Saturday, August 18, 2018
I try to explain to people that the motion to dismiss in a securities case is a unique animal; the complaints, and the briefing, are not like motions to dismiss in any other area of law. When it comes to securities cases – especially class actions – the motion to dismiss is really a mini motion for summary judgment. This week, in a case called Khoja v. Orexigen Therapeutics, No. 16-56069, the Ninth Circuit tried to draw a line in the sand, but as far as I’m concerned, did not go nearly far enough.
It all begins with the Private Securities Litigation Reform Act, which heightens the pleading requirements in securities cases. Among other things, the Act requires that plaintiffs “specify each statement alleged to have been misleading, the reason or reasons why the statement is misleading, and, if an allegation regarding the statement or omission is made on information and belief, the complaint shall state with particularity all facts on which that belief is formed.” Additionally, plaintiffs must “state with particularity facts giving rise to a strong inference that the defendant acted” intentionally or recklessly. 15 U.S.C. §78u-4.
As Hillary Sale has documented in the context of scienter pleading, and as I myself experienced over the course of my 1o-ish year-long career as a plaintiff-side securities litigator, over time, courts have incrementally raised the bar for what kinds of evidence they will accept as meeting these standards. To some extent, there’s a kind of one-way ratchet effect: plaintiffs hit upon the idea of using confidential witnesses to bolster the complaint’s allegations, and pretty soon they became de rigueur, to the point where the lack of such sources is viewed as a weakness, and so forth. Moreover, as Nancy Gertner pointed out in another context, the judicial habit of writing long opinions explaining the reasoning for dismissing a case – and often (though not always) writing shorter, brief orders when refusing to dismiss – creates a body of caselaw stacked against plaintiffs. Point being, securities complaints are subject to an extraordinary degree of scrutiny not present for many other kinds of actions.
One aspect of securities pleading that has grown worse over time is courts’ willingness to consider materials beyond the four corners of the complaint. Though the general rule is that on a motion to dismiss, courts can only consider the complaint itself, there are two significant exceptions. First, courts may consider “matters of which a court may take judicial notice,” and second, courts may consider “documents incorporated into the complaint by reference.” Tellabs, Inc. v. Makor Issues & Rights, Ltd., 551 U.S. 308, 322 (2007). The principles behind these rules are relatively uncontroversial. Judicial notice is supposed to be a relatively narrow category: Rule 201 permits notice of facts “not subject to reasonable dispute” because they “can be accurately and readily determined from sources whose accuracy cannot reasonably be questioned.” Meanwhile, the “incorporated by reference” doctrine was intended to prevent plaintiffs from cherry-picking from documents – like, selecting part of a quote and leaving out the qualifier so as to create a false impression of what the defendant said.
But these narrow exceptions to the general rule against looking beyond the complaint have been stretched to the breaking point in securities litigation. Today, it is common for defendants to submit hundreds of pages of supplemental material on a motion to dismiss. Using one or both of these exceptions, courts consider SEC filings, court filings, press releases, federal agency reports, news articles, analyst reports, stock prices, accounting standards, patent applications, complaints and briefs in other cases, to name a few examples (See, e.g., Wolfe v. Aspenbio Pharma, 2012 WL 4040344 (D. Colo. Sept. 13, 2012); Carlucci v. Han, 2012 WL 3242618 (E.D. Va. Aug. 7, 2012); In re XenoPort, Inc. Sec. Litig., 2011 WL 6153134 (N.D. Cal. Dec. 12, 2011); In re MBIA, Inc. Sec. Litig., 700 F. Supp. 2d 566 (S.D.N.Y. 2010); Wilamowsky v. Take-Two Interactive Software, Inc., 818 F. Supp. 2d 744 (S.D.N.Y. 2011); In re American Apparel, Inc. S’holder Litig., 2012 WL 1131684 (C.D. Cal. Jan. 13, 2012); In re Seracare Life Sciences, Inc., 2007 WL 935583 (S.D. Cal. Mar. 19, 2007); In re White Elec. Designs Corp. Sec. Litig., 416 F. Supp. 2d 754 (D. Ariz. 2006)). Honestly, the supply of extraneous material is endless.
Partly, this is because plaintiffs cite and identify many sources to satisfy the particularity standards (thus opening the door for defendants to introduce documents “referenced”), and partly, this is because the fraud on the market context – with its vague articulation of what counts as publicly available information – invites introduction of any document that might, in any sense, be deemed publicly available. And partly, this is because so much of these disputes play out in SEC filings - which are at least noticeable as publicly-filed regulatory documents - it’s easy for defendants to drop a Tolstoy novel’s worth of them into an appendix.
From my perspective, the problem, really, is less that these documents are introduced at all than the matters for which they are introduced, namely, what they purport to establish. And that’s where the Khoja case comes in.
