John C. Coffee Jr.
John C. Coffee Jr. ()

Like children on Christmas Eve, securities defense attorneys and corporate executives are waiting in hopeful anticipation for the U.S. Supreme Court’s coming decision in Halliburton v. Erica P. John Fund (Halliburton II), which may overrule the “fraud on the market” doctrine (FOTM) that was announced over a quarter century ago in Basic v. Levinson.1 Academics are divided, with probably the majority fearing the loss of general deterrence if the securities class action is substantially undercut. Conversely, a minority (including this author) believe it is remarkable that FOTM has survived as long as it has because it is extraordinarily ill-suited to the real world of securities fraud (as hereafter explained). A third more nervous group of spectators are the managing partners of litigation-oriented law firms, who know that FOTM’s potential abolition would likely imply a steep decline in securities litigation, which is the staple of their practice. Ironically, some of the securities defense attorneys eagerly awaiting FOTM’s demise may next year be learning how to litigate patent cases. Be careful then what you wish for, as you may get it.

Last week, several of the amicus curiae briefs in Halliburton II were filed with the court, and, in light of them, this column will first survey the arguments being made to the court and then turn to the likely options before the court. It will argue that there is a middle ground available to the court that does not necessitate overruling long-established precedent but that would curb many of the abuses in contemporary securities litigation: namely, to focus the class certification hearing on the likelihood of price distortion (based on event studies and other evidence), while dropping any inquiry into the irrelevant issue of market efficiency. This would enhance defendants’ leverage without burying the securities class action. As usual, both plaintiff attorneys and defense counsel will stoutly resist this compromise and hope for an all-or-nothing victory.

The Basics of ‘Basic’

Because some readers may be humble corporate practitioners (and not litigators), it is useful to begin with the “basics” by recalling that Basic v. Levinson established a “presumption of reliance,” which enables the putative class to gain class certification under Rule 23(b)(3) of the Federal Rules of Civil Procedure (which rule requires that common questions of law or fact “predominate” over individual questions). Absent this presumption, reliance, which is a requisite element of a Rule 10b-5 cause of action, would generally be an individual question that would preclude class certification. Basic made reliance a common question by announcing that “[a]n investor who buys or sells stock at the price set by the market does so in reliance on the integrity of that price.”2 It justified this result on the premise that “the market price of shares traded on well-developed markets reflects all publicly available information,” including “any public material misrepresentation.”3 The chain of logic thus flows from the premise that a material misrepresentation will be reflected in a security’s price to the conclusion that investors’ reliance on that misrepresentation “may be presumed for purposes of Rule 10b-5 action.”4

One can easily think of counterexamples. For example, an indexed investor is probably not relying on the “integrity” of the market price, because it knows that, if it broadly invests in the market, it will be buying both overvalued and undervalued stocks (and expects them to average out over the long run). Still, the greater problem in Basic‘s logic is its use of market efficiency to support the entirely valid idea that when the market incorporates fraudulent information into price, a “fraud on the market” results. In truth, whenever the market price is distorted by fraudulent material information, investors suffer an injury, whether they individually relied or not and whether the market is efficient or not. The efficiency of the market determines only the speed at which the material misinformation is incorporated into price.

Prior to Basic, the federal courts that had accepted the FOTM doctrine required the plaintiff to make a showing that “a lie, misleading statement, or omission has affected the price of the stock.”5 Today, such a showing would be described as directed at loss causation and would be postponed until trial or summary judgment pursuant to the court’s holding in its first Halliburton decision.6 Because the vast majority of certified securities class actions settle, this postponement effectively trivializes this central question of whether the market price was in fact distorted.

Despite the original focus of the FOTM doctrine on whether there was in fact a price distortion, courts after Basic v. Levinson shifted their focus from whether the market price has been distorted to whether the market was efficient. Very quickly, the operative interpretation of Basic became that if the market was efficient, then a “presumption of reliance” followed, the class would be certified, and the question of actual market distortion became submerged. To be sure, the issue of the loss causation was later rediscovered in 2005 by the court in Dura Pharmaceuticals v. Broudo,7 but loss causation has remained the caboose on the securities fraud train. Juries tend not to understand it, and on a motion for summary judgment, the burden is high and on the defendant to prove that no reasonable jury could believe that the specific misstatement at issue caused the stock drop.

