Today's oral argument before the Supreme Court in Tellabs, Inc. v. Makor Issues & Rights, Ltd. was as frustrating to many of the Justices as it was to this observer and a packed gallery gathered to hear the most significant case concerning the heightened pleading requirements for securities fraud claims brought under Section 10(b) of the Exchange Act and SEC Rule 10b-5 since Congress enacted the Private Securities Litigation Reform Act (PSLRA) more than a decade ago.
In response to the perceived need to curtail frivolous and abusive "strike" suits, Congress passed the PSLRA in 1995 which contained, among other provisions designed to limit meritless claims (including new lead-plaintiff requirements), a heightened burden of pleading securities fraud. A complaint will be dismissed under Rule 12(b)(6) if it fails to "state with particularity facts giving rise to a strong inference that the defendant acted with the required state of mind" -- viz., with scienter (intentionally or recklessly). The "strong inference of scienter" pleading requirement is at the heart of the issues raised in Tellabs.
Carter G. Phillips of Sidley & Austin argued for Petitioners, Kannon K. Shanmugam, assistant to the Solicitor General, argued for the United States as amicus curiae, and Professor Arthur R. Miller of Harvard Law School argued for Respondents. Jason Harrow of Scotusblog posted a preview of the Tellabs argument here and Gretchen Sund collected articles on today's argument from the Wall Street Journal and elsewhere in Scotusblog's Round-Up. The parties' merits briefs are collected here and many of the amici briefs are here and here. The transcript of the argument can be found here.
Although Congress clearly intended to raise the bar on the pleading standard, nothing in the text of the statute or the legislative history gives any indication that it was altering the burden of persuasion in securities cases, nor that it intended to make it functionally impossible for meritorious claims to proceed to the discovery phase. It only wanted to facilitate the gatekeeping function of judges in weeding out meritless cases without imposing an onerous and expensive discovery burden on innocent defendants.
At this morning's argument, so many clear opportunities were missed to give the Court the straightforward guidance it was clearly seeking that I feel compelled to answer the Court's unanswered questions. First, the Court asked both sides what the pleading standard should be in securities cases given the civil rules and the PSLRA. Notwithstanding the Government's argument that PSLRA functionally amended the civil rules to eliminate the usual inferences and constructions on motions to dismiss, it cannot fairly be read that way nor is there support for such a reading in the legislative history.
The pleading standard both internally consistent and faithful to the statutory text is -- taking the allegations set forth in the complaint as a whole, accepting them as true, and liberally construing them in the light most favorable to plaintiff -- could a reasonable person draw a strong inference from the alleged facts and circumstances that the defendant acted with the requisite scienter. This is essentially what the Seventh Circuit held in Tellabs:
[W]e will allow the complaint to survive if it alleges facts from which, if true, a reasonable person could infer that the defendant acted with the required intent.
Makor Issues v. Tellabs, 437 F.3d 588, 602 (7th Cir. 2006)(emphasis supplied). As the Court elaborated:
"Scienter is normally a factual question to be decided by a jury, but the complaint must at least provide a factual basis for its scienter allegations." If a reasonable person could not draw such an inference from the alleged facts, the defendants are entitled to dismissal . . .
Id., quoting In re Cerner Corp. Sec. Litig., 425 F.3d 1079, 1084-85 (8th Cir. 2005), and citing Adams v. Kinder-Morgan, Inc., 340 F.3d 1083, 1105 (10th Cir. 2003) ("we understand a 'strong inference' of scienter to be a conclusion logically based upon particular facts that would convince a reasonable person that the defendant knew a statement was false or misleading").
The focus on a "reasonable person" does not dilute the strong inference requirement; it does not transform the congressional requirement of a "strong inference" of scienter into a "reasonable inference" of scienter. Rather, it merely erects an objective standard for finding a "strong inference." Certainly, if an unreasonable or irrational person would draw a strong inference of scienter, that is not what Congress could possibly have had in mind. Instead, examining all of the allegations in the complaint, the court must decide whether collectively they establish facts from which a reasonable person could strongly infer that each defendant acted with scienter.
