Securities fraud class actions depend on the idea that investors rely on the integrity of market prices—that stock price reflects the market’s appraisal of all information available about the firm. Thus, an investor may be presumed to have relied on a material misrepresentation even if he was unaware of it because the misrepresentation would have affected market price. Without this fraud-on-the-market (FOTM) presumption, class actions would be impossible; they would raise too many questions as to whether individual class members should be presumed to have relied on a misrepresentation.

Defendant firms can challenge the FOTM presumption in court. In the 1988 case Basic Inc. v. Levinson in which the Court first articulated the FOTM theory, it also ruled that the defendant may show that the misrepresentation did not affect market price and that the plaintiff thus cannot be presumed to have relied on the misrepresentation. But virtually all such cases settle out of court, long before the presumption can be challenged in the courtroom, because of the potential for massive liability in class action suits. A judgment against a defendant firm could result in damages wholly out of proportion with genuine investor losses, and firms would rather settle for a negotiated amount than risk an enormous loss in court.

This concern—or the lack of it—was central to the Court’s recent decision in the Amgen case. Amgen argued that early in the legal process, when the court considers a motion to certify class status, defendants should be permitted to introduce evidence as to a lack of price impact—and that Basic itself so held. That way, the materiality of an alleged misrepresentation—and class reliance thereon—can be determined before the target company decides whether to settle or proceed with the case. The Supreme Court disagreed. As a result of the 6–3 decision, defendant corporations will be induced to settle many cases that would never succeed at trial.

WHY NOT WAIT FOR TRIAL?

The Court majority reasoned that if the plaintiff cannot prove materiality at trial, the action will fail for all class members. As for the argument that materiality will often never be tried because defendant firms will opt to settle, the Court responded that in many types of civil cases, the facts are never tried because of the pressure to settle. Why should FOTM cases be treated differently?

The problem with this reasoning is that materiality is not the point. The point is reliance—the presumption that members of the plaintiff class relied on the integrity of the market price (though not necessarily the accuracy thereof) when they decided to purchase the security in question. If a misrepresentation affected market price, it is presumed to be material and absent class members may be presumed to have relied on it. In other words, reliance cannot be presumed in the absence of price impact. But a misrepresentation may nonetheless be material (as the Court has defined the word) even if it does not affect market price.

It is telling that the Court describes the FOTM presumption as a substantive doctrine of federal securities fraud law that can be invoked by any plaintiff even outside the context of a class action. What the Court fails to appreciate is that the plaintiff can prove fraud without resorting to the doctrine. In other words, the doctrine is a sufficient, but not necessary, condition for proving fraud. The presumption may be substantive in some sense, but it is not necessarily about the merits.

Thus, arguing that a defendant should get the chance to rebut the presumption of reliance by showing lack of price impact is not the same as suggesting that the plaintiff should be required to prove his case at certification in order to get the chance to prove his case at trial. Rather, the point is that the action should not be tried as a class action unless price impact is shown, because reliance by absent class members cannot be presumed. An individual plaintiff may nevertheless be able to prove his case by proving reliance in fact—as the plaintiff in Amgen said it would do if the action was not certified as a class action.

The Court fails to appreciate that the plaintiff can prove fraud without resorting to the fraud-on-the-market doctrine. And the action should not be tried as a class, because reliance cannot be presumed for absent members.

Ironically, on the very day that Amgen was announced, a federal district court ruled in Gamco Investors Inc. v. Vivendi S.A. that the defendant (Vivendi) had successfully rebutted the presumption of reliance by Gamco, a member of the plaintiff class. In that case, Vivendi was found to have misrepresented material facts relating to its liquidity. When the truth came out, the price of Vivendi stock fell and litigation ensued. Vivendi was found liable in the class action, which did in fact go to trial. Although Vivendi argued that the class should not have been certified because some members of the class could be shown not to have relied on the facts in question, the court ruled that issues of individual reliance could be addressed with issues of individual damages following trial of the class action. At trial, Vivendi introduced evidence showing that Gamco may have known about Vivendi’s liquidity problems and that Gamco’s investment strategy—which was to identify undervalued companies by a proprietary formula that eschewed any reliance on liquidity—made it more likely to invest in Vivendi because of such problems.

The problem with this rationale is that Gamco presumably would have preferred to buy at the lowest possible price. Thus, if it did know that Vivendi was seriously short on cash, it must have thought—incorrectly—that the market had already adjusted the price of Vivendi stock accordingly. But as the court noted, every investor may be said to rely on market price in this sense, thus making the presumption irrebuttable.

