Before I started writing full-time at National Review in 2020, I spent over two decades in private law practice, the bulk of it defending investor class actions under the federal securities laws. So, this morning’s U.S. Supreme Court decision in Goldman Sachs Group Inc. v. Arkansas Teacher Retirement System adds to a body of law I know well. Allow me to explain why Goldman Sachs is a positive step and modestly good news for the defense bar, even though defenses against class certification remain an uphill battle in practice.
Having Some Class
Some background in this area is necessary to understand what exactly Goldman Sachs changes. Decisions on whether or not to certify a case as a class action are procedurally unusual: The decision belongs to a judge, not a jury, but unlike deciding whether a complaint can go forward, the judge does not just presume that the plaintiff’s factual allegations are all true and test them only against the law. Instead, the parties can submit evidence. In investor class actions, that frequently involves battles of expert economists analyzing the movements of stock prices. The goal of class-certification rules, as with any rule of law, is to make it possible and workable for a court to actually tell apart the cases that should be class actions from those that shouldn’t. If one side or the other will always win, it’s not really a rule. And there is a lot of money at stake in deciding which cases are proper class actions, because most cases will settle if they are certified as class actions, even if the defendant would stand a good chance at trial.
Investor class actions have long been one of the mainstays of the federal bar. For years, they were easily the largest category of class actions in federal court, although that distinction has been eroded since the 2005 expansion of federal jurisdiction over national class actions. Yet the main statute — Section 10(b) of the Securities Exchange Act of 1934 — does not say anything about private damages lawsuits for false or misleading statements to the stock market, let alone whether or how investors are supposed to show that they relied on those statements when they bought a stock. Section 10(b) was written as a criminal and regulatory statute, not a source of civil lawsuits. Rule 10b-5, the SEC rule that implements it, was drafted in an afternoon in 1942 without a moment’s discussion of private-damages lawsuits.
For decades, the courts have just been making up as they go along. Then–Justice Rehnquist in 1975 described Section 10(b) cases as “a judicial oak which has grown from little more than a legislative acorn.” But stare decisis in reading statutes is strong, and because Congress for half a century has gradually written rules about how to decide Section 10(b) lawsuits, the courts have treated that as congressional consent to let judge-made rules fill the gaps. Legitimacy-minded conservative judges have come up with a variety of next-best-thing limiting principles, such as looking at what similar statutes say, to keep this from becoming an exercise in pure judicial creative writing.
What Moves Markets
In a common-law fraud case, a plaintiff can only sue over false or misleading statements that the plaintiff actually relied upon to his detriment. But most statements to the markets are not read by most investors. Also, a lawsuit about what each investor relied on cannot proceed as a class action, because it presents issues individual to each investor.
To help plaintiffs pursue their lawsuits as class actions, the Court has established a series of legal fictions unique to Section 10(b) cases. Indeed, they are found almost nowhere else in either statutes or the common law. Under a framework beginning with the 1988 Basic v. Levinson decision, written by that master of judicial invention Justice Harry Blackmun, investors can be presumed to rely on an efficient stock market to interpret the statements of companies and incorporate them into the stock’s price. Under Basic, proof that a stock’s price regularly reacts to new information can show that the market for that stock is efficient, whether or not the stock price reacted to the particular statements at issue in the case. If the plaintiff shows that the market for the stock is efficient, the Basic presumption of reliance applies, and the class will be certified.
On this initial step, there may be a battle of expert economists to dispute whether the market for a particular stock is actually efficient. That may be a big issue if the security in question is thinly traded or esoteric, but it is usually easy for a plaintiff to prove in the case of stocks that are heavily traded on public markets, to the point where defendants will often not contest market efficiency. One of my biggest wins in practice was a case in 2011–12 in which we defended the former CEO of Freddie Mac and persuaded the court that the market for one $3.7 billion class of Freddie Mac’s preferred stock — which crashed when the government took over Freddie Mac and Fannie Mae in September 2008 — was not efficient because the price did not regularly react to news. It was the first time a court had ever decided, in a contested hearing, that a stock traded on the New York Stock Exchange did not trade in an efficient market. In most cases, however, that is not the real battleground.
Are Defendants Defenseless?
Suppose that a stock’s price normally reacts to news. Does that mean a court will always certify a class? Or does it matter if the stock’s price didn’t move at all when the allegedly fraudulent statements were made? In a trio of cases over the past decade and a half, the Court has said that the plaintiff does not need to show at the class-certification stage that the statements would normally have been material to an investor, but that a defendant could rebut the presumption by showing that the stock’s price did not actually react to the statements involved in the case.
