U.S. Securities & Exchange Commission
SEC Seal
Home | Previous Page
U.S. Securities and Exchange Commission

Speech by SEC Commissioner:
Statement at Open Meeting and Dissent Regarding the Adoption of Amendments to Regulation SHO (the "Alternative Uptick Rule")


Commissioner Troy A. Paredes

U.S. Securities and Exchange Commission

Washington, D.C.
February 24, 2010

Thank you, Chairman Schapiro.

I want to join my colleagues in thanking the staff from the Division of Trading and Markets, the Division of Risk, Strategy, and Financial Innovation, and the Office of the General Counsel for their hard work throughout this rulemaking. I appreciate your professionalism and commitment as the Commission has considered whether to reinstitute a price test designed to restrict short selling.

Today, the Commission is adopting the “alternative uptick rule,” including a circuit breaker feature. When the price of a security has decreased by 10% or more from the prior day’s closing price, short selling in that security will be subject to the alternative uptick rule. In particular, when the 10% circuit breaker is triggered, short selling in that security will be prohibited at a price that is less than or equal to the current national best bid. The price test restriction will remain in place for the rest of the day on which the circuit breaker is triggered, as well as for the entire following day.

I am not able to support the rule amendments before the Commission and respectfully dissent. These remarks are not an exhaustive treatment of the rulemaking’s shortcomings but highlight a number of the key reasons I cannot support the rule change. My comments center around two related themes. First, the essential rationale behind the rule is that the short sale restriction, if implemented, will bolster investor confidence. This claim is rooted in conjecture and is too speculative to form a properly cognizable basis for adopting the alternative uptick rule. Indeed, a more thorough analysis indicates that the rule amendments are just as likely, if not more so, to erode investor confidence instead of boosting it. Second, there is an insubstantial empirical basis to support the Commission in adopting the rule, especially in light of the rigorous economic analysis that led the SEC to repeal the “original” uptick rule in 2007 after years of study. The Commission bears the burden to justify its rules. It has not done so in this instance.


Three observations provide the background for my subsequent remarks that more specifically target the short sale price test the Commission is adopting.

First, short selling contributes to the effective functioning of our securities markets. Much attention has focused on concerns with short selling. Among the concerns that have been expressed is the potential for short selling to be used in abusive or manipulative ways and the worry that selling pressure, particularly if spurred by fear and uncertainty, may contribute to the downward mispricing of securities and market instability.

However worrisome these or other short selling concerns may be in isolation, it is important to recognize that regulation already targets abuse and manipulation, as well as to emphasize the benefits of short selling to ensure that the tradeoffs of subjecting short selling to still more regulation are balanced appropriately. Short selling makes significant contributions to the effective operation of securities markets, benefitting all market participants and the economy overall. Short selling contributes positively to liquidity, capital formation, and the efficient allocation of risk. Short selling can facilitate buying by allowing investors going long to hedge their positions; and short selling can encourage market participation by leading to improved price discovery. Short selling helps ensure that securities prices are not systematically biased higher than the fundamentals warrant, as could be the case if prices did not reflect the less optimistic views of short sellers. Price discovery matters because investors would be less willing to invest if the contrarian views of short sellers were not fully incorporated into securities prices. Furthermore, when price discovery is compromised, we run the risk that our securities markets allocate capital inefficiently.

Because it is essential to the dynamics of price discovery and market efficiency, short selling ultimately bolsters investor confidence. Market prices are meaningful in that they aggregate into a single number available information and the myriad views of diverse market participants. Because of short selling, investors can be confident that securities prices reflect a range of perspectives — optimistic, pessimistic, and everything in between. If contrarian views are not fully incorporated into prices, investors are unable to rely on market prices as reflecting the overall assessment of the market as to what a company is “worth.” Consequently, a restriction on short selling that compromises market efficiency and the accurate pricing of securities risks eroding investor confidence.

