When is an opinion a false or misleading statement?  If a company official says “I think the deal is fair,” is it a false statement just because the deal is objectively unfair?  Or only if the official also did not subjectively believe the deal was fair when he voiced that opinion?

With the Sixth Circuit’s opinion in Indiana State District Council of Laborers v. Omnicare, 719 F.3d 498 (6th Cir. 2013), a circuit split has developed around the question of what is required to demonstrate that a statement of opinion is false or misleading.  This issue is key to many securities class actions, which often hinge upon the truth or falsity of opinions, not facts.

People routinely express opinions that may be incorrect, or about which they may later change their minds.  But those opinions are not lies if they accurately reflect what the speaker thinks.  Thus, even if a deal is not fair by objective standards, the statement “I think the deal is fair,” is not false unless the speaker did not actually think that the deal was fair.  An opinion is thus “true” if it reflects what the speaker actually thought at the time that he spoke – regardless of whether it is objectively mistaken, differs from the opinions of others, appears to be unreasonable, or is later changed.  So, the truth or falsity of any statement of opinion necessarily turns on the speaker’s actual belief.

This analysis of what makes an opinion a false or misleading statement seems self-evident.  Thus, when the Supreme Court considered the issue of whether and when a fairness opinion can be an actionable false statement, it assumed that an opinion was not false unless it the speaker did not hold the belief stated.  Va. Bankshares, Inc. v. Sandberg, 501 U.S. 1083, 1095-96 (1991) (“Because such a statement by definition purports to express what is consciously on the speaker’s mind, we interpret the jury verdict as finding that the directors’ statements of belief and opinion were made with knowledge that the directors did not hold the beliefs or opinions expressed”).  The threshold question faced by the Virginia Bankshares Court was whether or not a statement of belief or opinion could ever be actionable under the federal securities laws, or was instead per se immaterial.  The Court found that “knowingly false” statements of opinion could be challenged because they can be material and factual “as statements that the [speakers] do act for the reasons given or hold the belief stated and as statements about the subject matter of the reason or belief expressed.”  Id. at 1092.  Making an assumption that the jury had found that the statement was subjectively false, the Court concluded that objective falsity was also necessary for the plaintiff to prove a false or misleading statement under Section 14(a).  Id.

Summarizing the Virginia Bankshares holding in a concurring opinion, Justice Scalia emphasized that both subjective and objective falsity were required for liability:  “[T]he statement ‘In the opinion of the Directors, this is a high value for the shares’ would produce liability if in fact it was not a high value and the directors knew that.  It would not produce liability if in fact it was not a high value but the directors honestly believed otherwise.”  Id. at 1108-09 (Scalia, concurring).

Yet Virginia Bankshares is anything but a paragon of clarity.  It focused on the question of whether or not an opinion can ever be an actionable statement, and then backed into the question of subjective falsity by assuming that it existed in that case.  As a result, the circuit courts were slow to apply its holding.  Although the decision was rendered more than 20 years ago, its analysis of the standard for evaluating statements of opinion was not widely used until the last decade.

After Virginia Bankshares had been discussed and digested, however, most circuit courts began to cite to it for the rule that a statement of opinion is only actionable if it is both subjectively and objectively false or misleading.  See, e.g., In re Credit Suisse First Boston Corp., 431 F.3d 36, 47 (1st Cir. 2005) (citing Virginia Bankshares and holding that in order to challenge a statement of opinion plaintiffs must plead “facts sufficient to indicate that the speaker did not actually hold the opinion expressed,” or in other words, “subjective falsity”); Shields v. Citytrust Bancorp, Inc., 25 F.3d 1124, 1131 (2d Cir. 1994) (citing Virginia Bankshares to hold that “a statement of reasons, opinion or belief by such a person when recommending a course of action to stockholders can be actionable under the securities laws if the speaker knows the statement to be false.”); Nolte v. Capital One Fin. Corp., 390 F.3d 311, 315 (4th Cir. 2004) (“In order to plead that an opinion is a false factual statement under Virginia Bankshares, the complaint must allege that the opinion expressed was different from the opinion actually held by the speaker.”); Greenburg v. Crossroads Sys., Inc., 364 F.3d 657, 670 (5th Cir. 2004) (citing Virginia Bankshares for the proposition that a “statement of belief is only open to objection where the evidence shows that the speaker did not in fact hold that belief and the statement made asserted something false or misleading about the subject matter”); Helwig v. Vencor, Inc., 251 F.3d 540, 562 (6th Cir. 2001) (“‘Material statements which contain the speaker’s opinion are actionable under Section 10(b) of the Securities Exchange Act if the speaker does not believe the opinion and the opinion is not factually well-grounded.’”); Rubke v. Capitol Bancorp Ltd., 551 F.3d 1156, 1162 (9th Cir.2009) (fairness opinions “can give rise to a claim under Section 11 only if the complaint alleges with particularity that the statements were both objectively and subjectively false or misleading”).

