Argued September 22, 2009.
Decided August 20, 2010.
Appeal from United States District Court for the Southern District of Indiana, David F. Hamilton, J.
Peter A. Binkow, Attorney (argued), Charles B. Chokel (argued), Glancy Binkow Goldberg LLP, Los Angeles, CA, for Plaintiffs-Appellees.
Robert J. Kopecky, Attorney, Daniel E. Laytin, Attorney (argued), Kirkland Ellis LLP, Barrie L. Brejcha, Attorney, Baker McKenzie, Chicago, IL, Todd G. Vare, Attorney, Barnes Thornburg LLP, Indianapolis, IN, for Defendants-Appellants.
Before EASTERBROOK, Chief Judge, and BAUER and ROVNER, Circuit Judges.
EASTERBROOK, Chief Judge.
When a large, public company makes statements that are said to be false, securities-fraud litigation regularly proceeds as a class action. Class treatment is appropriate when issues common to class members predominate over those that affect them individually. Fed.R.Civ.P. 23(b)(3). Whether the statements are false is one common question. Whether the falsehoods are intentional ( i.e., whether each defendant acted with the required state of mind) is another. Whether the falsehoods affected the stock's price is a third. (If investors already know the truth, false statements won't affect the price.) Whether the magnitude of any effect shows that the false information was "material" is a fourth. There will be some person-specific issues, such as when (and how many shares) a given investor purchased or sold. Timing of each person's transactions, in relation to the timing of the supposedly false statements, determines how much a given investor lost (or gained) as a result of the fraud. But these questions can be resolved mechanically. A computer can sort them out using a database of time and quantity information.
The canonical elements of a claim under § 10(b) of the Securities Exchange Act of 1934, 15 U.S.C. § 78j(b), and the SEC's Rule 10b-5, 17 C.F.R. § 240.10b-5, are falsehood in connection with the purchase or sale of securities, scienter, materiality, reliance, causation, and loss. See Dura Pharmaceuticals, Inc. v. Broudo, 544 U.S. 336, 341-42, 125 S.Ct. 1627, 161 L.Ed.2d 577 (2005). Reliance usually shows how the false statements caused the loss. Until Basic Inc. v. Levinson, 485 U.S. 224, 108 S.Ct. 978, 99 L.Ed.2d 194 (1988), defendants tried to fend off class certification by contending that each investor was bound to have received different information about the company, and that many investors would not have read the supposedly false statements at all. Each investor's fund of information differs from every other investor's. But Basic concluded that the price of a well-followed and frequently traded stock reflects the public information available about a company.
When someone makes a false (or true) statement that adds to the supply of available information, that news passes to each investor through the price of the stock. And since all stock trades at the same price at any one time, every investor effectively possesses the same supply of information. The price both transmits the information and causes the loss. This approach, dubbed the fraud-on-the-market doctrine, supplants "reliance" as an independent element by establishing a more direct method of causation. See Asher v. Baxter International Inc., 377 F.3d 727, 731 (7th Cir. 2004); Eckstein v. Balcor Film Investors, 8 F.3d 1121, 1129 (7th Cir. 1993). When a company's stock trades in a large and efficient market, the contestable elements of the Rule 10b-5 claim reduce to falsehood, scienter, materiality, and loss. Because each investor's loss usually can be established mechanically, common questions predominate and class certification is routine, if a suitable representative steps forward. See In re Mexico Money Transfer Litigation, 267 F.3d 743, 746-7 (7th Cir. 2001); see also Hal S. Scott, The Impact of Class Actions on Rule 10b-5, 38 U. Chi. L. Rev. 337 (1971).
Conseco is a large, publicly traded financial-services holding company. It changed its name to CNO Financial Group in May 2010; we use the name it had when the events that led to this suit occurred. In 2001 and 2002 Conseco's stock was doing poorly, and it filed for bankruptcy late in 2002. (Its subsidiaries, such as Bankers Life Casualty and Colonial Penn Life Insurance, did not enter bankruptcy.) Conseco was reorganized and today is profitable.
