By Yan Cao, Allie Schwartz and Janko Cizel

Law360 (June 12, 2020, 5:29 PM EDT )

Yan Cao

Allie Schwartz

Janko Cizel

A number of securities class actions have been filed since the outbreak of the COVID-19 pandemic, and more are expected. These class actions will include as a part of the class period the current COVID-19 crisis, which has seen some of the most extreme swings in U.S. stock market prices in the past 20 years.

Thus far, most of these swings have been concentrated in March. Indeed, the CBOE Market Volatility Index — the VIX or "fear index" — which measures the 30-day forward-looking volatility of the S&P 500, experienced a spike in March as severe as the one during the height of the financial crisis.

Historically, extreme stock price movements have led to regulatory measures such as automatic trading halts and/or restrictions on short-selling activity on exchanges. The U.S. Securities and Exchange Commission first implemented a rule for triggering an automatic marketwide circuit breaker to curb excessive selling in periods of market stress, following the 1987 Black Monday stock market crash.

Before the COVID-19 pandemic, this marketwide circuit breaker was triggered only once, on Oct. 27, 1997. By contrast, in March 2020 alone, this circuit breaker was triggered on four separate days.

Within the past decade, the SEC has also implemented a number of additional mechanisms to automatically restrict certain trading and short-selling activity in specific stocks, and some of these restrictions are still enforced today.

Automatic trading halts and other trading restrictions were adopted in an attempt to mitigate excessive market volatility and to prevent potentially destabilizing short-selling activity. A number of academic studies have found that these restrictions can be effective in addressing some of the stated objectives, but may have the unintended consequence of potentially hampering price discovery and thus, affecting market efficiency.

This article explains how certain current trading restrictions work, documents their prevalence in the period of extreme market volatility in March, and discusses the potential implications of these restrictions for analyzing market efficiency in securities litigation.

Types of Automatic Trading Restrictions

Trading halts and other trading restrictions were put in place by regulators to curb certain trading and/or short-selling activity during periods of extreme stock price volatility.[1]

In the current regulatory regime, some trading restrictions can halt trading automatically on entire exchanges while others affect trading in individual securities. Trading restrictions also vary in duration and the type of affected activity (i.e., whether short selling is involved).

Described below are the marketwide and security-specific trading restrictions in place today.

Marketwide Circuit Breakers

Marketwide circuit breakers were adopted by the SEC following the crash of 1987.[2] There are currently three levels of marketwide circuit breakers. Level 1 is triggered by a 7% decline in the S&P 500 from the prior day's close; Level 2 is triggered by a 13% decline; and Level 3 is triggered by a 20% decline.

Level 1 and 2 triggers result in 15-minute halts in trading if they occur before 3:25 p.m. If the circuit breaker is triggered after 3:25 p.m., trading continues for the rest of the day unless a Level 3 halt is reached. Level 3 halts will stop trading for the remainder of the day, no matter the time.[3]

The Limit Up-Limit Down

The limit up-limit down, or LULD, rule was implemented to address extraordinary volatility in individual securities by preventing trades outside certain bounds.[4] The SEC approved the rules guiding the LULD rule in 2012 for all national market system, or NMS, securities, except options.[5]

The LULD mechanism replaced the single-stock circuit breaker mechanism[6] that was put in place in response to the flash crash of May 6, 2010.[7] Under the LULD mechanism, the market for a security enters a limit state if the national best bid hits the lower price band or the national best offer hits the upper price band.

Price bands are set at a certain percentage level above and below the reference price, which is based on the average transaction price over the last five minutes.

The LULD trading pause is triggered if a security does not exit the limit state within 15 seconds. The LULD trading pause is also triggered if a security enters a straddle state, which occurs when the national best bid is below the lower price band or the national best offer exceeds the upper price band.