But in Omnicare the Sixth Circuit reversed course, departing not only from numerous decision by other circuits, but also from its own precedent, ­­in holding that plaintiffs did not need to prove that defendants knew their opinion was false in an action brought under Section 11.  719 F.3d at 505-07.  The Omnicare court recognized its departure from holdings by the Second Circuit and the Ninth Circuit, which had specifically found that proof of subjective falsity was necessary in cases brought under Section 11.  See Fait v. Regions Financial Corp., 655 F.3d 105 (2011); Rubke, 551 F.3d at 1162.  But the court distinguished all the cases that required proof of the subjective falsity of opinions in a Section 10(b) context, including its own holding in Helwig v. Vencor – arguing that those decisions were only applicable in the context of a Section 10(b) action, in which proof of scienter was required.

In regard to Virginia Bankshares, the Omnicare court reasoned that the Supreme Court had not been squarely faced with whether a plaintiff must plead the subjective falsity of an opinion, contending that the Court’s comments regarding subjective falsity were both dicta and tied to the question of scienter.  “The Virginia Bankshares discussion, therefore, has very limited application to § 11; a provision which the Court has already held to create strict liability.”   Id. at 507.  Extending the “dicta” of Virginia Bankshares to a Section 11 case would be “most dangerous,” the Court ruled: “it would be most unwise for this Court to add an element to § 11 claims based on little more than a tea-leaf reading in a § 14(a) case.”  Id.

The Omnicare court was not the first court to observe that subjective falsity feels like a scienter requirement.  Other courts have noted that the question of subjective falsity in a Section 10(b) case may “essentially merge” with the scienter requirement, such that if a complaint demonstrates subjective falsity, it will also have adequately shown scienter.  See, e.g., Credit Suisse, 431 F.3d at 47.  But the Omnicare court missed the point when it held that proof of subjective falsity is only necessary in a case where there is a requirement that scienter be shown.  Like Section 11, Section 14(a) does not require proof that a speaker knew the statement was false.  In Virginia Bankshares, the court explicitly declined to address the question of scienter – so its discussion was necessarily centered around the element of falsity.  And the subsequent cases that have examined subjective falsity in the context of Section 10(b) have discussed it in terms of falsity, not scienter.  Simply put, an opinion is not false if it is genuinely believed by the speaker.  This question is separate from the inquiry as to the speakers’ intent in making the statement.  To conflate the two elements is an analytic mistake and legal error.

This distinction is most vital in cases such as Omnicare, brought under statutes for which proof of scienter is not required.  But it is also important in Section 10(b) cases, where scienter can be established by showing either actual knowledge or some form of recklessness, while subjective falsity requires plaintiffs to show a speaker knew his opinion was false. Preserving this focused inquiry is especially important when plaintiffs seek to prove scienter “holistically” or through doctrines such the core operations inference or corporate scienter, which cause the scienter analysis to be more and more removed from a showing that a speaker had actual knowledge of fraud as to a particular statement.

Analysis of the hypothetical statement “I think the deal is fair” illustrates these points.  Assume the deal was not objectively fair, but the speaker genuinely believed it was.  In that case, the statement was true and is not actionable – a scienter analysis is not even required.  If the falsity analysis got it wrong, however, we would then delve into a scienter analysis with a looser recklessness test, which would look at a variety of factors other than the speaker’s actual belief in his statement.  The error is worsened when the recklessness analysis is performed with analytic tools like the core operations inference and corporate scienter theory, which have the capacity to devolve into assumptions about the speaker’s intent based on company-wide factors, and under a holistic analysis that tends to look at scienter generally, rather than focusing on the speaker’s state of mind as to a particular challenged statement.  As a result, a speaker could be held liable for saying that he thought the deal was fair, even though that is what he actually thought, because of a judgment that the general factors present at the company suggest overall recklessness. The result is not just analytic impurity, but injustice – such that some cases that should be dismissed would not be.

It took the better part of two decades for the crucial holding of Virginia Bankshares to be well understood and widely applied by the circuit courts.  The Omnicare case represents a step backward in this analysis.  Hopefully, other circuit courts will decline to follow the Sixth Circuit in its abrupt wrong turn, and the Supreme Court will eventually clarify its Virginia Bankshares decision.