This securities-fraud suit, against some of Conseco's managers during 2001-02 (claims against Conseco itself were discharged in bankruptcy), contends that the managers made unduly rosy statements that led investors to pay too much for the shares. Before the bankruptcy began, Conseco was listed on the New York Stock Exchange and included in the Standard Poor's 500 Index. Average daily trading volume was four million shares. Average market capitalization exceeded $2 billion. These facts comfortably qualify under Basic: Conseco was larger, more widely followed by analysts, and traded more frequently than Basic Inc. had been in the 1980s. A financial economist concluded, in an expert report that the district judge credited, that the market for Conseco's shares was efficient, as Basic employs that term, and that investors therefore can use the fraud-on-the-market doctrine as a replacement for person-specific proof of reliance and causation. The judge certified a class. 2009 WL 761157, 2009 U.S. Dist. LEXIS 24810 (S.D.Ind. Mar. 20, 2009).
Defendants have vigorously resisted class certification. That's not surprising, because certification substantially increases the settlement value of a securities suit. What do surprise are the arguments defendants advance, arguments that if accepted would end the use of class actions in securities cases. Defendants contend that even a firm as large, and as widely followed, as Conseco was in 2001-02 does not qualify for fraud-on-the-market treatment under Basic. They also contend that, before certifying a class, the district judge must determine that the contested statements actually caused material changes in stock prices. In other words, they insist that before a class can be certified plaintiffs must prove everything (except falsity) required to win on the merits. And defendants further contend that, even if the evidence shows scienter, materiality, causation, and loss, individual damages questions still predominate and prevent class certification. A more thoroughgoing challenge to class treatment of securities litigation is hard to imagine. Defendants find some support in a recent decision of the fifth circuit. See Oscar Private Equity Investments v. Allegiance Telecom, Inc., 487 F.3d 261 (5th Cir. 2007). The district court declined to follow Oscar Private Equity. A motions panel granted defendants' petition for leave to take an interlocutory appeal so that this court could address the subject before the district court holds what could be a complex trial, or the defendants are induced to settle to curtail the substantial risk. See 28 U.S.C. § 1292(e); Fed.R.Civ.P. 23(f).
Defendants don't ask in so many words that we jettison the fraud-on-the-market doctrine. A court of appeals can't revise principles established by the Supreme Court. But they see an opening in the fact that although, in many securities-fraud cases, the false statements (or the material omissions) propel the stock's price upward, for Conseco the statements were designed to slow the rate of fall. Conseco's price was declining throughout the class period, eventually reaching zero when the bankruptcy court cancelled the shares and converted the debt investors' claims to equity. Defendants say that this makes a difference. Their opening sally is that plaintiffs must be using a "materialization-of-risk theory" to show causation, and that this is incompatible with the fraud-on-the-market approach.
Although "materialization of risk" runs like a mantra through the parties' briefs, we do not think that it has any significance. The phrase appears in a few decisions, e.g., Ray v. Citigroup Global Markets, Inc., 482 F.3d 991, 995 (7th Cir. 2007); In re Omnicom Group, Inc. Securities Litigation, 597 F.3d 501, 513 (2d Cir. 2010), to describe particular claims, but it is not a legal doctrine or anything special as a matter of fact. When an unduly optimistic false statement causes a stock's price to rise, the price will fall again when the truth comes to light. Likewise when an unduly optimistic statement stops a price from declining (by adding some good news to the mix): once the truth comes out, the price drops to where it would have been had the statement not been made. If a firm that is losing money says "we expect to lose $100 million next quarter" when the managers actually expect the loss to be $200 million, that statement will keep the price higher than it ought to be, and when the next quarterly results show the real $200 million loss the price will adjust (unless the managers try a new, and larger, falsehood). The parties are wont to call the bad outcome (the $200 million loss) a "materialization of the risk" that the loss would exceed $100 million. But it should be clear that this is just a mirror image of the situation for the same figures in black ink, rather than red. If the firm projects a $200 million profit, when the managers actually expect $100 million, then the eventual disclosure of the expected result could be called a "materialization of the risk" that the real profit would be less than the managers' optimistic number of $200 million. The phrase adds nothing to the analysis. Whether the numbers are black or red, the fraud lies in an intentionally false or misleading statement, and the loss is realized when the truth turns out to be worse than the statement implied.