Once triggered, the trading pause remains in place for five minutes, after which the market reopens with an auction on the primary listing market.[8] If the reopening auction cannot be priced within the applicable auction price collars due to continuing market imbalances, a trading pause may continue beyond the initial five minutes.[9]

The percentages applied to create the price bands differ depending on whether the security is Tier 1 or Tier 2.[10] Tier 1 NMS securities are those included in the S&P 500, Russell 1000, and select exchange-traded products. Tier 2 NMS securities are all other NMS securities, except options.[11]

Automatic Short-Sale Restrictions

Automatic short-sale restrictions were implemented by the SEC in early 2010 under Rule 201, also known as the alternative uptick rule.[12] The SEC implemented this rule in an attempt to "promote market stability and preserve investor confidence"[13] as well as to "address erosion of investor confidence in [its] markets generally."[14]

SEC Rule 201 restricts the price at which a security may be sold short when its price decreases by 10% or more from the previous day's closing price.[15] Specifically, once triggered, the rule prevents the execution or display of short-sale orders at a price that is below or equal to the current national best bid.[16]

If triggered, SEC Rule 201 short-sale restriction on the stock remains effective for the remainder of the day and the following day.[17] SEC Rule 201 applies to all NMS securities, except options.[18]

Triggers of Automatic Trading Restrictions During the COVID-19 Crisis

Due to the extraordinary level of market volatility during the COVID-19 crisis, specifically in March, Level 1 marketwide circuit breakers have been triggered four times: on March 9, March 12, March 16 and March 18.[19] To put this into context, as noted above, the last instance of a marketwide circuit breaker being triggered was on Oct. 27, 1997.[20]

In addition, since March 2020, there has been a large increase in the number of LULD trading pauses and SEC Rule 201 short-sale restrictions triggered. During March alone, over 28% of stocks listed on the New York Stock Exchange or the Nasdaq Stock Market were subject to security-specific LULD trading pauses, as compared to 1.4% of listed stocks subject to such restrictions in January.[21]

In March, almost 90% of stocks listed on NYSE or Nasdaq were subject to SEC Rule 201 short-sale restrictions, as compared to 16% of listed stocks subject to such restrictions in January.

Similarly, the number of stocks subject to LULD trading pauses or SEC Rule 201 short-sale restrictions on a typical trading day increased substantially from January to March. On a typical trading day in March, 5.1% of stocks listed on the NYSE or Nasdaq were subject to LULD trading pauses, as compared to 0.1% in January. On a typical trading day in March, more than 38% of stocks listed on the NYSE or Nasdaq, on average, were subject to SEC Rule 201 short-sale restrictions as compared to 2.9% on average in January.

Stocks subject to LULD trading pauses or SEC Rule 201 short-sale restrictions tended to be subject to such restrictions over longer periods of time in March as compared to January. Conditional on being subject to at least one LULD trading pause in a given month, a stock was subject to such restriction on four days on average in March, as compared to two days on average in January.

Conditional on being subject to at least one SEC Rule 201 short-sale restriction in a given month, a stock was subject to such restriction on 11 days on average in March, as compared to four days on average in January.

Trading Restrictions and Assessment of Market Efficiency

Under the U.S. Supreme Court ruling in Basic v. Levinson,[22] plaintiffs can invoke the fraud-on-the-market presumption of classwide reliance if they demonstrate that the stock traded in an efficient market.

In an efficient market, security prices fully and rapidly reflect public information.[23] The underlying rationale for why prices in an efficient market fully reflect all publicly available information is that competition among investors and the ability to trade on public information would quickly eliminate opportunities to profit on such information.

In the past, courts have ruled that the fraud-on-the-market presumption of reliance requires that the stock traded in an efficient market throughout the entire class period.[24] Defendants have challenged plaintiffs' presumption of market efficiency on the grounds of trading impediments that can affect market efficiency.