Second, a demanding regulatory regime already governs short selling. Since 2008, we have taken a number of important steps to remedy abusive and manipulative short selling. These regulatory enhancements include the elimination of the options market maker exception from the Regulation SHO close-out requirement;1 Rule 10b-21, an antifraud rule that gives additional bite to Regulation SHO’s “locate” requirement and otherwise curtails fails to deliver;2 the hard close-out requirement of Rule 204, which superseded Interim Final Temporary Rule 204T;3 and regulatory efforts to ensure more robust public disclosure of short selling volume on a daily basis and individual short sale transactions on a delayed basis, along with more frequent disclosure of fails-to-deliver data.4

These regulatory protections targeting short selling are complemented by Rule 10b-5 under the Securities Exchange Act of 1934. As a general antifraud provision under the federal securities laws, Rule 10b-5 broadly regulates manipulation and deception, including a short seller’s spreading of false rumors in an attempt to drive down a security’s price. Fraudulent rumor mongering has long been prohibited.

Many have viewed “naked” short selling as a particular abuse for regulators to target while otherwise allowing non-naked short selling to occur. It is important to observe that an empirical study by the SEC’s Office of Economic Analysis has shown that fails to deliver have “decreased significantly” and that “[t]here has been a large downward trend in fails since July 2008.”5 These findings should allay concerns over naked short selling. Furthermore, it is worth clarifying that, whatever value one may find in a short sale price test, such as the alternative uptick rule, such a restriction on short selling is not intended to curb, nor does the Commission claim that it will curb, naked shorting. One should not blame the repeal of the uptick rule for naked short selling.

Third, the SEC’s decision in 2007 to repeal the uptick rule was grounded in the Commission’s best assessment of the available empirical research. Careful empirical analysis by SEC economists and outside academics had found that the short sale price tests then in effect did not effectively serve their intended purposes but did impose costs on markets.6 Before deciding to repeal the uptick rule, the SEC engaged in a multi-year review of the potential impact of doing so and conducted an extensive pilot program that allowed for detailed economic study of the effects of the rule’s repeal. The SEC had run a real-life experiment to ascertain the impact of the rule change. Rarely, if ever, does the agency have the benefit of such results before changing its rules.

Notwithstanding the market turmoil of 2008, it is not apparent that the economic findings that supported the SEC’s decision to repeal the uptick rule in 2007 are no longer valid. I am unaware, for example, of any empirically-demonstrated link between the turbulent market conditions of 2008 and the repeal of the uptick rule, and it is important to distinguish correlation from causation. The adopting release itself states: “[C]oncurrent with the subprime mortgage crises and credit crisis in 2007, U.S. markets experienced increased volatility and steep price declines, particularly in the stocks of certain financial issuers. We are not aware, however, of any empirical evidence that the elimination of short sale price test restrictions contributed to the increased volatility in the U.S. markets.” It is worth acknowledging a study by SEC economists, the findings of which indicate that selling pressure in September 2008 before the short sale ban went into effect, an especially tumultuous time for the markets, primarily came from “long sellers,” as compared to short sellers, and that short sellers tended to be contrarian during this period.7 It also is worth observing that financial sector stocks continued to fall during the September 2008 short sale ban, a more restrictive constraint on short selling than the alternative uptick rule the Commission is adopting.

With the benefit of this background — which on its own counsels against the adoption of the alternative uptick rule — I will now turn to some specific shortcomings with the Commission’s action today.

The Costs and Benefits of the Alternative Uptick Rule

The short sale restriction the Commission is adopting is costly. The adopting release discusses the potential costs associated with the rule, and the Commission received extensive comment throughout the rulemaking recommending against the agency’s various proposals for constraining short selling. At bottom, to the extent the alternative uptick rule, when triggered, limits short selling, there is a considerable risk that market quality will deteriorate as, for example, liquidity is adversely impacted, trading strategies are distorted, and markets become less efficient. Furthermore, as I already suggested, restrictions on short selling can frustrate capital formation by suppressing buying interest if price discovery is impeded and hedging becomes more costly. These results are inconsistent with the goal of bolstering investor confidence.