For example, one of plaintiffs' allegations is that defendants concealed $900 million in guarantees that Conseco had written. It makes sense to say that the risk — that Conseco would have to honor some of these promises — materialized when the guarantees were drawn on. But the fraud (if there was any, a subject on which we express no view) was the omission from public filings of information about the guarantees, at a time when the omission of this news made other statements misleading or incomplete, not the materialization of the risk that the beneficiaries would draw on the guarantees. Fraud depends on the state of events when a statement is made, not on what happens later. See Pommer v. Medtest Corp., 961 F.2d 620, 623 (7th Cir. 1992); Jordan v. Duff Phelps, Inc., 815 F.2d 429, 440 (7th Cir. 1987); In re Burlington Coat Factory Securities Litigation, 114 F.3d 1410, 1429 n. 16 (3d Cir. 1997) (Alito, J.); Media General Inc. v. Tomlin, 387 F.3d 865, 869 (D.C. Cir. 2004).
Conseco was a large, well-followed firm, whose stock traded actively in a liquid market. It comfortably meets Basic's requirements. Plaintiffs' expert verified that the price of Conseco's stock changed rapidly, and in the expected direction, in response to new information. Defendants did not present a contrary analysis; they just tried (and failed) to poke holes in plaintiffs' analysis. That Conseco's stock was falling during the class period is irrelevant; fraud could have affected the speed of the fall. If a firm says that it lost $100 million, when it actually lost $200 million — and analysts had expected it to announce that it lost only $50 million — then the announcement will cause the stock's price to fall. But the fall won't be as much as the truth would have produced. People who buy the stock after the announcement, and before the truth comes out, pay too much; they will lose money when the rest of the bad news emerges. This is no different in principle from a firm's announcement of a $200 million profit, when the truth is $100 million; only the signs on the numbers differ.
That the class includes short sellers (many investors were long at some times and short at others) also is irrelevant. A person buys stock (goes long) because he thinks the current price too low and expects it to rise; a person sells short (sells today and promises to cover in the market and deliver the shares in the future) because he thinks the price too high and expects it to fall. These positions are mirror images. If a long can participate in a class, so can a short. Both the long and the short are affected by news that influences the price they pay or receive. It may turn out that the shorts do not suffer compensable losses — that, indeed, the shorts' gains should be subtracted from the longs' losses, and only the net treated as damages — but this does not imply that the class definition is defective. See Kohen v. Pacific Investment Management Co., 571 F.3d 672 (7th Cir. 2009); Fry v. UAL Corp., 84 F.3d 936, 938-39 (7th Cir. 1996).
Defendants' insistence that short sellers don't rely on the market price suggests that they misunderstand the efficient capital market hypothesis, which underlies the fraud-on-the-market doctrine. There are three versions of the efficient capital market hypothesis: weak, semi-strong, and strong. See generally Donald C. Langevoort, Theories, Assumptions, and Securities Regulation: Market Efficiency Revisited, 140 U. Pa. L. Rev. 851 (1992). The weak version is that prices incorporate information in a way that prevents the historical pattern of prices from being used to predict changes in price. In other words, it is not possible to identify any trading rule that beats the market. Everyone can observe historical prices; if information were there, sophisticated traders would use it, prices would adjust, and the past prices would cease to be informative. This implies that only someone with new information can make a trading profit. The semi-strong version adds that the value of new information is itself reflected in prices quickly after release, so that only the first recipient of this information (or someone with inside information) makes a profit; everyone else might as well ignore the information and rely on the prices. The strong version adds a claim that the price set in this way is right, in the sense that it accurately reflects the firm's value.
Many economists think that the strong form of the hypothesis has been refuted, but the weak and semi-strong forms are widely accepted. See Tarun Chordia, Richard Roll Avanidhar Subrahmanyam, Evidence on the speed of convergence to market efficiency, 76 J. Fin. Econ. 271 (2005). And the fraud-on-the-market doctrine rests on the semi-strong form. See Sanjai Bhagat Roberta Romano, Event Studies and the Law: Part II: Empirical Studies of Corporate Law, 4 Am. L. Econ. Rev. 380, 397-400 (2002). See also Zohar Goshen Gideon Parchomovsky, The Essential Role of Securities Regulation, 55 Duke L.J. 711 (2006). Short sellers play a role in aligning prices with information under any version of the efficient capital market hypothesis. That the resulting price may be inaccurate does not detract from the fact that false statements affect it, and cause loss, whether or not any given investor reads and relies on the false statement. That's all that Basic requires.
