پوششی بر روی محدودیت های فروش کوتاه مدت آشکار در سال 2008

On July 15, 2008, the US Securities and Exchange Commission announced temporary restrictions on naked short sales of the stocks of 19 financial firms. The restrictions offer a unique empirical setting to test Miller’s (1977) conjecture that short-sale constraints result in overpriced securities and low subsequent returns. Consistent with Miller’s overpricing hypothesis, we find evidence of a positive (negative) market reaction to the announcement (expiration) of the short-sale restrictions. Announcement returns are higher for firms that appear to be subject to more naked short selling in the days immediately preceding the announcement of the restrictions. The restrictions are successful in eliminating naked short sales for the restricted stocks, but naked short sales increase dramatically for a closely matched sample of financial firms during the restricted period. We also find that the restrictions negatively impact various measures of liquidity, including bid-ask spreads and trading volume. From a public policy perspective, our findings suggest that, at a minimum, policymakers should pause when considering further short sale restrictions.

In the third quarter of 2008, a wave of international markets introduced restrictions on short selling.2 With a financial crisis sweeping through the world’s capital markets, the restrictions were intended to protect firms from short sellers, who were singled out as exacerbating the quick and steep stock price declines many firms were experiencing. Although short sales have long been controversial, the actions of market regulators stand in contradiction to the general consensus in academic research that short-sale constraints reduce market efficiency, damage market quality, and result in security mispricing. For example, Miller (1977) argues that short-sale constraints can lead to overpriced securities and low subsequent returns, and Diamond and Verrecchia (1987) predict that short-sale constraints can damage market quality. We leverage the first of several recent short-sale restrictions to test the impact of short-sale constraints on stock prices and market quality in an empirical setting not available to prior studies. We find that the restrictions temporarily inflated the prices and negatively impacted the market quality of the stocks they were designed to protect. The recent wave of short-sale regulations began on July 15, 2008, when the US Securities and Exchange Commission (SEC) announced an Emergency Order restricting naked short sales of the stocks of 19 publicly traded financial firms. The Emergency Order was in effect from July 21 to August 12, 2008. On announcing the Emergency Order, the SEC pointed to “a substantial threat of sudden and excessive fluctuations of securities prices generally and disruption in the functioning of the securities markets that could threaten fair and orderly markets.”3 We exploit the Emergency Order as a natural experiment to study the impact of short-sale constraints on stock prices and market quality by comparing stock returns and market quality changes for the restricted firms with a matched sample of unrestricted firms. Miller’s (1977) overpricing hypothesis suggests that short-sale constraints dampen the incorporation of negative information into prices, resulting in overpriced securities. Lamont (2004, p. 1) notes, “[Short sale] constraints are difficult to measure, however, and researchers have struggled to find appropriate data to test the overpricing hypothesis.” Previous tests of the overpricing hypothesis have employed a variety of proxies for short-sale constraints, including short interest or the change in short interest (Figlewski, 1981, Desai et al., 2002 and Asquith et al., 2005), the introduction of options (Figlewski and Webb, 1993; Danielson and Sorescu, 2001), stock loan supply and fees (D’Avolio, 2002, Geczy et al., 2002 and Jones and Lamont, 2002), breadth of ownership (Chen et al., 2002, Asquith et al., 2005 and Nagel, 2005), and legal threats, investigations, or lawsuits (Lamont, 2004). An alternative approach, found in Charoenrook and Daouk (2005) and Bris et al. (2007), leverages cross-country differences in short-sale regulations to test the relation between short-sale restrictions and stock returns. Although tests of the overpricing hypothesis generally support the notion that short-sale constraints result in overpriced securities and low subsequent returns, each of these proxies previously used for short-sale constraints has limitations. For example, Chen et al. (2002) note that using short interest to proxy for short-sale constraints is problematic for at least two reasons. First, most stocks have little or no short interest outstanding at any given time, and, second, low short interest may reflect the high transaction costs of shorting and result in the underestimation of the overpricing magnitude. The period surrounding the Emergency Order offers a unique empirical setting for examining the overpricing hypothesis and the impact of short-sale constraints on market quality. Ours is the first study to our knowledge to benefit from a setting in which (1) the effective date of the short-sale constraints is known in advance with certainty, (2) the period in which short sales are constrained is bracketed by periods during which short sales are unconstrained, and (3) an industry-matched sample of firms not subject to the short-sale constraints and trading under a common market structure with the restricted sample is available. Also, unlike subsequent policy actions related to short sales, our results suggest that market participants did not anticipate the announcement of the Emergency Order. For these reasons, the Emergency Order represents the most powerful setting available to date for examining the impact of short-sale constraints on stock prices and market quality. At least two factors bias against finding a significant impact related to the Emergency Order. First, although the Emergency Order limited short selling, short sales were still permitted during the restricted period. Only naked short sales were restricted, and a number of entities received exceptions, including registered market makers, block positioners, and other market makers selling short as part of market making and hedging activities related to market making.4 Second, 8 of the 19 financial firms affected by the Emergency Order have international listings that are not subject to the oversight of the SEC. For many of these stocks, the naked short-sale constraints were less binding because the Emergency Order preceded the restrictive short-sale regulations subsequently enacted in many international markets. Our findings are three-fold. First, despite the possible bias against finding a significant impact, we report a positive market reaction to the announcement of the Emergency Order. Specifically, using Fama and French’s (1993) three-factor model to control for size and book-to-market factors known to influence stock returns, we find average cumulative abnormal returns (CARs) of 12.9% for the restricted sample in the four trading days between the announcement and implementation of the restrictions. However, over the 7 trading days ending 4 days after the expiration of the Emergency Order, the gains associated with the announcement period were nearly entirely reversed, with the restricted sample stocks experiencing a mean CAR of –10.9%. In fact, the mean CAR for the restricted sample from the announcement through 4 days after the expiration is only 0.7%, suggesting that the gains associated with the announcement were almost entirely reversed within a few days after the expiration of the Emergency Order. These results are consistent with Miller’s (1977) prediction that short-sale constraints can lead to overpriced securities and low subsequent returns. Second, using failures to deliver to proxy for naked short-sale activity, we report evidence consistent with a significant spike in naked short sales on the announcement date of the Emergency Order. Interestingly, this spike is more pronounced for a matched sample of financial firms not subject to the naked short-sale restrictions. During the restricted period, failures to deliver are nearly nonexistent for the restricted sample, yet remain at elevated levels for the matched sample when compared with the period prior to the announcement of the Emergency Order. Multivariate analysis suggests that the announcement period CARs are positively related to preannouncement naked short-sale activity for stocks subject to the restriction, whereas the expiration period CARs are unrelated to preannouncement naked short-sale activity for both the restricted and matched samples. Third, we find that market quality for the affected stocks deteriorated during the restricted period. Consistent with the predictions of Diamond and Verrecchia (1987), we find that the short-sale restrictions had a negative impact on the liquidity of the affected stocks as bid-ask spreads widened, trading volume decreased, and fewer trades were executed during the 17 trading day restricted period as compared with the 17 trading days prior to the announcement of the Emergency Order. Contrary to the notion that short-sale restrictions reduce volatility, we find no evidence that volatility decreased during the restricted period. In fact, consistent with the theoretical models of Abreu and Brunnermeier (2002) and Scheinkman and Xiong (2003), we find that daily return volatility increased during the restricted period, albeit not significantly more than for a matched sample of financial firms. Our results are consistent with Bris (2008) and Kolasinski et al. (2009), who report evidence that the Emergency Order negatively impacted market quality and efficiency for the stocks subject to the restrictions. From a public policy perspective, our findings suggest that restrictions on short selling damage market quality, resulting in overpriced securities, higher bid-ask spreads, and lower trading volume. Recently, the SEC indicated that it is considering reinstating the price tests on short sales that were eliminated in 2007.5 As policymakers continue to examine the rules and regulations governing short sales, this study provides valuable information that can improve their decision making in this area. The remainder of this paper is structured as follows. Section 2 details the SEC’s Emergency Order. Section 3 describes our sample. Section 4 reports our results on the impact of the Emergency Order on stock prices and market quality, and Section 5 summarizes and concludes.

