A recurring question arising in class action securities litigation is what constitutes a “corrective disclosure” for purposes of satisfying the requirements for pleading loss causation. In the Amedisys securities class action litigation, the district court had examined the five partial disclosures on which the plaintiff sought to rely to establish loss causation and held that none of the five alone was sufficient to meet the loss causation pleading requirement. On that basis, the district court had granted the defendants’ motion to dismiss.
However, in an October 2, 2014 opinion (here), the Fifth Circuit held that a “corrective disclosure” does not need to involve a single disclosure; rather, it held, the truth can be gradually revealed through a series of disclosures and whether the disclosures taken collectively satisfy the loss causation pleading requirement is to be determined from the totality of the circumstances. Noting that sometimes the whole can be greater than the sum of the parts, the appellate court held here that the five partial disclosures taken collectively constituted a corrective disclosure and satisfied the requirements to plead loss causation.
Background
Amedisys provides home health care services to patients with chronic health problems. In their securities class action complaint, the plaintiffs allege that the company made misrepresentations regarding its practices in connection with Medicare reimbursement. The plaintiffs contend that the truth about Ameidys’s misrepresentations became known through a series of five partial disclosures. The plaintiffs allege that as the truth leaked out about the company’s Medicare reimbursement practices, its share price declined.
The five partial disclosures on which the plaintiff relied are as follows: (1) an August 12, 2008 online report by Citron Research raising questions about the company’s Medicare billing practices; (2) the September 3, 2009 resignation of the company’s CEO and its CIO, who were reported to have left “to pursue other interests”; (3) an April 26, 2010 Wall Street Journal article reporting a detailed expert analysis of the company’s Medicare reimbursement data, and stating that the company might be “taking advantage of the Medicare reimbursement system”; (4) the announcement between May and September 2010 of investigations of the company by the Senate Finance Committee, the SEC and the DoJ; (5) the company’s July 12, 2010 announcement of disappointing operating results.
Between August 11, 2008 and September 28, 2010, the company’s share price declined from $66.07 per share to $24.02 per share, a drop of 63.6%.
The defendants moved to dismiss the plaintiff’s complaint. The district court granted the defendants’ motion to dismiss on the ground that the plaintiff had failed to adequately plead loss causation. The district court reviewed each of the five partial disclosures on which the plaintiff relied and found that each one was insufficient to constitute a corrective disclosure for purposes of pleading loss causation. The plaintiffs appealed.
The October 2 Opinion
In an October 2, 2014 opinion written by Judge James Rodney Gilstrap for a three-judge panel, the Fifth Circuit reversed the district court and remanded the case to the district court for further proceedings, expressly holding that the plaintiff has adequately pled loss causation.
The appellate court opened its analysis by examining the question of the extent to which fraud must become known to the market before it can constitute a corrective disclosure. The court said that the plaintiff must prove when the “relevant truth” about the fraud began to leak out, causing the plaintiff’s economic loss, which begs the question about the meaning of “relevant.” The Court said the test for “relevant truth simply means that the truth disclosed must make the existence of the actionable fraud more probable than it would be without that alleged fact.” In other words, the relevant truth need not reveal the fraud, it only need make the existence of the fraud more probable.
The Court then said that a corrective disclosure need not be contained in a single disclosure; rather, the Court said, “the truth can be gradually perceived in the marketplace through a series of partial disclosures.” With respect to several of the five disclosures on which the plaintiffs relied, the court said that even any one of the specific disclosures did not make actionable fraud more probable than not, “it must be considered with the totality of all such partial disclosures.”
The Court then reviewed the five disclosures on which the plaintiff sought to rely and said that the disclosures “collectively constitute and culminate in a corrective disclosure that adequately pleads loss causation.” This holding can “best be understood by simply observing that the whole is great than the sum of the parts” In summing up, the Court said “when this series of events is viewed together and with the context of Amedisys’s poor second quarter earnings, it is plausible that the market, which was unaware of Amedisys’s alleged Medicare fraud, had become aware of the fraud and incorporated that information into the price of Amedisys’s stock.”
With respect to the plaintiff’s attempt to rely on the disclosure of the governmental investigations, the appellate court noted that in general the commencement of a governmental investigation of suspected fraud does not standing alone constitute a corrective disclosure. However the court said the disclosure of the governmental investigations of the company’s Medicare billing practices “must be viewed together with the totality of the other alleged partial disclosures.” The court added that the district court erred in “imposing an overly rigid rule that government investigations can never constitute a corrective disclosure in the absence of a discovery of actual fraud.”
The court’s observation with respect to the Wall Street Journal article is also noteworthy. The defendants had attempted to argue that the article could not represent a disclosure because the article’s content was based on information that was already publicly available. The court noted that while the Medicare data on which the article was based may have been publicly available, expert analysis was required to understand its significance. The court noted that “it is plausible that … the efficient market was not aware of the hidden meaning of the Medicare data that required expert analysis [which] may not be readily digestible in the marketplace.”
Discussion
At a very basic level, the Fifth Circuit’s ruling in this case is noteworthy because it is the Fifth Circuit’s ruling. The Firth Circuit is not exactly known as the most plaintiff-friendly of courts.
As for the substance of the ruling, the court’s opinion is noteworthy for its conclusion that a series of disclosures can satisfy the loss causation requirement even if no single one of the disclosures standing alone would be sufficient. The court’s observation that the whole can be greater than the sum of its parts is significant and consistent with the realistic understanding that sometimes the truth leaks out gradually rather than coming out all at once.