Defendants say that, before certifying a class, a court must determine whether false statements materially affected the price. But whether statements were false, or whethr) the effects were large enough to be called material, are questions on the merits. Although we concluded in Szabo v. Bridgeport Machines, Inc., 249 F.3d 672 (7th Cir. 2001), that a court may take a peek at the merits before certifying a class, Szabo insisted that this peek be limited to those aspects of the merits that affect the decisions essential under Rule 23. If something about "the merits" also shows that individual questions predominate over common ones, then certification may be inappropriate. Falsehood and materiality affect investors alike, however. It is possible to certify a class under Rule 23(b)(3) even though all statements turn out to have only trivial effects on stock prices. Certification is appropriate, but the class will lose on the merits. Defendants have approached this case as if class certification is proper only when the class is sure to prevail on the merits. That would resurrect the one-way-intervention model that was ditched by the 1966 amendments to Rule 23. Under the current rule, certification is largely independent of the merits (save for the situation covered in Szabo), and a certified class can go down in flames on the merits. The possibility that individual hearings will be required for some plaintiffs to establish damages does not preclude certification. See Pella Corp. v. Saltzman, 606 F.3d 391 (7th Cir. 2010); Arreola v. Godinez, 546 F.3d 788 (7th Cir. 2008).
We could stop here, but for Oscar Private Equity. The fifth circuit earlier held that, when truthful and false statements are made simultaneously, plaintiffs must establish how much of the price movement can be attributed to the false statements. See Greenberg v. Crossroads Systems, Inc., 364 F.3d 657, 666-67 (5th Cir. 2004). Otherwise they can't establish loss causation, which Dura Pharmaceuticals holds is one element of a securities-fraud claim. In Oscar Private Equity the fifth circuit held that proof of loss causation is essential not only to success on the merits but also to class certification. The majority in Oscar Private Equity stated that Basic entitles each circuit to "tighten the requirements" for class certification ( 487 F.3d at 265) and that the fifth circuit would use this authority to curtail the ability of plaintiffs to put pressure on defendants to settle. Id. at 266-70. The right way to show loss causation, the fifth circuit held, is to establish that when the issuer announces the truth, "the market reacted to the corrective disclosure." Id. at 262.
Unlike the fifth circuit, we do not understand Basic to license each court of appeals to set up its own criteria for certification of securities class actions or to "tighten" Rule 23's requirements. Rule 23 allows certification of classes that are fated to lose as well as classes that are sure to win. To the extent it holds that class certification is proper only after the representative plaintiffs establish by a preponderance of the evidence everything necessary to prevail, Oscar Private Equity contradicts the decision, made in 1966, to separate class certification from the decision on the merits. See Eisen v. Carlisle Jacquelin, 417 U.S. 156, 94 S.Ct. 2140, 40 L.Ed.2d 732 (1974).
Congress has been concerned about the potential for class certification to create pressure for settlement, and some studies have concluded that these settlements reflect the limits of insurance rather than the strength of the plaintiffs' claims. See Janet Cooper Alexander, Do the Merits Matter? A Study of Settlements in Securities Class Actions, 43 Stan. L. Rev. 497 (1991); Reinier Kraakman, Hyun Park Steven Shavell, When Are Shareholder Suits in Shareholder Interests?, 82 Geo. L.J. 1733 (1994); Roberta Romano, The Shareholder Suit: Litigation Without Foundation?, 7 J.L. Econ. Org. 55 (1991). But the means that Congress chose to deal with settlement pressure were to require more at the pleading stage and to ensure that litigation occurs in federal court under these special standards, rather than state court under looser ones. The pleading requirement is one aspect of the Private Securities Litigation Reform Act, discussed in Tellabs, Inc. v. Makor Issues Rights, Ltd., 551 U.S. 308, 127 S.Ct. 2499, 168 L.Ed.2d 179 (2007), and the federal-forum rule is part of the Securities Litigation Uniform Standards Act, discussed in Merrill Lynch, Pierce, Fenner Smith Inc. v. Dabit, 547 U.S. 71, 126 S.Ct. 1503, 164 L.Ed.2d 179 (2006). We do not think it appropriate for the judiciary to make its own further adjustments by reinterpreting Rule 23 to make likely success on the merits essential to class certification in securities-fraud suits.
