Let’s Talk About Corporate Scienter

A plaintiff alleging claims under Section 10(b) of the Securities Exchange Act must show that the defendant acted with “scienter,” which usually means either an intent to mislead investors, or reckless indifference to whether investors would be misled.

Since corporations, as well as natural persons, can be Section 10(b) defendants, there is often a question as to how a corporation’s “state of mind” can be determined for Section 10(b) purposes.  For example, the Third Circuit brushed up against this issue in Pamcah-UA Local 675 Pension Fund v. BT Group PLC, 2021 WL 3415060 (Aug. 5, 2021), and in In re Cognizant Tech. Solutions Corp. Securities Litigation, 2020 WL 3026564 (D.N.J. June 5, 2020), the court tried to develop a Section 10(b) corporate-scienter taxonomy.  My very first article, Slouching Towards Monell: The Disappearance of Vicarious Liability Under Section 10(b), was about how courts identify corporate scienter.  But I’m still finding that a lot of judicial opinions demonstrate confusion on this subject, which inspires me to post about it now.  To be sure, this is not an issue unique to 10(b) actions – it comes up in other areas of law – but Section 10(b) has some specific challenges, so I’m focusing on 10(b) here.

[More under the jump]

The biggest conceptual difficulty in the 10(b) context is that most of the time, the issue of corporate scienter is litigated at the motion to dismiss stage.  That means that the issue of how one pleads corporate scienter is often conflated with the substantive question of what it actually means for a corporation to harbor scienter.  Once these two inquiries are collapsed, the conversation gets very confusing very quickly.  So, in my view, it’s really important to distinguish between what we mean when we say that a corporation acted with intent (or recklessness), and then – separately – to discuss, if a corporation did so act, what kinds of facts a plaintiff should be expected to plead in order to satisfy the PSLRA’s requirement that the complaint raise a “strong inference” of scienter.

So, what actually is it?  What does it mean to say a corporation had scienter?

In general, there are two theoretical possibilities:

(1) A natural person harbored the relevant scienter (intent to mislead, or reckless indifference to it), and that person’s scienter – for whatever reason – can be imputed to the corporation, or

(2) No natural person harbored the relevant scienter, but something about the corporation’s functioning suggests that the corporation as a whole behaved in a manner that we would judge to have acted recklessly or intentionally.

Every single court agrees that (1) is sufficient to show corporate scienter.

Very few courts have accepted (2).  There is some precedent for it – and I discuss some of those cases in Slouching Towards Monell – but not a lot.

If we’re in Category (2), what does that mean, exactly?  Well, in the scenarios where courts have accepted the idea, usually there was something so dysfunctional about corporate operations as a whole that, looking at them, you’d say – “wow, that’s reckless” – even if no human person actually harbored that state of mind.  Like, imagine a situation where someone in corporate communications puts out a statement that a drug is safe.  Separately, someone else within the corporation – who has no responsibility for communications at all – has information showing the drug is not safe.  The two don’t talk to each other.  Neither one of them may individually harbor scienter, and in general, courts will not aggregate their knowledge to impute scienter to the corporation.  But if, say, lines of communication were so incredibly poor, internal controls were so incredibly sloppy, that anyone looking in from the outside would recognize that it was simply irresponsible to let the communications people do their thing, without mechanisms for channeling accurate information to them before they spoke – then we might say, the corporation as a whole was reckless even if no individual natural person was.

Again, few courts have really endorsed that theory of corporate scienter, but there are some cases where it’s happened.

That’s a Category (2) definition of what it means for a corporation to harbor scienter.

Now let’s return to Category (1).  You start by identifying an actual human being who genuinely intended to mislead investors or was genuinely reckless as to whether investors were misled.  But we also need to impute their state of mind to the corporation.  What justifies that, equating a person with the corporation?  And here’s where there tends to be some ambiguity.

One question that comes up is whether anyone in the org chart can have their scienter imputed to the corporation, or whether instead they have to occupy a certain management-level set of responsibilities.  For example, what if the person was an officer of a subsidiary, but not the parent company responsible for corporate communications?  In Pugh v. Tribune Co., 521 F.3d 686 (7th Cir. 2008), the Seventh Circuit said no, subsidiary employees will not be deemed to be agents of the parent, and their scienter will not be imputed to the parent, but not all courts have approached the problem that way.

Another, separate, question is whether it’s enough for someone to sit quietly, silently in their office while intending to defraud people, or whether instead they need to actually do something to cause the false statement to issue before their scienter will be imputed to the corporation.  Almost all courts agree they need to actually do something, i.e., they need to have in some way caused the false statements at issue before their scienter will be imputed to the corporation, but if they need to do something, what do they need to do? 

Certainly, if they speak directly to the public on the corporation’s behalf, while harboring scienter, every court will agree their state of mind will be imputed to the corporation, but what if they did something less than that?  What if, for example, they knowingly provided false information to someone for inclusion in a public statement?  Or knowingly approved a statement issued by someone else? 

