I. Introduction

On November 5, 2018, NYSE Group, Inc., on behalf of the other parties ^[1] to the National Market System Plan to Address Extraordinary Market Volatility (the “Plan”), filed with the Securities and Exchange Commission (“Commission”) pursuant to Section 11A of the Securities Exchange Act of 1934 (“Act”) ^[2] and Rule 608 thereunder,^[3] a proposal to amend the Plan ^[4] to, among other things, amend Section VIII of the Plan to transition the Plan from operating on a pilot to a permanent basis. The proposal represents the eighteenth amendment to the Plan, and reflects proposed changes unanimously approved by the Participants (“Eighteenth Amendment”). A copy of the Plan is ( printed page 16087) attached hereto as Exhibit A. The proposed Eighteenth Amendment was published for comment in the Federal Register on December 26, 2018.^[5] The Commission received three comment letters regarding the amendment.^[6] This order approves the Eighteenth Amendment to the Plan as proposed.

II. Background

On May 6, 2010, the U.S. equity markets experienced a severe disruption.^[7] Among other things, the prices of a large number of individual securities suddenly declined by significant amounts in a very short time period, before suddenly reversing to prices consistent with their pre-decline levels. The Commission was concerned that events such as those that occurred on May 6, 2010 could seriously undermine the integrity of the U.S. markets. Accordingly, Commission staff worked with the exchanges and FINRA (“SROs”) to develop policy responses that would help prevent a recurrence of the May 6 market disruption. Initially, the SROs developed a single-stock circuit breakers pilot program, implemented through a series of rule filings, to pause trading during periods of extraordinary volatility in all NMS Stocks, except rights and warrants.

As a replacement to the single-stock circuit breaker pilot, the Participants filed the Plan with the Commission on April 5, 2011 to create a market-wide limit up-limit down (“LULD”) mechanism intended to address extraordinary market volatility in “NMS Stocks,” as defined in Rule 600(b)(47) of Regulation NMS under the Exchange Act.^[8] The Plan sets forth procedures that provide for market-wide limit up-limit down requirements to prevent trades in individual NMS Stocks from occurring outside of the specified Price Bands.^[9] These limit up-limit down requirements are coupled with Trading Pauses, as defined in Section I(Y) of the Plan, to accommodate more fundamental price moves (as opposed to erroneous trades or momentary gaps in liquidity). The limit up-limit down mechanism is intended to reduce the negative impacts of sudden, unanticipated price movements in NMS Stocks, such as those experienced on May 6, 2010, thereby protecting investors and promoting a fair and orderly market.

The Plan was approved in May 2012 on a pilot basis to “allow the Participants and the public to gain valuable practical experience with Plan operations during the pilot period” and to assess “whether further modifications of the Plan are necessary or appropriate prior to final approval.” ^[10] After two amendments,^[11] the initial date of Plan operations was April 8, 2013.^[12] Since that date, the Plan has been amended fourteen times ^[13] and the pilot period has been extended six times.^[14] The most recent substantive changes to the Plan were made through the Tenth,^[15] Twelfth,^[16] and Thirteenth ^[17] Amendments. On May 28, 2015, the Participants submitted a Supplemental Joint Assessment, in which the Participants provided additional analysis required under Appendix B.^[18]

III. Description of the Proposal

In the Eighteenth Amendment, the Participants propose to: (i) Amend Section VIII of the Plan to transition the Plan from operating on a pilot to a permanent basis; (ii) adopt a mechanism for periodic review and assessment of the Plan; (iii) eliminate the doubling of the Percentage Parameters between 9:30 a.m. and 9:45 a.m.; and (iv) eliminate the doubling of the Percentage Parameters between 3:35 p.m. and 4:00 p.m., or in the case of an early scheduled close, during the last 25 minutes of trading before the early scheduled close, for Tier 2 NMS Stocks with a Reference Price above $3.00.

