Saturday, April 11, 2020
I first blogged about this case back when the Supreme Court’s Halliburton Co. v. Erica P. John Fund, Inc. (Halliburton II), 573 U.S. 258 (2014) decision was new, and now we finally have some answers.
In Halliburton II, the Court held that while securities class action plaintiffs get the benefit of a presumption that any material information – including false information – impacts the price of stock that trades in an efficient market, defendants may, at the class certification stage, attempt to rebut that presumption with evidence that there was no such price impact.
The complicating factor in all of this is that, per Amgen Inc. v. Connecticut Ret. Plans & Tr. Funds, 568 U.S. 455 (2013) and Erica P. John Fund v. Halliburton Co., 563 U.S. 804 (2011) (Halliburton I), defendants cannot rebut that evidence by demonstrating either a lack of materiality or a lack of loss causation. This, of course, severely ties defendants’ hands, because those are the usual proxies for price impact. (See prior blog posts for further discussion here and here and here and here.)
The Goldman case has an interesting twist, though. The basic allegation is that Goldman falsely represented that it behaved with honesty and integrity toward its clients, and its lies were revealed when the SEC filed an enforcement action alleging that Goldman’s transactions involving CDOs were riddled with undisclosed conflicts of interest. The SEC’s complaint triggered a price drop, and Goldman investors suffered as a result.
Goldman, however, argued it could rebut the presumption of price impact by demonstrating that at multiple times throughout the class period, the media reported on its conflicts, with no market reaction. Therefore, argued Goldman, it was clear that the initial lies did not impact prices, and the price drop at the end of the class period was not due to the revelation of the truth – all of which was known to the market – but simply due to the SEC’s enforcement action itself.
As I previously argued, much of this was really a disguised attempt to relitigate the materiality of the initial statements, but in a 2018 appeal, the Second Circuit remanded to the district court to give Goldman the opportunity to prove lack of price impact by a preponderance of evidence. See Ark. Teachers Ret. Sys. v. Goldman Sachs Grp., Inc., 879 F.3d 474 (2d Cir. 2018). The district court recertified the class, Goldman appealed again, and the Second Circuit’s affirmance, 2-1, issued earlier this week, engages more closely with the substance of Goldman’s argument.
Okay, that sets the stage. What actually happened here?
(More under the jump)
There was a lot of expert evidence going back and forth but the main thrust Goldman’s argument was that (1) truth was revealed throughout the class period, (2) without causing a stock price reaction, which (3) meant that the false statements had never affected price in the first place, and (4) the price drop at the end of the period was not due to any new information.
But the district court concluded that the SEC’s complaint was so detailed and credible as compared to earlier revelations that there remained the possibility that it was those new facts, and not solely the enforcement action itself, that contributed to the price drop at the end of the period. Which in turn meant the earlier disclosures did not reveal the full truth, which in turn explained why there was no market reaction to those earlier disclosures, which meant the presumption of price impact remained intact. On appeal, the majority declined the invitation to second-guess the district court’s interpretation of the expert evidence.
But there was more to it.
A lot of the jousting in this case concerned the validity of the price maintenance theory of fraud on the market. My earlier posts about Halliburton discuss this theory, but the idea is that a lot of frauds aren’t designed to push stock prices up; they’re designed to keep stock prices level, typically because something bad has happened and the company doesn’t want to admit the truth. So, it lies, investors believe there are no changes, and the stock price remains unmoved when it otherwise would have fallen. In this situation, there’s no price movement at the time of the lie to assess, which is why litigants are forced to reason backward by watching the market reaction upon the revelation of the truth to infer the effects of the earlier fraud. And that’s why everything’s so confuzzled: Price reaction upon revelation of the truth is a much messier factual inquiry than an examination of the effects of the initial lie.
Nonetheless, that’s how most frauds are designed, and in In re Vivendi, S.A. Sec. Litig., 838 F.3d 223 (2d Cir. 2016), the Second Circuit explicitly endorsed the price maintenance theory of fraud.
The price maintenance theory underlay the plaintiffs’ claims in Goldman: There was no upward movement when Goldman made its statements, so plaintiffs claimed that the statements maintained Goldman’s stock price at artificially-inflated levels, until the truth was disclosed.
Goldman argued that price maintenance claims require false statements about some kind of concrete operational results, and cannot be pursued based on the kind of vague, aspirational statements at issue in this case. In response, the majority agreed with my earlier assessment that all of this was backdoor attempt to import a prohibited materiality inquiry into the class certification process:
Goldman is not formally asking for a materiality test. But its “special circumstances” test would commandeer the inflation-maintenance theory by essentially requiring courts to ask whether the alleged misstatements are, in Goldman’s words, “immaterial as a matter of law.” This is the precise question posed by materiality.
