ایا همه سرمایه گذاران فردی برابر هستند؟ مدارک و شواهد از سرمایه گذار تجاری فردی در سراسر رویدادهای شکایت های قانونی اوراق بهادار

This study examines the trading behavior of individual (retail) investors around securities litigation disclosure events. We hypothesize and find that less informed investors (low-income investors and investors in non-professional occupations) purchase earlier in the class period relative to more informed investors who tend to sell later in the class period. We also find that more informed investors achieve higher returns, or at least incur fewer losses, than less informed investors during the class period. Less informed investors are, therefore, more likely to suffer greater class action damages than more informed investors from their presumed reliance on allegedly false information.

This study examines the trading of individual investors in response to disclosures about securities class action litigation. We test hypotheses about the purchase and sale of stocks by individual (retail) investors around the start and end of the litigation class period.1 In particular, we examine how different categories of individual investors trade when allegedly false information is introduced into the market and, later, corrected as a curative disclosure and, more specifically, whether some investors (e.g., more informed investors) are able to spot false information earlier than others (e.g., less informed investors), thereby potentially cutting their losses prior to the curative disclosure. This topic is important for at least two reasons. First, we build upon a developing literature on variation in stock market response to litigation events across different categories of investors. This literature, thus far, has focused mostly on the responses of professionals such as securities analysts, short sellers and corporate insiders to litigation and related events. One result is that some groups (e.g., short sellers, certain institutional investors) are more astute than others (e.g., financial analysts) in predicting the accounting conditions that might lead to a restatement or an allegation of securities fraud (section 2 reviews the literature). Differences in incentives across these groups with respect to adverse information, however, may evince this result. This provides a second reason for our study, in that differences in incentives should have less influence within a retail investor sample than across different professional groups. By examining the trading responses of individuals to litigation events, we are able to ascertain whether some results that hold for professional investors also hold for individual investors. We focus on a single class of disclosures, litigation disclosures and not disclosures in general, because litigation disclosures typically have reasonably similar and, mostly, significant adverse consequences for the named company. These can reflect claims of substantial out-of-pocket damages by investors who have allegedly suffered losses (e.g., In re Cendant, 2000). Yet, such consequences can be difficult to predict ex ante. 2 A litigation disclosure setting, in our view, increases the chances, relative to disclosures in general, that certain categories of individual investors, such as more informed or less informed investors, might respond differently. We focus on litigation disclosures, also, because by their nature they are infrequent events for a company not subject to regular forecasting or updating by analysts and managers as is the case with earnings. Our study also pertains to the behavioral notion that some investors realize their losses reluctantly (the disposition effect) (Odean, 1998; Feng and Seasholes, 2005; Dhar and Zhu, 2006). Differences in the response of individual investors to the correction of heretofore allegedly false positive information may offer further clues about this behavioral effect. Based on the class action companies in the same data set as used by Barber and Odean, 2000 and Barber and Odean, 2001, Dhar and Zhu (2006) and others3, we find that not only do less informed investors (proxied by low-income investors and those in non-professional occupations) tend to maintain their stock holdings through the end of the class period, consistent with delayed recognition of losses, but also relative to more informed investors, less informed investors tend to buy earlier in the class period at which time the class action companies disclose allegedly false positive information to the market. Conversely, we find that more informed investors tend to sell later in the class period but before the end of the period thereby, presumably, cutting their losses and/or profiting from the earlier price run up which may be, in part, a response to the false information.4 Table 2. Sample Characteristics for Discount Brokerage and Securities Class Action Samples Sample Characteristic/Yeara 1991 1992 1993 1994 1995 1996 Panel A: Discount brokerage sample b Number of buy trades 184,358 174,466 168,790 141,570 167,134 179,417 Number of sell trades 133,845 132,541 146,768 120,209 151,427 154,251 Average trade size (buys) $12,941 $13,166 $13,114 $12,698 $13,004 $12,815 Average trade size (sells) $13,147 $12,927 $12,953 $12,977 $13,043 $12,853 Total number of stocks traded 5,930 6,010 6,514 6,804 7,206 7,096 Total number of investors 42,109 41,222 39,712 35,850 35,800 34,262 Panel B: Securities class action sample Number of buy trades 29,607 32,129 31,570 28,264 44,557 50,394 Number of sell trades 23,204 24,685 27,500 24,597 37,564 41,114 Average trade size (buys) $11,522 $12,510 $13,136 $13,367 $16,183 $15,500 Average trade size (sells) $14,947 $16,120 $15,693 $16,056 $19,683 $19,674 Total number of stocks traded 591 621 569 604 645 658 Total number of investors 16,500 17,972 17,443 16,021 18,671 18,938 a Year of the date of the end of the class period. b Same as Kumar and Lee (2006), Table 1, panel A. Table options We examine further these findings by testing whether more informed investors do indeed make higher returns, and/or suffer fewer losses, than less informed investors during the class period. Based on percentage returns, we find that the returns on purchase trades executed by more informed investors are significantly higher than those by less informed investors. Consistent with the disposition effect, the difference is particularly strong for outstanding purchase trades that are not liquidated before the end of the class period. But a positive difference also exists for round-trip trades before the end of the class period where the disposition effect is less important. This result supports the view that it is the less informed rather than the more informed who suffer higher damages in class action securities litigation. The remainder of the study is organized as follows. Section 2 summarizes the literature relating to investor behavior around securities class actions, and develops and states the principal hypotheses. Section 3 describes the data and sample used to test the hypotheses. The findings are discussed in section 4. Section 5 summarizes the results and presents the major conclusions.