That may sound like a clear answer, but it is not. In asking whether a stock’s price reacted to news, it is easy enough for courts to assess the testimony of economists who use well-known measuring techniques. The economist reviews when the news first hit the market, whether the stock’s price moved soon after, if it moved in the same direction as the news (i.e., it didn’t drop on good news or rise on bad), if its movement was large enough to be statistically significant, and if the movement could be explained by other factors (e.g., the stock moved along with the market or with comparable stocks in its industry, or the price movement could be attributed to other information about the company released at the same time). Only the last part of that — how much to attribute the price’s movement to other news at the same time — involves a matter of more judgment than statistical proof.
Where this really gets complicated is when the plaintiffs say that the price would have dropped by some particular amount (which constitutes the damages they claim), but that the defendant’s statements prevented it from moving at all. They then file suit over the price dropping later and just assume that this drop was caused by something that a previous statement concealed — even if the previous statement was just some generic corporate statement of being a good company with good policies. At some point, that becomes an unfalsifiable hypothetical. How is a defendant supposed to disprove a negative? And worse, how is a court of law supposed to tell apart the cases where a defendant has disproved a negative from the cases where the defendant hasn’t?
The courts could just reject the “inflation maintenance” theory entirely as being too speculative, and limit suits to cases where the price actually moved when the statements at issue were made. It has done things like this occasionally, such as barring suits by people who say that they would have bought or sold the stock, but didn’t. The Court was not asked to decide the validity of the “inflation maintenance” theory in Goldman Sachs, and left that for another day.
The courts could also stick the burden of proof on the plaintiffs, on the theory that asking a litigant to prove a negative is at least a little less unrealistic than asking one to disprove a negative. But the Court in Goldman Sachs concluded that the logic of its prior decisions had left that burden on the defendant. There was a 6–3 split on that point; Justice Amy Coney Barrett’s opinion for the Court found that Basic and later cases had essentially created a special rule for the presumption of reliance, while Justice Neil Gorsuch’s dissent (joined by Justices Clarence Thomas and Samuel Alito) wanted to apply the framework of Rule 301 of the Federal Rule of Evidence, under which the party who benefits from an evidentiary presumption always bears the ultimate burden of persuasion.
Making the Merits Matter
So much for doctrine. The Rule 301 argument was always something of a long shot, given how Basic and its progeny were written. But if a defendant is allowed to disprove a negative — i.e., show that the price not moving was not caused by what the defendant said — how does that work? An expert can show that the price didn’t react, but how does an expert show what the price would have done?
That is where the facts of the Goldman Sachs case come in. The investors said that they were defrauded by not knowing of various bad business practices at Goldman, mainly before the 2008 credit crisis. But to identify statements that supposedly concealed those practices, the plaintiffs pointed only to vague, generic positive statements about the investment bank:
We have extensive procedures and controls that are designed to identify and address conflicts of interest. . . . Our clients’ interests always come first. . . . Integrity and honesty are at the heart of our business.
Goldman Sachs argued, quite reasonably, that this is not the sort of thing that moves markets. Everybody expects companies to say these things even when they are not true. The Second Circuit, however, concluded that the Supreme Court’s prior decisions had precluded it from considering the nature of the statements at class certification, since the materiality of the statements was a merits issue and not a requirement of proof at class certification. But if the judge can’t look at the price or the statements, what evidence is left?
Justice Barrett’s opinion allowed courts to look at the nature of the statements in order to give at least some tether to reality to the speculative inquiry into what the stock price would have done if the statements had not been made:
The generic nature of a misrepresentation often will be important evidence of a lack of price impact, particularly in cases proceeding under the inflation-maintenance theory. . . . Plaintiffs typically . . . point to a negative disclosure about a company and an associated drop in its stock price; allege that the disclosure corrected an earlier misrepresentation; and then claim that the price drop is equal to the amount of inflation maintained by the earlier misrepresentation.
But that final inference — that the back-end price drop equals front-end inflation — starts to break down when there is a mismatch between the contents of the misrepresentation and the corrective disclosure. That may occur when the earlier misrepresentation is generic (e.g., “we have faith in our business model”) and the later corrective disclosure is specific (e.g., “our fourth quarter earnings did not meet expectations”). Under those circumstances, it is less likely that the specific disclosure actually corrected the generic misrepresentation, which means that there is less reason to infer front-end price inflation — that is, price impact — from the back-end price drop.
This is still something of a half-measure, as it keeps courts in the business of speculation rather than verifiable fact. But it requires courts, when deciding on class certification in “inflation maintenance” cases, to look for a closer connection between the original statements and the corrective disclosures. That offers courts some judicially manageable way to separate the good cases from the bad ones.
In practice, that is a bigger win for the securities defense bar than the ultimate question of who has the burden of proof or persuasion. The burden matters a lot on a question such as “Did other news on the same day cause the price to move?” It matters much less in practice to a court weighing two sets of statements and comparing them, even when the experts may supplement that showing by various comparative tests of the kinds of statements that do — and don’t — move markets. In that sense, Goldman Sachs is good news for helping judges ferret out the weakest and most implausible investor class actions.