The Court’s ruling with respect to the potential relevance of the disclosure of governmental investigations is also significant. Among other things, it shows that it is not the case that the disclosure of a governmental investigation is never relevant to the loss causation inquiry in the absence of disclosure of actual fraud. The court’s analysis suggests that the existence of a governmental investigation can be taken into account as part of the totality of circumstances as part of the loss causation analysis.
By the same token the court’s analysis of the significance of the Wall Street Journal article is also interesting, particularly the court’s analysis of the fact that the Medicare data on which the article was based is publicly available. The appellate court’s analysis highlights the fact that information may be publicly available but it may not be understood, or at least that its significance may not be understood. While this observation makes sense, it does raise interesting questions about one of the basic assumptions of the fraud on the market theory – that is, that in an efficient market at any given point in time, the company’s share price reflects all of the publicly available information about the company. The example of the publicly available but not full understood Medicate information suggests that even in an efficient market a company’s share price may not in fact reflect all of the information that is publicly available about the company. Specifically, when there is publicly available information that is not understood by the marketplace, the company’s share price may not reflect that information. This is an interesting point when considering the theoretical underpinnings of the fraud on the market theory.
The court’s observation that the partially revelatory disclosures must be viewed collectively, in totality and in context suggests the possibility of a wide-ranging inquiry that potentially could encompass broad time frames and a multitude of statements or disclosures. While there is nothing in the court’s opinion that would necessarily sanction the approach, there might be some reason to be concerned that plaintiffs in other cases who lack a single obvious corrective disclosure on which to rely may try to satisfy the loss causation requirement by trying to build up a mosaic of many disclosures to create the desired picture. The danger with this type of approach is that it could encourage some plaintiffs to try to bootstrap a collection of unrelated or innocuous statements in order to try to argue that the totality of the statements taken collectively satisfy the loss causation pleading requirement.
It is probably worth noting that the court’s analysis took place in the context of a massive but gradual share price decline that accompanied the piecemeal revelation of the Medicare billing problems at the company. The existence of the steep price decline and its connection to the partial revelations helps to explain the appellate court’s conclusion here. It could be that another court might not be as receptive to the kinds of arguments raised here if the share price decline had not been as steep or if it did not track with the purported pattern of disclosures. Plaintiffs trying to satisfy the loss causation requirement in the absence of these factors may find it harder to argue that, in their case, the whole is great than the sum of the parts. In other cases where these factors are lacking a court might conclude that the whole is no more sufficient that the insufficient separate parts.
Special thanks to John Browne of the Bernstein Litowitz firm for sending me a copy of the opinion. Bernstein Litowitz represents the plaintiff-appellant in the case and John briefed and argued the case in the Fifth Circuit.
Ninth Circuit Affirms Nvidia Securities Suit Dismissal, Holds Item 303 Disclosure Duties Are Not Actionable: In an October 2, 2014 opinion written by Judge Beverly Reid O’Connell for a three-judge panel (here), the Ninth Circuit affirmed the district court’s dismissal of the Nvidia securities class action lawsuit, concluding that the plaintiffs had not adequately pled scienter. In a particularly interesting part of its opinion, the Ninth Circuit, joining the Third Circuit on this point, held that the district court did not err in failing to consider plaintiffs’ allegations of scienter in the context of Item 303 of Regulation S-K, because, the appellate court held, Item 303’s disclosure duty is not actionable under Section 10(b) and Rule 10b-5.
Among other things, Item 303 requires reporting companies to disclosure “known trends or uncertainties” in the Management’s Discussion and Analysis of Financial Condition and Results of Operations (MD&A) section of their SEC filings The plaintiff had asserted that Item 303 requires disclosure of specified information and that, if the information is material, failure to disclose the required information constitutes a material omission for purposes of Section 10(b) and Rule 10b-5.
In rejecting this argument, the Ninth Circuit cited with approval from a prior holding on the same issue by the Third Circuit, in which that court had said “Item 303’s disclosure requirement varies considerably from the general test for securities fraud materiality set out by the Supreme Court in Basic Inc. v. Levinson.” The Ninth Circuit added that “Management’s duty to disclose under Item 303 is much broader than what is required under the standard pronounced in Basic.” The court held that Item 303 does not create a duty to disclose for purposes of Section 10(b) and Rule 10b-5
For disussion of an interesting example where the Second Circuit arguably took a contrary position with respect to Item 303 disclosure duties in the Ikanos Communications securities litigation, refer here.The Ninth Circuit distinguised the Ikanos Communications case, primarily based on the fact that the Ikanos Communications, unlike the Nvidia case, arose under Section 11 of the ’33 Act rather than under Section 10(b) and Rule 10b-5.
I note as an aside that while Item 303 may not, as the Ninth Circuit held, create a duty to disclose, it does create an opportunity to disclose. Companies interested in avoiding the unwanted attention of plaintiffs’ lawyers can try to use the Management Discussion and Analysis section of their periodic filings as an opportunity to “bespeak caution.” I have long argued that the use of precautionary disclosure in the MD&A can be an important part of securities litigation risk management, because it provides a way for companies to try to avoid securities litigation altogether or to put themselves in a better position to defend themselves if securities litigation does arise.
Securities litigation loss prevention is a subject and dear to my heart but I am afraid it is a topic that is not always given the attention it deserves. While it may be a topic for another day, I would welcome the opportunity to reinvigorate the discussion of steps that well-advised companies can take to try to reduce their risks of securities litigation.