First –and perhaps most usefully – in Khoja, the Ninth Circuit simply made clear that it is important for district courts to think carefully about why they are considering a document and what its relevance is. In my experience, when faced with hundreds of exhibits produced by defendants (and often little pushback from plaintiffs, because of page limitations and because fighting too hard about authentic documents appears too defensive), district courts often simply say they are generally noticing the documents, without explaining what, precisely, they are considering and why. The Khoja decision demonstrates that this is insufficient: If the district court is relying on a document submitted by the defendants to dismiss a complaint, it should explain its reasoning.
Next, Khoja makes clear that when it comes to judicial notice, courts should articulate what, exactly, they are noticing. Identifying the document is not sufficient: That a document was filed with a public agency may be noticeable; that the document demonstrated the agency had knowledge of a particular fact by a particular date may also be noticeable. But the truth of facts asserted within those documents, or the implications of those documents, are not judicially noticeable facts. Courts need to make the appropriate distinctions.
Here’s where I think the Ninth Circuit did not go far enough. Usually for these kinds of documents – even ones never cited by plaintiffs – courts notice them for the purpose of making assumptions about what “the market” knew and when it was known, or how the stock price reacted. This is the nub of the problem. Some nominally “public” documents – like obscure court filings – may or may not have had an impact on the market, depending on whether they came to traders’ attention: courts cannot make assumptions without further evidence. Further, a document purporting to “reveal the truth” to the market may seem straightforward, but it’s difficult to say how the market interpreted the document without resort to even further evidence of analyst reports, news articles, and so forth, all of which may not be before the court. And price impact? Without expert analysis, what counts as impact (or what may have been confounded by other news) is impossible to tell. The motion to dismiss is not the appropriate stage at which to resolve these disputes.
There is also often a danger that courts will accept the facts in these documents as truthful. SEC reports of insider stock sales, for example; yes, they were filed, but they cannot be assumed (on a motion to dismiss) to represent a complete and accurate portrait of the defendants’ trading history. (The story is different if the plaintiffs themselves cite the document for that purpose - discussed below).
Okay, that’s judicial notice. A separate and distinct basis for a court to consider an extraneous document is when the plaintiffs themselves incorporated the document by reference. In this scenario, the Ninth Circuit set different ground rules. First, the complaint must “referextensively to the document or the document forms the basis of the plaintiff’s claim” to be considered at all. Second, according to the Ninth Circuit, unlike in judicial notice, the contents of an incorporated document may be assumed to be truthful, but “it is improper to assume the truth of an incorporated document if such assumptions only serve to dispute facts stated in a well-pleaded complaint.”
I don’t think the court is wrong here, exactly, but the standard should be articulated more precisely. Certainly, not every document relied on by plaintiffs can be assumed to be truthful (what if the plaintiffs alleged the document was a false press release?) What the court is grappling with is the scenario where plaintiffs use a document as the basis for an allegation, so that the plaintiffs themselves implicitly accept it as truthful. For example, the plaintiffs may rely on a news article, or a government report, that purports to describe the “true facts” at the company at the time the defendants made the allegedly fraudulent statements. A tempting rule of thumb is that the plaintiffs should accept the bitter with the sweet: if they themselves are alleging the document has accurately described the facts, then defendants should also be able to rely on it for that purpose.
But some of these documents may be long, or complex, or contain different anecdotes and explanations for different time periods. It is overinclusive to say that if plaintiffs rely on one specific description in one part of a document, then all other parts of the document should also be treated as truthful for 12(b)(6) purposes, including for the purpose of injecting entirely new facts into the motion to dismiss.
My experience with this comes from the mortgage-crisis era cases that dominated the latter part of my career. We would sometimes rely on something like the FCIC Report – which is 662 pages long – or even Michael Lewis’s The Big Short, which is only a few hundred pages long. We might cite a total of 10 pages, but we were occasionally met with an argument that the entire document should be considered by the court, and assumed to be truthful in all respects – even though we, as plaintiffs, were only basing our allegations on limited excerpts.
In my view, this is difficult balancing act for courts: Cherry picking may be unfair – if plaintiffs rely on a story, necessary parts of the story should not be ignored. But at the same time, unrelated sections of the document should not be assumed to be true for 12(b)(6) purposes merely because the plaintiffs relied on a few, separate portions.
What it comes down to is, the complaint is a highly selective set of allegations – no doubt. But so are documents submitted by defendants in rebuttal. When the defendants’ documents add necessary context to the specific facts being alleged by plaintiffs, so that those facts are rebutted or recontextualized, it is appropriate to consider them on a motion to dismiss. Past that – that’s why we have summary judgment and, eventually, trial.