2006]MARKET EFFICIENCY1
Market Efficiency, Crashes, and
Securities Litigation
Bradford Cornell[*]
James C. Rutten[†]
In BasicInc. v. Levinson the United States Supreme Court effectively affirmed the efficient market hypothesis by ruling that a plaintiff who purchased securities on an open and developed market can be presumed to have relied on the integrity of the market price and in that way to have relied indirectly on false public statements allegedly affecting that price. Although the Court confined its analysis to the question of reliance and explicitly avoided any question of damages, Justice White worried in his dissent that one could not be separated from the other. In this Article, we argue that Justice White’s concerns were well founded. In light of theoretical and empirical research in finance, we show that the failure to understand the differing implications of the efficient market hypothesis for proving reliance and assessing damages introduces a significant plaintiff’s bias in securities class action litigation. Furthermore, although Congress attempted to address this bias in passing the Private Securities Litigation Reform Act of 1995, the Act is likely to be ineffective in this regard.
I.Introduction
II.Assessing Market Efficiency for Reliance Purposes
III.Reliance Versus Damages: Ex Ante and Ex Post
IV.Crashes and News: Why Markets Can Overreact
V.The Failure of the PSLRA To Address the Problem of Crashes
VI.Conclusion
I.Introduction
In Basic Inc. v. Levinson, the United States Supreme Court made it easier for plaintiffs alleging securities fraud under section 10(b) of the Securities Exchange Act of 1934 and Securities and Exchange Commission (SEC) Rule 10b-5 to prove the essential element of reliance.[1] The Court held that under the so-called fraud-on-the-market theory, a plaintiff who purchased securities on an “open and developed” market can be presumed to have relied on the integrity of the market price and in that way to have relied, indirectly, on allegedly false or misleading public statements of the defendants.[2] The Court explained:
“The fraud on the market theory is based on the hypothesis that, in an open and developed securities market, the price of a company’s stock is determined by the available material information regarding the company and its business.... Misleading statements will therefore defraud purchasers of stock even if the purchasers do not directly rely on the misstatements.... The causal connection between the defendants’ fraud and the plaintiffs’ purchase of stock in such a case is no less significant than in a case of direct reliance on misrepresentations.”
....
... [W]here materially misleading statements have been disseminated into an impersonal, well-developed market for securities, the reliance of individual plaintiffs on the integrity of the market price may be presumed.[3]
Although the Court confined its analysis to the question of reliance and explicitly avoided any question of damages,JusticeWhite worried in his dissent that one could not be separated from the other.[4] He recognized that if, in the Court’s words, “‘The market is acting as the unpaid agent of the investor, informing him that given all the information available to it, the value of the stock is worth the market price,’” then damages and reliance are bound together.[5] Justice White recognized that if market prices incorporate all public information and thereby reflect true value, then when a corrective disclosure occurs, the ensuing decline in the sock price will reveal the true value of the company, and the entire amount of the decline will be the plaintiff’s damages (all else being equal).[6] Justice White concluded: “[A]nswers to the question of the proper measure of damages in a fraud-on-the-market case are essential for proper implementation of the fraud-on-the-market presumption.”[7]
In the years since Basic, Justice White’s words have been borne out, for securities plaintiffs have come to use the efficient market hypothesis not only as a means of proving reliance, but also as the theoretical pillar supporting their damages estimates.[8] Plaintiffs, in calculating damages, routinely spurn discounted cash flow analyses and other fundamental techniques for valuing securities in favor of simply analyzing stock price movements. Their approach finds substantial support in the finance literature, which frequently argues that calculating the value of securities based on traditional cash flow or earnings analyses is less reliable than calculating value based on asset-pricing models. As one author of this Article, Cornell, and R. Gregory Morgan opined: “[C]alculating the value [of a security] by analyzing asset value, earnings data, and other similar information is inherently speculative. The market model approach attempts to place the value [calculation] on firmer ground by calculating ... value ... on the basis of security price data.”[9]
There are, to be sure, numerous technical issues that arise when attempting to infer damages from stock price movements. As described originally by Cornell and Morgan and confirmed recently by the Supreme Court, these include netting out movements attributable to market and industry factors, nonfraud related disclosures, and information leaks.[10] Such issues have been addressed by the courts in numerous cases.[11] Despite these complexities, however, courts have viewed inferring damages from stock price movements as inherently more reliable than applying more traditional valuation approaches of the type Daniel Fischel criticizes.[12] Indeed, notwithstanding the difficulties that inhere in measuring damages from stockprice movements, few if any courts since Basic have rejected the efficient market hypothesis and analysis of stock price movements in favor of more traditional valuation tools.[13]