One notable example is IBEW Local 90 Pension Fund v. Deutsche Bank AG in the U.S. District Court for the Southern District of New York, where defendants successfully argued that plaintiffs failed to establish market efficiency for Deutsche Bank's Global Registered Shares partly by ignoring short-sale restrictions and other trading disruptions that affected the markets at the height of the financial crisis.[25]

The court denied class certification on the basis that plaintiffs failed to demonstrate that the market for the at-issue securities was efficient, noting that "an analysis of market efficiency that ignores … the fact that the Class Period encompasses an extraordinary financial crisis directly impacting trading conditions and the firm at issue, is fatally flawed."[26]

Similarly, in the case In re: PolyMedica Corp. Securities Litigation, on remand, the U.S. District Court for the District of Massachusetts denied class certification on the basis of plaintiffs' failure to demonstrate market efficiency and noted that "[defendant's] evidence suggests significant barriers to short selling, a mechanism which is … relevant to information efficiency."[27]

Given the extreme level of market distress observed in March, and the frequency of automatic trading restrictions being triggered, it is worth assessing whether similar issues could arise when experts try to establish market efficiency for stocks affected by these trading restrictions.

Specifically, for stocks subject to trading interruptions and constraints on short-sale activity, economic evidence may be needed to demonstrate that such restrictions did not prevent the stock price from quickly and fully impounding new value-relevant information. In particular, if trading interruptions or short-sale restrictions occurred during the class period, additional analysis may be needed to assess whether these restrictions affected price discovery for the purposes of establishing reliance.[28]

Academic research shows that trading restrictions can hamper price discovery and thus, potentially prevent the stock prices from quickly and fully impounding new value-relevant information.[29] For instance, studies show that trading halts may have an undesirable effect on market volatility and liquidity.[30]

While the occurrence of trading halts may create a chance for the market to change course from its extreme movement,[31] it may inadvertently cause traders to concentrate their trading prior to the expected triggering of the trading halt, causing additional price volatility and increasing bid-ask spreads, both in anticipation of and after the halt.[32]

This empirical observation is consistent with the notion of the magnet effect.[33] The academic literature has used this term to describe the mechanism where traders of the stock whose price is close to the circuit breaker limit may concentrate their trades prior to the expected triggering of the circuit breaker. This may in turn increase price variability and the probability of the price crossing the circuit breaker bound.[34]

Similarly, academic research has also examined constraints on short selling and whether these can potentially impact price discovery and market efficiency.[35] The 2008 financial crisis provided a particularly effective setting for examining short-sale constraints, because the SEC instituted a broad ban on short selling in stocks of financial companies that lasted from Sept.18, 2008, to Oct.8, 2008,[36] and European countries took more far-reaching actions.

A number of articles have empirically studied the effect of short-sale bans on publicly traded stocks, and have shown evidence of short-sale constraints leading to distortions in stock prices and preventing information from being fully and quickly reflected in stock prices, thus impacting market efficiency.[37] At the same time, academic research has also found evidence of short-selling behavior exacerbating downward price movements in periods of extreme market stress, showing a potential need for these measures.[38]

Finally, as discussed above, automatic restrictions on short-selling activity under SEC Rule 201 have become much more prevalent since the COVID-19 crisis began in March 2020. A recent academic study examining the impact of short-sale restrictions under SEC Rule 201 suggests that "short-sale restrictions harm the price discovery process and lead to a reduction in market efficiency."[39] Parties may need to consider this evidence in the context of market efficiency for the affected stocks.

Conclusion

Market volatility associated with the current COVID-19 crisis has triggered a large number of automatic trading restrictions, affecting trading behavior of stocks on U.S. exchanges. Academic literature examining trading restrictions has shown that, notwithstanding the benefits of these mechanisms, trading halts and short-sale constraints could potentially impede market efficiency. Parties may need to take this evidence into account when examining market efficiency for the stocks affected by these trading restrictions.

---

Yan Cao, Ph.D., is a vice president, Allie Schwartz, Ph.D., is a principal and Janko Cizel, Ph.D., is a manager at Cornerstone Research Inc.

The opinions expressed are those of the author(s) and do not necessarily reflect the views of the organization, or Portfolio Media Inc., or any of its or their respective affiliates. This article is for general information purposes and is not intended to be and should not be taken as legal advice.

[1] Note that in addition to automatic trading restrictions, the regulators and exchanges can impose certain discretionary trading restrictions. For example, the SEC has the authority to impose discretionary trading halts and short-sale restrictions. See, e.g., SEC Halts Short Selling of Financial Stocks to Protect Investors and Markets, U.S. Sec. & Exch. Comm'n, Sept. 18, 2008, https://www.sec.gov/news/press/2008/2008-211.htm; Investor Bulletin: Trading Suspensions – What Happens When They End?, U.S. Sec. & Exch. Comm'n, Mar. 4, 2008, sec.gov/oiea/investor-alerts-and-bulletins/ib_tradingsuspensions.