The adopting release points to the experience with former Rule 10a-1, the original uptick rule, in downplaying the potential costs of the alternative uptick rule in terms of its adverse impact on market quality. The release repeatedly notes that Rule 10a-1 did not meaningfully deteriorate market quality, deducing from this that neither will the alternative uptick rule. One explanation, however, is that Rule 10a-1 did not appreciably impact market quality because Rule 10a-1’s effect on short selling was not particularly restrictive. This creates a tension that the adopting release does not resolve. If the alternative uptick rule is restrictive enough to “prevent short selling from being used as a tool to exacerbate a declining market in a security,” as the adopting release claims, then one must account properly for the adverse market consequences that could attend when shorting is constrained. The price test, when triggered, applies to all short selling that does not benefit from one of the limited exceptions; the rule does not only constrain short selling that one might reasonably term abusive or manipulative. To the extent the price test keeps a security’s price from dropping, this may signal that the price is inefficiently inflated and not that the short selling restriction has deterred abusive or manipulative trading. On the other hand, if costly market impacts are unlikely to occur because the alternative uptick rule, in practice, will only modestly constrain short selling, then one has to question whether the alternative uptick rule will nonetheless be binding enough to prevent price declines caused by short selling.

In my view, the Commission has not met its burden of substantiating the claim that a short sale price test that will “prevent” short selling from exacerbating a security’s price decline will nonetheless not unduly impact liquidity, price discovery, and hedging activity. And if the alternative uptick rule, when triggered, does not appreciably prevent short selling from driving down the price of a security further, then the rule’s purported benefits will not materialize and there is no basis for its adoption.

Aside from its potential impact on markets, the new rule will impose substantial implementation and compliance costs on broker-dealers and trading centers. I am troubled that the Commission’s cost-benefit analysis may underestimate these costs. As it is, the Commission estimates that it will cost a total of nearly $29,000,000 for trading centers to establish policies and procedures pursuant to the rule; that the cost for all trading centers to ensure that their policies and procedures remain current will be approximately $7,700,000 annually; that trading centers, on the whole, will incur ongoing compliance costs of over $42,000,000 each year; that it will cost over $350,000,000 for all broker-dealers to establish appropriate policies and procedures under the broker-dealer and riskless principal provisions of the rule; that the yearly cost thereafter for broker-dealers to ensure that these policies and procedures are current will total over $96,000,000; that the aggregate ongoing annual cost for broker-dealers to ensure compliance with the policies and procedures will total over $530,000,000; that the aggregate cost imposed on broker-dealers to implement the “short exempt” marking requirements will total between approximately $595,000,000 and $750,000,000; and that the annual ongoing cost for all broker-dealers subject to the “short exempt” marking requirements will total over $480,000,000.

The regulatory question is whether the benefits of the alternative uptick rule justify these costs. The Commission has not demonstrated that they do. I want to build on what has already been stated by focusing on the foundation that underpins this rulemaking — namely, the claim that the rule will boost investor confidence. The adopting release provides, for example: “Although in recent months there has been an increase in stability in the securities markets, we remain concerned that excessive downward price pressure on individual securities accompanied by the fear of unconstrained short selling can undermine investor confidence in our markets generally.” The release also asserts that the short sale restriction will “help restore investor confidence during times of substantial uncertainty because, once the circuit breaker has been triggered for a particular security, . . . the security’s continued price decline will more likely be due to long selling and the underlying fundamentals of the issuer,” instead of other factors.

Investor confidence undergirds our securities markets. However, it is an infirm basis of support for this rule.

Investor confidence is too elusive of a concept to justify this rulemaking. Investor confidence is about investor psychology; and human psychology is notoriously difficult to predict and is not uniform across market actors. Is there an appreciable deficit of investor confidence today because short selling is not subject to a price test? There is no way to know. Accordingly, there is no way to know the extent to which the implementation of the alternative uptick rule will boost investor confidence when a circuit breaker is triggered. Given the centrality of investor confidence to this rulemaking, there is, thus, no reasonable basis for concluding that the claimed benefits justify the rule’s costs.

The adopting release defines investor confidence in a highly-stylized and confined way so as to facilitate its claim that the alternative uptick rule will shore up investor confidence. The release states in a footnote:

We note that investor confidence may include a number of different elements, such as investor perceptions about fundamental market risk, investor optimism about the economy, or investor trust in the fairness of financial markets as influenced by applicable regulatory protections. Although the latter can be directly influenced by Commission actions, the Commission does not have control over fundamental market risk and economic optimism. Thus, as used here, the term “investor confidence” refers to investor trust in the fairness of financial markets.

The Commission offers no objective measure for evaluating the extent to which regulatory protections promote “investor trust in the fairness of financial markets.” The self-fulfilling logic would seem to be that merely by regulating, the Commission can claim to shore up investor confidence, thus justifying its decision to regulate. This proves too much.

The definition suffers two other flaws. First, the release’s definition of investor confidence fails to recognize that investors are not monolithic. Different investors are likely to be confident to different degrees and for different reasons. Consider, for example, comments submitted in this rulemaking by institutional investors, as well as broker-dealers, trading centers, and other market participants, urging the Commission to refrain from instituting a new short sale price test on the grounds that new restrictions on short selling would do more harm than good to our markets. It is unclear from the adopting release how, if at all, the Commission has factored in the confidence of these investors and other market participants in claiming that adopting the price test will bolster investor confidence. Second, the definition of investor confidence is deficient in that it omits any meaningful consideration of the link between market quality and the willingness of investors to participate actively in securities markets. To the extent “regulatory protections” adversely impact liquidity, cause prices to become less efficient, or make it more costly to manage risk by hedging, investors do not benefit. It is hard to claim that investors, under these circumstances, would have faith in the integrity of the market.

Because of its indeterminacy, investor confidence, when relied upon as the predominant regulatory justification as it is in this rulemaking, affords the Commission too much discretion to regulate, unchecked by the discipline of more exacting and rigorous analysis. More is needed to justify regulating our securities markets than the unfalsifiable assertion that the alternative uptick rule will shore up investor confidence, particularly given the empirical data, including studies that this agency’s Office of Economic Analysis has conducted, that counsel against imposing a short sale price test.

Turning to look at the release’s investor confidence rationale from a more practical perspective, the claim that the alternative uptick rule will, in fact, bolster investor confidence is highly questionable, at best.

First, investor confidence is at risk of being undermined if the short selling restriction adversely impacts market quality, such as by interfering with price discovery. As I have discussed, I question the Commission’s conclusion that the price test can both “prevent short selling from being used as a tool to exacerbate a declining market in a security” yet only minimally deteriorate market quality. If markets become less efficient, investors should protect themselves against the prospect that securities are inaccurately priced. Among other things, investors may reassess the prices at which they are willing to buy to account for the risk that prices are not fully informed when the views of short sellers are not fully reflected in securities prices.

Second, it is unlikely that investor confidence will be bolstered if a steep price decline is experienced after the alternative uptick rule is triggered for a particular security. Investor confidence may erode if the price test fails to ensure that prices do not continue to drop. Investors, for example, may become even more concerned about the company than its fundamentals warrant if the price drop continues. In addition, investors may question the Commission’s decision to adopt the price test if the restriction is not seen as working. Investor confidence will be at risk if the alternative uptick rule does not prevent price declines as the Commission claims it will. For the Commission will have fostered a gap between the reality of what the price test can achieve and what investors expect.

Studies focusing on former Rule 10a-1 and its repeal suggest that the alternative uptick rule may not relieve downward pressure on prices to a measurable degree. Experience with the September 2008 short sale ban also is instructive. When the ban — a more restrictive limit on short selling than the alternative uptick rule — was in place, financial sector stocks continued to decline. Furthermore, to the extent long sellers primarily drive down prices,8 impeding short selling will have a limited impact on price levels.