Given that market efficiency has come to play such a central role in both proving reliance and estimating damages, the manner in which courts assess efficiency becomes paramount. Assessing market efficiency must be an exercise in approximation from an economic point of view, because a paradox identified originally by Sanford Grossman and Joseph Stiglitz demonstrates that a market can never be truly efficient: If a market is perfectly efficient such that prices reflect all public information, including a company’s prospects, there would be no incentive for any investor to analyze public information when making investment decisions, but if investors do not analyze public information, there would be no mechanism by which such information would come to be reflected in the market prices, and the market would not be efficient.[14] It follows, therefore, that even the most efficient markets, such as those for stocks traded heavily on the New York Stock Exchange (NYSE), must be sufficiently inefficient so that at the margin sophisticated investors have an incentive to analyze public information.[15] As Jonathan Macey and Geoffrey Millerput it, “[M]arket professionals will only search out and analyze new information about a firm [when] an additional $1.00 expenditure on searching and analyzing is expected to produce [at least] an additional $1.00 in trading profits.”[16]
Just as the idea of a perfectly efficient market makes no economic sense, the notion of a completely inefficient market, in which information is not disseminated at all and price bears no relation to value, likewise makes no sense. Even in the market for used cars, information is disseminated and price is related to value in the sense that people know that a one-year-oldPorsche 911 is worth more than a ten-year-old Ford pickup, and the two vehicles are priced accordingly. In fact, the basic premise of free market economics, for which there is immense empirical support, is that price reflects the interplay of supply and demand. If this is true, then no securities market will be completely “inefficient” in the sense that price floats freely unrelated to supply and demand as influenced by information about company fundamentals.
The foregoing demonstrates that efficiency is not an on-off switch, but rather exists along a continuum. The relevant legal question therefore becomes: How efficient must a market be for a court to conclude that it is “efficient” for purposes of showing reliance or estimating damages? In this regard, the Court in Basic provided only vague guidance, speaking of “open and developed” markets and leaving to later decisions to hash out what “open and developed” means and to figure out how to assess efficiency.[17] Moreover, the Court did not distinguish between efficiency in the reliance context and efficiency in the damages context.[18]
Against this backdrop, this Article puts forth two arguments. First, in the reliance context, we argue that the standard for efficiency should be relatively low and tied to the fundamental reliance question. This question is whether but for the defendant’s statements the plaintiff would have purchased the security at the price at which the plaintiff did purchase it.[19] In the reliance context, we argue that a market should be deemed “efficient” if an investor would be justified in relying on the integrity of the market price,i.e., if an investor knowing full well that the market cannot be perfectly efficient and that, accordingly, the market price may not reflect the true value of the stock, nevertheless would be justified in treating the market price as if it reflected true value. Reliance on the integrity of the market price invariably will result in reliance on a defendant’s statements because, as noted in the used car example above, even grossly inefficient securities markets incorporate information to somedegree such that, but for the fraud, the plaintiff would not have purchased the stock at the price at which the plaintiff did purchase it. The precise extent to which the price was affected by the fraud is irrelevant to the reliance question. Accordingly, in the reliance context, the important question is when itis appropriate to presume that the plaintiff relied on the integrity of the market price, and thus to presume that the plaintiff relied indirectly on the defendant’s statements.[20] Reliance on the integrity of the market price is sensibly presumed, we argue, if the market bears enough hallmarks of efficiency that investors, mindful of the costs they would incur if they went out and conducted their own research into stock values, reasonably could decide instead to treat the market’s prices as indicative of fair value.
Second, in the damages context, we argue for a much stricter standard for efficiency that is again tied to the fundamental issue. This issue, unlike in the reliance context, is the extent to which the alleged fraud affected the stock price. Damages cannot accurately be measured by reference to the decline in the stock price unless the market is perfectly efficient such that it reacts perfectly to fraudulent statements and the later revelation of the true facts. Yet as Grossman and Stiglitz demonstrate, perfect efficiency is not possible.[21] Although damages in securities fraud cases, as in other types of cases, need not be measured accuratelyandonly approximated,[22] even approximating damages by reference to the decline in the stock price would require the market to approximate perfect efficiency because even minor inefficiencies are magnified significantly by selection bias.[23] Unfortunately, courts have failed to appreciate the major impact that selection bias can have on estimates of damages even when deviations from market efficiency are relatively infrequent.