[2] Stock Market Crash of 1987, Fed. Res. Hist., Nov. 22, 2013, https://www.federalreservehistory.org/essays/stock_market_crash_of_1987. See also Michael A. Goldstein, Circuit Breakers, Trading Collars, and Volatility Transmission across Markets: Evidence from NYSE Rule 80A,50 Fin. Times 459 (2015).

[3] SEC Release No. 34-67090, May 31, 2012, at 4. See also SEC Release No. 34-85560, Apr. 9, 2019.

[4] SEC Release No. 34-67090, May 31, 2012, at 5-6.

[5] SEC Release No. 34-67090, May 31, 2012, at 2.

[6] SEC Release No. 34-67090, May 31, 2012, at 5-6. The single-stock circuit breaker (SSCB) mechanism that preceded the LULD trading pauses was triggered by trades occurring at or outside a 10% price band around a reference price during a five-minute period. While it was in effect, the SSCB was triggered on multiple occasions by erroneous trades or reporting errors. The SEC believed that the implementation of the LULD mechanism — which is triggered by movement in national best bids and offers as opposed to single trades under SSCB — would reduce the number of unnecessary trading halts (e.g., trading halts triggered by erroneous trades). SEC Release No. 34-67091, May 31, 2012, at 41-42.

[7] SEC Release No. 34-64735, June 23, 2011, at 3.

[8] Limit Up-Limit Down: Frequently Asked Questions, Nasdaq (2020), https://www.nasdaqtrader.com/content/MarketRegulation/LULD_FAQ.pdf (last visited May 12, 2020).

[9] Enhancement to Reopening Process after LULD Trading Pause, NYSE Trader Update (2017), https://www.nyse.com/publicdocs/nyse/markets/nyse/NYSE_Group_LULD_12_testing.pdf.

[10] Limit Up-Limit Down Plan: Overview, LULD Plan, http://www.luldplan.com/ (last visited May 12, 2020).

[11] Id.

[12] SEC Approves Short Selling Restrictions, U.S. Sec. & Exch. Comm'n, Feb. 24, 2010, https://www.sec.gov/news/press/2010/2010-26.htm.

[13] Id.

[14] SEC Release No. 34-61595, Feb. 26, 2010, at 2.

[15] EC Approves Short Selling Restrictions, U.S. Sec. & Exch. Comm'n, Feb. 24, 2010, https://www.sec.gov/news/press/2010/2010-26.htm; SEC Release No. 34-61595, Feb. 26, 2010, at 1.

[16] SEC Release No. 34-61595, Feb. 26, 2010, at 1.

[17] Id.

[18] Id. at 45-46.

[19] Count as of May 7, 2020. See Timeline of Events Related to the COVID-19 Pandemic, Fraser, https://fraser.stlouisfed.org/timeline/covid-19-pandemic (last visited May 7, 2020).

[20] On Oct. 27, 1997, the Dow Jones index fell by 554 points, triggering two marketwide trading halts that day. The Oct. 27, 1997, trading halts were the first marketwide trading halts since the adoption of marketwide circuit breaker rules after the October 1987 market crash. See Michael A. Goldstein, Circuit Breakers, Trading Collars, and Volatility Transmission across Markets: Evidence from NYSE Rule 80A,50 Fin. Times 459 (2015).

[21] Based on the information from Refinitiv Eikon, there are approximately 4,400 common stocks that are currently listed on either the NYSE or Nasdaq. Information on LULD trading halts and SEC Rule 201 short-sale restrictions covers stocks listed on Nasdaq or the NYSE, and is sourced from the respective exchanges.

[22] Basic, Inc. v. Levinson , 485 U.S. 224 (1988) ("Basic").

[23] Eugene F. Fama, Efficient Capital Markets: A Review of Theory and Empirical Work, 25 J. Fin. 383 (1970). The discussion here focuses on the definition of "semi-strong form" market efficiency, which is the most relevant in the securities litigation setting.

