If the approval is conditioned on satisfying underwriting or creditworthiness conditions and they are not met, the institution reports the action taken as a denial. If, however, the conditions involve submitting additional information about underwriting or creditworthiness that the institution needs to make the credit decision, and the institution has sent a written notice of incompleteness under Regulation B, 12 CFR 1002.9(c)(2), and the applicant did not respond within the period of time specified in the notice, the institution reports the action taken as file closed for incompleteness. See comment 4(a)(8)(i)-6. If the conditions are solely customary commitment or closing conditions and the conditions are not met, the institution reports the action taken as approved but not accepted. If all the conditions (underwriting, creditworthiness, or customary commitment or closing conditions) are satisfied and the institution agrees to extend credit but the covered loan is not originated, the institution reports the action taken as application approved but not accepted. If the applicant expressly withdraws before satisfying all underwriting or creditworthiness conditions and before the institution denies the application or closes the file for incompleteness, the institution reports the action taken as application withdrawn. If all underwriting and creditworthiness conditions have been met, and the outstanding conditions are solely customary commitment or closing conditions and the applicant expressly withdraws before the covered loan is originated, the institution reports the action taken as application approved but not accepted. ii. Customary commitment or closing conditions. Customary commitment or closing conditions include, for example: a clear-title requirement, an acceptable property survey, acceptable title insurance binder, clear termite inspection, a subordination agreement from another lienholder, and, where the applicant plans to use the proceeds from the sale of one home to purchase another, a settlement statement showing adequate proceeds from the sale. iii. Underwriting or creditworthiness conditions. Underwriting or creditworthiness conditions include, for example: conditions that constitute a counter-offer, such as a demand for a higher down- payment; satisfactory debt-to-income or loan-to-value ratios, a determination of need for private mortgage insurance, or a satisfactory appraisal requirement; or verification or confirmation, in whatever form the institution requires, that the applicant meets underwriting conditions concerning applicant creditworthiness, including documentation or verification of income or assets. 14. Action taken--pending applications. An institution does not report any covered loan application still pending at the end of the calendar year; it reports that application on its loan/application register for the year in which final action is taken. Paragraph 4(a)(8)(ii) 1. Action taken date--general. A financial institution reports the date of the action taken. 2. Action taken date--applications denied and files closed for incompleteness. For applications, including requests for a preapproval, that are denied or for files closed for incompleteness, the financial institution reports either the date the action was taken or the date the notice was sent to the applicant. 3. Action taken date--application withdrawn. For applications withdrawn, the financial institution may report the date the express withdrawal was received or the date shown on the notification form in the case of a written withdrawal. 4. Action taken date--approved but not accepted. For a covered loan approved by an institution but not accepted by the applicant, the institution reports any reasonable date, such as the approval date, the deadline for accepting the offer, or the date the file was closed. Although an institution need not choose the same approach for its entire HMDA submission, it should be generally consistent (such as by routinely using one approach within a particular division of the institution or for a category of covered loans). 5. Action taken date--originations. For covered loan originations, including a preapproval request that leads to an origination by the financial institution, an institution generally reports the closing or account opening date. For covered loan originations that an institution acquires from a party that initially received the application, the institution reports either the closing or account opening date, or the date the institution acquired the covered loan from the party that initially received the application. If the disbursement of funds takes place on a date later than the closing or account opening date, the institution may use the date of initial disbursement. For a construction/permanent covered loan, the institution reports either the closing or account opening date, or the date the covered loan converts to the permanent financing. Although an institution need not choose the same approach for its entire HMDA submission, it should be generally consistent (such as by routinely using one approach within a particular division of the institution or for a category of covered loans). Notwithstanding this flexibility regarding the use of the closing or account opening date in connection with reporting the date action was taken, the institution must report the origination as occurring in the year in which the origination goes to closing or the account is opened. 6. Action taken date--loan purchased. For covered loans purchased, a financial institution reports the date of purchase. Paragraph 4(a)(9) 1. Multiple properties with one property taken as security. If a covered loan is related to more than one property, but only one property is taken as security (or, in the case of an application, proposed to be taken as security), a financial institution reports the information required by Sec. 1003.4(a)(9) for the property taken as or proposed to be taken as security. A financial institution does not report the information required by Sec. 1003.4(a)(9) for the property or properties related to the loan that are not taken as or proposed to be taken as security. For example, if a covered loan is secured by property A, and the proceeds are used to purchase or rehabilitate (or to refinance home purchase or home improvement loans related to) property B, the institution reports the information required by Sec. 1003.4(a)(9) for property A and does not report the information required by Sec. 1003.4(a)(9) for property B. 2. Multiple properties with more than one property taken as security. If more than one property is taken or, in the case of an application, proposed to be taken as security for a single covered loan, a financial institution reports the covered loan or application in a single entry on its loan/application register and provides the information required by Sec. 1003.4(a)(9) for one of the properties taken as security that contains a dwelling. A financial institution does not report information about the other properties taken as security. If an institution is required to report specific information about the property identified in Sec. 1003.4(a)(9), the institution reports the information that relates to the property identified in Sec. 1003.4(a)(9). For example, Financial Institution A originated a covered loan that is secured by both property A and property B, each of which contains a dwelling. Financial Institution A reports the loan as one entry on its loan/ application register, reporting the information required by Sec. 1003.4(a)(9) for either property A or property B. If Financial Institution A elects to report the information required by Sec. 1003.4(a)(9) about property A, Financial Institution A also reports the information required by Sec. 1003.4(a)(5), (6), (14), (29), and (30) related to property A. For aspects of the entries that do not refer to the property identified in Sec. 1003.4(a)(9) (i.e., Sec. 1003.4(a)(1) through (4), (7), (8), (10) through (13), (15) through (28), (31) through (38)), Financial Institution A reports the information applicable to the covered loan or application and not information that relates only to the property identified in Sec. 1003.4(a)(9). 3. Multifamily dwellings. A single multifamily dwelling may have more than one postal address. For example, three apartment buildings, each with a different street address, comprise a single multifamily dwelling that secures a covered loan. For the purposes of Sec. 1003.4(a)(9), a financial institution reports the information required by Sec. 1003.4(a)(9) in the same manner described in comment 4(a)(9)-2. 4. Loans purchased from another institution. The requirement to report the property location information required by Sec. 1003.4(a)(9) applies not only to applications and originations but also to purchased covered loans. 5. Manufactured home. If the site of a manufactured home has not been identified, a financial institution complies by reporting that the information required by Sec. 1003.4(a)(9) is not applicable. Paragraph 4(a)(9)(i) 1. General. Section 1003.4(a)(9)(i) requires a financial institution to report the property address of the location of the property securing a covered loan or, in the case of an application, proposed to secure a covered loan. The address should correspond to the [[Page 66328]] property identified on the legal obligation related to the covered loan. For applications that did not result in an origination, the address should correspond to the location of the property proposed to secure the loan as identified by the applicant. For example, assume a loan is secured by a property located at 123 Main Street, and the applicant's or borrower's mailing address is a post office box. The financial institution should not report the post office box, and should report 123 Main Street. 2. Property address--format. A financial institution complies with the requirements in Sec. 1003.4(a)(9)(i) by reporting the following information about the physical location of the property securing the loan. i. Street address. When reporting the street address of the property, a financial institution complies by including, as applicable, the primary address number, the predirectional, the street name, street prefixes and/or suffixes, the postdirectional, the secondary address identifier, and the secondary address, as applicable. For example, 100 N Main ST Apt 1. ii. City name. A financial institution complies by reporting the name of the city in which the property is located. iii. State name. A financial institution complies by reporting the two letter State code for the State in which the property is located, using the U.S. Postal Service official State abbreviations. iv. Zip Code. A financial institution complies by reporting the five or nine digit Zip Code in which the property is located. 3. Property address--not applicable. A financial institution complies with Sec. 1003.4(a)(9)(i) by indicating that the requirement is not applicable if the property address of the property securing the covered loan is not known. For example, if the property did not have a property address at closing or if the applicant did not provide the property address of the property to the financial institution before the application was denied, withdrawn, or closed for incompleteness, the financial institution complies with Sec. 1003.4(a)(9)(i) by indicating that the requirement is not applicable. Paragraph 4(a)(9)(ii)(B) 1. General. A financial institution complies by reporting the five-digit Federal Information Processing Standards (FIPS) numerical county code. Paragraph 4(a)(9)(ii)(C) 1. General. Census tract numbers are defined by the U.S. Census Bureau. A financial institution complies with Sec. 1003.4(a)(9)(ii)(C) if it uses the boundaries and codes in effect on January 1 of the calendar year covered by the loan/application register that it is reporting. Paragraph 4(a)(10)(i) 1. Applicant data--general. Refer to appendix B to this part for instructions on collection of an applicant's ethnicity, race, and sex. 2. Transition rule for applicant data collected prior to January 1, 2018. If a financial institution receives an application prior to January 1, 2018, but final action is taken on or after January 1, 2018, the financial institution complies with Sec. 1003.4(a)(10)(i) and (b) if it collects the information in accordance with the requirements in effect at the time the information was collected. For example, if a financial institution receives an application on November 15, 2017, collects the applicant's ethnicity, race, and sex in accordance with the instructions in effect on that date, and takes final action on the application on January 5, 2018, the financial institution has complied with the requirements of Sec. 1003.4(a)(10)(i) and (b), even though those instructions changed after the information was collected but before the date of final action. However, if, in this example, the financial institution collected the applicant's ethnicity, race, and sex on or after January 1, 2018, Sec. 1003.4(a)(10)(i) and (b) requires the financial institution to collect the information in accordance with the amended instructions. Paragraph 4(a)(10)(ii) 1. Applicant data--completion by financial institution. A financial institution complies with Sec. 1003.4(a)(10)(ii) by reporting the applicant's age, as of the application date under Sec. 1003.4(a)(1)(ii), as the number of whole years derived from the date of birth as shown on the application form. For example, if an applicant provides a date of birth of 01/15/1970 on the application form that the financial institution receives on 01/14/ 2015, the institution reports 44 as the applicant's age. 2. Applicant data--co-applicant. If there are no co-applicants, the financial institution reports that there is no co-applicant. If there is more than one co-applicant, the financial institution reports the age only for the first co-applicant listed on the application form. A co-applicant may provide an absent co- applicant's age on behalf of the absent co-applicant. 3. Applicant data--purchased loan. A financial institution complies with Sec. 1003.4(a)(10)(ii) by reporting that the requirement is not applicable when reporting a purchased loan for which the institution chooses not to report the income. 4. Applicant data--non-natural person. A financial institution complies with Sec. 1003.4(a)(10)(ii) by reporting that the requirement is not applicable if the applicant or co-applicant is not a natural person (for example, a corporation, partnership, or trust). For example, for a transaction involving a trust, a financial institution reports that the requirement to report the applicant's age is not applicable if the trust is the applicant. On the other hand, if the applicant is a natural person, and is the beneficiary of a trust, a financial institution reports the applicant's age. 5. Applicant data--guarantor. For purposes of Sec. 1003.4(a)(10)(ii), if a covered loan or application includes a guarantor, a financial institution does not report the guarantor's age. Paragraph 4(a)(10)(iii) 1. Income data--income relied on. When a financial institution evaluates income as part of a credit decision, it reports the gross annual income relied on in making the credit decision. For example, if an institution relies on an applicant's salary to compute a debt- to-income ratio but also relies on the applicant's annual bonus to evaluate creditworthiness, the institution reports the salary and the bonus to the extent relied upon. If an institution relies on only a portion of an applicant's income in its determination, it does not report that portion of income not relied on. For example, if an institution, pursuant to lender and investor guidelines, does not rely on an applicant's commission income because it has been earned for less than 12 months, the institution does not include the applicant's commission income in the income reported. Likewise, if an institution relies on the verified gross income of the applicant in making the credit decision, then the institution reports the verified gross income. Similarly, if an institution relies on the income of a cosigner to evaluate creditworthiness, the institution includes the cosigner's income to the extent relied upon. An institution, however, does not include the income of a guarantor who is only secondarily liable. 