The U.S. Court of Appeals for the Ninth Circuit recently amended an opinion in Loos v. Immersion Corp., No. 12-15100 (9th Cir. Sept. 11, 2014) (available here: Loos v Immersion Corp), affirming dismissal of a securities class action for failure to allege facts supporting loss causation. The change to the opinion was small, adding a footnote saying that the court did “not mean to suggest that the announcement of an investigation can never form the basis of a viable loss causation theory,” and noting that “[t]o the extent an announcement contains an express disclosure of actual wrongdoing, the announcement alone might suffice” to support loss causation for the wrongdoing acknowledged. Because the opinion, even as amended, is founded on specious analysis of the content of these types of public disclosures, how the market functions, and the nature of loss causation, it is worthy of discussion here.
In substance, the Loos opinion, even as amended, continues to accept the questionable notion that a drop in stock market price can be “caused” by a revelation of prior false or misleading public statements only if the price reaction follows news explicitly showing that the earlier disclosures were inaccurate. That approach adopts a simplistic view of the content of information – and of the reasons underlying the market price reaction – that bears little relation to reality. It represents, in effect, an illegitimate fact-finding exercise by the courts on the issue of causation long before the evidence is collected and evaluated. It prevents the judicial process, which includes fact-finding by a jury in most such cases, from running its course.
To see why this is true, let us look at the factual allegations in Loos as stated by the appellate court.
Immersion Corp. develops and licenses “haptic” technology, which is the means by which electronic devices vibrate to provide tactile feedback to users. The complaint centers on Immersion’s quarterly disclosures of its business activity, and in particular its quarterly revenues and profits. In successive quarters in 2007, Immersion reported its first profits, and, more importantly, increasing revenue growth. The profitability trend stalled in Q1 2008, but the company reporting continuing revenue growth nonetheless, with increased revenues in Q1, Q2 and Q3 2008. The reported revenue growth ended in Q4 2008, but continued revenue growth was reported in the company’s medical device segment. In May 2009, the company reported that in Q1 2009, revenue was stagnant, and reported medical device revenue dropped significantly.
In July 2009, the company issued a press release revealing “an internal investigation into certain previous revenue transactions in its Medical line of business,” and that it had “not yet determined the impact, if any” to its historical financial statements. The stock price dropped 23% following this news release.
In August 2009, Immersion said that its previous financial statements “should no longer be relied upon” because of irregularities with certain revenue transactions in its medical device business.
In February 2010, the company disclosed errors in its recording of revenue in the medical line of business and a restatement of earnings for 2006, 2007, 2008 and Q1 2009.
The critical issue addressed in the appellate opinion is one of loss causation: whether the stock price drop in July 2009 following the announcement of an internal investigation into previous revenue disclosures could have been caused by investor awareness of previously undisclosed overstatements of company revenue. The court ruled it could not because the July 2009
announcement of an investigation does not “reveal” fraudulent practices to the market. Indeed, at the moment an investigation is announced, the market cannot possibly know what the investigation will ultimately reveal. While the disclosure of an investigation is certainly an ominous event, it simply puts investors on notice of a potential future disclosure of fraudulent conduct. Consequently, any decline in a corporation’s share price following the announcement of an investigation can only be attributed to market speculation about whether fraud has occurred.
Slip. Op. at 19. In reaching this result, the Loos court endorsed and followed the reasoning of the Eleventh Circuit in Meyer v. Greene, 710 F.3d 1189 (11th Cir. 2013).
In this view, inaccuracies in earlier disclosures are only “revealed” by a new disclosure that what was previously stated to be a “fact” is no longer a “fact.” It reflects an assumption that investors can only have a binary view of disclosed information: either it is true or false, and until it is known (to some unstated degree of certainty) to be false, investors can only “speculate” over whether it is true or false. That approach is wholly artificial because it ignores the reality that people, including investors, reason their way to conclusions about the truth or falsity of information based on less than certain evidence. When investors reevaluate the price of a security based on new information that raises the prospect that previous information about a company may no longer be correct, that is not “speculation,” it is the human process of determining, as quickly as possible, the long-term implications of the new facts on the value of the company’s securities.