If one nonetheless insists that the alternative uptick rule will, in practice, prevent short selling from driving down prices, then one has to appreciate that the cost to be paid for this result will likely be a deterioration of market quality and a concomitant erosion of investor confidence, as discussed above. Given this dynamic, it is difficult to see how the net effect on investor confidence would justify the price test.

Third, the adopting release states that investors will be more confident during a period of “substantial uncertainty” if they know that a price drop is more likely the product of long selling and the issuer’s fundamentals and not short selling or other factors. I am doubtful. It is at least as reasonable to conclude that investors will become increasingly uneasy if, instead of being able to blame a price decline on the lack of an uptick rule, investors focus more intently on the poor outlook for the company and perhaps the economy more generally. Although “investor perceptions about fundamental market risk” and “investor optimism about the economy” are excluded from the Commission’s chosen definition of investor confidence, they influence investor behavior and sentiment nonetheless and must be taken into account.

Finally, the adopting release argues that it is appropriate to prefer long sellers over short sellers in certain instances. In particular, the release explains that, when triggered, the alternative uptick rule “will allow long sellers, who will be able to sell at the bid, to sell first in a declining market for a particular security.” According to the release, it will “help restore investor confidence” to give long sellers “preferred access to bids” when a security’s price is falling. The adopting release further asserts that “placing long sellers ahead of short sellers in the execution queue under [such] circumstances . . . will help promote capital formation, since investors should be more willing to hold long positions if they know that they may have a preferred position over short sellers when they wish to sell.”

The adopting release implies that investors presently are discouraged from participating in the market if, as long sellers, they may have to sell after short sellers when a security’s price is dropping. The release does not substantiate the assertion that investor confidence is undermined or that investors otherwise are deterred from holding securities because short sellers may sell first. Moreover, other than asserting it is the case, the release offers no basis for the claim that allowing long sellers to sell at the bid when short sellers cannot will promote capital formation.

Indeed, it seems more likely that restricting short selling to allow long sellers to sell first will frustrate capital formation. Investors, for example, will be less likely to buy securities if short selling constraints increase the cost of hedging. In addition, if limits on short selling lead to artificially inflated prices because the contrarian views of short sellers are not fully integrated into the market, investors may be less willing to go long or at least will insist on paying less when buying. One has to consider the impact of a short sale restriction on the incentives and interests of buyers and not just those of sellers. Over time, capital formation is best promoted when investors can hedge their exposures and when investors can rely on the integrity of a security’s price as reflecting available information and the complete range of market perspectives.

The bottom line is this: When the costs and benefits are appropriately identified and balanced, the Commission is not justified in adopting the alternative uptick rule. It is especially important to underscore that the central rationale relied on to justify the rule — that the implementation of a new short sale price test will bolster investor confidence — is too speculative and unsubstantiated to serve as a proper basis for the Commission’s decision to adopt the short sale restriction. Moreover, one must recognize that, to a large degree, the available empirical data, including studies by SEC economists, do not support the decision to adopt the alternative uptick rule. One also should take notice of the extensive number of comments the Commission received advising the agency not to reinstitute a price test.

Administrative agencies, including the SEC, are built on independence and expertise.9 Independence allows administrative agencies the room they need to exercise their expert judgment. As the Commission has considered whether to subject short selling to a new price test, many have questioned the agency’s independence, asking whether the Commission was going to bend too far in response to external pressures to constrain short selling. Such questions will undoubtedly persist given that the agency is proceeding to adopt the alternative uptick rule notwithstanding the strong arguments against doing so.

Circuit Breaker

The alternative uptick rule goes into effect for a security when the security’s price falls 10% or more in a day from the prior day’s closing price. The price test remains in effect for the rest of the day on which the 10% circuit breaker is triggered plus the entire next day. The price test would not be re-triggered unless the security experienced another 10% intra-day drop.