2. Income data--co-applicant. If two persons jointly apply for a covered loan and both list income on the application, but the financial institution relies on the income of only one applicant in evaluating creditworthiness, the institution reports only the income relied on. 3. Income data--loan to employee. A financial institution complies with Sec. 1003.4(a)(10)(iii) by reporting that the requirement is not applicable for a covered loan to, or an application from, its employee to protect the employee's privacy, even though the institution relied on the employee's income in making the credit decision. 4. Income data--assets. A financial institution does not include as income amounts considered in making a credit decision based on factors that an institution relies on in addition to income, such as amounts derived from annuitization or depletion of an applicant's remaining assets. 5. Income data--credit decision not made. Section 1003.4(a)(10)(iii) requires a financial institution to report the gross annual income relied on in processing the application if a credit decision was not made. For example, assume an institution received an application that included an applicant's self-reported income, but the application was withdrawn before a credit decision that would have considered income was made. The financial institution reports the income information relied on in processing the application at the time that the application was withdrawn or the file was closed for incompleteness. 6. Income data--credit decision not requiring consideration of income. A financial institution complies with Sec. 1003.4(a)(10)(iii) by reporting that the requirement is not applicable if the application did not or would not have required a credit decision that considered income under the financial institution's policies and procedures. For example, if the financial institution's policies and procedures do not consider income for a streamlined refinance program, the institution reports that the requirement is not [[Page 66329]] applicable, even if the institution received income information from the applicant. 7. Income data--non-natural person. A financial institution reports that the requirement is not applicable when the applicant or co-applicant is not a natural person (e.g., a corporation, partnership, or trust). For example, for a transaction involving a trust, a financial institution reports that the requirement to report income data is not applicable if the trust is the applicant. On the other hand, if the applicant is a natural person, and is the beneficiary of a trust, a financial institution is required to report the information described in Sec. 1003.4(a)(10)(iii). 8. Income data--multifamily properties. A financial institution complies with Sec. 1003.4(a)(10)(iii) by reporting that the requirement is not applicable when the covered loan is secured by, or application is proposed to be secured by, a multifamily dwelling. 9. Income data--purchased loans. A financial institution complies with Sec. 1003.4(a)(10)(iii) by reporting that the requirement is not applicable when reporting a purchased covered loan for which the institution chooses not to report the income. 10. Income data--rounding. A financial institution complies by reporting the dollar amount of the income in thousands, rounded to the nearest thousand ($500 rounds up to the next $1,000). For example, $35,500 is reported as 36. Paragraph 4(a)(11) 1. Type of purchaser--loan-participation interests sold to more than one entity. A financial institution that originates a covered loan, and then sells it to more than one entity, reports the ``type of purchaser'' based on the entity purchasing the greatest interest, if any. For purposes of Sec. 1003.4(a)(11), if a financial institution sells some interest or interests in a covered loan but retains a majority interest in that loan, it does not report the sale. 2. Type of purchaser--swapped covered loans. Covered loans ``swapped'' for mortgage-backed securities are to be treated as sales; the purchaser is the entity receiving the covered loans that are swapped. 3. Type of purchaser--affiliate institution. For purposes of complying with Sec. 1003.4(a)(11), the term ``affiliate'' means any company that controls, is controlled by, or is under common control with, another company, as set forth in the Bank Holding Company Act of 1956 (12 U.S.C. 1841 et seq.). 4. Type of purchaser--private securitizations. A financial institution that knows or reasonably believes that the covered loan it is selling will be securitized by the entity purchasing the covered loan, other than by one of the government-sponsored enterprises, reports the purchasing entity type as a private securitizer regardless of the type or affiliation of the purchasing entity. Knowledge or reasonable belief could, for example, be based on the purchase agreement or other related documents, the financial institution's previous transactions with the purchaser, or the purchaser's role as a securitizer (such as an investment bank). If a financial institution selling a covered loan does not know or reasonably believe that the purchaser will securitize the loan, and the seller knows that the purchaser frequently holds or disposes of loans by means other than securitization, then the financial institution should report the covered loan as purchased by, as appropriate, a commercial bank, savings bank, savings association, life insurance company, credit union, mortgage company, finance company, affiliate institution, or other type of purchaser. 5. Type of purchaser--mortgage company. For purposes of complying with Sec. 1003.4(a)(11), a mortgage company means a nondepository institution that purchases covered loans and typically originates such loans. A mortgage company might be an affiliate or a subsidiary of a bank holding company or thrift holding company, or it might be an independent mortgage company. Regardless, a financial institution reports the purchasing entity type as a mortgage company, unless the mortgage company is an affiliate of the seller institution, in which case the seller institution should report the loan as purchased by an affiliate institution. 6. Purchases by subsidiaries. A financial institution that sells a covered loan to its subsidiary that is a commercial bank, savings bank, or savings association, should report the covered loan as purchased by a commercial bank, savings bank, or savings association. A financial institution that sells a covered loan to its subsidiary that is a life insurance company, should report the covered loan as purchased by a life insurance company. A financial institution that sells a covered loan to its subsidiary that is a credit union, mortgage company, or finance company, should report the covered loan as purchased by a credit union, mortgage company, or finance company. If the subsidiary that purchases the covered loan is not a commercial bank, savings bank, savings association, life insurance company, credit union, mortgage company, or finance company, the seller institution should report the loan as purchased by other type of purchaser. The financial institution should report the covered loan as purchased by an affiliate institution when the subsidiary is an affiliate of the seller institution. 7. Type of purchaser--bank holding company or thrift holding company. When a financial institution sells a covered loan to a bank holding company or thrift holding company (rather than to one of its subsidiaries), it should report the loan as purchased by other type of purchaser, unless the bank holding company or thrift holding company is an affiliate of the seller institution, in which case the seller institution should report the loan as purchased by an affiliate institution. 8. Repurchased covered loans. See comment 4(a)-5 regarding reporting requirements when a covered loan is repurchased by the originating financial institution. 9. Type of purchaser--quarterly recording. For purposes of recording the type of purchaser within 30 calendar days after the end of the calendar quarter pursuant to Sec. 1003.4(f), a financial institution records that the requirement is not applicable if the institution originated or purchased a covered loan and did not sell it during the calendar quarter for which the institution is recording the data. If the financial institution sells the covered loan in a subsequent quarter of the same calendar year, the financial institution records the type of purchaser on its loan/ application register for the quarter in which the covered loan was sold. If a financial institution sells the covered loan in a succeeding year, the financial institution should not record the sale. 10. Type of purchaser--not applicable. A financial institution reports that the requirement is not applicable for applications that were denied, withdrawn, closed for incompleteness or approved but not accepted by the applicant; and for preapproval requests that were denied or approved but not accepted by the applicant. A financial institution also reports that the requirement is not applicable if the institution originated or purchased a covered loan and did not sell it during that same calendar year. Paragraph 4(a)(12) 1. Average prime offer rate. Average prime offer rates are annual percentage rates derived from average interest rates, points, and other loan pricing terms offered to borrowers by a representative sample of lenders for mortgage loans that have low- risk pricing characteristics. Other pricing terms include commonly used indices, margins, and initial fixed-rate periods for variable- rate transactions. Relevant pricing characteristics include a consumer's credit history and transaction characteristics such as the loan-to-value ratio, owner-occupant status, and purpose of the transaction. To obtain average prime offer rates, the Bureau uses a survey of lenders that both meets the criteria of Sec. 1003.4(a)(12)(ii) and provides pricing terms for at least two types of variable-rate transactions and at least two types of non- variable-rate transactions. An example of such a survey is the Freddie Mac Primary Mortgage Market Survey[supreg]. 2. Bureau tables. The Bureau publishes on the FFIEC's Web site (http://www.ffiec.gov/hmda), in tables entitled ``Average Prime Offer Rates-Fixed'' and ``Average Prime Offer Rates-Adjustable,'' current and historic average prime offer rates for a wide variety of closed-end transaction types. The Bureau calculates an annual percentage rate, consistent with Regulation Z (see 12 CFR 1026.22 and part 1026, appendix J), for each transaction type for which pricing terms are available from the survey described in comment 4(a)(12)-1. The Bureau uses loan pricing terms available in the survey and other information to estimate annual percentage rates for other types of transactions for which direct survey data are not available. The Bureau publishes on the FFIEC's Web site the methodology it uses to arrive at these estimates. A financial institution may either use the average prime offer rates published by the Bureau or may determine average prime offer rates itself by employing the methodology published on the FFIEC Web site. A financial institution that determines average prime offer rates itself, however, is responsible for correctly determining the rates in accordance with the published methodology. [[Page 66330]] 3. Rate spread calculation--annual percentage rate. The requirements of Sec. 1003.4(a)(12)(i) refer to the covered loan's annual percentage rate. A financial institution complies with Sec. 1003.4(a)(12)(i) by relying on the annual percentage rate for the covered loan, as calculated and disclosed pursuant to Regulation Z, 12 CFR 1026.18 or 1026.38 (for closed-end mortgage loans) or 1026.40 (for open-end credit lines of credit), as applicable. 4. Rate spread calculation--comparable transaction. The rate spread calculation in Sec. 1003.4(a)(12)(i) is defined by reference to a comparable transaction, which is determined according to the covered loan's amortization type (i.e., fixed- or variable-rate) and loan term. For covered loans that are open-end lines of credit, Sec. 1003.4(a)(12)(i) requires a financial institution to identify the most closely comparable closed-end transaction. The tables of average prime offer rates published by the Bureau (see comment 4(a)(12)-2) provide additional detail about how to identify the comparable transaction. i. Fixed-rate transactions. For fixed-rate covered loans, the term for identifying the comparable transaction is the transaction's maturity (i.e., the period until the last payment will be due under the closed-end mortgage loan contract or open-end line of credit agreement). If an open-end credit plan has a fixed rate but no definite plan length, a financial institution complies with Sec. 1003.4(a)(12)(i) by using a 30-year fixed-rate loan as the most closely comparable closed-end transaction. Financial institutions may refer to the table on the FFIEC Web site entitled ``Average Prime Offer Rates-Fixed'' when identifying a comparable fixed-rate transaction. ii. Variable-rate transactions. For variable-rate covered loans, the term for identifying the comparable transaction is the initial, fixed-rate period (i.e., the period until the first scheduled rate adjustment). For example, five years is the relevant term for a variable-rate transaction with a five-year, fixed-rate introductory period that is amortized over thirty years. Financial institutions may refer to the table on the FFIEC Web site entitled ``Average Prime Offer Rates-Variable'' when identifying a comparable variable- rate transaction. If an open-end line of credit has a variable rate and an optional, fixed-rate feature, a financial institution uses the rate table for variable-rate transactions. iii. Term not in whole years. When a covered loan's term to maturity (or, for a variable-rate transaction, the initial fixed- rate period) is not in whole years, the financial institution uses the number of whole years closest to the actual loan term or, if the actual loan term is exactly halfway between two whole years, by using the shorter loan term. For example, for a loan term of ten years and three months, the relevant term is ten years; for a loan term of ten years and nine months, the relevant term is 11 years; for a loan term of ten years and six months, the relevant