Class Certification and Class Impact

As I have noted in my previous posts, I am blogging on the four class action cases the Supreme Court is reviewing this term - (1) AT&T Mobility LLC v. Concepcion; (2) Erica P. John Fund, Inc. v. Halliburton Co., No. 09-1403; (3) Smith v. Bayer, 09-1205; (4) Wal-Mart Stores, Inc. v. Dukes, No. 10-277. As many of you know, Wal-Mart has just been decided, and I am furiously reading it. 

In the meantime, I want to continue in chronological order by discussing Erica P. John Fund v. Halliburton Co., which was decided a couple of weeks ago in a terse, unanimous opinion by Justice Roberts.  As others have noted, the opinion itself is very narrow, and did not reach beyond the immediate issue before the Court.  But the opinion hints at a rising trend in class actions that, in my view, must be stopped.

The issue in Halliburton concerned the certification of a class of investors who were allegedly defrauded by the defendant, Halliburton, in violation of federal securities law.  The plaintiff class alleged that Halliburton made some fraudulent statements that caused the price of Halliburton stock to increase, and thus the investors were damaged by paying an fraudulently-inflated price for their stock. 

Now, fraud, even in the securities context, is nothing new, and the Supreme Court has looked to common law fraud in interpreting the requirements of securities fraud.  One element of a securities fraud claim is reliance - that the investor relied on the fraudulent statement to purchase the stock.  Suppose that an investor got a tip from her uncle to buy a share of Halliburton stock shortly after the fraudulent statements were made.  Further suppose that neither the investor nor the uncle heard the statements.  In one sense the investor is damaged because she bought stock at an artificially-inflated price.  But she did not rely on the fraudulent statement, because she would have bought the stock even in the absence of the statement.  Thus, no reliance, no fraud.

The element of reliance poses a problem for certifying a securities fraud class action, because whether an investor relied on the fraudulent statement is unique to each individual investor, resulting in a multiplicity of individual trials and making the class action seem less than worthwhile. To use the class action parlance, the issue of reliance is an "individual" issue that would tend to "predominate" in the litigation, making the class action not necessarily a "superior" procedure to individual litigation.

But the Supreme Court decided in Basic v. Levinson that the plaintiffs can presume reliance if the security was traded on an efficient market.  Under the theory, if the market is efficient, then the prices of securities reflect all available information, including the fraud.  Thus the investors relied upon the fraud by relying on the price.  I am not sure how this washes (what if you only rely upon price, not having heard of the fraud), but the "fraud-on-the-market" presumption has the added bonus of turning the reliance issue from an individual one to a common one.  Since reliance is now collectively presumed, you don't have individual issues predominating. 

The Fifth Circuit in Halliburton, relying on an earlier case, Oscar Private Equity Invs. v. Allegiance Telecom, Inc., did a clever end-run around the fraud-on-the-market presumption.  First, the Fifth Circuit relied upon some language in Basic suggesting that the presumption can be rebutted if the defendant can show that "market price would not have been affected by their misrepresentation."  Thus, if the defendant can show a lack of loss causation by "sever[ing] the link" between the statement and the price of the security, then no presumption.  Second, the Fifth Circuit pulled the old switcheroo by noting that a defendant will invariably "move to make 'any showing that severs the link,'" and since the burden of persuasion rests on the plaintiffs to prove certification elements, then the plaintiffs have to prove loss causation in order to get the fraud-on-the-market presumption.  Since loss causation is already an element of a securities fraud claim, the Fifth Circuit's approach would effectively require the plaintiffs to prove their claim before they can certify a class.  Indeed, the Fifth Circuit had exactly that result in mind when it noted that the class action "bestows upon plaintiffs extraordinary leverage, and its bite should dictate the process that precedes it."

Au contraire, according to the Supremes.  Justice Roberts opinion notes, correctly, that loss causation is a separate element of a securities fraud claim, and one that is not necessarily linked to reliance.  After all, one can show a lack of loss causation by showing that, even if the fraudulent statements had an effect on the stock price, other causes prevented a price inflation.  Since loss causation is a separate element and unrelated to reliance, the Court put the kibosh on requiring the plaintiffs to prove loss causation to claim the fraud-on-the-market presumption.

In closing the Court rejected an argument by Halliburton that the Fifth Circuit was really requiring "price impact" - that "the statement affected the market in the first place."  The opinion essentially says that loss causation, which is a familiar enough concept, does not equal price impact.

But the resort to "price impact" is in line with a move by some Courts of Appeals to require some common proof of classwide impact (sometimes referred to as "transaction causation") to certify a class, particularly in the antitrust and consumer fraud contexts.  In other words, rather than establish a presumption that can be later rebutted, courts are requiring plaintiffs to guarantee that every class member was injured. In fact, courts have been very hesitant to apply fraud-on-the-market presumptions outside the securities context.

I am writing on proof of classwide impact this summer, but I worry that this trend towards making plaintiffs prove classwide impact will have the same effect as the Fifth Circuit's end-run around the fraud-on-the-market presumption.  It would, in effect, saddle the plaintiffs with a proof requirement without the class certification needed to properly invest in the issue.  The function of the class action in equalizing the stakes between the plaintiffs and the defendant, and thus equalizing the incentives to invest in the litigation, is a point I stress in my recent article on mass torts.  Although the Supremes did not address the issue here, I hope they similarly put on kibosh on this requirement should it come up.