The bigger problem is with the idea that individual reliance may be addressed together with individual damages following trial. Gamco demonstrates that this approach may work sometimes, but only in extraordinary cases. Damages can be addressed individually because the burden is on class members to file and prove their claims. But unless the defendant has reason to believe that a specific class member follows some sort of contrarian investment strategy—as was apparently the case in Gamco—there is no reason for such investors to identify themselves and every reason not to do so. To be sure, the defendant has a strong incentive to identify such investors. But it does not appear to make any difference for purposes of certification.

The Amgen Court went out of its way to emphasize that the presumption may be rebutted at trial or on motion for summary judgment. The Court could have ruled that the presumption should be irrebuttable. But it did not do so. So given that the presumption of reliance is rebuttable, why must a defendant wait? If a case can be resolved early on, why not do so? The defendant ought not to be required to defend itself from the collective claims of thousands of purchasers if the action is doomed to fail.

To be sure, the Amgen Court suggests (several times) that materiality may be addressed by motion for summary judgment. But the argument that a misrepresentation cannot be material because it did not affect market price would almost certainly be met with the response that materiality itself is a matter of fact that depends on whether it would be important to a reasonable investor and not whether it affects market price. Indeed, the Court has repeatedly rejected any bright-line test for materiality. Similarly, a motion based on the argument that investors suffered no loss would likely be met with the response that proof of loss is also a matter for trial, as the Court ruled in its 2011 Halliburton decision.

Arguably, the Court cannot have it both ways. It cannot reject efforts to rebut the presumption of reliance for lack of price impact on the ground that it would require the plaintiff to prove materiality before trial, and then hide behind a definition of materiality as something other than price impact. But that is the law according to the Court.

WHY CLASS ACTIONS?

The Amgen Court rejected the judicial economy argument on the dubious grounds that if class action status is denied, individual class members would be tempted to file again for another chance at proving materiality. In effect, the Amgen Court ruled that all such cases should get their day in court—if only to bind the masses that fail to opt out of the class. The flaw in this logic is that big investors often opt out anyway to avoid statutory restrictions on class actions and to negotiate for a better settlement than they might achieve as a member of the class (where the terms of any settlement must be approved by the court).

This increasingly common opt-out strategy raises the question why we need securities fraud class actions at all. In the case of Amgen, more than 80 percent of its stock is held by institutions. So what is to prevent those institutions from joining together in an ordinary non-class action to recover their losses? It may be that the Amgen Court ruled as it did in part because of a worry that small investors would be excluded from recovery. Indeed, the statement by the Amgen plaintiff that it would continue its suit even if class action status was denied could be seen as a veiled threat. Otherwise, it would seem to suggest that the plaintiff did not much care whether class action status was granted, which may explain why the Court went on at some length about the importance of private securities litigation and why it was reluctant to tinker with the FOTM presumption that it created 25 years ago. As Justice Ruth Bader Ginsburg stated in her opinion for the majority, “Congress, the Executive Branch, and this Court … have recognized that meritorious private actions to enforce federal antifraud securities laws are an essential supplement to criminal prosecutions and civil enforcement actions” (my emphasis).

Yet most legal scholars agree that class actions do little to compensate investors while at the same time foisting excessive liability onto defendants. To be specific, diversified investors (like the institutions that own most of Amgen) are just as likely to sell an overpriced stock as to buy one. Over time, losses are offset by gains. Moreover, because the defendant company pays, its stock price falls by some additional amount, creating a feedback effect that increases the potential damage award. (See “The End of Securities Fraud Class Action?” Summer 2006.) In other words, holders compensate buyers for losses that are magnified by the class action itself. Given that portfolio investors (such as institutions) typically hold more shares than they buy during the fraud period, they often lose more because of a class action than they gain from any settlement. To be sure, the settlement may be covered by insurance. But if so, the company will pay higher premiums in the future.

But wait, there’s more. Because the plaintiffs in a class action can theoretically recover their entire loss, they may be better off than they would have been if there had been no fraud, in which case they would have suffered whatever loss came from the timely disclosure of bad news. Thus, the only genuine loss suffered by investors in the aggregate comes from the excess price decrease attributable to enforcement costs (such as fines and defense expenses) and increases in the cost of capital (as a result of loss in trust in management). In other words, the standard measure of damages in a federal securities fraud action includes losses that investors would suffer even in the absence of fraud. To be sure, plaintiffs seldom (if ever) recover in full. But the strength of the case derives from the potential for damages, and defendants must assume the worst. Settlements are negotiated in the shadow of the law. As a result, too many cases are filed seeking damages in excess of true economic loss, creating an excessive deterrent effect.