A circuit breaker can mitigate the costs associated with restricting short selling because the restriction is not in place all the time. For this reason, the circuit breaker feature improves the rule before us as compared to a price test that would always be in effect. That said, the 10% threshold does not change the overall conclusion that the asserted benefits do not justify the rule’s costs.

The adopting release repeatedly relies on the circuit breaker feature in discounting — in a way that I believe is inappropriate — the risk that the alternative uptick rule will adversely impact market quality. For example, the release states that, because the circuit breaker is “narrowly-tailored” to respond to a sharp price decline, the extent to which the price test could erode the quality of our securities markets is “limited.” The release amplifies the point, indicating that short selling will not be restricted in those securities for which the circuit breaker has not been triggered. Based on data for the period April 9, 2001 to September 30, 2009, the adopting release estimates that, on an average day, the 10% circuit breaker would have been triggered for approximately 4% of the securities covered by the rule and approximately 6% of stocks would have been subject to the alternative uptick rule.

That the circuit breaker might mitigate how costly the rule is, however, does not mean that the price test is justified. The regulatory test is not whether the alternative uptick rule, when combined with the circuit breaker, is less costly than if there were no circuit breaker and the price test always restricted short selling. One cannot infer that the benefits of the rule before the Commission outweigh the costs because a more restrictive approach to regulating short selling would have been even more costly. Moreover, by emphasizing that the circuit breaker feature will allow short selling to occur unimpeded by a price test in the vast majority of instances, the adopting release obscures the key question — that is, whether the benefits outweigh the costs in those instances when the circuit breaker is triggered. The alternative uptick rule is security-specific, only applying to a security that has experienced a 10% intra-day price drop. Accordingly, the cost-benefit analysis should be security-specific as well. If the costs of the rule exceed the benefits when the 10% threshold has been met and the price test is binding, the recognition that short selling is not restricted in other securities for which the circuit breaker has not been triggered should be of no particular consequence in determining whether the price test, when triggered, is warranted.

I would like to make three additional points concerning the circuit breaker.

First, I disagree with the adopting release’s characterization of the 10% threshold as “narrowly-tailored.” Rather, I find the 10% threshold to be indiscriminate. A narrowly-tailored rule would, in my view, account for the extent to which a price decline was the product of the market’s assessment of the company’s fundamentals. To my knowledge, the Commission did not undertake any serious study to determine the extent to which 10% intra-day price declines historically are attributable to new information finding its way into the market and are not the product of abusive or manipulative short selling. In other words, a 10% daily price drop, while perhaps steep and rapid, might generally reflect the revaluing of a company by investors that improves market efficiency. In fact, the release cites empirical studies suggesting that short sellers tend to be contrarian and driven by their assessment of available information, reducing the likelihood that securities will be overpriced.

The Commission could have helped guard against the prospect that the circuit breaker will constrain the kind of short selling that promotes market efficiency by keying the circuit breaker threshold off a security’s opening price instead of the prior day’s closing price. This would have allowed after-hours news and events to be incorporated into a security’s price without the corresponding price drop counting toward a 10% decline that could trigger the short sale restriction.

Second, the circuit breaker, when triggered, has the distinct potential to bias investor psychology and judgment. By definition, the triggering of the circuit breaker signals that the Commission has determined that there is a particular problem that warrants additional regulation of the trading in a company’s securities. Thus, securities for which the circuit breaker has been triggered may experience a “stigmatizing effect,” as the adopting release notes. For example, because the price decline, once it reaches 10%, results in a unique consequence — a limit on short selling — untoward concern among investors may be stirred. Likewise, because it triggers more regulation, a price decline may become exceedingly salient in investors’ thinking, skewing their decision making. The consequences of such biased investor judgment are uncertain, but may include more selling pressure precisely because the circuit breaker was triggered, whether or not the fundamentals warrant it. In short, the mere fact that the circuit breaker was triggered is information that investors may find meaningful.