As a result, the FOTM doctrine operates in both an underinclusive and overinclusive fashion. On the underinclusive side, courts have generally refused to certify Rule 10b-5 classes in cases involving stocks traded in thin markets, in IPO cases, and in cases of securities (such as bonds) that are not traded on exchanges. Worse yet, since Cammer v. Bloom8 in 1989, courts have used a primitive and ill-fitting list of arbitrary factors to determine whether the market was efficient. As a result, in those areas where a fraud remedy is most needed (for example, in “pump and dump” and similar cases involving OTC stocks and the lower rungs of Nasdaq), class certification will be generally unavailable on the grounds that the market is not efficient, even though price distortion may seem clear.

On the overinclusive side, defendants are even more dissatisfied because class certification is virtually automatic in Rule 10b-5 cases, at least once it is determined that the market is efficient. The defense that the misstatement was immaterial or that the stock drop was attributable to other, unrelated factors cannot be heard until late in the litigation and thus tends to be nullified by the defendant’s normal reluctance to gamble on a jury trial in a potential billion dollar case. To be sure, defendants are hardly disarmed, as the Private Securities Litigation Reform Act gives them effective weapons. But if plaintiffs can plead a securities fraud claim with the requisite particularity, the reality is that the issue of whether the alleged misstatement in fact distorted the market price tends to fall by the wayside.

In short, both sides have good reason to be dissatisfied with the FOTM doctrine in its actual operation (although plaintiffs are desperately dependent on some means of satisfying their need to make reliance a “common” issue under Rule 23(b)(3)).

What Might the High Court Do?

The court granted certiorari on two issues in Halliburton II:

(1) Whether this court should overrule or modify the holding of Basic v. Levinson, to the extent that it recognizes a presumption of classwide reliance derived from the fraud on the market 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 of reliance and prevent class certification by introducing evidence that the alleged misrepresentation did not distort the market price of its stock.

In my judgment, the court is more likely to resolve Halliburton II on the second above question by permitting the defendant to rebut the presumption at class certification in some fashion.

Why would the court prefer the second route? Although four Justices expressed skeptical views about the FOTM doctrine in 2012 in Amgen v. Conn. Ret. Plans & Trust Funds,9 they (or some of them) may still be reluctant to overrule a long-established precedent. Certainly that has been the position of Chief Justice John Roberts (who sided with the majority in Amgen and expressed no criticism about the FOTM doctrine). Indeed, legal realists might suggest that the current “conservative” majority knows that if the next president is a Democrat, they will likely eventually be in the minority. If they reverse settled precedents (including Basic v. Levinson), they can expect a future “liberal” majority to do the same to them. Thus, stare decisis may have an even stronger appeal to them today than it had in the past.

Nonetheless, the possibility of a complete reversal of Basic v. Levinson exists. In an amicus brief authored by Stanford Law professor Joseph Grundfest and several Wachtell, Lipton partners, the argument is made that, as a judicially created implied cause of action, Rule 10b-5 must be read and shaped in the light of the most analogous express cause of action in the Securities Exchange Act of 1934, which the court has previously said is §18(a) of the 1934 Act.10 Because §18(a) expressly requires proof of actual reliance,11 they argue that Rule 10b-5 thus also requires plaintiffs to prove actual reliance.

Although several amici briefs argue that this could be done without offending stare decisis, those justices who consider fidelity to stare decisis to be an important principle may still be troubled about overruling Basic. A minority of the court is certain to protest, and, if the decision is seen as illegitimate, a later different court might not respect it.

A further and still unrecognized problem surrounds using §18(a) as the template for Rule 10b-5. If Rule 10b-5 can reach only as far as §18(a), Rule 10b-5 faces even greater trimming in the future because §18(a) reaches only documents filed with the SEC.12 Thus, it does not apply as a practical matter to private, non-reporting companies or to oral statements, press releases and other communications that are not filed with the SEC. This limitation would radically curb the reach of Rule 10b-5 (which the court has previously held to cover even oral promises made on a face-to-face basis13).

The §18(a) argument may thus prove too much. One cannot logically limit this thesis that Rule 10b-5 cannot reach further than §18(a) to only the issue of reliance. If §18(a) sets the outer boundaries on the scope of Rule 10b-5, many decisions will have to be reversed (Basic, itself, dealt with false press releases). The legislative history of §10(b) reveals, above all, that it was intended as a catch-all device. Tommy “the Cork” Corcoran told Congress on behalf of the 1934 Act’s draftsmen that the §10(b) essentially said: “Thou shall not devise any other cunning devices.”14 Cabining Rule 10b-5 within the limits of §18(a) runs counter to that purpose and to the court’s express statement in SEC v. Zandford that §10(b) “should be ‘construed not technically and restrictively, but flexibly to achieve its remedial purpose.’”15 Still, some on the court will be attracted to the bright-line quality of the §18(a) argument.