The particular step that the fifth circuit took in Oscar Private Equity would do more than just "tighten" the requirements for class certification. It would make certification impossible in many securities suits, because when true and false statements are made together it is often impossible to disentangle the effects with any confidence. The court suggested (perhaps required) that this' be done by showing what happened when the truth was announced. Yet truth can come out, and affect the market price, in advance of a formal announcement. Suppose a lie on September 1 increases a stock's price by $1 a share. Market professionals (broker-ages, investment banks, and arbitrageurs, among others) regularly conduct their own investigations to discover why a stock's price has moved, net of general market movements. These investigations may turn up the truth. Suppose that by October 1 professional investors had discounted the issuer's statement as probably false. These investors would trade with each other until they were satisfied by the price, which would quickly lose the $1 it gained because of the fraud. If the issuer then formally announced the truth on November 1, the stock's price would not budge. The announcement was no news at all; the truth was reflected in the price by November 1. This is the truth-on-the-market corollary to the fraud-on-the-market doctrine. See Asher, 377 F.3d at 735; Wielgos v. Commonwealth Edison Co., 892 F.2d 509, 516 (7th Cir. 1989); Flamm v. Eberstadt, 814 F.2d 1169, 1179-80 (7th Cir. 1987); In re Apple Computer Securities Litigation, 886 F.2d 1109, 1115-16 (9th Cir. 1989). (Similarly, if the truth had been known on September 1, the false statement would not have affected the price in the first place.)
But the fact that investors who bought between October 1 and November 1 could not establish loss causation (or loss, period) would not imply that investors who purchased between September 1 and October 1 also were uninjured. After a class has been certified, and other elements of the claim have been established, the court will need to pin down when the stock's price was affected by any fraud. That decision, like the other issues, can be made on a class-wide basis, because it affects investors in common. It gets the cart before the horse to insist that it be made before any class can be certified. If the data are so ambiguous that the decision can't be made at all, then the class loses outright (plaintiffs bear the burden of persuasion, after all), but to repeat a point already made: The chance, even the certainty, that a class will lose on the merits does not prevent its certification.
Oscar Private Equity represents a go-it-alone strategy by the fifth circuit. It is not compatible with this circuit's decisional law ( Asher, Eckstein, Flamm, and others), and we disapprove its holding. It has not been adopted by any other circuit, and it has been rejected implicitly by some. See, e.g., In re Salomon Analyst Metromedia Litigation, 544 F.3d 474, 479, 483 (2d Cir. 2008).
Just as plaintiffs need not establish loss causation before a class can be certified, so they need not establish that the false statements or misleading omissions are material. Although several circuits have thought materiality a condition to class certification, e.g., In re PolyMedica Corp. Securities Litigation, 432 F.3d. 1, 8 n. 11 (1st Cir. 2005); In re Salomon Analyst Metromedia Litigation, 544 F.3d at 481, that conclusion misreads Basic. These circuits observe that footnote 27 in Basic, 485 U.S. at 248, 108 S.Ct. 978, lists materiality as an element in the fraud-on-the-market doctrine. All note 27 does, however, is state that the court of appeals deemed materiality essential; the Justices did not adopt it as a precondition to class certification. Note 27 observes that the court of appeals had listed "public misrepresentations" as another element, but all this could have meant is that the complaint must allege that there were public misrepresentations. Falsehood and materiality are issues on the merits; whether a statement is materially false is a question common to all class members and therefore may be resolved on a class-wide basis after certification. See Burlington Coat Factory Securities Litigation, 114 F.3d at 1419 n. 8. A complaint must support allegations of falsehood and scienter in the way required by the PSLRA, as understood in Tellabs, but proof can await motions for summary judgment and trial.
The district court assured itself that the market for Conseco's stock was thick enough to transmit defendants' statements to investors by way of the price. That finding supports use of the fraud-on-the-market doctrine as a replacement for individual reading and reliance on defendants' statements. The district court did not commit a legal error, or abuse its discretion, in deciding that the fraud-on-the-market doctrine should not be conscripted to serve some other function, and its decision therefore is affirmed.