IV. Summary of Comments Received

The Commission received three comment letters regarding the amendment.^[19] All three commenter letters support approval of the Eighteenth Amendment.^[20] Two commenters specifically support the proposal to transition the plan from a pilot to operating on a permanent basis, subject to periodic review and assessment.^[21] Two commenters support the proposal to eliminate the doubling of percentage parameters,^[22] with both commenters providing specific rationales for eliminating doubling of parameters between 9:30 a.m. and 9:45 a.m.,^[23] and one commenter providing specific support for the elimination of double-wide Price Bands at the close for certain securities.^[24]

Beyond addressing the proposals in the Eighteenth Amendment, one commenter urges the Commission to add representatives of non-SRO experts, including advisers to registered funds and broker-dealers, to the operating committee of the Limit Up-Limit Down Plan and other NMS plans.^[25] Another commenter recommends that the Commission, after adopting the proposal, adopt the recommendation of the Equity Market Structure Advisory Committee (“EMSAC”) ^[26] to review clearly erroneous execution (“CEE”) rules to promote certainty of execution so that all trades executed within the Limit Up-Limit Down Plan bands stand.^[27]

V. Discussion and Commission Findings

The Commission finds that the Eighteenth Amendment, as proposed, is consistent with the requirements of the Act and the rules and regulations thereunder. Specifically, the Commission finds that the Eighteenth Amendment is consistent with Section 11A of the Act ^[28] and Rule 608 thereunder ^[29] in that, as discussed below, the proposal is appropriate in the public interest, for the protection of investors and the maintenance of fair and orderly markets, and that it removes impediments to, and perfects the mechanism of, a national market system.

Proposal for Plan To Operate on a Permanent Basis

The Plan was originally approved on a pilot basis to allow the public, the Participants, and the Commission to assess the operation of the Plan and whether the Plan should be modified prior to consideration of approval on a permanent basis.^[30] The Plan has been operating on a pilot basis since its inception.^[31] The Participants are now proposing to make the Plan permanent, with procedures to help ensure regular monitoring of the LULD mechanism.^[32]

In support of their proposal for permanence, the Participants state that during the pilot period they collected and provided to the Commission and the public with a significant amount of data on the Plan's performance to aid in an assessment of its operations.^[33] The Participants state that the data collected during the pilot period and studies conducted by the Participants and the Commission's Division of Economic and Risk Analysis (“DERA”) show that the Plan has been beneficial to the markets by serving to dampen price volatility.^[34] The Participants cite a DERA analysis that, depending on the methodology employed, found evidence that the LULD mechanism reduced extraordinary transitory volatility relative to the Single Stock Circuit Breaker (“SSCB”) mechanism that was in place prior to the LULD mechanism.^[35] The Participants also rely on the results of the Supplemental Joint Assessment by the Participants that found that the number of trades that were cancelled decreased under the Plan and that the Plan's parameters were successful in preventing trades from occurring outside of the Price Bands, thus avoiding the types of mispriced trades that resulted in the Flash Crash.^[36]

The Participants further state that recent amendments approved by the Commission have improved the operation of the Plan.^[37] Amendment No. 10 changed the manner in which Reference Prices were determined in situations where a security opened for trading on a quote rather than a trade. Prior to implementation of Amendment No. 10, Participants state that Reference Prices in these situations triggered Limit States and Trading Pauses at inaccurate price levels.^[38] After implementation of Amendment No. 10, data provided by the Participants in the Transmittal Letter showed the number of Trading Pauses dropped significantly.^[39] A White Paper written by DERA confirmed these findings.^[40]

Additionally, the Participants note that the implementation of Amendment Nos. 12 and 13 in November 2017 modified the operation of the Plan to address issues that were uncovered by market events on August 24, 2015.^[41] These changes, which were made alongside coordinated changes by Primary Listing Exchanges to their reopening auction processes, were designed to avoid repeated Trading Pauses by improving the accuracy of reopening prices.^[42] To achieve this, the Plan was amended to prohibit trade resumption until a Primary Listing Exchange conducted a reopening auction, a feature that was designed to concentrate liquidity in the reopening auctions.^[43] The Primary Listing Exchanges also harmonized aspects of their reopening auction processes that provided for gradual extension of auction time frames accompanied by a gradual widening of auction price ranges with each auction extension.^[44] The Participants state that since these changes, although there has not been an event like August 24, 2015, there has been stable price continuity at the open and following reopenings after a Trading Pause, and the amended Plan has worked well during normal market conditions as well as the volatile market activity that occurred in February 2018.^[45]