term is ten years. If a loan term includes an odd number of days, in addition to an odd number of months, the financial institution rounds to the nearest whole month, or rounds down if the number of odd days is exactly halfway between two months. The financial institution rounds to one year any covered loan with a term shorter than six months, including variable-rate covered loans with no initial, fixed-rate periods. For example, if an open-end covered loan has a rate that varies according to an index plus a margin, with no introductory, fixed-rate period, the transaction term is one year. iv. Amortization period longer than loan term. If the amortization period of a covered loan is longer than the term of the transaction to maturity, Sec. 1003.4(a)(12)(i) requires a financial institution to use the loan term to determine the applicable average prime offer rate. For example, assume a financial institution originates a closed-end, fixed-rate loan that has a term to maturity of five years and a thirty-year amortization period that results in a balloon payment. The financial institution complies with Sec. 1003.4(a)(12)(i) by using the five-year loan term. 5. Rate-set date. The relevant date to use to determine the average prime offer rate for a comparable transaction is the date on which the covered loan's interest rate was set by the financial institution for the final time before closing or account opening. i. Rate-lock agreement. If an interest rate is set pursuant to a ``lock-in'' agreement between the financial institution and the borrower, then the date on which the agreement fixes the interest rate is the date the rate was set. Except as provided in comment 4(a)(12)-5.ii, if a rate is reset after a lock-in agreement is executed (for example, because the borrower exercises a float-down option or the agreement expires), then the relevant date is the date the financial institution exercises discretion in setting the rate for the final time before closing or account opening. The same rule applies when a rate-lock agreement is extended and the rate is reset at the same rate, regardless of whether market rates have increased, decreased, or remained the same since the initial rate was set. If no lock-in agreement is executed, then the relevant date is the date on which the institution sets the rate for the final time before closing or account opening. ii. Change in loan program. If a financial institution issues a rate-lock commitment under one loan program, the borrower subsequently changes to another program that is subject to different pricing terms, and the financial institution changes the rate promised to the borrower under the rate-lock commitment accordingly, the rate-set date is the date of the program change. However, if the financial institution changes the promised rate to the rate that would have been available to the borrower under the new program on the date of the original rate-lock commitment, then that is the date the rate is set, provided the financial institution consistently follows that practice in all such cases or the original rate-lock agreement so provided. For example, assume that a borrower locks a rate of 2.5 percent on June 1 for a 30-year, variable-rate loan with a 5-year, fixed-rate introductory period. On June 15, the borrower decides to switch to a 30-year, fixed-rate loan, and the rate available to the borrower for that product on June 15 is 4.0 percent. On June 1, the 30-year, fixed-rate loan would have been available to the borrower at a rate of 3.5 percent. If the financial institution offers the borrower the 3.5 percent rate (i.e., the rate that would have been available to the borrower for the fixed-rate product on June 1, the date of the original rate-lock) because the original agreement so provided or because the financial institution consistently follows that practice for borrowers who change loan programs, then the financial institution should use June 1 as the rate-set date. In all other cases, the financial institution should use June 15 as the rate-set date. iii. Brokered loans. When a financial institution has reporting responsibility for an application for a covered loan that it received from a broker, as discussed in comment 4(a)-4 (e.g., because the financial institution makes a credit decision prior to closing or account opening), the rate-set date is the last date the financial institution set the rate with the broker, not the date the broker set the borrower's rate. 6. Compare the annual percentage rate to the average prime offer rate. Section 1003.4(a)(12)(i) requires a financial institution to compare the covered loan's annual percentage rate to the most recently available average prime offer rate that was in effect for the comparable transaction as of the rate-set date. For purposes of Sec. 1003.4(a)(12)(i), the most recently available rate means the average prime offer rate set forth in the applicable table with the most recent effective date as of the date the interest rate was set. However, Sec. 1003.4(a)(12)(i) does not permit a financial institution to use an average prime offer rate before its effective date. 7. Rate spread--not applicable. If the covered loan is an assumption, reverse mortgage, a purchased loan, or is not subject to Regulation Z, 12 CFR part 1026, a financial institution complies with Sec. 1003.4(a)(12) by reporting that the requirement is not applicable. If the application did not result in an origination for a reason other than the application was approved but not accepted by the applicant, a financial institution complies with Sec. 1003.4(a)(12) by reporting that the requirement is not applicable. 8. Application approved but not accepted or preapproval request approved but not accepted. In the case of an application approved but not accepted or a preapproval request that was approved but not accepted, Sec. 1003.4(a)(12) requires a financial institution to report the applicable rate spread. Paragraph 4(a)(13) 1. HOEPA status--not applicable. If the covered loan is not subject to the Home Ownership and Equity Protection Act of 1994, as implemented in Regulation Z, 12 CFR 1026.32, a financial institution complies with Sec. 1003.4(a)(13) by reporting that the requirement is not applicable. If an application did not result in an origination, a financial institution complies with Sec. 1003.4(a)(13) by reporting that the requirement is not applicable. Paragraph 4(a)(14) 1. Determining lien status for applications and covered loans originated and purchased. i. Financial institutions are required to report lien status for covered loans they originate [[Page 66331]] and purchase and applications that do not result in originations (preapproval requests that are approved but not accepted, preapproval requests that are denied, applications that are approved but not accepted, denied, withdrawn, or closed for incompleteness). For covered loans purchased by a financial institution, lien status is determined by reference to the best information readily available to the financial institution at the time of purchase. For covered loans that a financial institution originates and applications that do not result in originations, lien status is determined by reference to the best information readily available to the financial institution at the time final action is taken and to the financial institution's own procedures. Thus, financial institutions may rely on the title search they routinely perform as part of their underwriting procedures--for example, for home purchase loans. Regulation C does not require financial institutions to perform title searches solely to comply with HMDA reporting requirements. Financial institutions may rely on other information that is readily available to them at the time final action is taken and that they reasonably believe is accurate, such as the applicant's statement on the application or the applicant's credit report. For example, where the applicant indicates on the application that there is a mortgage on the property or where the applicant's credit report shows that the applicant has a mortgage--and that mortgage will not be paid off as part of the transaction--the financial institution may assume that the loan it originates is secured by a subordinate lien. If the same application did not result in an origination--for example, because the application was denied or withdrawn--the financial institution would report the application as an application for a subordinate-lien loan. ii. Financial institutions may also consider their established procedures when determining lien status for applications that do not result in originations. For example, assume an applicant applies to a financial institution to refinance a $100,000 first mortgage; the applicant also has an open-end line of credit for $20,000. If the financial institution's practice in such a case is to ensure that it will have first-lien position--through a subordination agreement with the holder of the lien securing the open-end line of credit-- then the financial institution should report the application as an application for a first-lien covered loan. 2. Multiple properties. See comment 4(a)(9)-2 regarding transactions involving multiple properties with more than one property taken as security. Paragraph 4(a)(15) 1. Credit score--relied on. Except for purchased covered loans, Sec. 1003.4(a)(15) requires a financial institution to report the credit score or scores relied on in making the credit decision and information about the scoring model used to generate each score. A financial institution relies on a credit score in making the credit decision if the credit score was a factor in the credit decision even if it was not a dispositive factor. For example, if a credit score is one of multiple factors in a financial institution's credit decision, the financial institution has relied on the credit score even if the financial institution denies the application because one or more underwriting requirements other than the credit score are not satisfied. 2. Credit score--multiple credit scores. When a financial institution obtains or creates two or more credit scores for a single applicant or borrower but relies on only one score in making the credit decision (for example, by relying on the lowest, highest, most recent, or average of all of the scores), the financial institution complies with Sec. 1003.4(a)(15) by reporting that credit score and information about the scoring model used. When a financial institution obtains or creates two or more credit scores for an applicant or borrower and relies on multiple scores for the applicant or borrower in making the credit decision (for example, by relying on a scoring grid that considers each of the scores obtained or created for the applicant or borrower without combining the scores into a composite score), Sec. 1003.4(a)(15) requires the financial institution to report one of the credit scores for the applicant or borrower that was relied on in making the credit decision. In choosing which credit score to report in this circumstance, a financial institution need not use the same approach for its entire HMDA submission, but it should be generally consistent (such as by routinely using one approach within a particular division of the institution or for a category of covered loans). In instances such as these, the financial institution should report the name and version of the credit scoring model for the score reported. 3. Credit score--multiple applicants or borrowers. In a transaction involving two or more applicants or borrowers for which the financial institution obtains or creates a single credit score, and relies on that credit score in making the credit decision for the transaction, the institution complies with Sec. 1003.4(a)(15) by reporting that credit score for either the applicant or first co- applicant. Otherwise, a financial institution complies with Sec. 1003.4(a)(15) by reporting a credit score for the applicant that it relied on in making the credit decision, if any, and a credit score for the first co-applicant that it relied on in making the credit decision, if any. To illustrate, assume a transaction involves one applicant and one co-applicant and that the financial institution obtains or creates two credit scores for the applicant and two credit scores for the co-applicant. Assume further that the financial institution relies on the lowest, highest, most recent, or average of all of the credit scores obtained or created to make the credit decision for the transaction. The financial institution complies with Sec. 1003.4(a)(15) by reporting that credit score and information about the scoring model used. Alternatively, assume a transaction involves one applicant and one co-applicant and that the financial institution obtains or creates three credit scores for the applicant and three credit scores for the co-applicant. Assume further that the financial institution relies on the middle credit score for the applicant and the middle credit score for the co- applicant to make the credit decision for the transaction. The financial institution complies with Sec. 1003.4(a)(15) by reporting both the middle score for the applicant and the middle score for the co-applicant. 4. Transactions for which no credit decision was made. If a file was closed for incompleteness or the application was withdrawn before a credit decision was made, the financial institution complies with Sec. 1003.4(a)(15) by reporting that the requirement is not applicable, even if the financial institution had obtained or created a credit score for the applicant or co-applicant. For example, if a file is closed for incompleteness and is so reported in accordance with Sec. 1003.4(a)(8), the financial institution complies with Sec. 1003.4(a)(15) by reporting that the requirement is not applicable, even if the financial institution had obtained or created a credit score for the applicant or co-applicant. Similarly, if an application was withdrawn by the applicant before a credit decision was made and is so reported in accordance with Sec. 1003.4(a)(8), the financial institution complies with Sec. 1003.4(a)(15) by reporting that the requirement is not applicable, even if the financial institution had obtained or created a credit score for the applicant or co-applicant. 5. Transactions for which no credit score was relied on. If a financial institution makes a credit decision without relying on a credit score for the applicant or borrower, the financial institution complies with Sec. 1003.4(a)(15) by reporting that the requirement is not applicable. 6. Purchased covered loan. A financial institution complies with Sec. 1003.4(a)(15) by reporting that the requirement is not applicable when the covered loan is a purchased covered loan. 7. Non-natural person. When the applicant and co-applicant, if applicable, are not natural persons, a financial institution complies with Sec. 1003.4(a)(15) by reporting that the requirement is not applicable. Paragraph 4(a)(16) 1. Reason for denial--general. A financial institution complies with Sec. 1003.4(a)(16) by reporting the principal reason or reasons it denied the application, indicating up to four reasons. The financial institution should report only the principal reason or reasons it denied the application, even if there are fewer than four reasons. For example, if a financial institution denies the application because of the applicant's credit history and debt-to- income ratio, the financial institution need only report these two principal reasons. The reasons reported must be specific and accurately describe the principal reason or reasons the financial institution denied the application. 