In our post in the immediate wake of the Supreme Court’s decision in Amgen Inc. v. Connecticut Retirement Plans, we concluded that rather than being a new threat to the defense of securities class actions, Amgen basically endorsed the status quo: In holding that plaintiffs do not need to establish that allegedly false statements were material to the market before they can gain class certification, the Amgen Court reinforced the rule that is already followed in most courts.  At the same time, we promised to dive deeper in a future post, to discuss the effect of Amgen in those circuits that had previously entertained disputes over materiality in determining whether to certify a class.

This was a weighty task.  In the absence of clear guidance from the Supreme Court, the law on class certification has developed in myriad, complex, and contradictory ways across the circuit courts. The precise legal effect of the Amgen decision will therefore vary from one circuit to the next, as the law in many circuits has not been fully developed, and other circuits have developed distinct doctrines in their effort to find a principled way to implement the fraud-on-the-market presumption from Basic v. Levinson.  Yet a survey of these circuit court decisions, and the way that they have been interpreted in the district courts, merely reinforces our conclusion that Amgen will have relatively little negative practical impact on defendants – in any jurisdiction.  Far from being a “crushing blow” to the defense bar that will “make it easier” for plaintiffs to maintain securities class actions, as many commentators have claimed, the Amgen decision seems to close very few strategic doors to the defense, no matter where a case is litigated.

This is true for three primary reasons.  First, most of the arguments that defendants have made to dispute materiality at class certification are the same as arguments that can be made – and often are, with greater success – on a motion to dismiss, because they challenge fundamental flaws in the complaint that point to plaintiffs’ inability to make sufficient allegations of falsity and loss causation.  Second, the only circuit that required plaintiffs to “prove” materiality before a class would be certified was the Fifth Circuit, whose holdings in this regard had already been largely neutralized by the Supreme Court’s 2011 ruling in Erica P. John Fund, Inc. v. Halliburton.  Finally, in those circuits that had allowed defendants a chance to rebut materiality and thus defeat class certification – in particular, the Second and Third Circuits – the bar was already set so high that this opportunity seemed to be largely illusory.

As an initial matter, it is necessary to identify those circuits that have holdings which are incompatible with the Amgen opinion.  While many circuit courts have yet to grapple explicitly with the issue, the Amgen ruling is in line with the approach articulated by the Ninth Circuit (see that court’s decision in Amgen, 660 F.3d 1170 (9th Cir. 2011)) and the Seventh Circuit (see Schleicher v. Wendt, 618 F.3d 679, 687 (2010)), and the emerging doctrine of the First Circuit (see, e.g., In re Boston Scientific Corp., 604 F. Supp.2d 275 (D. Mass. 2009)) and the Fourth Circuit (see, e.g., In re Red Hat, Inc., Sec. Litig., 261 F.R.D. 83 (E.D. N.C. 2009)).  On the other hand, as discussed below, Amgen raises questions when examined in connection with the extreme approach formerly taken by the Fifth Circuit (see Oscar Private Equity Investments v. Allegiance Telecom, Inc., 487 F.3d 261, 267 (5th Cir. 2007)) as well as the more moderate doctrines articulated by the Second Circuit (see In re Salomon Analyst Metromedia Litigation, 544 F.3d 474, 484 (2d Cir. 2008)) and the Third Circuit (see In re DVI, Inc. Securities Litigation, 639 F. 3d 623, 631 (3d Cir. 2011)).

Until 2011, the Fifth Circuit took the most aggressive approach toward class certification, requiring not only proof that an alleged misstatement “actually moved” the market in order to invoke the fraud-on-the-market presumption, but also requiring plaintiffs to prove loss causation.  This doctrine was summarily rejected by a unanimous Supreme Court in Halliburton, which found that the requirement that plaintiffs prove loss causation to gain class certification was “not justified by Basic or its logic.”  (See 131 S.Ct. 2179 (2011)). It is not clear what, if anything, remained of the Fifth Circuit’s doctrine after Halliburton.  Arguably, a requirement survived that plaintiffs make a showing prior to class certification that either the misrepresentation or the corrective disclosure had an impact on stock price, but that is unclear, because the Fifth Circuit’s justification for this requirement was closely tied to its demand that plaintiffs prove loss causation – and neither the Fifth Circuit nor its district courts have made rulings on this basis since the Halliburton decision.

By contrast, the rule in the Second Circuit survived Halliburton, but was overturned by Amgen. According to the Second Circuit, plaintiffs were required to make “some showing” of materiality in order to trigger the fraud-on-the-market presumption, either through showing an impact of information on the stock price, or simply by arguing that there was a “substantial likelihood” that the misrepresented or omitted information “would have been viewed by the reasonable investor as having significantly altered the total mix of information made available.” (See Salomon, 544 F.3d at 485).  At that point, the burden shifted to the defendants to rebut the presumption of reliance, with evidence that “severs the link” between the alleged misrepresentation and the price of the stock.