The Commission notes that the analysis presented by the Participants, in addition to other analyses, demonstrates that the Plan has operated effectively in accomplishing its stated goal of addressing extraordinary market volatility.^[46] For example, the analysis presented in the Supplemental Joint Assessment demonstrates that the Plan has been effective in reducing volatility by showing that the Plan has reduced the frequency of multiple cancellation events that occur compared to the period during which the SSCB mechanism was in effect, as well as the time period before the SSCB mechanism was in effect.^[47] The Commission notes that this analysis is also consistent with other analyses. One of the DERA White Papers cited by the Participants also found that the Plan's mechanism reduced extraordinary transitory volatility relative to the SSCB mechanism, as well as the time period before the SSCB mechanism was in effect.^[48] Both the Supplemental Joint Assessment ^[49] and a DERA White Paper ^[50] demonstrate that over 90% of Limit States resolve themselves in less than five seconds. Alternatively, the Commission notes that other analysis has found that the LULD mechanism increased the number of trading pauses and cancelled trades in Tier 2 securities compared to the SSCB mechanism.^[51] However, since this study focused on the time period before the implementation of Amendment 10, the results could be driven by bad Reference Prices that resulted from opening auctions with no trades. Both a DERA White Paper ^[52] and the Transmittal Letter from the Participants,^[53] present analysis that demonstrates that Amendment 10 reduced the number of Trading Pauses that occurred during the trading day.

The Participants have worked together with the Plan Advisory Committee to identify instances where improvements to the Plan were necessary, and developed and implemented amendments to the Plan to modify the operation of the LULD mechanism to help ensure its continued effectiveness over time.^[54] As a result of these efforts and based on analyses of the Plan's operation, the Commission believes that the LULD mechanism effectively addresses extraordinary market volatility, and therefore is approving the Plan on a permanent basis.

The Commission recognizes, however, that the market is dynamic and constantly evolving and that the Participants will continue to study the Plan. As a result, certain features or parameters used in the LULD mechanism may require modifications over time for the mechanism to remain effective. For example, the occurrence of CEE events and long-lasting Straddle States, i.e. Straddle States that last longer than five minutes,^[55] demonstrate that the parameters for Price Bands set forth in the Plan need to continue to be monitored in order to ensure their ( printed page 16090) calibration is appropriate.^[56] The Participants acknowledge the need for ongoing review of these and other types of potential issues, and have proposed a process that will include quarterly, annual, and ad hoc reports that will facilitate an ongoing assessment of the Plan's effectiveness.^[57]

Proposed Mechanism for Periodic Review and Assessment

The Participants state that the proposed ongoing review and assessment procedures are designed to ensure that the Plan will be monitored continually in a data-driven manner.^[58] Pursuant to this periodic reporting and assessment mechanism, the Participants propose to provide the Commission, and make publicly available, three categories of reports concerning the Plan's ongoing operation: (1) An annual report produced in consultation with the Advisory Committee assessing the Plan's performance,^[59] which would include an update on the Plan's operations,^[60] an analysis of any amendments to the Plan implemented during the period covered by the report,^[61] and an analysis of potential material emerging issues that may directly impact the operation of the Plan; ^[62] (2) quarterly reports providing basic statistics that could be used to identify trends in the performance and impact of the Plan on market activity; ^[63] and (3) upon Commission request, an ad hoc report on the effectiveness of LULD following a significant market event.^[64]

The Commission believes that a process for the ongoing evaluation of the Plan is critical for its permanent approval. Markets evolve, and the Commission believes that a process for assessing the effectiveness of the Plan over time will help ensure that the Plan continues to achieve its objective of reducing extraordinary volatility. In order to assess its effectiveness and identify appropriate modifications to the Plan, data and analysis of the ongoing functions of the LULD mechanism must be produced, reviewed and considered. In addition to the Participants, Advisory Committee members and the Commission having access to data and analyses regarding the Plan's performance, making such information available to the public will promote a robust public dialogue regarding the Plan's effectiveness.