2. Reason for denial--preapproval request denied. Section 1003.4(a)(16) requires a financial institution to report the principal reason or reasons it denied the application. A request for a preapproval under a preapproval program as defined by Sec. 1003.2(b)(2) is an application. If a financial institution denies a preapproval request, the financial institution complies with Sec. 1003.4(a)(16) by reporting the reason or reasons it denied the preapproval request. 3. Reason for denial--adverse action model form or similar form. If a financial institution [[Page 66332]] chooses to provide the applicant the reason or reasons it denied the application using the model form contained in appendix C to Regulation B (Form C-1, Sample Notice of Action Taken and Statement of Reasons) or a similar form, Sec. 1003.4(a)(16) requires the financial institution to report the reason or reasons that were specified on the form by the financial institution, which includes reporting the ``Other'' reason or reasons that were specified on the form by the financial institution, if applicable. If a financial institution chooses to provide a disclosure of the applicant's right to a statement of specific reasons using the model form contained in appendix C to Regulation B (Form C-5, Sample Disclosure of Right to Request Specific Reasons for Credit Denial) or a similar form, or chooses to provide the denial reason or reasons orally under Regulation B, 12 CFR 1002.9(a)(2)(ii), the financial institution complies with Sec. 1003.4(a)(16) by entering the principal reason or reasons it denied the application. 4. Reason for denial--not applicable. A financial institution complies with Sec. 1003.4(a)(16) by reporting that the requirement is not applicable if the action taken on the application, pursuant to Sec. 1003.4(a)(8), is not a denial. For example, a financial institution complies with Sec. 1003.4(a)(16) by reporting that the requirement is not applicable if the loan is originated or purchased by the financial institution, or the application or preapproval request was approved but not accepted, or the application was withdrawn before a credit decision was made, or the file was closed for incompleteness. Paragraph 4(a)(17)(i) 1. Total loan costs--not applicable. Section 1003.4(a)(17)(i) does not require financial institutions to report the total loan costs for applications, or for transactions not subject to Regulation Z, 12 CFR 1026.43(c), and 12 CFR 1026.19(f), such as open-end lines of credit, reverse mortgages, or loans or lines of credit made primarily for business or commercial purposes. In these cases, a financial institution complies with Sec. 1003.4(a)(17)(i) by reporting that the requirement is not applicable to the transaction. 2. Purchased loans--applications received prior to the integrated disclosure effective date. For purchased covered loans subject to this reporting requirement for which applications were received by the selling entity prior to the effective date of Regulation Z, 12 CFR 1026.19(f), a financial institution complies with Sec. 1003.4(a)(17)(i) by reporting that the requirement is not applicable to the transaction. 3. Revised disclosures. If the amount of total loan costs changes because a financial institution provides a revised version of the disclosures required under Regulation Z, 12 CFR 1026.19(f), pursuant to Regulation Z, 12 CFR 1026.19(f)(2), the financial institution complies with Sec. 1003.4(a)(17)(i) by reporting the revised amount, provided that the revised disclosure was provided to the borrower during the same reporting period in which closing occurred. For example, in the case of a financial institution's quarterly submission made pursuant to Sec. 1003.5(a)(1)(ii), if the financial institution provides a corrected disclosure to reflect a refund made pursuant to Regulation Z, 12 CFR 1026.19(f)(2)(v), the financial institution reports the corrected amount of total loan costs only if the corrected disclosure was provided prior to the end of the quarter in which closing occurred. The financial institution does not report the corrected amount of total loan costs in its quarterly submission if the corrected disclosure was provided after the end of the quarter, even if the corrected disclosure was provided prior to the deadline for timely submission of the financial institution's quarterly data. However, the financial institution reports the corrected amount of total loan costs on its annual loan/application register. Paragraph 4(a)(17)(ii) 1. Total points and fees--not applicable. Section 1003.4(a)(17)(ii) does not require financial institutions to report the total points and fees for transactions not subject to Regulation Z, 12 CFR 1026.43(c), such as open-end lines of credit, reverse mortgages, or loans or lines of credit made primarily for business or commercial purposes, or for applications or purchased covered loans. In these cases, a financial institution complies with Sec. 1003.4(a)(17)(ii) by reporting that the requirement is not applicable to the transaction. 2. Total points and fees cure mechanism. For covered loans subject to this reporting requirement, if a financial institution determines that the transaction's total points and fees exceeded the applicable limit and cures the overage pursuant to Regulation Z, 12 CFR 1026.43(e)(3)(iii) and (iv), a financial institution complies with Sec. 1003.4(a)(17)(ii) by reporting the correct amount of total points and fees, provided that the cure was effected during the same reporting period in which closing occurred. For example, in the case of a financial institution's quarterly submission, the financial institution reports the revised amount of total points and fees only if it cured the overage prior to the end of the quarter in which closing occurred. The financial institution does not report the revised amount of total points and fees in its quarterly submission if it cured the overage after the end of the quarter, even if the cure was effected prior to the deadline for timely submission of the financial institution's quarterly data. However, the financial institution reports the revised amount of total points and fees on its annual loan/application register. Paragraph 4(a)(18) 1. Origination charges--not applicable. Section 1003.4(a)(18) does not require financial institutions to report the total borrower-paid origination charges for applications, or for transactions not subject to Regulation Z, 12 CFR 1026.19(f), such as open-end lines of credit, reverse mortgages, or loans or lines of credit made primarily for business or commercial purposes. In these cases, a financial institution complies with Sec. 1003.4(a)(18) by reporting that the requirement is not applicable to the transaction. 2. Purchased loans--applications received prior to the integrated disclosure effective date. For purchased covered loans subject to this reporting requirement for which applications were received by the selling entity prior to the effective date of Regulation Z, 12 CFR 1026.19(f), a financial institution complies with Sec. 1003.4(a)(18) by reporting that the requirement is not applicable to the transaction. 3. Revised disclosures. If the total amount of borrower-paid origination charges changes because a financial institution provides a revised version of the disclosures required under Regulation Z, 12 CFR 1026.19(f), pursuant to Regulation Z, 12 CFR 1026.19(f)(2), the financial institution complies with Sec. 1003.4(a)(18) by reporting the revised amount, provided that the revised disclosure was provided to the borrower during the same reporting period in which closing occurred. For example, in the case of a financial institution's quarterly submission made pursuant to Sec. 1003.5(a)(1)(ii), if the financial institution provides a corrected disclosure to reflect a refund made pursuant to Regulation Z, 12 CFR 1026.19(f)(2)(v), the financial institution reports the corrected amount of origination charges only if the corrected disclosure was provided prior to the end of the quarter in which closing occurred. The financial institution does not report the corrected amount of origination charges in its quarterly submission if the corrected disclosure was provided after the end of the quarter, even if the corrected disclosure was provided prior to the deadline for timely submission of the financial institution's quarterly data. However, the financial institution reports the corrected amount of origination charges on its annual loan/application register. Paragraph 4(a)(19) 1. Discount points--not applicable. Section 1003.4(a)(19) does not require financial institutions to report the discount points for applications, or for transactions not subject to Regulation Z, 12 CFR 1026.19(f), such as open-end lines of credit, reverse mortgages, or loans or lines of credit made primarily for business or commercial purposes. In these cases, a financial institution complies with Sec. 1003.4(a)(19) by reporting that the requirement is not applicable to the transaction. 2. Purchased loans--applications received prior to the integrated disclosure effective date. For purchased covered loans subject to this reporting requirement for which applications were received by the selling entity prior to the effective date of Regulation Z, 12 CFR 1026.19(f), a financial institution complies with Sec. 1003.4(a)(19) by reporting that the requirement is not applicable to the transaction. 3. Revised disclosures. If the amount of discount points changes because a financial institution provides a revised version of the disclosures required under Regulation Z, 12 CFR 1026.19(f), pursuant to Regulation Z, 12 CFR 1026.19(f)(2), the financial institution complies with Sec. 1003.4(a)(19) by reporting the revised amount, provided that the revised disclosure was provided to the borrower during the same reporting period in which closing occurred. For example, in the case of a financial institution's quarterly submission made pursuant to Sec. 1003.5(a)(ii), if the financial institution provides a corrected [[Page 66333]] disclosure to reflect a refund made pursuant to Regulation Z, 12 CFR 1026.19(f)(2)(v), the financial institution reports the corrected amount of discount points only if the corrected disclosure was provided prior to the end of the quarter in which closing occurred. The financial institution does not report the corrected amount of discount points in its quarterly submission if the corrected disclosure was provided after the end of the quarter, even if the corrected disclosure was provided prior to the deadline for timely submission of the financial institution's quarterly data. However, the financial institution reports the corrected amount of discount points on its annual loan/application register. Paragraph 4(a)(20) 1. Lender credits--not applicable. Section 1003.4(a)(20) does not require financial institutions to report lender credits for applications, or for transactions not subject to Regulation Z, 12 CFR 1026.19(f), such as open-end lines of credit, reverse mortgages, or loans or lines of credit made primarily for business or commercial purposes. In these cases, a financial institution complies with Sec. 1003.4(a)(20) by reporting that the requirement is not applicable to the transaction. 2. Purchased loans--applications received prior to the integrated disclosure effective date. For purchased covered loans subject to this reporting requirement for which applications were received by the selling entity prior to the effective date of Regulation Z, 12 CFR 1026.19(f), a financial institution complies with Sec. 1003.4(a)(20) by reporting that the requirement is not applicable to the transaction. 3. Revised disclosures. If the amount of lender credits changes because a financial institution provides a revised version of the disclosures required under Regulation Z, 12 CFR 1026.19(f), pursuant to Regulation Z, 12 CFR 1026.19(f)(2), the financial institution complies with Sec. 1003.4(a)(20) by reporting the revised amount, provided that the revised disclosure was provided to the borrower during the same reporting period in which closing occurred. For example, in the case of a financial institution's quarterly submission made pursuant to Sec. 1003.5(a)(1)(ii), if the financial institution provides a corrected disclosure to reflect a refund made pursuant to Regulation Z, 12 CFR 1026.19(f)(2)(v), the financial institution reports the corrected amount of lender credits only if the corrected disclosure was provided prior to the end of the quarter in which closing occurred. The financial institution does not report the corrected amount of lender credits in its quarterly submission if the corrected disclosure was provided after the end of the quarter, even if the corrected disclosure was provided prior to the deadline for timely submission of the financial institution's quarterly data. However, the financial institution reports the corrected amount of lender credits on its annual loan/application register. Paragraph 4(a)(21) 1. Interest rate--disclosures. Section 1003.4(a)(21) requires a financial institution to identify the interest rate applicable to the approved application, or to the covered loan at closing or account opening. For covered loans or applications subject to the disclosure requirements of Regulation Z, 12 CFR 1026.19(e) or (f), a financial institution complies with Sec. 1003.4(a)(21) by reporting the interest rate disclosed on the applicable disclosure. For covered loans for which disclosures were provided pursuant to both 12 CFR 1026.19(e) and 12 CFR 1026.19(f), a financial institution reports the interest rate disclosed pursuant to 12 CFR 1026.19(f). A financial institution may rely on the definitions and commentary to the sections of Regulation Z relevant to the disclosure of the interest rate pursuant to 12 CFR 1026.19(e) or 12 CFR 1026.19(f). 2. Applications. In the case of an application, Sec. 1003.4(a)(21) requires a financial institution to report the applicable interest rate only if the application has been approved by the financial institution but not accepted by the borrower. In such cases, a financial institution reports the interest rate applicable at the time that the application was approved by the financial institution. A financial institution may report the interest rate appearing on the disclosure provided pursuant to 12 CFR 1026.19(e) or (f) if such disclosure accurately reflects the interest rate at the time the application was approved. For applications that have been denied or withdrawn, or files closed for incompleteness, a financial institution reports that no interest rate was applicable to the application. 