In a significant decision issued yesterday in In re Omnicom Group, Inc. Securities Litigation, the United States Court of Appeals for the Second Circuit continued to develop its demanding standards of loss causation in cases brought under Section 10(b) of the Securities Exchange Act. The Court of Appeals affirmed the dismissal on summary judgment, on loss causation grounds, of securities fraud claims against Omnicom Group, Inc. (“Omnicom”) and its management.

The decision, written by Circuit Judge Ralph K. Winter and joined by Circuit Judges Wilfred Feinberg and José A. Cabranes, is notable in at least two respects. First, it further explains one of the two principal pathways to proving loss causation set forth by the Second Circuit in its landmark loss causation ruling, Lentell v. Merrill Lynch & Co., Inc., namely, the materialization of a foreseeable risk concealed by the alleged fraud. In Omnicom, the Court of Appeals made clear that this theory of loss causation cannot be satisfied merely by evidence of a stock price drop resulting from fall-out from an alleged fraud, such as negative publicity or the resignation of a director. Rather, Omnicom explains that in order to establish loss causation, a plaintiff must prove that some unknown facts about the fraud itself—and not collateral facts—caused a decline in the stock price at issue. This may prove to be particularly relevant in the relatively common circumstances where a company announces an SEC or internal investigation, or delays a public report. In those instances, a company’s stock price often drops due to market uncertainty, even if the announcement does not identify the specific matter under investigation or causing the delay. Omnicom suggests plaintiffs may have difficulty using such a stock drop as evidence of loss causation, and may instead be limited to situations where there is a distinct stock drop when the matter is fully disclosed.

Second, Omnicom is one of the relatively few decisions on loss causation in the context of summary judgment, and suggests that substantial proof is required to carry plaintiff’s burden of proving loss causation. In Omnicom, the court did not hesitate to affirm summary judgment notwithstanding that the parties had both proffered conflicting expert testimony on the topic of loss causation. Of course, loss causation is quite frequently litigated in the contexts of motions to dismiss and motions for class certification; the holding of Omnicom plainly has application there as well.

Relevant Background

The Omnicom plaintiffs’ fraud claims were based on allegations of fraudulent accounting in connection with Omnicom’s 2001 Seneca transaction. Plaintiffs allege that, in 2000, Omnicom, a large global marketing and advertising holding company, began to sustain losses on investments in several Internet marketing and advertising companies (the “Internet Companies”). Shortly thereafter, Omnicom entered into a transaction in the first quarter of 2001 with private equity firm Pegasus Partners II, L.P. (“Pegasus”) that created a new company, Seneca, the stock of which would be owned by both Omnicom and Pegasus. As part of this transaction, Omnicom transferred to Seneca its interests in the Internet Companies. Omnicom reported that it would not incur any gain or loss from the transaction because it was exchanging the Internet Companies for cash and stock of equivalent value.

Several news articles written between May 2001 and May 2002 reported on the Seneca transaction, and suggested that it was an attempt to move the value of the deteriorating Internet Companies off of Omnicom’s own balance sheet. The articles observed, for example, that the transaction was “a way for Omnicom to get struggling stocks off of its books” and stated that Omnicom’s CEO was “cleaning up the mess” by “getting . . . assets off Omnicom’s books by shoveling them into . . . Seneca.” (Slip op. at 5 & n.1-2 (internal citations omitted)). Omnicom’s stock price, however, did not experience any statistically significant drop following the publication of these reports.

On June 5, 2002, Omnicom filed a Form 8-K disclosing the resignation of Robert Callander, an outside director who was also Chair of the Audit Committee. Between June 6 and 12, analyst reports and news articles speculated that Callander resigned over concerns relating to the accounting for the Seneca transaction. Then, on June 12, the Wall Street Journal published an article reporting, among other things, that Callander had “resigned amid questions about how the company handled a series of soured Internet investments,” and had “questioned whether something wasn’t being disclosed to the board about the initial off-loading of the problematic investments.” (Slip op. at 8 (internal citations omitted).) The June 12 article also raised questions about Omnicom’s accounting practices, quoting one accounting professor as saying that Seneca “raises a red flag,” and more generally suggesting that Omnicom may be employing unduly aggressive accounting practices. (Id. at 9.)