As proposed, the Participants will provide the Commission and make available publicly quarterly reports, including basic statistics that can be used to identify trends in the performance of the LULD mechanism and its impact on market activity. In addition, the Participants will provide the Commission, and make available publicly on the LULD website,^[65] an Annual Report containing an analysis of the Plan's operation, including an examination of the parameters for Price Bands set forth in the Plan. The Annual Report will also include an analysis of the impact of any amendments to the Plan on the operation on the LULD mechanism. Finally, the Annual Report will discuss and analyze the LULD mechanism's performance during any significant market event that occurred during the period covered by the Annual Report, as well as any analyses performed on issues raised in the previous Annual Report. The Participants intend to submit the first Annual Report no later than March 31, 2020.^[66] The Participants will also provide to the Commission upon request, and make publicly available, a report analyzing the Plan's operation during a significant market event to the extent it is not reported in the Annual Report. In addition to these reports, the Participants will provide the Commission upon request within 30 days, data that is not otherwise publicly available and is substantially similar to the data they are required to provide under the current Plan.^[67]

The Commission believes the ongoing review and assessment requirements proposed by the Participants will both facilitate a robust, data-driven assessment of the Plan's effectiveness and provide the Commission and the public sufficient transparency of the effectiveness of the LULD mechanism necessary to help ensure the Plan remains designed to achieve its objective.

Proposal To Amend Calculation of Percentage Parameters

The Participants propose to (i) eliminate the doubling of the Percentage Parameters between 9:30 a.m. and 9:45 a.m.; and (ii) eliminate the doubling of the Percentage Parameters between 3:35 p.m. and 4:00 p.m., or in the case of an early scheduled close, during the last 25 minutes of trading before the early scheduled close, for Tier 2 NMS Stocks with a Reference Price above $3.00.

A. Elimination of Double-Wide Percentage Parameters at the Open

Currently under the Plan, between 9:30 a.m. and 9:45 a.m. (“the Open”), the Price Bands are calculated by applying double the Percentage Parameters. The Percentage Parameters are doubled to accommodate higher volatility at the Open.^[68] The ( printed page 16091) Participants propose to eliminate the double-wide Percentage Parameters at the Open.^[69] The Participants make two arguments for narrowing Price Bands at the Open.

First, the Participants argue that the current contraction of price bands at 9:45 causes unnecessary Limit States and Trading Pauses. In support of this argument, the Participants provide data that shows there is a disproportionate number of Limit States and Trading Pauses that occur at or shortly after 9:45 a.m., which is the only time during the trading day that the Price Bands contract.^[70] Furthermore, the Participants present evidence that the contraction of Price Bands at 9:45 causes Limit States and Trading Pauses at 9:45 a.m. that are not due to market volatility.^[71] Second, the Participants argue that narrower Price Bands at the Open would prevent erroneous trades during this time period by pausing trading at the narrower Price Bands rather than allowing such trades to execute at erroneous prices.^[72] The Participants present evidence that there are a disproportionate number of erroneous trades at the Open when the Price Bands are double-wide.^[73]

While the Participants present evidence that narrowing the Price Bands at the Open could be beneficial to the market, the Participants also present data analyzing the potential negative impact of narrowing Price Bands at the Open. This data shows that if double-wide Percentage Parameters are eliminated at the Open, the number of Limit State and Trading Pauses could quadruple in NMS Stocks and could triple in ETPs.^[74] The Participants argue that this projected increase in Limit States and Trading Pauses may not occur, however, because they and the Advisory Committee anticipate that market participants will quickly adapt systems to quote within the new, narrower Price Bands.^[75]