Similarly, the Third Circuit ruled in 2011 that district courts may consider evidence to rebut the presumption of reliance, and thereby defeat class certification.  The court reasoned that evidence that a corrective disclosure did not affect the market price could defeat class certification in one of two ways.  If the statement was material, it could show that the market was not efficient in absorbing information (an acknowledged prerequisite to the Basic presumption).  On the other hand, if the market was efficient but the corrective disclosure did not affect the stock price, the Third Circuit held that it could demonstrate that the challenged statement was immaterial as a matter of law. (See In re DVI, 639 F. 3d at 638).

Before Halliburton, the Fifth Circuit’s standard undoubtedly made it difficult for plaintiffs to gain class certification – not only because loss causation was explicitly incorporated into the class certification inquiry, but because plaintiffs bore the burden of proving loss causation before certification was granted.  As noted above, much or all of this hurdle was removed by Halliburton, leaving little or nothing for Amgen to resolve.  By contrast, the standard used by in the Second and Third Circuits placed the burden on the defendants to rebut the presumption of materiality.  Although this gave defendants the opportunity to present evidence about materiality, including expert testimony, the courts implementing this standard generally found that the defendants had failed to meet their burden, and granted class certification despite this evidence.  In ruling that defendants could not present evidence to rebut materiality until summary judgment or trial, the Amgen Court eliminated the possibility that this materiality evidence will be considered at the class certification stage – but this ruling will likely have little practical effect in the Second and Third Circuits, where that opportunity did not seem to give defendants any real advantage.

What is most striking in the pre-Amgen cases that considered materiality on class certification was that nearly all of the arguments that defendants advanced could have been advanced – and often, already had been advanced – at the motion-to-dismiss stage.  If the courts had already accepted these arguments on a motion to dismiss, in most cases the issue would not have reached class certification.  On the other hand, if the courts had already rejected these contentions once at the motion-to-dismiss stage, they did not seem to be any more willing to accept them when they were reframed on class certification.  For example, defendants in these cases sought to defeat materiality by contending that the defendant company had disclosed the truth to the market, rendering the allegedly false statement material.  This argument is easily recast, especially in the case of alleged omissions, to contend that plaintiffs failed to adequately allege of the existence of a false or misleading statement in the first place. Similarly, at class certification, defendants advanced arguments that the plaintiffs had failed to connect the alleged misstatements with any corrective disclosure that revealed the truth to the market, or that they were unable to point to a drop in the stock price following the corrective disclosure – failures that the courts already routinely recognize as fatal to adequate pleading of loss causation, which can be adjudicated on motions to dismiss.

Indeed, the only arguments advanced by defendants in these cases that were unique to the class certification procedure were those that used expert testimony to assert that a price drop was not due to an alleged corrective disclosure, but rather to other negative information that was released simultaneously, or general adverse market conditions.  These contentions involved difficult factual distinctions that the district courts were reluctant to make, particularly in the Second Circuit, where the defendants carried the burden of disproving materiality.

In sum, the Amgen decision seems to foreclose very little in terms of defense strategy.  It may eliminate the use of expert testimony regarding materiality on class certification motions – although such testimony can still be relevant regarding the “efficiency” of the market for the company’s stock.  And it will likely foreclose defendants’ efforts to contend, based largely on expert testimony, that a stock price drop was not the result of a corrective disclosure, but of other factors present at the same time – although defendants should still be able to offer evidence and expert testimony to define the proper contours of the class period based on market events and the timing of corrective disclosures.  But most of the defense arguments that have been used to oppose class certification – whether they are phrased in terms of materiality, falsity, or loss causation – will continue to be available, and effective, through motions to dismiss.

More significantly, the Amgen decision suggests room for doubt on the larger question of reliance – the most fundamental and problematic issue for plaintiffs in obtaining certification of a securities class action.  The decision all but invites the defense bar to use its creativity to find the right argument to advance in the right case, to engage district and circuit courts that are already struggling with the fraud-on-the-market doctrine, and in turn, to tempt the Supreme Court into reconsidering the wisdom of Basic (which it has shown a clear inclination to do).  While Amgen may have closed the door on a few defense strategies – which were rarely successful, in any case, in defeating class certification – it has simultaneously opened the window for defense counsel to find new ways to illustrate the shortcomings of the Basic presumption, and thus to mount a much more serious challenge to ability of the plaintiffs’ bar to bring securities class actions.