3. Adjustable rate--interest rate unknown. Except as provided in comment 4(a)(21)-1, for adjustable-rate covered loans or applications, if the interest rate is unknown at the time that the application was approved, or at closing or account opening, a financial institution reports the fully-indexed rate based on the index applicable to the covered loan or application. For purposes of Sec. 1003.4(a)(21), the fully-indexed rate is the index value and margin at the time that the application was approved, or, for covered loans, at closing or account opening. Paragraph 4(a)(22) 1. Prepayment penalty term--not applicable. Section 1003.4(a)(22) does not require financial institutions to report the term of any prepayment penalty for transactions not subject to Regulation Z, 12 CFR part 1026, such as loans or lines of credit made primarily for business or commercial purposes, or for reverse mortgages or purchased covered loans. In these cases, a financial institution complies with Sec. 1003.4(a)(22) by reporting that the requirement is not applicable to the transaction. 2. Transactions for which no prepayment penalty exists. For covered loans or applications that have no prepayment penalty, a financial institution complies with Sec. 1003.4(a)(22) by reporting that the requirement is not applicable to the transaction. A financial institution may rely on the definitions and commentary to Regulation Z, 12 CFR 1026.32(b)(6)(i) or (ii) in determining whether the terms of a transaction contain a prepayment penalty. Paragraph 4(a)(23) 1. General. For covered loans that are not purchased covered loans, Sec. 1003.4(a)(23) requires a financial institution to report the ratio of the applicant's or borrower's total monthly debt to total monthly income (debt-to-income ratio) relied on in making the credit decision. For example, if a financial institution calculated the applicant's or borrower's debt-to-income ratio twice--once according to the financial institution's own requirements and once according to the requirements of a secondary market investor--and the financial institution relied on the debt- to-income ratio calculated according to the secondary market investor's requirements in making the credit decision, Sec. 1003.4(a)(23) requires the financial institution to report the debt- to-income ratio calculated according to the requirements of the secondary market investor. 2. Transactions for which a debt-to-income ratio was one of multiple factors. A financial institution relies on the ratio of the applicant's or borrower's total monthly debt to total monthly income (debt-to-income ratio) in making the credit decision if the debt-to- income ratio was a factor in the credit decision even if it was not a dispositive factor. For example, if the debt-to-income ratio was one of multiple factors in a financial institution's credit decision, the financial institution has relied on the debt-to-income ratio and complies with Sec. 1003.4(a)(23) by reporting the debt- to-income ratio, even if the financial institution denied the application because one or more underwriting requirements other than the debt-to-income ratio were not satisfied. 3. Transactions for which no credit decision was made. If a file was closed for incompleteness, or if an application was withdrawn before a credit decision was made, a financial institution complies with Sec. 1003.4(a)(23) by reporting that the requirement is not applicable, even if the financial institution had calculated the ratio of the applicant's total monthly debt to total monthly income (debt-to-income ratio). For example, if a file was closed for incompleteness and was so reported in accordance with Sec. 1003.4(a)(8), the financial institution complies with Sec. 1003.4(a)(23) by reporting that the requirement is not applicable, even if the financial institution had calculated the applicant's debt-to-income ratio. Similarly, if an application was withdrawn by the applicant before a credit decision was made, the financial institution complies with Sec. 1003.4(a)(23) by reporting that the requirement is not applicable, even if the financial institution had calculated the applicant's debt-to-income ratio. 4. Transactions for which no debt-to-income ratio was relied on. Section 1003.4(a)(23) does not require a financial institution to calculate the ratio of an applicant's or borrower's total monthly debt to total monthly income (debt-to-income ratio), nor does it require a financial institution to rely on an applicant's or borrower's debt-to-income ratio in making a credit decision. If a financial institution made a credit decision without relying on the applicant's or borrower's debt-to-income ratio, the financial institution complies with [[Page 66334]] Sec. 1003.4(a)(23) by reporting that the requirement is not applicable since no debt-to-income ratio was relied on in connection with the credit decision. 5. Non-natural person. A financial institution complies with Sec. 1003.4(a)(23) by reporting that the requirement is not applicable when the applicant and co-applicant, if applicable, are not natural persons. 6. Multifamily dwellings. A financial institution complies with Sec. 1003.4(a)(23) by reporting that the requirement is not applicable for a covered loan secured by, or an application proposed to be secured by, a multifamily dwelling. 7. Purchased covered loans. A financial institution complies with Sec. 1003.4(a)(23) by reporting that the requirement is not applicable when reporting a purchased covered loan. Paragraph 4(a)(24) 1. General. Section 1003.4(a)(24) requires a financial institution to report, except for purchased covered loans, the ratio of the total amount of debt secured by the property to the value of the property (combined loan-to-value ratio) relied on in making the credit decision. For example, if a financial institution calculated a combined loan-to-value ratio twice--once according to the financial institution's own requirements and once according to the requirements of a secondary market investor--and the financial institution relied on the combined loan-to-value ratio calculated according to the secondary market investor's requirements in making the credit decision, Sec. 1003.4(a)(24) requires the financial institution to report the combined loan-to-value ratio calculated according to the requirements of the secondary market investor. 2. Transactions for which a combined loan-to-value ratio was one of multiple factors. A financial institution relies on the total amount of debt secured by the property to the value of the property (combined loan-to-value ratio) in making the credit decision if the combined loan-to-value ratio was a factor in the credit decision even if it was not a dispositive factor. For example, if the combined loan-to-value ratio is one of multiple factors in a financial institution's credit decision, the financial institution has relied on the combined loan-to-value ratio and complies with Sec. 1003.4(a)(24) by reporting the combined loan-to-value ratio, even if the financial institution denies the application because one or more underwriting requirements other than the combined loan-to- value ratio are not satisfied. 3. Transactions for which no credit decision was made. If a file was closed for incompleteness, or if an application was withdrawn before a credit decision was made, a financial institution complies with Sec. 1003.4(a)(24) by reporting that the requirement is not applicable, even if the financial institution had calculated the ratio of the total amount of debt secured by the property to the value of the property (combined loan-to-value ratio). For example, if a file is closed for incompleteness and is so reported in accordance with Sec. 1003.4(a)(8), the financial institution complies with Sec. 1003.4(a)(24) by reporting that the requirement is not applicable, even if the financial institution had calculated a combined loan-to-value ratio. Similarly, if an application was withdrawn by the applicant before a credit decision was made and is so reported in accordance with Sec. 1003.4(a)(8), the financial institution complies with Sec. 1003.4(a)(24) by reporting that the requirement is not applicable, even if the financial institution had calculated a combined loan-to-value ratio. 4. Transactions for which no combined loan-to-value ratio was relied on. Section 1003.4(a)(24) does not require a financial institution to calculate the ratio of the total amount of debt secured by the property to the value of the property (combined loan- to-value ratio), nor does it require a financial institution to rely on a combined loan-to-value ratio in making a credit decision. If a financial institution makes a credit decision without relying on a combined loan-to-value ratio, the financial institution complies with Sec. 1003.4(a)(24) by reporting that the requirement is not applicable since no combined loan-to-value ratio was relied on in making the credit decision. 5. Purchased covered loan. A financial institution complies with Sec. 1003.4(a)(24) by reporting that the requirement is not applicable when the covered loan is a purchased covered loan. Paragraph 4(a)(25) 1. Amortization and maturity. For a fully amortizing covered loan, the number of months after which the legal obligation matures is the number of months in the amortization schedule, ending with the final payment. Some covered loans do not fully amortize during the maturity term, such as covered loans with a balloon payment; such loans should still be reported using the maturity term rather than the amortization term, even in the case of covered loans that mature before fully amortizing but have reset options. For example, a 30-year fully amortizing covered loan would be reported with a term of ``360,'' while a five year balloon covered loan would be reported with a loan term of ``60.'' 2. Non-monthly repayment periods. If a covered loan or application includes a schedule with repayment periods measured in a unit of time other than months, the financial institution should report the covered loan or application term using an equivalent number of whole months without regard for any remainder. 3. Purchased loans. For a covered loan that was purchased, a financial institution reports the number of months after which the legal obligation matures as measured from the covered loan's origination. 4. Open-end line of credit. For an open-end line of credit with a definite term, a financial institution reports the number of months from origination until the account termination date, including both the draw and repayment period. 5. Loan or application without a definite term. For a covered loan or application without a definite term, such as a reverse mortgage, a financial institution complies with Sec. 1003.4(a)(25) by reporting that the requirement is not applicable. Paragraph 4(a)(26) 1. Types of introductory rates. Section 1003.4(a)(26) requires a financial institution to report the number of months, or proposed number of months in the case of an application, from closing or account opening until the first date the interest rate may change. For example, assume an open-end line of credit contains an introductory or ``teaser'' interest rate for two months after the date of account opening, after which the interest rate may adjust. In this example, the financial institution complies with Sec. 1003.4(a)(26) by reporting the number of months as ``2.'' Section 1003.4(a)(26) requires a financial institution to report the number of months based on when the first interest rate adjustment may occur, even if an interest rate adjustment is not required to occur at that time and even if the rates that will apply, or the periods for which they will apply, are not known at closing or account opening. For example, if a closed-end mortgage loan with a 30-year term has an adjustable-rate product with an introductory interest rate for the first 60 months, after which the interest rate is permitted, but not required to vary, according to the terms of an index rate, the financial institution complies with Sec. 1003.4(a)(26) by reporting the number of months as ``60.'' Similarly, if a closed-end mortgage loan with a 30-year term is a step-rate product with an introductory interest rate for the first 24 months, after which the interest rate will increase to a different known interest rate for the next 36 months, the financial institution complies with Sec. 1003.4(a)(26) by reporting the number of months as ``24.'' 2. Preferred rates. Section 1003.4(a)(26) does not require reporting of introductory interest rate periods based on preferred rates unless the terms of the legal obligation provide that the preferred rate will expire at a certain defined date. Preferred rates include terms of the legal obligation that provide that the initial underlying rate is fixed but that it may increase or decrease upon the occurrence of some future event, such as an employee leaving the employ of the financial institution, the borrower closing an existing deposit account with the financial institution, or the borrower revoking an election to make automated payments. In these cases, because it is not known at the time of closing or account opening whether the future event will occur, and if so, when it will occur, Sec. 1003.4(a)(26) does not require reporting of an introductory interest rate period. 3. Loan or application with a fixed rate. A financial institution complies with Sec. 1003.4(a)(26) by reporting that the requirement is not applicable for a covered loan with a fixed rate or an application for a covered loan with a fixed rate. 4. Purchased loan. A financial institution complies with Sec. 1003.4(a)(26) by reporting that requirement is not applicable when the covered loan is a purchased covered loan with a fixed rate. Paragraph 4(a)(27) 1. General. Section 1003.4(a)(27) requires reporting of contractual features that would allow payments other than fully amortizing payments. Section 1003.4(a)(27) defines the [[Page 66335]] contractual features by reference to Regulation Z, 12 CFR part 1026, but without regard to whether the covered loan is consumer credit, as defined in Sec. 1026.2(a)(12), is extended by a creditor, as defined in Sec. 1026.2(a)(17), or is extended to a consumer, as defined in Sec. 1026.2(a)(11), and without regard to whether the property is a dwelling as defined in Sec. 1026.2(a)(19). For example, assume that a financial institution originates a business- purpose transaction that is exempt from Regulation Z pursuant to 12 CFR 1026.3(a)(1), to finance the purchase of a multifamily dwelling, and that there is a balloon payment, as defined by Regulation Z, 12 CFR 1026.18(s)(5)(i), at the end of the loan term. The multifamily dwelling is a dwelling under Sec. 1003.2(f), but not under Regulation Z, 12 CFR 1026.2(a)(19). In this example, the financial institution should report the business-purpose transaction as having a balloon payment under Sec. 1003.4(a)(27)(i), assuming the other requirements of this part are met. Aside from these distinctions, financial institutions may rely on the definitions and related commentary provided in the appropriate sections of Regulation Z referenced in Sec. 1003.4(a)(27) of this part in determining whether the contractual feature should be reported. Paragraph 4(a)(28). 1. General. A financial institution reports the property value relied on in making the credit decision. For example, if the institution relies on an appraisal or other valuation for the property in calculating the loan-to-value ratio, it reports that value; if the institution relies on the purchase price of the property in calculating the loan-to-value ratio, it reports that value. 