As noted above, commenters support narrowing the Price Bands at the Open. One commenter argues that narrowing the Price Bands should reduce volatility and not result in a significant increase in Limit States and Trading Pauses as market participants will adjust their quotes to be within the narrower Price Bands.^[76] Other commenters similarly argue that narrowing Price Bands at the Open would promote continuous trading by helping reduce the number of extraneous halts that occur shortly after 9:45 a.m., although these same commenters recognize that there could be an increase in the number of Trading Pauses between 9:30 a.m. and 9:45 a.m.^[77] These commenters also argue that band contraction at 9:45 a.m. has been shown to harm price discovery.^[78]

Calibration of the Price Bands requires the balancing of dual objectives: Preventing extraordinary volatility and facilitating price discovery. On one hand, if Price Bands are too wide, there is potential for extraordinary volatility resulting in trades at prices far away from a security's fundamental value, ultimately harming investors that are party to the trade. On the other hand, if Price Bands are too narrow, there is a potential for increased Trading Pauses that could impede price discovery for a security, also resulting in investor harm.^[79]

By proposing to narrow the Price Bands at the Open, the Participants (and the market participants commenting in favor of the proposal) believe a better balance can be achieved in favor of preventing extraordinary volatility that could result in erroneous trades at the Open. As the Participants demonstrate, the wider Price Bands currently employed have resulted in a number of trades that qualify as clearly erroneous executions under current SRO rules. Preventing trades that qualify as clearly erroneous executions protects investors that may have traded at bad prices. Preventing these trades also promotes better liquidity provision, as liquidity providers would be certain that executed trades will stand and that their hedging trades will not need to be unwound at potential losses.

The trade-off, however, is that there could be more Limit States and Trading Pauses during this most volatile period of the trading day,^[80] potentially impeding price discovery. Indeed, the Participants' historical analysis demonstrates that the number of Limit States could quadruple for NMS Stocks and triple for ETPs.^[81] The Participants believe, however, that the benefits of narrower Price Bands may be achieved without resulting in an increase in Limit States and Trading Pauses, arguing that their historical analysis is only theoretical and the number of Limit States and Trading Pauses overall will decrease at the Open because they expect that market participants will adjust their quoting behavior to narrower price bands.^[82]

In approving this proposal, the Commission recognizes the dual objectives served by the Price Bands. While the Commission acknowledges that narrowing the Price Bands during the most volatile period of the trading day ^[83] could potentially harm the price discovery process, the Commission recognizes the benefits of preventing extraordinary volatility discussed above,^[84] and believes that the amendment is an appropriate resolution regarding the balance of these dual objectives. The Commission also notes that no commenters opposed the proposed rebalancing of the dual objectives of preventing extraordinary volatility and facilitating price discovery.

In approving this proposal, the Commission notes that the Participants have committed to analyzing the performance of narrower Price Bands at the Open in a future Annual Report.^[85] The Commission looks forward to reviewing that analysis. The Commission notes that an analysis of anticipated adjustments to quoting behavior prior to implementation of the proposed changes would not have been practical. As part of their future analysis, the Commission is particularly interested in whether the data demonstrate a change in quoting behavior by market participants, as argued by the Participants and commenters, and if there is no change in quoting behavior, the extent to which Trading Pauses and Limit states negatively impact price discovery and whether the Participants continue to believe that the narrower Price Bands at the Open remain warranted.

B. Elimination of Double-Wide Percentage Parameters at the Close

Similar to the Percentage Parameters in place at the Open, between 3:35 p.m. and 4:00 p.m., or in the case of an early scheduled close, during the last 25 minutes of trading before the early scheduled close (“the Close”), the Percentage Parameters are doubled to accommodate increased volatility that may occur at the Close.^[86] The Participants are proposing to eliminate double-wide Percentage Parameters at the Close for Tier 2 NMS Stocks 2 with a Reference Price above $3.00.^[87] This would result in narrowing the Price Bands from 20% to 10% at the Close for these securities.^[88]

