Whether last week’s Supreme Court decision in Halliburton Co. v. Erica P. John Fund, Inc., No. 13-317 (June 23, 2014) (Halliburton II) was a victory for plaintiffs or for defendants remains to be seen. At issue is the decision’s effect on the range of circumstances under which the fraud-on-the-market presumption of reliance continues to be available. Without this presumption, first endorsed by the court in Basic v. Levinson, 485 U.S. 224 (1988), it is impossible for a Rule 10b-5 action for damages against an issuer for a misstatement that inflates its share prices in secondary markets to proceed on a class basis. If such an action cannot proceed on a class basis, it is much less likely to be brought.
The decision certainly contains some good news for plaintiffs, at least compared to how it might have turned out. Chief Justice Roberts’ opinion forthrightly dismisses Halliburton’s argument that developments in financial economics since 1988 have rendered Basic’s premises outmoded. It similarly dismisses Halliburton’s argument that experience has shown the policy considerations driving Basic to have been misguided. Given that Robert’s opinion had the backing of six of the Court’s nine justices and that its reasoning leaned heavily on the doctrine of stare decisis, another frontal assault on Basic is unlikely for the foreseeable future.
Whether the decision contains good news for defendants as well is harder to say. The Court unanimously agreed that the “defendants must be afforded an opportunity before class certification to defeat the presumption through evidence that an alleged misrepresentation did not actually affect the market price of the stock.”[1] Whether this right of rebuttal is of genuine value to defendants depends very much on the standards that courts apply to determine whether a defendant has established that the misstatement had no affect on share price.
One plausible approach would be for the courts to impose the same burden of proof on defendants seeking to rebut the presumption of reliance by showing that the disclosure correcting the misstatement[2] did not have a negative effect on price as the burden imposed on plaintiffs seeking to establish loss causation by showing that the corrective disclosure did have such an effect. If the courts adopt this approach, however, the grant to defendants of this right of rebuttal will turn out to be a largely empty gesture.
As a general matter, the problem for determining whether or not a corrective disclosure had an impact on price is that its impact cannot be directly observed. One can directly observe only the total price change on the day of the corrective disclosure, which is the sum of both the disclosure’s impact, if any, and the impacts of all the other bits of news that day affecting investor views of the future prospects of the firm. In response to this problem, courts generally require a plaintiff seeking to establish loss causation to introduce expert testimony based on an event study that meets the 95% confidence standard.[3] An event study is a method from empirical finance that provides a probabilistic estimate related to whether the corrective disclosure in fact affected price. Simplifying a bit, this standard is met when the issuer’s market-adjusted observed price change on the day of the corrective disclosure is sufficiently negative that the change is greater than the market-adjusted changes on 95% of the other trading days over the preceding year. This permits the expert to reject with 95% confidence the proposition that the observed change on the day of the corrective disclosure was solely due to this day’s other bits of news and thus not due in any part to the disclosure.
To impose the same standard on a defendant seeking to rebut the presumption of reliance would require an event study showing a market-adjusted price change the day of the corrective disclosure that is sufficiently positive that the change is greater than the changes on 95% of the other trading days over the last year. Such a study would correspondingly permit an expert to reject with 95% confidence the proposition that the observed change on the corrective disclosure day was at least in part due to the corrective disclosure.
An expert has considerable discretion in constructing an event study. This discretion permits the plaintiff’s and defendant’s respective experts in a given case to offer studies with significantly different results. But for reputable experts, professional constraints and Daubert concerns place limits on this difference. Any situation that would permit a reputable defendant’s expert to introduce an event study meeting this 95% confidence standard for rebuttal purposes is one where it would be essentially impossible to find a reputable plaintiffs’ expert who could introduce an event study meeting the 95% standard for loss causation purposes. It is unlikely that any sensible plaintiff’s attorney would invest in such a case even if the defendant did not have a rebuttal right because both sides know the action is bound to be dismissed at summary judgment.[4] So, if courts adopt this standard, Halliburton II’s grant of the rebuttal right changes little.