As any student of civil procedure knows from law school, there is a continuum of evidentiary proof along which various burdens of proof must be satisfied or else the claim will be adjudicated against the party bearing the burden. If one imagines a probability chart running from 0% on the left end to 100% on the right end, the party with the burden (the plaintiff on most issues) has the initial burden of going forward, and the burden of producing enough evidence that reasonable minds can find in his favor. Too little evidence (let's say in the range of 0-25% probability) and the court will enter an adverse judgment against him. An overwhelming amount of factual evidence in his favor (in the 80-100% range) and the court will direct a verdict or enter judgment in his favor. If the totality of evidence is in the middle range where reasonable minds can differ and a verdict for or against the plaintiff would be rational, the decision will be left to the factfinder. On a preponderance of the evidence standard, the plaintiff must have more than 50% of the credible evidence in his favor or the factfinder must decide against him.
Justices Stevens and Kennedy asked all three counsel what percentage of probability would constitute a "strong inference." Mr. Phillips answered that strong inference means greater than a preponderance of the evidence (more likely than not) and greater than the showing of probable cause in an affidavit for a search warrant. Mr. Shanmugam indicated that the requirement imposes the "very high burden" of alleging facts that give rise to a "high likelihood" that the defendant acted with scienter. He also argued that it would be "more consistent with Congress's intent" to apply this "high burden" of a "strong inference" requirement at the proof stage -- in effect, raising the standard of proof in civil securities litigation -- as well as at the pleading stage. Professor Miller opined that "strong inference" is not stronger than "more likely than not" and may be around the 40 percent mark. He agreed that "strong inference" was higher than "probable cause" in a criminal case, but did not directly answer Justice Scalia's question about whether it was greater or lesser than "clear and convincing" evidence.
The Court would have benefitted from more analytical responses. The statutory requirement of a "strong" inference clearly means something more than a "reasonable" inference. So, it has to be more than the evidentiary realm where reasonable minds could differ but the totality of the evidence does not amount to a preponderance (i.e., in the 25-50% range). In fact, without an explicit indication that Congress meant to impose a superburden or to raise the burden of persuasion in securities cases, "strong" must mean "preponderance" -- no more, no less.
The strong inference requirement cannot logically mean the "very high likelihood" burden for which Messrs. Phillips and Shanmugam argued, or the "greater than two-thirds" hyper-probability that Justice Scalia mentioned. If it did, it would create a logical anomaly and practical conundrum. Any complaint that got past the motion to dismiss stage would carry the court's determination that the inferences of guilty knowledge and behavior were overwhelmingly strong. If, nevertheless, the jury returned a verdict for the defendants on a preponderance of the evidence instruction, the court would have to enter j.n.o.v. for plaintiffs because a defense verdict could not be rational given the overwhelming inferences of intentionally or recklessly deceptive or manipulative conduct the court previously found. Thus, the court's decision on the motion to dismiss would become, practically speaking, outcome determinative. Congress surely did not mean this result.
Finally, several members of the Court expressed the view that there was no serious Seventh Amendment concern so long as Congress intended there must be a strong inference alleged at the pleading stage, a strong inference shown at the summary judgment stage, and a strong inference proven at the proof stage. However, none of the advocates pointed out that the appropriate 7th Amendment concern is whether the court, rather than the jury, is making factual determinations or drawing inferences of fact. Such derogation of the jury's prerogative is the necessary and ineluctible consequence of Petitioner's argument. However, the 7th Amendment violation is avoided by adopting Respondent's analysis and affirming the Seventh Circuit's holding.
Unfortunately, I didn't hear sufficiently persuasive answers to satisfy the Court's concerns reflected in their questions. I predict a 7-2 majority (or at best 6-3) for reversal. It would be more faithful to the text of the PSLRA, and to the policies animating its adoption, for the Court to affirm the 7th Circuit's rationale and decision. I hope I am pleasantly surprised, but I am not holding my breath.