Goldman’s authority for what constitutes an impermissibly “general statement” provides further evidence that its “special circumstances” test is really a means for smuggling materiality into Rule 23. Its brief contains a table of nearly a dozen cases holding that “general statements . . . about business principles and conflicts controls are too general to cause a reasonable investor to rely upon them.” But every one of these cases is the dismissal of a securities claim under Rule 12(b)(6)…
Goldman’s alternative argument was that for price maintenance theory to work, there has to be some evidence of inflation to begin with. I.e., if plantiffs are arguing that the stock dropped to its unmanipulated levels upon revelation of the truth, there must be some evidence that the stock price was artificially high first, and if it didn’t come from the false statements – because they merely “maintained” preexisting inflation – where did it come from?
For example, in Vivendi, the company’s liquidity position deteriorated, so that statements about its financial position that once were true, and led the market to have confidence in the company, increasingly became untrue as the problems worsened.
The Second Circuit held that all that matters is that the price was inflated – regardless of how it got that way initially – and we can infer such inflation from the fact that the price dropped upon revelation of the truth. Nothing more is required.
Still, it’s unsatisfying not to at least have an idea of where the inflation came from. To be sure, there’s precedent for ignoring that issue; in Glickenhaus & Co. v. Household Int’l, Inc., 787 F.3d 408 (7th Cir. 2015), the Seventh Circuit held:
Actual inflation could have resulted from prior misrepresentations or just from the company's fundamentals having deteriorated without investors knowing about it. How the stock became inflated in the first place is irrelevant because each subsequent false statement prevented the price from falling to its true value and therefore caused the price to remain elevated.
All of which is fair, but in Glickenhaus, the allegation was essentially that the quality of the loans in a mortgage company’s portfolio was far worse than previously understood and the company used accounting tricks to cover up that fact. Like Vivendi, it’s not hard to see how the market became misled in the first instance; the company had previously been reporting its financial condition and business model, so either those statements were false originally, or things got worse over time but the statements didn’t, and either way, the market price ended up diverging from the company’s true value.
But the Goldman story is a bit less clear. According to the plaintiffs’ briefing, their theory was that “Goldman’s stock traded at a premium compared to its peers because of the firm’s reputation for integrity and its ability to manage conflicts among its various departments,” and that when Goldman “abandoned the practices that had earned it that reputation, the justification for that premium evaporated.”
But we still don’t know how the market came to understand Goldman in this manner. Was it prior statements by Goldman, or more like a long course-of-conduct thing? If the latter, it does seem odd that relatively shallow statements about integrity were deemed to be the sole basis for “maintaining” that much deeper understanding of Goldman’s business model, such that a stock price drop upon the revelation of the truth was attributed to the false statements (and the SEC enforcement action), and not to a wholesale reevaluation of Goldman’s business model based on impressions that had been developed independent of Goldman’s self-praise.
But that’s a question of what Donald Langevoort would call counterfactuals: Namely, if Goldman’s statements “maintained” its price, should we compare that “maintained” price to what would have happened if Goldman told the truth (“we actually exploit conflicts to our clients’ disadvantage!”) or to what would have happened if Goldman had remained silent and allowed the market to draw its own (wrong) conclusions?
In Vivendi, the Second Circuit said the former, because once a company chooses to speak, it has a duty to speak truthfully and completely. So, even if Goldman’s stock price was high initially because it had developed – entirely fairly – a reputation for integrity, and even if it would have stayed that high, unjustifiedly, all by itself without Goldman saying a word, once Goldman decided to secretly behave without integrity and lie about it, no matter how anodyne the lie – and no matter how much of its reputation was still based on its actions rather than its words – Goldman became duty-bound to disclose the truth, and is responsible for the full damages associated with failing to do so.
But if that’s the theory, it’s still not clear what actual stock price effect the lie itself had, or is being presumed to have had. Very possibly, no effect at all! After all, if speaking and remaining silent would have the same effect on stock prices, that does seem to suggest the speech was … immaterial. And at this point, it doesn’t matter, because what’s deemed to be influencing stock prices – where the damages are actually coming from – is not the lie itself, but Goldman’s own silence, now rendered fraudulent by its decision to speak.
To be sure, this isn’t necessarily a problem limited to price-maintenance cases; it’s true of many situations where the lie is a fairly minimal statement, and the truth is some kind of underlying massive scheme. What’s fooling the market in these cases is not so much the lie itself, but an assumption of business regularity.
That’s an argument I’ve made in other spaces, i.e., that a lot of the time, it seems less like the statements are doing any work than investors’ simple assumptions about how companies ordinarily operate, with the statements simply confirming their impressions. (See Searching for Market Efficiency, and a related sort of discussion in this blog post about a district court Affiliated Ute decision – a case that the Ninth Circuit recently agreed to hear on interlocutory appeal.)
That said, in many cases, companies don’t really have the option of remaining silent. Either they’re required to disclose something by SEC rule, or they have such a long tradition of speaking on a subject that an unexplained silence would draw attention. In Goldman’s case, it appears that the company had been attesting to its integrity since well before the class period. Had it suddenly … stopped, the market would have probably noticed, and reacted. And if that’s the case, then for sure, yeah, the class period statements were maintaining its stock price.