2. Multiple property values. When a financial institution obtains two or more valuations of the property securing or proposed to secure the covered loan, the financial institution complies with Sec. 1003.4(a)(28) by reporting the value relied on in making the credit decision. For example, when a financial institution obtains an appraisal, an automated valuation model report, and a broker price opinion with different values for the property, it reports the value relied on in making the credit decision. Section Sec. 1003.4(a)(28) does not require a financial institution to use a particular property valuation method, but instead requires a financial institution to report the valuation relied on in making the credit decision. 3. Transactions for which no credit decision was made. If a file was closed for incompleteness or the application was withdrawn before a credit decision was made, the financial institution complies with Sec. 1003.4(a)(28) by reporting that the requirement is not applicable, even if the financial institution had obtained a property value. For example, if a file is closed for incompleteness and is so reported in accordance with Sec. 1003.4(a)(8), the financial institution complies with Sec. 1003.4(a)(28) by reporting that the requirement is not applicable, even if the financial institution had obtained a property value. Similarly, if an application was withdrawn by the applicant before a credit decision was made and is so reported in accordance with Sec. 1003.4(a)(8), the financial institution complies with Sec. 1003.4(a)(28) by reporting that the requirement is not applicable, even if the financial institution had obtained a property value. 4. Transactions for which no property value was relied on. Section 1003.4(a)(28) does not require a financial institution to obtain a property valuation, nor does it require a financial institution to rely on a property value in making a credit decision. If a financial institution makes a credit decision without relying on a property value, the financial institution complies with Sec. 1003.4(a)(28) by reporting that the requirement is not applicable since no property value was relied on in making the credit decision. Paragraph 4(a)(29) 1. Classification under State law. A financial institution should report a covered loan that is or would have been secured only by a manufactured home but not the land on which it is sited as secured by a manufactured home and not land, even if the manufactured home is considered real property under applicable State law. 2. Manufactured home community. A manufactured home community that is a multifamily dwelling is not considered a manufactured home for purposes of Sec. 1003.4(a)(29). 3. Multiple properties. See comment 4(a)(9)-2 regarding transactions involving multiple properties with more than one property taken as security. 4. Scope of requirement. A financial institution reports that the requirement is not applicable for a covered loan where the dwelling related to the property identified in Sec. 1003.4(a)(9) is not a manufactured home. Paragraph 4(a)(30) 1. Indirect land ownership. Indirect land ownership can occur when the applicant or borrower is or will be a member of a resident- owned community structured as a housing cooperative in which the occupants own an entity that holds the underlying land of the manufactured home community. In such communities, the applicant or borrower may still have a lease and pay rent for the lot on which his or her manufactured home is or will be located, but the property interest type for such an arrangement should be reported as indirect ownership if the applicant is or will be a member of the cooperative that owns the underlying land of the manufactured home community. If an applicant resides or will reside in such a community but is not a member, the property interest type should be reported as a paid leasehold. 2. Leasehold interest. A leasehold interest could be formalized in a lease with a defined term and specified rent payments, or could arise as a tenancy at will through permission of a land owner without any written, formal arrangement. For example, assume a borrower will locate the manufactured home in a manufactured home community, has a written lease for a lot in that park, and the lease specifies rent payments. In this example, a financial institution complies with Sec. 1003.4(a)(30) by reporting a paid leasehold. However, if instead the borrower will locate the manufactured home on land owned by a family member without a written lease and with no agreement as to rent payments, a financial institution complies with Sec. 1003.4(a)(30) by reporting an unpaid leasehold. 3. Multiple properties. See comment 4(a)(9)-2 regarding transactions involving multiple properties with more than one property taken as security. 4. Manufactured home community. A manufactured home community that is a multifamily dwelling is not considered a manufactured home for purposes of Sec. 1003.4(a)(30). 5. Direct ownership. An applicant or borrower has a direct ownership interest in the land on which the dwelling is or is to be located when it has a more than possessory real property ownership interest in the land such as fee simple ownership. 6. Scope of requirement. A financial institution reports that the requirement is not applicable for a covered loan where the dwelling related to the property identified in Sec. 1003.4(a)(9) is not a manufactured home. Paragraph 4(a)(31) 1. Multiple properties. See comment 4(a)(9)-2 regarding transactions involving multiple properties with more than one property taken as security. 2. Manufactured home community. For an application or covered loan secured by a manufactured home community, the financial institution should include in the number of individual dwelling units the total number of manufactured home sites that secure the loan and are available for occupancy, regardless of whether the sites are currently occupied or have manufactured homes currently attached. A financial institution may include in the number of individual dwelling units other units such as recreational vehicle pads, manager apartments, rental apartments, site-built homes or other rentable space that are ancillary to the operation of the secured property if it considers such units under its underwriting guidelines or the guidelines of an investor, or if it tracks the number of such units for its own internal purposes. For a loan secured by a single manufactured home that is or will be located in a manufactured home community, the financial institution should report one individual dwelling unit. 3. Condominium and cooperative projects. For a covered loan secured by a condominium or cooperative property, the financial institution reports the total number of individual dwelling units securing the covered loan or proposed to secure the covered loan in the case of an application. For example: i. Assume that a loan is secured by the entirety of a cooperative property. The financial institution would report the number of individual dwelling units in the cooperative property. ii. Assume that a covered loan is secured by 30 individual dwelling units in a condominium property that contains 100 individual dwelling units and that the loan is not exempt from Regulation C under Sec. 1003.3(c)(3). The financial institution reports 30 individual dwelling units. [[Page 66336]] 4. Best information available. A financial institution may rely on the best information readily available to the financial institution at the time final action is taken and on the financial institution's own procedures in reporting the information required by Sec. 1003.4(a)(31). Information readily available could include, for example, information provided by an applicant that the financial institution reasonably believes, information contained in a property valuation or inspection, or information obtained from public records. Paragraph 4(a)(32) 1. Affordable housing income restrictions. For purposes of Sec. 1003.4(a)(32), affordable housing income-restricted units are individual dwelling units that have restrictions based on the income level of occupants pursuant to restrictive covenants encumbering the property. Such income levels are frequently expressed as a percentage of area median income by household size as established by the U.S. Department of Housing and Urban Development or another agency responsible for implementing the applicable affordable housing program. Such restrictions are frequently part of compliance with programs that provide public funds, special tax treatment, or density bonuses to encourage development or preservation of affordable housing. Such restrictions are frequently evidenced by a use agreement, regulatory agreement, land use restriction agreement, housing assistance payments contract, or similar agreement. Rent control or rent stabilization laws, and the acceptance by the owner or manager of a multifamily dwelling of Housing Choice Vouchers (24 CFR part 982) or other similar forms of portable housing assistance that are tied to an occupant and not an individual dwelling unit, are not affordable housing income-restricted dwelling units for purposes of Sec. 1003.4(a)(32). 2. Federal affordable housing sources. Examples of Federal programs and funding sources that may result in individual dwelling units that are reportable under Sec. 1003.4(a)(32) include, but are not limited to: i. Affordable housing programs pursuant to Section 8 of the United States Housing Act of 1937 (42 U.S.C. 1437f); ii. Public housing (42 U.S.C. 1437a(b)(6)); iii. The HOME Investment Partnerships program (24 CFR part 92); iv. The Community Development Block Grant program (24 CFR part 570); v. Multifamily tax subsidy project funding through tax-exempt bonds or tax credits (26 U.S.C. 42; 26 U.S.C. 142(d)); vi. Project-based vouchers (24 CFR part 983); vii. Federal Home Loan Bank affordable housing program funding (12 CFR part 1291); and viii. Rural Housing Service multifamily housing loans and grants (7 CFR part 3560). 3. State and local government affordable housing sources. Examples of State and local sources that may result in individual dwelling units that are reportable under Sec. 1003.4(a)(32) include, but are not limited to: State or local administration of Federal funds or programs; State or local funding programs for affordable housing or rental assistance, including programs operated by independent public authorities; inclusionary zoning laws; and tax abatement or tax increment financing contingent on affordable housing requirements. 4. Multiple properties. See comment 4(a)(9)-2 regarding transactions involving multiple properties with more than one property taken as security. 5. Best information available. A financial institution may rely on the best information readily available to the financial institution at the time final action is taken and on the financial institution's own procedures in reporting the information required by Sec. 1003.4(a)(32). Information readily available could include, for example, information provided by an applicant that the financial institution reasonably believes, information contained in a property valuation or inspection, or information obtained from public records. 6. Scope of requirement. A financial institution reports that the requirement is not applicable if the property securing the covered loan or, in the case of an application, proposed to secure the covered loan is not a multifamily dwelling. Paragraph 4(a)(33) 1. Agents. If a financial institution is reporting actions taken by its agent consistent with comment 4(a)-4, the agent is not considered the financial institution for the purposes of Sec. 1003.4(a)(33). For example, assume that an applicant submitted an application to Financial Institution A, and Financial Institution A made the credit decision acting as Financial Institution B's agent under State law. A covered loan was originated and the obligation arising from a covered loan was initially payable to Financial Institution A. Financial Institution B purchased the loan. Financial Institution B reports the origination and not the purchase, and indicates that the application was not submitted directly to the financial institution and that the transaction was not initially payable to the financial institution. Paragraph 4(a)(33)(i) 1. General. Section 4(a)(33)(i) requires a financial institution to indicate whether the applicant or borrower submitted the application directly to the financial institution that is reporting the covered loan or application. The following scenarios demonstrate whether an application was submitted directly to the financial institution that is reporting the covered loan or application. i. The application was submitted directly to the financial institution if the mortgage loan originator identified pursuant to Sec. 1003.4(a)(34) was an employee of the reporting financial institution when the originator performed the origination activities for the covered loan or application that is being reported. ii. The application was also submitted directly to the financial institution reporting the covered loan or application if the reporting financial institution directed the applicant to a third- party agent (e.g., a credit union service organization) that performed loan origination activities on behalf of the financial institution and did not assist the applicant with applying for covered loans with other institutions. iii. If an applicant contacted and completed an application with a broker or correspondent that forwarded the application to a financial institution for approval, an application was not submitted to the financial institution. Paragraph 4(a)(33)(ii) 1. General. Section 1003.4(a)(33)(ii) requires financial institutions to report whether the obligation arising from a covered loan was or, in the case of an application, would have been initially payable to the institution. An obligation is initially payable to the institution if the obligation is initially payable either on the face of the note or contract to the financial institution that is reporting the covered loan or application. For example, if a financial institution reported an origination of a covered loan that it approved prior to closing, that closed in the name of a third-party, such as a correspondent lender, and that the financial institution purchased after closing, the covered loan was not initially payable to the financial institution. 2. Applications. A financial institution complies with Sec. 1003.4(a)(33)(ii) by reporting that the requirement is not applicable if the institution had not determined whether the covered loan would have been initially payable to the institution reporting the application when the application was withdrawn, denied, or closed for incompleteness. Paragraph 4(a)(34) 1. NMLSR ID. Section 1003.4(a)(34) requires a financial institution to report the Nationwide Mortgage Licensing System and Registry unique identifier (NMLSR ID) for the mortgage loan originator, as defined in Regulation G, 12 CFR 1007.102, or Regulation H, 12 CFR 1008.23, as applicable. The NMLSR ID is a unique number or other identifier generally assigned to individuals registered or licensed through NMLSR to provide loan originating services. For more information, see the Secure and Fair Enforcement for Mortgage Licensing Act of 2008, title V of the Housing and Economic Recovery Act of 2008 (S.A.F.E. Act), 12 U.S.C. 5101 et seq., and its implementing regulations (12 CFR part 1007 and 12 CFR part 1008). 