Third, although several commenters suggested that the Commission could adopt a circuit breaker approach, an important caution is in order so that one does not overstate the degree to which these commenters support the Commission’s action today. A careful reading of the comment letters indicates that, to a large extent, these commenters urged the Commission not to adopt any new short sale price test but agreed that, if the Commission were to restrict short selling, a circuit breaker would be preferable to a price test that was in effect all the time. It would be a stretch to claim that these commenters support the rule the Commission is adopting.


The rule provides for certain exceptions that limit the reach of the price test; specified short sale transactions are not subject to the alternative uptick rule, even when the circuit breaker is triggered. The exceptions allow short sales that do not raise the concerns the alternative uptick rule targets to occur unfettered by the price test.

I welcome the exceptions that are part of the rule, but I am troubled that other exceptions are not granted. Most notably, the Commission has decided not to adopt an exception for bona fide market making in either equities or options. And the exception afforded domestic and international arbitrage, as defined by the rule, is limited; the Commission has chosen not to except from the price test a broader set of bona fide hedging activities.

I disagree with the Commission’s decision to limit the exceptions in this way. Short sales that facilitate legitimate market making and hedging activities are not intended to drive down the price of a security. Thus, restricting such transactions is at odds with the goals of this rulemaking. In fact, a market maker exception and a more expansive hedging exception would, on balance, more likely advance the rulemaking’s objectives, not undercut them. The risk is that a short sale restriction that impedes market making will result in less liquidity, greater volatility, and wider bid-ask spreads, none of which is conducive to boosting investor confidence. Furthermore, a short sale restriction that makes it more costly for investors to manage their risk by hedging can hinder the ability of companies to raise capital. When it becomes more costly to hedge, investors can be discouraged from buying securities in the first place.


Having considered the core elements of the rule before us today, I want to turn to what may develop in the future. I am unsettled by the prospect that, over time, the short selling constraint will be made more restrictive.

The concern that I have expressed throughout — that the short sale restriction will deteriorate the quality of our securities markets and hinder capital formation — becomes more acute if the temptation is to limit short selling further when the price test adopted today does not achieve the desired outcome. Assume, for example, that the price at which a company trades continues to fall after the circuit breaker is triggered and that short sellers are blamed. One can imagine that there will be calls for a more binding price test if the alternative uptick rule that is implemented seems not to work and worry persists that investor confidence may erode. This is troubling, as a more restrictive price test would harm market quality even more. Indeed, with a relatively small ratcheting up of the increment above the national best bid that is required to execute a short sale, a short sale price test can, in effect, morph into a short sale ban. Recent experience teaches that short sale bans come at a high cost.

Whether or not there is pressure to force short selling to occur at a higher increment above the bid, there may be pressure to lower the 10% circuit breaker threshold that triggers the price test or to extend the price test’s duration additional days once it is triggered. The claim that investor confidence may be eroded when short selling is perceived as driving down the price of a security has motivated this rulemaking. In this view, why wouldn’t a single-day price drop of, say, 8% or even 4% warrant a restriction on short selling? If investors believe that short selling puts undue downward pressure on prices, why would they be willing to tolerate up to a 10% intra-day loss in value before the alternative uptick rule applies? Furthermore, is it going to bolster investor confidence to allow the price of a security to fall over a number of days — perhaps more than a total of 10% — without triggering the alternative uptick rule because the price never falls 10% or more in a single day? Similarly, will investors calling for a new short sale price test become uneasy if, after the circuit breaker is triggered for a security and the price test is lifted, the price falls appreciably? According to the adopting release’s logic, the fact that the circuit breaker has been triggered suggests that there is a problem that warrants a limit on short selling. Investors may wonder why they have lost the purported protection of the alternative uptick rule until the price again falls 10% in one day.