Several amici briefs elaborately develop the argument that the contemporary class action fails both as a means of compensation and deterrence. There is a wealth of data supporting this assessment, and they amass it well. But their suggestion that deterrence should be left to the SEC is less convincing. SEC settlements have long been a small fraction of private class action settlements,16 and the SEC’s “fair funds” procedure involves the same circularity problem of investors paying investors as does the private class action. More importantly, this is not a court that likes to discuss policy arguments. It prefers to discover a bright line rule (such as a “presumption against extraterritoriality”).17

Rebutting the Presumption

All this brings us to the second question on which the court granted certiorari: How can defendants rebut the presumption of investor reliance?

This is the juncture at which the court needs to abandon its flawed assumption that the FOTM doctrine is underpinned by market efficiency. Put simply, market efficiency does not demonstrate price distortion; nor is price distortion limited to, or provable only in, efficient markets. This point is made in several of the amici briefs and has been accepted for some time in the academic literature.18

What then does indicate a price distortion showing that all investors shared a “common” injury? The short answer is a market movement seemingly caused by an alleged misrepresentation or omission. In the simplest case, this could be shown by an event study: Did the stock price fall quickly once the material misrepresentation was corrected? This is the standard methodology used today to show loss causation in response to Dura Pharmaceuticals. But it should apply also with regard to stocks traded in inefficient markets, where the price response may be slower and more gradual, but still evident.

Here also is where the most important debate is occurring among the amici briefs. One “conservative” brief, filed for petitioners, expressly argues that “the event study is the best available tool to examine market distortion and show reliance.”19 In contrast, the amicus brief submitted by the Securities Industry and Financial Markets Association (SIFMA) and written by Paul, Weiss argues strenuously against this so-called “middle ground” (in its words), expressly rejecting the view that price movement shows distortion because “proof of price movement upon corrective disclosure does not necessarily demonstrate that the alleged misrepresentation affected the market price.”20 This then may be the real issue in Halliburton II: not whether Basic will be modified, but whether the court will insist on proof of price distortion or insist on actual reliance.

“Liberal” law professors are concerned about any change in the law that would mandate event studies. In a forthcoming article, Prof. Donald Langevoort argues that in the case of smaller companies that omit to disclose material information over a sustained period, it can be extremely difficult to show price distortion (or when it began).21 Ideally, some alternative means is needed to supplement the event study.22 Consider, for example, this hypothetical: a crowdfunding offering is made to small investors, but the promoter sells five million shares, rather than the one million shares originally disclosed. All investors were thus diluted materially, but there was no trading prices and an event study is here impossible. This is a classic “negative value” action in which litigation is only feasible on a classwide basis.

A possible substitute here to an event study might be a restoration of the old “fraud created the market” doctrine, which presumed class-wide reliance on a showing that the misrepresentation or omission was so material that the securities would not have been “marketable” if it had been disclosed.23 Alternatively, class-wide reliance could be presumed on a showing that the misrepresentation was made in a manner likely to reach all investors (such as in offering documents) and of a magnitude that could not have been ignored by them. Neither rule will fully fill the void, but the issue for the near term is how much fraud will be sheltered after Basic is trimmed or reversed.

Change in the law is likely coming. Liberals may want to leave Basic untouched, but its logic is weak and it serves both sides poorly. Conservatives want to abolish the securities class action by insisting on an actual reliance standard. If rationality prevails (never a safe bet), some on the court will search for a middle ground. That lies in certifying the securities class action only after an inquiry into whether investors bought or sold in a distorted market.

John C. Coffee Jr. is the Adolf A. Berle Professor of Law at Columbia University Law School and Director of its Center on Corporate Governance.


1. 485 U.S. 224 (1988).

2. Id. at 247, 250.

3. Id. at 246-47 and n.24.

4. Id. at 247.

5. See, e.g., Flamm v. Eberstadt, 814 F.2d 1169, 1179 (7th Cir. 1987); see also Jeffrey L. Oldham, “Taking ‘Efficient Markets’ Out of the Fraud-on-the-Market Doctrine After the Private Securities Litigation Reform Act,” 97 Nw. U.L. Rev. 995, 1006-11 (2003); Donald Langevoort, “Theories, Assumption and Securities Regulation: Market Efficiency Revisited,” 140 U. Pa. L. Rev. 851, 899 (1992).