2. Mortgage loan originator without NMLSR ID. An NMLSR ID for the mortgage loan originator is not required by Sec. 1003.4(a)(34) to be reported by a financial institution if the mortgage loan originator is not required to obtain and has not been assigned an NMLSR ID. For example, certain individual mortgage loan originators may not be required to obtain an NMLSR ID for the particular transaction being reported by the financial institution, such as a commercial loan. However, some mortgage loan originators may have obtained an NMLSR ID even if they are not required to obtain one for that particular transaction. If a mortgage loan originator has been assigned an NMLSR ID, a financial institution complies with Sec. 1003.4(a)(34) by reporting the mortgage loan originator's NMLSR ID regardless of [[Page 66337]] whether the mortgage loan originator is required to obtain an NMLSR ID for the particular transaction being reported by the financial institution. In the event that the mortgage loan originator is not required to obtain and has not been assigned an NMLSR ID, a financial institution complies with Sec. 1003.4(a)(34) by reporting that the requirement is not applicable. 3. Multiple mortgage loan originators. If more than one individual associated with a covered loan or application meets the definition of a mortgage loan originator, as defined in Regulation G, 12 CFR 1007.102, or Regulation H, 12 CFR 1008.23, a financial institution complies with Sec. 1003.4(a)(34) by reporting the NMLSR ID of the individual mortgage loan originator with primary responsibility for the transaction as of the date of action taken pursuant to Sec. 1003.4(a)(8)(ii). A financial institution that establishes and follows a reasonable, written policy for determining which individual mortgage loan originator has primary responsibility for the reported transaction as of the date of action taken complies with Sec. 1003.4(a)(34). Paragraph 4(a)(35) 1. Automated underwriting system data--general. A financial institution complies with Sec. 1003.4(a)(35) by reporting, except for purchased covered loans, the name of the automated underwriting system (AUS) used by the financial institution to evaluate the application and the result generated by that AUS. The following scenarios illustrate when a financial institution reports the name of the AUS used by the financial institution to evaluate the application and the result generated by that AUS. i. A financial institution that uses an AUS, as defined in Sec. 1003.4(a)(35)(ii), to evaluate an application, must report the name of the AUS used by the financial institution to evaluate the application and the result generated by that system, regardless of whether the AUS was used in its underwriting process. For example, if a financial institution uses an AUS to evaluate an application prior to submitting the application through its underwriting process, the financial institution complies with Sec. 1003.4(a)(35) by reporting the name of the AUS it used to evaluate the application and the result generated by that system. ii. A financial institution that uses an AUS, as defined in Sec. 1003.4(a)(35)(ii), to evaluate an application, must report the name of the AUS it used to evaluate the application and the result generated by that system, regardless of whether the financial institution intends to hold the covered loan in its portfolio or sell the covered loan. For example, if a financial institution uses an AUS developed by a securitizer to evaluate an application and intends to sell the covered loan to that securitizer but ultimately does not sell the covered loan and instead holds the covered loan in its portfolio, the financial institution complies with Sec. 1003.4(a)(35) by reporting the name of the securitizer's AUS that the institution used to evaluate the application and the result generated by that system. Similarly, if a financial institution uses an AUS developed by a securitizer to evaluate an application to determine whether to originate the covered loan but does not intend to sell the covered loan to that securitizer and instead holds the covered loan in its portfolio, the financial institution complies with Sec. 1003.4(a)(35) by reporting the name of the securitizer's AUS that the institution used to evaluate the application and the result generated by that system. iii. A financial institution that uses an AUS, as defined in Sec. 1003.4(a)(35)(ii), that is developed by a securitizer to evaluate an application, must report the name of the AUS it used to evaluate the application and the result generated by that system, regardless of whether the securitizer intends to hold the covered loan it purchased from the financial institution in its portfolio or securitize the covered loan. For example, if a financial institution uses an AUS developed by a securitizer to evaluate an application and the financial institution sells the covered loan to that securitizer but the securitizer holds the covered loan it purchased in its portfolio, the financial institution complies with Sec. 1003.4(a)(35) by reporting the name of the securitizer's AUS that the institution used to evaluate the application and the result generated by that system. iv. A financial institution, which is also a securitizer, that uses its own AUS, as defined in Sec. 1003.4(a)(35)(ii), to evaluate an application, must report the name of the AUS it used to evaluate the application and the result generated by that system, regardless of whether the financial institution intends to hold the covered loan it originates in its portfolio, purchase the covered loan, or securitize the covered loan. For example, if a financial institution, which is also a securitizer, has developed its own AUS and uses that AUS to evaluate an application that it intends to originate and hold in its portfolio and not purchase or securitize the covered loan, the financial institution complies with Sec. 1003.4(a)(35) by reporting the name of its AUS that it used to evaluate the application and the result generated by that system. 2. Definition of automated underwriting system. A financial institution must report the information required by Sec. 1003.4(a)(35)(i) if the financial institution uses an automated underwriting system (AUS), as defined in Sec. 1003.4(a)(35)(ii), to evaluate an application. In order for an AUS to be covered by the definition in Sec. 1003.4(a)(35)(ii), the system must be an electronic tool that has been developed by a securitizer, Federal government insurer, or a Federal government guarantor. For example, if a financial institution has developed its own proprietary system that it uses to evaluate an application and the financial institution is also a securitizer, then the financial institution complies with Sec. 1003.4(a)(35) by reporting the name of that system and the result generated by that system. On the other hand, if a financial institution has developed its own proprietary system that it uses to evaluate an application but the financial institution is not a securitizer, then the financial institution is not required by Sec. 1003.4(a)(35) to report the use of that system and the result generated by that system. In addition, in order for an AUS to be covered by the definition in Sec. 1003.4(a)(35)(ii), the system must provide a result regarding both the credit risk of the applicant and the eligibility of the covered loan to be originated, purchased, insured, or guaranteed by the securitizer, Federal government insurer, or Federal government guarantor that developed the system being used to evaluate the application. For example, if a system is an electronic tool that provides a determination of the eligibility of the covered loan to be originated, purchased, insured, or guaranteed by the securitizer, Federal government insurer, or Federal government guarantor that developed the system being used by a financial institution to evaluate the application, but the system does not also provide an assessment of the creditworthiness of the applicant--such as, an evaluation of the applicant's income, debt, and credit history--then that system does not qualify as an AUS, as defined in Sec. 1003.4(a)(35)(ii). A financial institution that uses a system that is not an AUS, as defined in Sec. 1003.4(a)(35)(ii), to evaluate an application does not report the information required by Sec. 1003.4(a)(35)(i). 3. Reporting automated underwriting system data--multiple results. When a financial institution uses one or more automated underwriting systems (AUS) to evaluate the application and the system or systems generate two or more results, the financial institution complies with Sec. 1003.4(a)(35) by reporting, except for purchased covered loans, the name of the AUS used by the financial institution to evaluate the application and the result generated by that AUS as determined by the following principles. To determine what AUS (or AUSs) and result (or results) to report under Sec. 1003.4(a)(35), a financial institution follows each of the principles that is applicable to the application in question, in the order in which they are set forth below. i. If a financial institution obtains two or more AUS results and the AUS generating one of those results corresponds to the loan type reported pursuant to Sec. 1003.4(a)(2), the financial institution complies with Sec. 1003.4(a)(35) by reporting that AUS name and result. For example, if a financial institution evaluates an application using the Federal Housing Administration's (FHA) Technology Open to Approved Lenders (TOTAL) Scorecard and subsequently evaluates the application with an AUS used to determine eligibility for a non-FHA loan, but ultimately originates an FHA loan, the financial institution complies with Sec. 1003.4(a)(35) by reporting TOTAL Scorecard and the result generated by that system. If a financial institution obtains two or more AUS results and more than one of those AUS results is generated by a system that corresponds to the loan type reported pursuant to Sec. 1003.4(a)(2), the financial institution identifies which AUS result should be reported by following the principle set forth below in comment 4(a)(35)-3.ii. ii. If a financial institution obtains two or more AUS results and the AUS generating one of those results corresponds to the purchaser, insurer, or guarantor, if any, the financial institution complies with Sec. 1003.4(a)(35) by reporting that AUS name [[Page 66338]] and result. For example, if a financial institution evaluates an application with the AUS of Securitizer A and subsequently evaluates the application with the AUS of Securitizer B, but the financial institution ultimately originates a covered loan that it sells within the same calendar year to Securitizer A, the financial institution complies with Sec. 1003.4(a)(35) by reporting the name of Securitizer A's AUS and the result generated by that system. If a financial institution obtains two or more AUS results and more than one of those AUS results is generated by a system that corresponds to the purchaser, insurer, or guarantor, if any, the financial institution identifies which AUS result should be reported by following the principle set forth below in comment 4(a)(35)-3.iii. iii. If a financial institution obtains two or more AUS results and none of the systems generating those results correspond to the purchaser, insurer, or guarantor, if any, or the financial institution is following this principle because more than one AUS result is generated by a system that corresponds to either the loan type or the purchaser, insurer, or guarantor, the financial institution complies with Sec. 1003.4(a)(35) by reporting the AUS result generated closest in time to the credit decision and the name of the AUS that generated that result. For example, if a financial institution evaluates an application with the AUS of Securitizer A, subsequently again evaluates the application with Securitizer A's AUS, the financial institution complies with Sec. 1003.4(a)(35) by reporting the name of Securitizer A's AUS and the second AUS result. Similarly, if a financial institution obtains a result from an AUS that requires the financial institution to underwrite the loan manually, but the financial institution subsequently processes the application through a different AUS that also generates a result, the financial institution complies with Sec. 1003.4(a)(35) by reporting the name of the second AUS that it used to evaluate the application and the AUS result generated by that system. iv. If a financial institution obtains two or more AUS results at the same time and the principles in comment 4(a)(35)-3.i through .iii do not apply, the financial institution complies with Sec. 1003.4(a)(35) by reporting the name of all of the AUSs used by the financial institution to evaluate the application and the results generated by each of those systems. For example, if a financial institution simultaneously evaluates an application with the AUS of Securitizer A and the AUS of Securitizer B, the financial institution complies with Sec. 1003.4(a)(35) by reporting the name of both Securitizer A's AUS and Securitizer B's AUS and the results generated by each of those systems. In any event, however, the financial institution does not report more than five AUSs and five results. If more than five AUSs and five results meet the criteria in this principle, the financial institution complies with Sec. 1003.4(a)(35) by choosing any five among them to report. 4. Transactions for which an automated underwriting system was not used to evaluate the application. Section 1003.4(a)(35) does not require a financial institution to evaluate an application using an automated underwriting system (AUS), as defined in Sec. 1003.4(a)(35)(ii). For example, if a financial institution only manually underwrites an application and does not use an AUS to evaluate the application, the financial institution complies with Sec. 1003.4(a)(35) by reporting that the requirement is not applicable since an AUS was not used to evaluate the application. 5. Purchased covered loan. A financial institution complies with Sec. 1003.4(a)(35) by reporting that the requirement is not applicable when the covered loan is a purchased covered loan. 6. Non-natural person. When the applicant and co-applicant, if applicable, are not natural persons, a financial institution complies with Sec. 1003.4(a)(35) by reporting that the requirement is not applicable. Paragraph 4(a)(37) 1. Open-end line of credit. Section 1003.4(a)(37) requires a financial institution to identify whether the covered loan or the application is for an open-end line of credit. See comments 2(o)-1 and -2 for a discussion of open-end line of credit and extension of credit. Paragraph 4(a)(38) 1. Primary purpose. Section 1003.4(a)(38) requires a financial institution to identify whether the covered loan is, or the application is for a covered loan that will be, made primarily for a business or commercial purpose. See comment 3(c)(10)-2 for a discussion of how to determine the primary purpose of the transaction and the standard applicable to financial institution's determination of the primary purpose of the transaction. See comments 3(c)(10)-3 and -4 for examples of excluded and reportable business- or commercial-purpose transactions. 