In evaluating the current rule change before the Commission, one has to consider that the reasoning behind the rule presents the prospect of a more restrictive and costly constraint on short selling in the future.


The lynchpin of the rule is the claim that it will bolster investor confidence. This claim rests on an insubstantial basis. The claim that the rule will promote investor confidence is rooted in conjecture, and there are strong reasons to conclude that implementing the alternative uptick rule will, in fact, erode investor confidence by, for example, reducing liquidity and impeding price discovery. To be sure, if the new price test proves not to be particularly restrictive, presumably the alternative uptick rule will not deteriorate market quality to a measureable degree when triggered. On the other hand, if the price test is not particularly restrictive in practice, then the Commission is on infirm ground in claiming that the rule will “prevent” short selling from driving down a security’s price. In this case, the Commission, by claiming more than the alternative uptick rule can deliver, will have fostered expectations among investors that the price test cannot meet. This would not boost investor confidence with regard to a particular security, our securities markets generally, or this agency.

One also has to take notice of the economic studies, including those that SEC economists have conducted, that counsel against adopting the alternative uptick rule, even with the circuit breaker feature. The empirical findings suggest that the price test may do relatively little to prevent the price of a security from falling and that, to the extent short selling is restricted, the risk is that market quality and capital formation will be adversely impacted. The adopting release is too dismissive of this body of economic work. Although these studies do not specifically examine the impact of the alternative uptick rule, with or without a circuit breaker, these studies nonetheless are instructive in that, at a minimum, they demonstrate the qualitative consequences of efforts to constrain short selling.

Furthermore, the adopting release does not succeed in offering any persuasive data that make the affirmative case in support of the alternative uptick rule. At the very most, the Commission could claim that the available data are inconclusive. But in the context of this rulemaking, inconclusive data are an insufficient basis upon which to adopt a new price test.

It is troubling to consider the lack of empirical data that supports the Commission’s action today as compared to the rigorous empirical analysis that was undertaken to test the impact of repealing the original uptick rule before the Commission did so in 2007. Instead of adopting the short sale price test with a circuit breaker as a final rule, an alternative approach would have been to proceed on a pilot basis. The Commission could have instructed its new Division of Risk, Strategy, and Financial Innovation to fashion a serious pilot that would have allowed the SEC to undertake a meaningful empirical study of the impact of the rule before deciding to adopt it. The pilot study could have matched the analytical rigor of the pilot study the SEC conducted before committing to repeal former Rule 10a-1. Admittedly, in a pilot, some securities that experienced a 10% price drop would not be subject to the alternative uptick rule; otherwise, there would be no control group. And it would take time to generate data that could be tested empirically. However, if the findings of the pilot study were to warrant the price test, then there would be a proper, well-grounded basis for further restricting short selling and the agency could decide to do so then. If the empirical findings did not warrant the price test, then the Commission, by choosing not to adopt the short sale restriction, could avert the substantial risk that the agency is assuming today — namely, that the alternative uptick rule will do more harm than good. Moreover, a pilot could have reinforced confidence in the Commission as an independent agency that brought its expert judgment to bear in addressing concerns regarding our securities markets.

I would like to conclude how I began — by thanking the staff for their dedicated efforts throughout this rulemaking.

1 See http://www.sec.gov/rules/final/2008/34-58775.pdf.

2 See http://www.sec.gov/rules/final/2008/34-58774.pdf.

3 See http://www.sec.gov/rules/final/2009/34-60388.pdf.

4 See http://www.sec.gov/news/press/2009/2009-172.htm.

5 See http://www.sec.gov/spotlight/shortsales/oeamemo110409.pdf.

6 See generally http://www.sec.gov/news/studies/2007/regshopilot020607.pdf.

7 See http://www.sec.gov/comments/s7-08-09/s70809-369.pdf.

8 See id.

9 See generally Stephen Breyer, Breaking the Vicious Circle: Toward Effective Risk Regulation (1993); James M. Landis, The Administrative Process (1938).



Modified: 02/24/2010