In contrast to an extensive literature that has evolved in the past several decades on the response of the stock market and professional investors to financial news events (e.g., earnings announcements, dividends, restatements), comparatively little has been published on trading by individual investors in response to these and similar events. The unavailability of appropriate data is one factor that may have hampered these efforts. We use a unique data set of the trades by some 78,000 investors at a major discount brokerage company in 1991–1996 to study the buying and selling behavior of individual (retail) investors around events associated with securities class action litigation. Litigation events are an appropriate focus for study in that they typically have swift and dramatic impacts on market prices and trading volume. They are also irregular events not subject to periodic forecasts and updates as in the case of earnings so that some investors may anticipate or understand them better than others. This is less likely to be the case with earnings or other similar regular announcements. This study examines the buys and sells of individual investors at the start, end and during the litigation class period. Our merged data set allows us to examine trading for a substantial fraction of all federal securities class actions filed during 1991–1996. Given the irregular nature of litigation disclosures, we test the primary notion that individual investor trading in the class period varies on the basis of proxies of investor informedness. Our results support this notion in that we find that more informed investors (relative to less informed investors) tend to be net sellers close to the end of the class period. Conversely, less informed investors (relative to more informed investors) tend to be net buyers following the start of the class period, at which time the company allegedly first conveys false positive information to the market. These results are robust to two proxies of informedness, namely, investor income and occupation. They are also robust to company size, growth and possible trading in response to extreme good news or bad news earnings announcements. These findings imply that it is the less informed investor who is more likely “fooled” by allegedly false information. This is the group that, apparently, tends to buy earlier in the class period at the inflated price and tends to sell after the end of the class period following the price adjustment. We examine this implication by testing whether the returns to more informed investors in the class period exceed the returns to less informed investors. Our tests support this view. In fact, we find that the returns for purchase trades executed by more informed investors exceed those for purchase trades executed by less informed investors by more than 100 basis points. The difference is particularly strong for outstanding purchase trades not liquidated before the end of the class period. A further testable implication of this result is that it is the less informed investors rather than the more informed investors who suffer greater damages in class action securities litigation. The less informed investors buy earlier in class period and sell later, often after the corrective disclosure. The actual dollar shareholder loss suffered by an investor (more informed or less informed) from trading at an alleged inflated price that is subsequently corrected as a curative disclosure, however, is a more complex calculation than the dollar or percentage return we study in this paper, although it should be positively correlated with such return measures. This topic is best left for future research.