4(f) Quarterly Recording of Data 1. General. Section 1003.4(f) requires a financial institution to record the data collected pursuant to Sec. 1003.4 on a loan/ application register within 30 calendar days after the end of the calendar quarter in which final action is taken. Section 1003.4(f) does not require a financial institution to record data on a single loan/application register on a quarterly basis. Rather, for purposes of Sec. 1003.4(f), a financial institution may record data on a single loan/application register or separately for different branches or different loan types (such as home purchase or home improvement loans, or loans on multifamily dwellings). 2. Agency requirements. Certain State or Federal regulations may require a financial institution to record its data more frequently than is required under Regulation C. 3. Form of quarterly records. A financial institution may maintain the records required by Sec. 1003.4(f) in electronic or any other format, provided the institution can make the information available to its regulatory agency in a timely manner upon request. Section 1003.5--Disclosure and Reporting 5(a) Reporting to Agency 1. [Reserved] 2. [Reserved] 3. [Reserved] 4. [Reserved] 5. Change in appropriate Federal agency. If the appropriate Federal agency for a covered institution changes (as a consequence of a merger or a change in the institution's charter, for example), the institution must report data to the new appropriate Federal agency beginning with the year of the change. 6. Subsidiaries. An institution is a subsidiary of a bank or savings association (for purposes of reporting HMDA data to the same agency as the parent) if the bank or savings association holds or controls an ownership interest that is greater than 50 percent of the institution. 7. Transmittal sheet--additional data submissions. If an additional data submission becomes necessary (for example, because the institution discovers that data were omitted from the initial submission, or because revisions are called for), that submission must be accompanied by a transmittal sheet. 8. Transmittal sheet--revisions or deletions. If a data submission involves revisions or deletions of previously submitted data, it must state the total of all line entries contained in that submission, including both those representing revisions or deletions of previously submitted entries, and those that are being resubmitted unchanged or are being submitted for the first time. Depository institutions must provide a list of the MSAs or Metropolitan Divisions in which they have home or branch offices. 5(b) Disclosure Statement 1. Business day. For purposes of Sec. 1003.5(b), a business day is any calendar day other than a Saturday, Sunday, or legal public holiday. 2. Format of notice. A financial institution may make the written notice required under Sec. 1003.5(b)(2) available in paper or electronic form. 3. Notice--suggested text. A financial institution may use any text that meets the requirements of Sec. 1003.5(b)(2). The following language is suggested but is not required: Home Mortgage Disclosure Act Notice The HMDA data about our residential mortgage lending are available online for review. The data show geographic distribution of loans and applications; ethnicity, race, sex, age, and income of applicants and borrowers; and information about loan approvals and denials. These data are available online at the Consumer Financial Protection Bureau's Web site (www.consumerfinance.gov/hmda). HMDA data for many other financial institutions are also available at this Web site. 4. Combined notice. A financial institution may use the same notice to satisfy the requirements of both Sec. 1003.5(b)(2) and Sec. 1003.5(c). 5(c) Modified loan/application Register 1. Format of notice. A financial institution may make the written notice required under Sec. 1003.5(c)(1) available in paper or electronic form. 2. Notice--suggested text. A financial institution may use any text that meets the requirements of Sec. 1003.5(c)(1). The following language is suggested but is not required: [[Page 66339]] Home Mortgage Disclosure Act Notice The HMDA data about our residential mortgage lending are available online for review. The data show geographic distribution of loans and applications; ethnicity, race, sex, age, and income of applicants and borrowers; and information about loan approvals and denials. These data are available online at the Consumer Financial Protection Bureau's Web site (www.consumerfinance.gov/hmda). HMDA data for many other financial institutions are also available at this Web site. 3. Combined notice. A financial institution may use the same notice to satisfy the requirements of both Sec. 1003.5(c) and Sec. 1003.5(b)(2). 5(e) Posted Notice of Availability of Data 1. Posted notice--suggested text. A financial institution may post any text that meets the requirements of Sec. 1003.5(e). The Bureau or other appropriate Federal agency for a financial institution may provide a notice that the institution can post to inform the public of the availability of its HMDA data, or an institution may create its own notice. The following language is suggested but is not required: Home Mortgage Disclosure Act Notice The HMDA data about our residential mortgage lending are available online for review. The data show geographic distribution of loans and applications; ethnicity, race, sex, age, and income of applicants and borrowers; and information about loan approvals and denials. HMDA data for many other financial institutions are also available online. For more information, visit the Consumer Financial Protection Bureau's Web site (www.consumerfinance.gov/hmda). Section 1003.6--Enforcement 6(b) Bona Fide Errors 1. Bona fide error--information from third parties. An institution that obtains the property-location information for applications and loans from third parties (such as appraisers or vendors of ``geocoding'' services) is responsible for ensuring that the information reported on its HMDA/LAR is correct. 0 16. Effective January 1, 2019, in Supplement I to Part 1003: a. Under the heading Section 1003.5--Disclosure and Reporting, under the subheading 5(a) Reporting to Agency, paragraphs 1, 2, 3, and 4 are added, paragraph 5 is revised, and paragraphs 6, 7, and 8 are removed; b. Under the heading Section 1003.6--Enforcement, under the subheading 6(b) Bona Fide Errors, paragraph 1 is revised. The additions and revisions read as follows: Supplement I to Part 1003--Official Interpretations * * * * * Section 1003.5--Disclosure and Reporting 5(a) Reporting to Agency 1. Quarterly reporting--coverage. i. Section 1003.5(a)(1)(ii) requires that, within 60 calendar days after the end of each calendar quarter except the fourth quarter, a financial institution that reported for the preceding calendar year at least 60,000 covered loans and applications, combined, excluding purchased covered loans, must submit its loan/application register containing all data required to be recorded for that quarter pursuant to Sec. 1003.4(f). For example, if for calendar year 2019 Financial Institution A reports 60,000 covered loans, excluding purchased covered loans, it must comply with Sec. 1003.5(a)(1)(ii) in calendar year 2020. Similarly, if for calendar year 2019 Financial Institution A reports 20,000 applications and 40,000 covered loans, combined, excluding purchased covered loans, it must comply with Sec. 1003.5(a)(1)(ii) in calendar year 2020. If for calendar year 2020 Financial Institution A reports fewer than 60,000 covered loans and applications, combined, excluding purchased covered loans, it is not required to comply with Sec. 1003.5(a)(1)(ii) in calendar year 2021. ii. In the calendar year of a merger or acquisition, the surviving or newly formed financial institution is required to comply with Sec. 1003.5(a)(1)(ii), effective the date of the merger or acquisition, if a combined total of at least 60,000 covered loans and applications, combined, excluding purchased covered loans, is reported for the preceding calendar year by or for the surviving or newly formed financial institution and each financial institution or branch office merged or acquired. For example, Financial Institution A and Financial Institution B merge to form Financial Institution C in 2020. Financial Institution A reports 40,000 covered loans and applications, combined, excluding purchased covered loans, for 2019. Financial Institution B reports 21,000 covered loans and applications, combined, excluding purchased covered loans, for 2019. Financial Institution C is required to comply with Sec. 1003.5(a)(1)(ii) effective the date of the merger. Similarly, for example, Financial Institution A acquires a branch office of Financial Institution B in 2020. Financial Institution A reports 58,000 covered loans and applications, combined, excluding purchased covered loans, for 2019. Financial Institution B reports 3,000 covered loans and applications, combined, excluding purchased covered loans, for 2019 for the branch office acquired by Financial Institution A. Financial Institution A is required to comply with Sec. 1003.5(a)(1)(ii) in 2020 effective the date of the branch acquisition. iii. In the calendar year following a merger or acquisition, the surviving or newly formed financial institution is required to comply with Sec. 1003.5(a)(1)(ii) if a combined total of at least 60,000 covered loans and applications, combined, excluding purchased covered loans, is reported for the preceding calendar year by or for the surviving or newly formed financial institution and each financial institution or branch office merged or acquired. For example, Financial Institution A and Financial Institution B merge to form Financial Institution C in 2019. Financial Institution C reports 21,000 covered loans and applications, combined, excluding purchased covered loans, each for Financial Institution A, B, and C for 2019, for a combined total of 63,000 covered loans and applications reported, excluding purchased covered loans. Financial Institution C is required to comply with Sec. 1003.5(a)(1)(ii) in 2020. Similarly, for example, Financial Institution A acquires a branch office of Financial Institution B in 2019. Financial Institution A reports 58,000 covered loans and applications, combined, excluding purchased covered loans, for 2019. Financial Institution A or B reports 3,000 covered loans and applications, combined, excluding purchased covered loans, for 2019 for the branch office acquired by Financial Institution A. Financial Institution A is required to comply with Sec. 1003.5(a)(1)(ii) in 2020. 2. Change in appropriate Federal agency. If the appropriate Federal agency for a financial institution changes (as a consequence of a merger or a change in the institution's charter, for example), the institution must identify its new appropriate Federal agency in its annual submission of data pursuant to Sec. 1003.5(a)(1)(i) for the year of the change. For example, if an institution's appropriate Federal agency changes in February 2018, it must identify its new appropriate Federal agency beginning with the annual submission of its 2018 data by March 1, 2019 pursuant to Sec. 1003.5(a)(1)(i). For an institution required to comply with Sec. 1003.5(a)(1)(ii), the institution also must identify its new appropriate Federal agency in its quarterly submission of data pursuant to Sec. 1003.5(a)(1)(ii) beginning with its submission for the quarter of the change, unless the change occurs during the fourth quarter. For example, if the appropriate Federal agency for an institution required to comply with Sec. 1003.5(a)(1)(ii) changes during February 2020, the institution must identify its new appropriate Federal agency beginning with its quarterly submission pursuant to Sec. 1003.5(a)(1)(ii) for the first quarter of 2020. If the appropriate Federal agency for an institution required to comply with Sec. 1003.5(a)(1)(ii) changes during December 2020, the institution must identify its new appropriate Federal agency beginning with the annual submission of its 2020 data by March 1, 2021 pursuant to Sec. 1003.5(a)(1)(i). 3. Subsidiaries. A financial institution is a subsidiary of a bank or savings association (for purposes of reporting HMDA data to the same agency as the parent) if the bank or savings association holds or controls an ownership interest in the institution that is greater than 50 percent. 4. Retention. A financial institution may satisfy the requirement under Sec. 1003.5(a)(1)(i) that it retain a copy of its submitted annual loan/application register for three years by retaining a copy of the annual loan/application register in either electronic or paper form. 5. Federal Taxpayer Identification Number. Section 1003.5(a)(3) requires a financial institution to provide its Federal Taxpayer Identification Number with its data submission. If a financial institution obtains a new Federal Taxpayer Identification [[Page 66340]] Number, it should provide the new number in its subsequent data submission. For example, if two financial institutions that previously reported HMDA data under this part merge and the surviving institution retained its Legal Entity Identifier but obtained a new Federal Taxpayer Identification Number, then the surviving institution should report the new Federal Taxpayer Identification Number with its HMDA data submission. * * * * * Section 1003.6--Enforcement 6(b) Bona Fide Errors 1. Information from third parties. Section 1003.6(b) provides that an error in compiling or recording data for a covered loan or application is not a violation of the Act or this part if the error was unintentional and occurred despite the maintenance of procedures reasonably adapted to avoid such an error. A financial institution that obtains the required data, such as property-location information, from third parties is responsible for ensuring that the information reported pursuant to Sec. 1003.5 is correct. Dated: October 13, 2015. Richard Cordray, Director, Bureau of Consumer Financial Protection. [FR Doc. 2015-26607 Filed 10-27-15; 8:45 am] BILLING CODE 4810-AM-P
| Category | Regulatory Information | |
| Collection | Federal Register | |
| sudoc Class | AE 2.7: GS 4.107: AE 2.106: | |
| Publisher | Office of the Federal Register, National Archives and Records Administration | |
| Section | Rules and Regulations | |
| Action | Final rule; official interpretations. | |
| Dates | This rule is effective on January 1, 2018, except that the amendment to Sec. 1003.2 in amendatory instruction 3 is effective on January 1, 2017; the amendments to Sec. 1003.5 in amendatory instruction 8, the amendments to Sec. 1003.6 in amendatory instruction 10, the amendments to appendix A to part 1003 in amendatory instruction 12, and the amendments to supplement I to part 1003 in amendatory instruction 16 are effective on January 1, 2019; and the amendments to Sec. 1003.5 in amendatory instruction 9 are effective on January 1, 2020. See part VI for more information. | |
| Contact | Jaydee DiGiovanni, David Jacobs, Terry J. Randall, or James Wylie, Counsels; or Elena Grigera Babinecz, Courtney Jean, Joan Kayagil, Thomas J. Kearney, or Laura Stack, Senior Counsels, Office of Regulations, at (202) 435-7700. | |
| FR Citation | 80 FR 66128 | |
| RIN Number | 3170-AA10 | |
| CFR Associated | Banks; Banking; Credit Unions; Mortgages; National Banks; Savings Associations and Reporting and Recordkeeping Requirements |