In contrast, suppose that the lower courts instead decide that defendants can rebut the presumption of reliance simply by persuading the court that the plaintiff will not be able meet the burden concerning price effect that will be imposed when it is called upon to demonstrate loss causation. In this event, the effect of Halliburton II will be, in everything but name, to move up the loss causation inquiry to the class certification stage.[5] This is because the rebuttal inquiry would center around exactly the same issue as the loss causation inquiry – whether the plaintiff can meet the burden of proof to show that the corrective disclosure did have a negative impact on price.[6] A sensible plaintiff’s counsel often does bring a case where there is a reasonable hope that its expert testimony on loss causation will be persuasive, but where ultimately, at summary judgment or trial, this hope is not fulfilled. Accelerating this inquiry to the time of the class certification hearing can be of real value to a defendant. Most discovery typically occurs after the class certification hearing. If the defendant successfully rebuts the reliance presumption at the certification hearing and thereby blocks the suit from proceeding as a class action, the case usually ends and the defendant saves the very substantial costs of this discovery. There is precedent in the Second Circuit for using, in a somewhat different context, this lower standard for defendants in fraud-on-the-market cases seeking to show a lack of price impact.[7]
In sum, whether Halliburton II is a victory for plaintiffs or defendants very much depends on what courts end up requiring defendants to show in order to rebut the presumption of reliance. The first approach has the appeal of taking the Court at its word as to what the rebuttal requires and imposing a consistent approach for both plaintiffs and defendants in terms of determining price impact. The second has the appeal of reducing the costs of fraud-on-the-market litigations and doing so in a way that hinders cases with relatively weaker evidence of price impact more than ones with stronger such evidence. It is easy to imagine the courts taking either approach.
[1] Slip op., at 23.
[2] In fraud-on-the-market litigations, the focus is usually on the price change at the time of the disclosure correcting the misstatement. If the corrective disclosure causes the price to drop, this suggests both that the misstatement must have made the price higher than it otherwise would have been and that any plaintiff who purchases a security at a price inflated by the misstatement, and who still holds it when the corrective disclosure causes the inflation to disappear, has suffered a loss. Focusing instead on the price change at the time of the misstatement itself creates problems. A large portion of corporate misstatements are made in order to hide a truth that is less favorable than the market’s expectations for the issuer at the time. So, while the share price henceforth will be higher than it would have been but for the misstatement, the misstatement will not change the share price from what it had been right before the misstatement was made.
[3] See, e.g., In re REMEC Inc. Sec. Litig., 702 F. Supp. 2d 1202, 1266, 1275 (S.D. Cal. 2010); In re Imperial Credit Indus., Inc. Sec. Litig., 252 F. Supp. 2d 1005, 1015–16 (C.D. Cal. 2003) aff’d sub nom. Mortensen v. Snavely, 145 F. App’x 218 (9th Cir. 2005); Fener v. Operating Engineers Const. Indus. & Misc. Pension Fund (LOCAL 66), 579 F.3d 401, 409 (5th Cir. 2009).
[4] See Mary K. Warren & Sterling P.A. Darling, Jr., The Expanding Role of Event Studies in Federal Securities Litigation, 6 No. 6 Sec. Litig. Rep. 19 (2009).
[5] This result would be ironic because the Court’s holding in the first Halliburton case was that plaintiffs were not required to establish loss causation in order to obtain class certification. Erica John Fund, Inc.v. Halliburton Co., 563 U.S. , (2011) (slip op., at 6).
[6] It should be noted that a failure of an event study to meet the 95% standard does not mean that the corrective disclosure did not have any effect on price. It may have had an effect but the study simply did not have the statistical power to reliably detect it. As an example, for average companies in normal times, corrective disclosures with actual negative impacts on price of 5% will be accompanied by market-adjusted price changes that fail to meet the 95% standard about 20% of the time. Corrective disclosures with smaller actual impacts will be accompanied by market-adjusted price changes that fail to meet the standard much more often than this. See Edward G. Fox, Merritt B. Fox & Ronald Gilson, Economic Crisis and Share Price Unpredictability, available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2401712 at 36.
[7] This precedent arises from the Second Circuit’s 2008 decision, In re Salomon Analyst Metromedia Litigation, 544 F.3d 474 (2nd Cir. 2008). Because of this decision, the rule in the Second Circuit from 2008 until 2013 was that at the class certification stage, materiality is one of the elements that plaintiffs need to show in order to demonstrate that they are entitled to use the fraud-on-the-market presumption. This rule was overruled by Amgen Inc. v. Connecticut Retirement Plans & Trust Funds, 133 S. Ct. 1184 (2013). According to the Salomon decision, once the plaintiffs had established materiality, the defendant was entitled to rebut the presumption. The way the plaintiff would establish materiality at the class certification stage, according to the Salomon opinion, was to show that “a reasonable investor would think that the information would have ‘significantly altered the total mix of information.’” 544 F.3d at 483 (quoting in part Basic). The “plaintiffs do not bear the burden of showing an impact on price,” the court says, because, in accordance with Basic, “the effect on market price is presumed based on the materiality of the information.” Id. The way the defendant could rebut the presumption was to show no impact on price. In In re American International Group, Inc. Securities Litigation, 265 F.R.D. 157 (S.D.N.Y 2010), the court, applying this rule in Salomon, found that with respect to two corrective disclosures, the defendant had successfully rebutted the presumption of reliance because the plaintiff expert’s event studies did not show price changes sufficiently negative to meet the 95% confidence level. Id. at 185-187.
Merritt B. Fox is the Michael E. Patterson Professor of Law, NASDAQ Professor for Law and Economics of Capital Markets at Columbia Law School.