In particular, the Participants state that this proposed change is intended to dampen extreme price movements that may occur inside of the current Price Bands near the Close, noting that the current double-wide Percentage Parameters would accommodate price swings of as much as 40% when trading from the Upper Price Band to the Lower Price Band.^[89] The Participants state that the original concerns about volatility around the close were unfounded with respect to Tier 2 NMS Stocks.^[90] The Participants present data showing that only a de minimis number of trades actually occur outside of the regular 10% Percentage Parameter, and that therefore the doubling of the Percentage Parameters for Tier 2 NMS Stocks at the close is unwarranted.^[91] Further, the Participants present data that shows that the average number of Trading Pauses at the Close is nearly ten times lower than the average number of Trading Pauses for any other 25 minute period across the trading day.^[92]

However, the Participants acknowledge that if the double-wide Percentage Parameters at the Close were eliminated, the number of Trading Pauses would approximately triple based on their historical analysis, though the average number of Trading Pauses at the Close would still be lower than the average for any other 25 minute period across the trading day.^[93] Further, as with the proposal to eliminate double-wide Percentage Parameters at the Open, the Participants argue that this projected increase may not occur, because market participants may make behavioral changes to adjust to the new, narrower Price Bands, such that Trading Halts may not increase as projected.^[94]

The Participants state that there have been discussions around eliminating clearly erroneous rules when the Plan is in effect. They note that without the backstop of clearly erroneous rules, it is vital that the Price Bands are appropriately tailored to prevent trades that are so far from current market prices that they would be viewed as having been executed in error.^[95] The Participants state that permitting trading to occur within Price Bands that are as much as 20% above or below the Reference Price without the protections of the clearly erroneous rules would be detrimental to investors and the public interest.^[96]

Similar to the considerations around Price Bands at the Open noted above, the calibration of the Price Bands at the Close requires balancing dual objectives: preventing extraordinary volatility and facilitating price discovery. With respect to trading at the Close in particular, excessive Trading Pauses could impact the closing processes for securities in a manner that could harm price discovery at an important time of the trading day.^[97]

By proposing to narrow the Price Bands at the Close for Tier 2 NMS Stocks with a Reference Price above $3.00, the Participants (and the market participants commenting in favor of the proposal) believe a better balance can be achieved in favor of preventing extreme price movements and erroneous trades from occurring at the Close. Narrower bands, the Participants state, will prevent the potential for 40% price swings at the Close, which is consistent with the Plan's stated goal of preventing extraordinary volatility in NMS stocks.^[98] While their historical analysis shows that Trading Pauses could have tripled if narrower Price Bands as proposed were in place,^[99] the Participants argue that the number of Trading Pauses were de minimis and that the adjustment in market participant quoting behavior to the narrower price bands would result in even fewer Trading Pauses than the historical analysis demonstrated.^[100]

In approving this proposal to narrow the Price Bands at the Close for Tier 2 NMS Stocks with a Reference Price above $3.00, the Commission recognizes the dual objectives served by the Price Bands and believes that the Participants' proposal for narrower bands represents a different balance than that achieved by the current Plan. The Commission also notes that no commenters opposed the proposed rebalancing of the dual objectives of preventing extraordinary volatility and facilitating price discovery.

In approving this proposal, the Commission notes that the Participants have committed to analyzing the performance of narrower Price Bands at the Close in a future Annual Report. The Commission looks forward to reviewing that analysis. The Commission notes that an analysis of anticipated adjustments to quoting behavior prior to implementation of the proposed changes would not have been practical. As with the future analysis of the proposal concerning the narrower Price Bands at the Open, the Commission is particularly interested in whether the data demonstrate a change in quoting behavior by market participants, as argued by the Participants and commenters, and if there is no change in quoting behavior, whether the Participants continue to believe that the narrower Price Bands at the Close remain warranted. Furthermore, with respect to the analysis relating to the Close, the Commission is interested in an assessment of whether any increased Trading Pauses and Limit States negatively impacted closing auctions in affected securities.

For the reasons noted above, the Commission finds that the Eighteenth Amendment to the Plan is consistent with Section 11A of the Act ^[101] and Rule 608 thereunder.^[102]

IV. Conclusion

It is therefore ordered, pursuant to Section 11A of the Act ^[103] and Rule 608 thereunder,^[104] that the Eighteenth Amendment to the Plan (File No. 4-631) be, and it hereby is, approved.