Although one might argue that fraud cannot be over-deterred, fraud is a bit of a misnomer where neither the defendant corporation nor its directors, officers, or other agents have sold shares. And most securities fraud class actions involve trading in already outstanding shares—not offerings of new stock. Admittedly, investors may be misled in some sense by an over-optimistic press release, but it may be too strong to say that buyers are deceived. In Amgen, one of the supposedly offending statements was that an upcoming meeting of the Food and Drug Administration would not consider the safety of an Amgen drug. That statement was apparently wrong: the Federal Register had listed the matter on the FDA agenda. So it seems unlikely that such a misstatement would affect market price (as Amgen tried to argue). And even if the speaker intended to influence the market by downplaying the significance of news reports—as in the Matrixx case, decided by the Court in 2011—management is not necessarily in a good position to know what the effect will be on market price. What is material to the market may seem trivial to a company spokesperson, or vice versa. So the ultimate effect of the too-generous damages formula under existing law is that managers are reluctant to speak as freely as they might, and investors are left with less information.

The true loss suffered by investors in any meritorious case—from enforcement costs and increases in the cost of capital—is a loss that is suffered by the corporation and thus by all stockholders, not just buyers alone.

Moreover, the true loss suffered by investors in any meritorious case—from enforcement costs and increases in the cost of capital—is a loss that is suffered by the corporation and thus by all stockholders, not just buyers alone. It is a loss that should be recovered by the corporation, in a derivative action, from the individuals who lied to the market, rather than by individual buyers in a class action. Indeed, a derivative action is perfectly tailored to the genuine harm from securities fraud. Stockholders are made whole when the corporation recovers. In theory, the corporation and thus the stockholders are restored to where they would have been if there had been no fraud. And a derivative action is a more potent deterrent because the individual wrongdoers pay, even though the amount at stake may be smaller.

If this view of securities fraud litigation is correct (and it is), the threat that a large investor might sue individually becomes quite worrisome. If successful, such an action overcompensates the plaintiff investor, at the expense of other investors, for losses that should have been recovered by the corporation for the benefit of all stockholders.

REVISITING THE PRESUMPTION

Fortunately, the courts have the power to deal with this problem. As the late Judge Robert Bork observed, to permit individual recovery on such claims is to divert an asset of the corporation to the plaintiff stockholder to the exclusion of other stockholders. Since the rules require that a class claim for damages be superior to other means of resolving the dispute, it follows that if a claim can be handled as a derivative action, it must be so handled.

Indeed, it appears that the Court may be ready to reconsider the FOTM presumption. Last fall, the Court granted a second petition for certiorari in Halliburton to consider (1) whether it should overrule or substantially modify the holding of Basic to the extent that it recognizes a presumption of classwide reliance derived from the FOTM theory, and (2) whether, in a case where the plaintiff invokes the presumption of reliance to seek class certification, the defendant may rebut the presumption and prevent class certification by introducing evidence that the alleged misrepresentations did not distort the market price of its stock.

To be clear, Halliburton I held that loss causation need not be shown in order to gain class certification. Amgen held the same as to materiality (although the real issue in both cases was whether loss causation or materiality should be able to be disproven at the certification stage). But neither case focused on the operation of the presumption. To be more precise, both cases avoided questions as to how the presumption may be rebutted by focusing instead on the elements of the action, the substance of which may not be addressed in connection with certification under precedents governing class actions.

Halliburton II confronts the question of rebuttal head-on. But it also goes further in raising the question whether the FOTM presumption should be eliminated entirely. As argued here, there is no very good reason to preclude defendants from offering rebuttal evidence at the certification stage. But it is not at all clear that it makes sense to banish the idea of the efficient market from the courtroom. There is no serious doubt that information affects market prices and does so efficiently enough that traders are unable to beat the market any more often than is consistent with chance.

Moreover, even if the court were to overrule the FOTM presumption in connection with misrepresentations as established in Basic, the presumption would remain with regard to omissions of material facts. And almost every securities fraud class action can be characterized as an omission case—if only an omission to correct an earlier positive misstatement.

The problem with securities fraud class actions is not the FOTM theory and its presumption of reliance on market prices. Rather it is the idea that investors should be able to recover their losses from the issuer company. To the contrary, it is the company that should recover from the individual wrongdoers. So until we fix this fundamental flaw in the very concept of open market securities fraud, we should permit defendant companies to rebut the presumption of reliance at the certification stage.