[24] See, e.g., In re PolyMedica Corp. Sec. Litig. , 432 F.3d 1 (C.A.1 (Mass.) 2005) ("PolyMedica"). In PolyMedica, the appellate court ruled that the district court erred in certifying the class for the sub-period for which defendants argued that PolyMedica's stock did not trade in an efficient market. On remand, the district court denied to certify the class holding that "[plaintiff expert's] weak showing regarding market efficiency has been sufficiently rebutted by PolyMedica. PolyMedica's evidence suggests significant barriers to short selling, a mechanism which is both relevant to information efficiency and essential to fundamental value efficiency. Also, PolyMedica has demonstrated that its stock was serially correlated. Such a condition is fundamentally incompatible with the [market efficiency] standard the First Circuit announced." In re PolyMedica Corp. Sec. Litig., 453 F.Supp.2d 260 (D. Mass. 2006) ("PolyMedica (on remand)").

[25] IBEW Local 90 Pension Fund v. Deutsche Bank AG , 2013 WL 5815472 (S.D.N.Y. 2013) ("Deutsche Bank").

[26] Deutsche Bank.

[27] PolyMedica (on remand).

[28] See Basic at 226 ("We must also determine whether a person who traded a corporation's shares on a securities exchange after the issuance of a materially misleading statement by the corporation may invoke a rebuttable presumption that, in trading, he relied on the integrity of the price set by the market.").

[29] See, e.g., Nikolaus Hautsch & Akos Horvath, How Effective Are Trading Pauses?, 131 J. Fin. & Econ. 378 (2019); Pedro A. C. Saffi & Kari Sigurdsson, Price Efficiency and Short Selling,24 Rev. Fin. Stud. 821 (2011); Ryan L. Davis et al., Short-Sale Restrictions and Price Clustering: Evidence from SEC Rule 201, 54 J. Fin. Services Res. 345 (2018).

[30] See, e.g., Avanidhar Subrahmanyam, Circuit Breakers and Market Volatility: A Theoretical Perspective, 49 J. Fin. 237 (1994).

[31] Nikolaus Hautsch & Akos Horvath, How Effective Are Trading Pauses?, 131 J. Fin. & Econ. 378 (2019).

[32] Id. See also G. William Schwert, Stock Market Volatility: Ten Years after the Crash(Nat'l Bureau of Econ. Research, Working Paper No. 6381, 1998).

[33] Nikolaus Hautsch & Akos Horvath, How Effective Are Trading Pauses?, 131 J. Fin. & Econ. 378 (2019).

[34] See, e.g., Avanidhar Subrahmanyam, Circuit Breakers and Market Volatility: A Theoretical Perspective, 49 J. Fin. 237 (1994).

[35] Pedro A. C. Saffi & Kari Sigurdsson, Price Efficiency and Short Selling,24 Rev. Fin. Stud. 821 (2011); Ekkehart Boehmer et al., Shackling Short Sellers: The 2008 Shorting Ban,26 Rev. Fin. Stud. 1363 (2013).

[36] Boehmer et al., Shackling Short Sellers: The 2008 Shorting Ban,26 Rev. Fin. Stud. 1363 (2013).

[37] See, e.g., Pedro A. C. Saffi & Kari Sigurdsson, Price Efficiency and Short Selling,24 Rev. Fin. Stud. 821 (2011); Ekkehart Boehmer et al., Shackling Short Sellers: The 2008 Shorting Ban,26 Rev. Fin. Stud. 1363 (2013); Arturo Bris et al., Efficiency and the Bear: Short Sales and Markets around the World, 62 J. Fin. 1029 (2007).

[38] See, e.g., Markus K. Brunnermeier & Martin Oehmke, Predatory Short Selling, 18 Rev. Fin. 2153 (2014).

[39] Ryan L. Davis et al., Short-Sale Restrictions and Price Clustering: Evidence from SEC Rule 201, 54 J. Fin. Services Res. 345 (2018).

For a reprint of this article, please contact reprints@law360.com.