As we show below in discussing the markets for the securities of Intel Corporation (Intel) and Bristol-Myers Squibb Company (Bristol-Myers), it is highly doubtful that even the most open and developed markets so much as approximate perfect efficiency. Accordingly, we argue that while it is not inappropriate to analyze stockprice movements when assessing damages, even in the most efficient of markets, that analysis must be tempered with fundamental valuation techniques of the sort that have fallen out of favor in securities litigation. We show below that the courts’ tendency uncritically to apply the efficient market hypothesis when estimating damages and to import notions of efficiency appropriate for the reliance context into the very different context of damages has produced a pronounced bias in favor of plaintiffs. In short, we assert that Justice White’s concerns were well founded: Basic failed to treat the concept of efficiency with sufficient clarity and in so doing, blurred the implications of efficiency for reliance and for damages, creating a situation in which plaintiffs have been able to recover for price declines not proximately caused by fraud.
To develop our arguments, this Article proceeds as follows. Part II discusses judicial efforts since Basic to assess market efficiency. The pivotal cases in this regard are Cammer v. Bloom[24] and Krogman v. Sterritt.[25] Although Cammer and Krogman provide appropriate criteria for assessing efficiency, they do so only in the context of proving reliance. The Cammer and Krogman criteria have significantly different implications when assessing efficiency in the context of measuring damages.
Part III demonstrates that even the most efficient markets can behave very inefficiently, producing random crashes, i.e., price declines upon negative news that are grossly disproportionate to what the news reveals about company fundamentals. Part III demonstrates that uncritically applying efficient market theory in the damages context permits plaintiffs to use ex post selection bias to take unfair advantage of these crashes and to recover sums greatly in excess of actual damages.
Part IV examines recent financial research on market efficiency and explains some of the reasons that such crashes can occur. Part IV further demonstrates that markets should not be considered efficient when estimating damages, lest such estimates greatly overstate actual damages.
Part V explains that Congress attempted to address the problems crashes pose for damages estimates in the Private Securities Litigation Reform Act of 1995 (PSLRA),[26] but that it did so ineffectually.
Part VI summarizes our conclusions.
II.Assessing Market Efficiency for Reliance Purposes
Cammertried to resolve some of the questions left open in Basic and in so doing, became the seminal decision on market efficiency in the reliance context.[27] In Cammer, some of the defendants moved to dismiss the section10(b) complaint on reliance grounds.[28] The Cammerdefendants contended that the plaintiffs should not get the benefit of the fraud-on-the-market presumption because the company’s stock had not traded in an efficient market, in that it had not been listed on a national exchange but instead had traded over the counter.[29] The Cammer court eschewed such a bright-line rule and held that the critical inquiry was not where the securities were traded, but whether the market for the particular security was “open and developed” (and therefore, in the court’s view, inferentially efficient), as demonstrated by five criteria.[30]
First, the court held that “an average weekly trading volume ... in excess of a certain number of shares” would suggest efficiency.[31] The court explained:
The reason the existence of an actively traded market, as evidenced by a large weekly volume of stock trades, suggests there is an efficient market is because it implies significant investor interest .... Such interest, in turn, implies a likelihood that many investors are executing trades on the basis of newly available or disseminated corporate information.[32]
The court opined: “Turnover measured by average weekly trading of two percent or more of the outstanding shares would justify a strong presumption that the market for the security is an efficient one; one percent would justify a substantial presumption.”[33]
Second, the court stated that “it would be persuasive [if] a significant number of securities analysts followed and reported on a company’s stock.”[34] The court explained that “[t]he existence of such analysts would imply ... [that company information is] closely reviewed by investment professionals, who would in turn make buy/sell recommendations to client investors.”[35] The court stated that “[i]n this way the market price of the stock would be bid up or down to reflect the [company’s] financial information ... as interpreted by the securities analysts.”[36]
Third, the court noted that if “the stock had numerous market makers,” this would suggest efficiency.[37] The court explained: “The existence of market makers and arbitrageurs would ensure completion of the market mechanism; these individuals would react swiftly to company news and reported financial results by buying or selling stock and driving it to a changed price level.”[38]
Fourth, the court held that if “the [c]ompany was entitled to file an S-3 Registration Statement in connection with public offerings,” this would suggest efficiency because “it is the largest and most well known companies which register equity securities on Form S-3,” and “such companies are widely followed by professional analysts and investors in the market place.”[39]
Fifth, the court held that “it would be helpful to ... [have] empirical facts showing a cause and effect relationship between unexpected corporate events or financial releases and an immediate response in the stock price.”[40] As the court explained, “This, after all, is the essence of an efficient market and the foundation for the fraud on the market theory.”[41]