On July15,2008,theSECissuedanEmergencyOrderrestrictingnakedshortsalesof the commonstockof19financialfirms.IntheEmergencyOrder,theSECstatedthatthe restrictionsare‘‘inthepublicinterestandfortheprotectionofinvestorstomaintainfair and orderlysecuritiesmarkets,andtopreventsubstantialdisruptioninthesecurities markets.’’19 The generalconsensusinacademicresearchisthatshort-saleconstraints negativelyimpactmarketqualityandleadtooverpricedsecurities.Weexploitthe Emergency Ordertostudytheeffectofshort-saleconstraintsonstockpricesandmarket qualitybycomparingabnormalreturnsandmarketqualityfortherestrictedfirmswitha matchedsampleofunrestrictedfirms.TheEmergencyOrderrepresentsthemostpowerful setting availabletodateforstudyingtheimpactofshort-saleconstraintsonstockprices and marketquality. Consistentwith Miller’s(1977) overpricinghypothesis,wefindthatabnormalreturnsat the announcement(expiration)oftheshort-salerestrictionsarepositive(negative)forthe affected stocks.Additionalanalysissuggeststhattheannouncementperiod returnsarepositivelyrelatedtopre-announcementnakedshort-saleactivityforthestocks named intheEmergencyOrder.Severalproxiesforliquidityindicatethatthe short-salerestrictionsnegativelyimpactedthemarketqualityofthestocksoftherestricted firms. Recent actionsbyregulatorssuggestthattheyarenotfinishedinthisarea.On September18,2008,theSECannouncedatemporaryorderprohibitingshortsellingofall financialstocksundertheirjurisdiction.Awaveofinternationalmarketsfollowed, includingAustralia,Belgium,Canada,France,Germany,Ireland,Italy,Luxembourg,the Netherlands,Portugal,Russia,Singapore,Switzerland,Taiwan,andtheUnitedKingdom. Giventhecurrentfluidityofpublicpolicyrelatedtoshortsales,itisvitalthatresearchers assist policymakersbyprovidinginformationandinsightthatallowsinformeddecision making.Ourresultssuggestthatrestrictionsonshortsaleshaveanegativeimpacton market qualityandthepricediscoveryprocess.Inlightoftheseresults,regulatorsmay want totreadlightly.