Several courts have said that a corporation acts with scienter if someone in their organization approves false statements with scienter, or – with scienter – furnishes false information with the intention it be distributed to the public, see Southland Securities Corp. v. INSpire Insurance Solutions, Inc., 365 F.3d 353 (5th Cir. 2004); In re Omnicare, Inc. Sec. Litig., 769 F.3d 455 (6th Cir. 2014), but (and this is something I discuss in Slouching Towards Monell), in practice, courts get very uncomfortable when the person who did the furnishing is a low-level actor, and usually resist imputing the person’s state of mind to the corporation in those circumstances.  They might – going back to Pugh– say that the person was simply too low level or too far removed from management to matter, or they might say that a lower level person didn’t mean to defraud investors but only intended to defraud their bosses, which would mean they didn’t in fact harbor scienter at all, thus eliding the problem.  E.g., Pipefitters Local No. 636 Defined Benefit Plan v. Zale Corp., 499 Fed. Appx. 345 (5th Cir. 2012); Doshi v. General Cable Corp., 823 F.3d 1032 (6th Cir. 2016).

What if a person didn’t approve or furnish false information for public distribution, but instead (acting with intent or recklessness) simply failed to correct information that someone else distributed?  If the employees were in different parts of the organization and had nothing to do with each other, and the silent person had no responsibility for communications, that might look more like a failing Category (2) scenario, where you’re trying to infer corporate fault by cobbling together the knowledge of multiple employees.  But if our silent person was literally standing next to the speaker at the time and failed to correct them, that failure might be enough of an action on the part of the individual to justify imputing the single individual’s intent to the company.  See Barrie v Intervoice-Brite, 409 F.3d 653 (5th Cir. 2005).

Anyhoo, I go through this because I think that kind of runs the gamut of possibilities of what it means, definitionally, to say that a corporation acted with scienter.

But in most 10(b) cases, the question isn’t whether the corporation acted with scienter; the question is whether the plaintiff’s complaint raises, for 12(b)(6) purposes, a strong inference of scienter. 

And before you can even answer that question, you need to define your terms.  Which of the above possibilities, constitutes corporate scienter substantively?  Once you know that, you can then engage with the question whether the complaint contains enough facts to suggest that corporate scienter exists.

So, suppose you define corporate scienter to include Category (2).  Then, a complaint might plead corporate scienter by alleging sufficient facts to demonstrate that there was a false statement, that the true facts were known within the organization, and that internal controls were virtually nonexistent or extraordinarily poorly designed so that despite the facts being generally available, they may not have reached the person responsible for communications.

Suppose you define corporate scienter as Category (1), and you further define that the relevant person must only have approved the false statement, even if they did not make the statement themselves.  Then, the question becomes, what kinds of facts raise a strong inference that such a person exists, i.e., a person who approved the statement with scienter?

The complaint could identify the person, by name, and allege their role in approving the statement and their awareness of, or access to, truthful information.

But the complaint might also simply allege that the truth was so well known in the organization – including to all persons responsible for approving corporate communications – that there is a strong inference that there must have been at least one person who harbored the relevant scienter and took the relevant action.  Sure, in later stages of the case, the plaintiff will likely to have to identify who that person is and pin responsibility on them in order to prove the case, but for pleading purposes, the complaint may raise the specter that some such person exists without naming them.  And that’s exactly what the Seventh Circuit held in Makor Issues & Rights, Ltd. v. Tellabs, Inc., 513 F.3d 702, 708 (7th Cir. 2008); see also Teamsters Local 445 Freight Div. Pension Fund v. Dynex Capital Inc., 531 F.3d 190 (2d Cir. 2008). 

What’s significant here is that the facts necessary to allege Category (2) – that the corporation as a whole was reckless because the information was rattling around and never got to the right person – and the facts necessary to allege Category (1) – that the corporation was reckless because the facts were so well known that someone in authority must have known the truth and taken action to conceal it – may be very similar sets of facts.   But the inferences to be drawn from them are different: you either infer that someone responsible must have known the truth, or you infer that if no one responsible knew the truth, the corporation as a whole was at fault.

And this is why courts get confused about it.  They do not start with the definition of scienter; they start simply by asking whether the plaintiff has raised a strong inference of scienter, without defining terms.  And then, they might treat the failure to identify a specific, named person in the complaint as the “Agent With Scienter” as though the plaintiffs are arguing for a definition of corporate scienter that includes (2), and elide the possibility that even though there is no specific “Agent With Scienter” named in the complaint, plaintiffs may still have alleged sufficient facts to satisfy Category (1), with the relevant agent to be identified later in the case.

Courts might say, for example, they have doubts about the logic of Tellabs and Dynex – cases that permit corporate scienter to be alleged without naming an individual who harbored scienter – but when you read closely, it is not entirely clear whether these courts disagree as a pleading matter, which is to say, they think the PSLRA requires more specificity, or whether instead they disagree because think that there should be a more demanding standard than Tellabs and Dynex imply with respect to the actions a person must have taken before their scienter may be imputed to the company (i.e., approval of a false communication is not sufficient), or whether instead these courts simply equate failing to identify a scienter-harboring individual in the complaint with a definition of corporate scienter that does not hinge on individual-level scienter at all. 

Anyhoo, I mention all of this now because I’ve seen several recent examples of courts discussing corporate scienter while mixing up their terms, conflating pleading with proof, failing to articulate why some actors’ scienter will be imputed to the company and others’ won’t (typically lower level actors versus higher level ones), and treating the failure to identify a single person with scienter in the complaint as though plaintiffs are arguing that no such person need exist in order to establish 10(b) liability.

And.  I wish they wouldn’t do that.