6. Erica P. John Fund v. Halliburton, 131 S. Ct. 2179 (2011) (holding that loss causation need not be shown at class certification).

7. 544 U.S. 336 (2005).

8. 711 F. Supp. 1264 (D. N.J. 1989) (specifying list of factors to determine market efficiency, including whether the issuer qualified for SEC Form S-3).

9. 133 S. Ct. 1184 (2012).

10. In Musick, Peeler & Garrett v. Emp’rs Ins., 561 U.S. 286 294 (1993), the court described §18(a) as the express cause of action “most analogous to the private right of action that is of judicial creation.” It added in Central Bank of Denver v. First Interstate Bank of Denver, 511 U.S. 164, 178 (1994), that courts should use the closest express right “as the primary model for the § 10(b) action.”

11. For decisions so stating, see Ross v. A.H. Robins, 607 F.2d 545, 552 (2d Cir. 1979); Heit v. Weitzen, 402 F.2d 909, 916 (2d Cir. 1968).

12. Section 18(a) refers to “any statement in any application report or document filed pursuant to this title or any rule or regulation thereunder or any undertaking contained in a registration statement.” See 15 U.S.C. §78r(a).

13. See, e.g., Wharf (Holdings) v. United Intern. Holdings, 532 U.S. 588 (2001).

14. See Stock Exchange Regulation: Hearings on H.R. 7852 and H.R. 8720 Before the House Committee on Interstate and Foreign Commerce, 73rd Congress 115 (1934).

15. 535 U.S. 813, 819 (2002).

16. To give just the best known examples, the SEC settlement in Enron was $450 million, while the private class action settled for $7.24 billion; in WorldCom, the SEC settlement was $750 million, while the class settled for $6.19 billion; in Tyco Int’l, the SEC settled for $50 million, while the class settled for $3.2 billion, and in AOL Time Warner, the SEC received $308 million, while the class received $2.5 billion. This list could go on for another page.

17. In Morrison v. Nat’l Austl. Bank, 561 U.S. 286 (1953), the court jettisoned Judge Henry Friendly’s complex “conduct and effect” test after over 40 years of acceptance in favor of a simple bright line test, and that preference for the simple and absolute rule could resurface in Halliburton II.

18. The clearest example is the amicus brief filed on behalf of several law professors (including Adam Pritchard of Michigan Law School and Todd Henderson of Chicago Law School) written by Vinsom & Elkins (hereafter Vinsom & Elkin’s brief). See also articles cited supra at note 5.

19. See Vinsom & Elkin’s brief at 24.

20. See Brief of the Securities Industry and Financial Markets Association in Halliburton II at 22. It further argues that “price impact is not probative of class-wide reliance in an efficient market.” Id. at 26. This phrasing suggests that SIFMA (and its attorneys) suspect that the court might seek a middle ground.

21. See Donald Langevoort, “Judgment Day for Fraud-on-the-Market?: Reflections on Amgen and the Second Coming of Halliburton” (available at (Nov. 16, 2013).

22. Professors Lucian Bebchuk and Allen Ferrell of Harvard have recently proposed event studies as the answer to the problem of proving price distortion, but, unlike Prof. Langevoort, they do not seem sensitive to the problem of a long-term nondisclosure of material information. See Lucian Bebchuk and Allen Ferrell, “Rethinking ‘Basic’” (available on SSRN). (last revised January 2013). Under Basic v. Levinson, there is no obligation to disclose material information simply because it is material; it becomes subject to disclosure only when there is a “duty to disclose.” That complicates the use of event studies in omission cases.

23. In Shores v. Sklar, 647 F.2d 462 (5th Cir. 1981), the Fifth Circuit accepted the idea that class-wide reliance could be assumed if the fraud was so pervasive that, on full disclosure, the securities “would never have been issued or marketed.” Id. at 407. Several other circuits have also recognized in principle this “unmarketability” standard. See Ockerman v. May Zima & Co., 27 F.3d 1151, 1160 (6th Cir. 1994). Cf. Ross v. Bank South, 885 F.3d 723, 729 (11th Cir. 1989) (en banc) (fraud “must be so pervasive that it goes to the very existence of the bonds and the validity of their presence in the market”). So stated, this is a modest inroad, but better than nothing.