Over the next two days, a number of analyst reports and articles were published commenting on the events and accounting practices described in the June 12 article. Omnicom’s stock price dropped over 25 percent relative to trading prices and activity in the market and in the industry. (Slip op. at 12.) On June 13, shareholders filed the class action lawsuit, alleging that Omnicom and its managers misrepresented the loss in value of the Internet Companies and improperly failed to account for that loss.

At summary judgment, plaintiffs offered the expert report of Dr. Scott Hakala, who had conducted an event study and was prepared to testify that the June stock price drop was caused by investors’ reaction to partial disclosures about inappropriate accounting contained in the news articles. In a decision characterized by the Circuit as “thorough and well-reasoned,” Judge William H. Pauley of the United States District Court for the Southern District of New York held that plaintiffs had failed to proffer sufficient evidence that revelation of the alleged fraud—that is, the Seneca transaction—caused the stock price drop that damaged the class.

The Second Circuit Decision

The Second Circuit first observed that, having sought to establish reliance by invoking the fraudon- the-market theory, plaintiffs faced the “difficult task” of proving that losses sustained in 2002 were caused by a fraud that was the subject of media reports beginning in 2001. (Slip op. at 18.) In other words, given the assumption of efficient markets, the plaintiffs “must concede that the numerous public reports on the Seneca transaction were ‘promptly digested’ by the market and ‘reflected . . . in Omnicom’s stock price’ in 2001 while seeking to recover for a stock price decline a year later in 2002.” (Id.) The Second Circuit then evaluated whether plaintiffs could surmount this obstacle following either of the theories set forth by the Lentell court; that is, either (1) by proving that the market reacted negatively to a corrective disclosure, or (2) by showing that negative investor inferences drawn from Callander’s resignation and the news stories in June 2002 caused the loss and were a foreseeable materialization of the risk concealed by the fraudulent statement.

As to the former, the Court concluded that plaintiffs had failed to prove that their losses were caused by any new information in the June 12 article, and thus could not establish a corrective disclosure of the fraud. None of the information in the June news articles revealed a “thenundisclosed fact with regard to the specific misrepresentations alleged in the complaint concerning the Seneca transaction.” (Slip op. at 20.) The Court thus rejected as “unsustainable on this record” Dr. Hakala’s opinion that investors first became aware of the alleged fraud in June 2002. Although the Court did not engage in a formal Daubert analysis, it found that no reasonable jury verdict could be sustained for plaintiffs on the basis of Dr. Hakala’s testimony. The Court also found that citations to the opinions of accounting experts in the news articles do not reveal new facts about the Seneca transaction; rather, these references are simply a “negative characterization of already-public information.” (Id. at 21.)

Second, and perhaps most significantly, the Court next rejected the argument that Callander’s resignation and the ensuing negative media attention were foreseeable risks of the fraudulent Seneca transaction causing the June 2002 stock price drop. The Court explained that while an accounting fraud may lead both to a director’s resignation and to negative media stories, “it is generally the facts underlying the fraud and resignation that causes a compensable investor’s loss.” (Slip op. at 25.) Because the facts here were known a year before the resignation, and the resignation did not add to the public knowledge any new material fact about the Seneca transaction, plaintiffs had “at best shown that Callander’s resignation and resulting negative press stirred investors’ concerns that other unknown problems were lurking in Omnicom’s past.” (Id. at 26.) Such a reaction, however, is “far too tenuously connected – indeed, by a metaphoric thread” to the alleged fraudulent transaction to support liability. (Id.)

Explaining the “materialization of unknown risk” theory first articulated in Lentell, the Court explained that a broader interpretation of this theory would undermine the very purpose of the securities laws:

The securities laws require disclosure that is adequate to allow investors to make judgments about a company’s intrinsic value. Firms are not required by the securities laws to speculate about distant, ambiguous, and perhaps idiosyncratic reactions by the press or even by directors. To hold otherwise would expose companies and their shareholders to potentially expansive liabilities for events later alleged to be frauds, the facts of which were known to the investing public at the time but did not affect share price, and thus did no damage at that time to investors.

(Slip op. at 26.) The Second Circuit thus held that a stock price drop caused by public concern over new negative developments relating to previously-known facts does not constitute the materialization of an unknown risk for purposes of establishing loss causation.