Before the July 2009 disclosure of an internal investigation, investors had no reason to believe that Immersion’s historical revenue disclosures could be inaccurate, including its oft-noted increases in revenue generally and particularly in revenue from the medical segment of its business. That changed on July 1, 2009. Investors then were presented with a new state of the world for the company in which there was: (i) some probability that there were no errors, but that there could be significant drains on company performance arising out of the investigation, likely regulatory investigations and private litigation; and (ii) some probability that the prior revenue disclosures were false, which could change future projected performance of the company, in addition to the costs and dislocations from investigations and litigation. The market needed to reach a judgment on these respective probabilities, and how they would impact future enterprise value. That is a complex calculation which depends on how costly the investigation and litigation process will be (in dollars and lost focus on business activities); how likely it is that revenues would be restated; if so, by how much; and how that would translate into likely future earnings of the company. (Keep in mind that, at least theoretically from the standpoint of an economist, a company’s stock market price should be based on the market’s estimate of present value of future earnings.)
The 23% drop in stock price was the result of that type of analysis on a marketwide basis. While it is hard, and perhaps impossible, to determine what portion of that price decline is based on the expectation of restated revenues, undoubtedly some part of the decline is tied to that prospect. I say “undoubtedly” because investors know from experience that company management and boards of directors are reluctant to commence an internal investigation, and all the negative side effects it can have, even if it shows no misconduct occurred. The very fact that an investigation was commenced suggests the Board of Directors found something in the nature of “probably cause” that a problem existed. As a result, even without total certainty that revenues were previously overstated, it is more than plausible that some portion of the decline in stock price is attributable to expected future disclosure of a revised revenue pattern going back some period of time. If loss causation is to have any meaning at all, it has to include that portion of the price drop attributable to revised market estimates of likely company revenue in light of the newly disclosed prospect of false historical revenue disclosures.
Why is that so? Why is it wrong to insist, as the Loos and Meyer courts do, that loss causation can only be proved after a disclosure reveals definitively, or at least to a high degree of probability, that a fraud occurred? Because that is not how markets work, and taking that approach inevitably will prevent recovery of losses plainly resulting from market estimates that fraud occurred. The Loos court adopts a market model that is backward-looking, when markets are always forward-looking.
At all times, investors are looking to see if prices need to be adjusted because of future events of varying likelihood. That plainly is true for something like consumer tastes. If new data shows that consumers are moving towards a preference for smartphones with larger screens, investors will bid up the price of stock for companies producing larger-screen phones, and punish those producing smaller screens, long before those preferences are actually reflected in smartphone sales. Even though those stock price changes occur before the consumer purchases actually change, there can be no doubt that the lost value in stock of companies selling small-screen phones is, at least partially, “caused” by the expected changes in consumer product preference. The market functions by “speculating” about the future (in the Loos court’s words), although that would more properly be described as “revising expectations” of future company earnings. That same process applies to market expectations about future revelations of corporate misconduct.
If, as the Ninth and Eleventh Circuits seem to be saying, loss causation from a stock price decline can only be proved when the decline follows a definitive or near-definitive “revelation” that a fraud occurred, investors will be precluded from recovering identifiable losses incurred from stock price declines that can be proximately tied to the reasoned expectation of a future announcement of a fraud. Stock price declines that might be shown by a preponderance of evidence to have occurred because they arose out of revised market expectations based on clues that a fraud could be revealed will not be compensated. And when the definitive evidence of fraud is “revealed,” the market reaction may well by negligible because the market already correctly predicted that it would occur and bid the stock price down before the announcement, thereby negating any compensation for the loss. To be sure, the prospect of developing evidence showing that stock price declines are caused by expectations of future fraud revelations, rather than concerns about the impact of investigations and litigation apart from any fraud, may be slim. But whether losses demonstratively associated with clues of future fraud disclosures are proximately caused by the fraud is a decision that the ultimate trier-of-fact should make. It is not a decision properly made by the court as a matter of law to deprive investors of an opportunity to pursue relief in a private cause of action.
September 16, 2014
Contact Straight Arrow privately here, or leave a public comment below: