Class actions have dual natures. They start out as only individual cases, but they can become massive, collective cases where the rights of absent parties are adjudicated all at once. In most respects, class certification provides the clearest dividing line between when a case is a just a case and when it is a full-blown class action. When that line gets blurred, however, strange results follow.

New York Court of Appeal Decision Requires Needless Notice of Individual Settlements in Putative Class ActionsConsider the New York Court of Appeals’ recent decision in Desrosiers v. Perry Ellis Menswear, LLC. This decision came from consolidated appeals that both raised this question: Under CPLR 908, must notice of an individual settlement be given to the class even if no class is certified? In both cases, the plaintiffs styled their complaints as class actions, but both cases were settled on an individual basis. Even though neither case had a class certified and even though neither settlement purported to affect the rights of absent class members, the cases came to the Court of Appeals from orders requiring that notice be given to class members before the cases could be closed.

The Court of Appeals affirmed, holding that CPLR 908 requires in such circumstances that notice go out to all members of an uncertified class. In other words, in New York, once a case has been styled as a class action, the defendant cannot enter into an individual settlement with the named plaintiff without providing notice to the entire class.

The court reached this conclusion by finding a textual ambiguity. CPLR 908 states:

A class action shall not be dismissed, discontinued, or compromised without the approval of the court. Notice of the proposed dismissal, discontinuance, or compromise shall be given to all members of the class in such manner as the court directs.

The court found the terms “class action” and “the class” to be ambiguous because the terms could mean “a certified class action” or something else such as “a case whose complaint contains class allegations.” It opted to adopt the second meaning, drawing on two sources: federal cases interpreting a previous version of a federal rule and a previous New York lower-court appellate decision from 1982. Neither compelled this conclusion—in particular, even the previous version of Federal Rule 23 was interpreted by most courts to make notice optional in the district court’s discretion.

This case creates serious practical problems for individual settlements.

  • First, who gets notice? Even when drafted by excellent counsel, initial class definitions in complaints are seldom clear enough to specifically identify an exact group of actual people. Some definitions are fail-safe, such as “all persons injured by defendant’s defective product.” Others define unascertainable classes, such as “all persons who relied on defendant’s television advertisements to purchase a product.” In a certified class action, the class definition receives judicial scrutiny and narrowing, which almost always results in changes to the class definition that make it narrower, clearer, and more workable. A major reason for individual settlement in many putative class actions is that the actual members of the class are unknown, impossible to identify in a feasible, cost-effective and efficient way, or not reasonably locatable for purposes of notice.
  • Second, notice costs money. Mailers and advertisements impose real costs that raise the price of litigation, and imposing these costs on individual settlements either raises the cost of settlement (to the defendant’s detriment) or reduces the amount available to the named plaintiff (to the plaintiff’s detriment).
  • Third, what is notice supposed to accomplish? The putative class members’ rights are not affected by an individual settlement, and they would have no standing to object to an individual settlement.
  • Fourth, and most troubling, sending notice of a settlement invites copycats. This ruling basically requires the defendant to run an advertisement that says to any list of people a plaintiff may name as a class “here is how much I paid someone like you for suing me.” That is not a public policy that promotes settlement.

It is not clear that this needless notice rule benefits any party in litigation. We expect that defendants will now refuse to settle putative class actions on an individual basis under this rule.  Why would they settle? The costs of notice and the burden of identifying every class member are incentives to settle putative class actions individually. If a defendant cannot get those benefits, it is has a much stronger incentive to fight. The court may have been thinking that by imposing requirements traditionally reserved for class settlements and litigation of certified classes on purely individual settlements, it has given defendants a stronger incentive to settle on a class-wide basis in order to get class-wide peace, since notice is going to have to be given to the class anyways. But that policy would only make sense if class settlement were a proper judicial or legislative goal in and of itself, which it is not.

The defendant also has all the more reason to remove cases in light of this decision. Desrosiers is limited to New York state court. While the language of CPLR 908 is based on the old federal Rule 23(e), the federal rule was clarified in 2003 to make doubly clear that notice is only required in a certified class or where the proposed settlement would bind class members.

Last month, in Arkansas Teachers Retirement System v. Goldman Sachs Group, Inc., the Second Circuit vacated the Southern District of New York’s order certifying a class in a Rule 10b-5 securities fraud class action. At issue was whether defendants had rebutted the so-called “fraud on the market” presumption of reliance that applies in securities fraud class actions by showing that the alleged misrepresentations did not have any impact on the price of the defendant company’s stock. In reversing the district court, the Second Circuit held that the district court had improperly required defendants to come forward with evidence that “conclusively” proved the “complete absence” of price impact, when, according to Second Circuit, a “preponderance of the evidence” will do. Moreover, the Second Circuit criticized the district court for failing to consider certain aspects of defendants’ price impact evidence, making clear that the court should have considered any evidence that could sever the link between an alleged misstatement and a stock price movement. In so doing, Goldman is likely to become a helpful tool for defendants challenging class certification in 10b-5 cases.

What is the “fraud on the market” presumption?

Just as it is with common-law fraud claims, reliance is an essential element of a securities fraud claim brought under Rule 10b-5. Plaintiffs must prove that they relied upon an alleged misrepresentation when determining whether to purchase or hold a given security. Without such reliance — in other words, where the evidence shows that the plaintiff would have made the same investment decision even if the alleged misrepresentation had not been made — you cannot say that a plaintiff’s loss is the result of the alleged misrepresentation.

As a general matter, reliance is an inherently individualized inquiry, which is why the run-of-the-mill fraud claim is usually not susceptible to class treatment  (see e.g., McLaughlin v. Am. Tobacco Co., 522 F.3d 215 (2d Cir. 2008). But securities fraud claims under 10b-5 are different.

In Basic Inc. v. Levinson, 485 U.S. 224 (1988), the Supreme Court adopted a presumption that the price of a stock traded in an efficient market reflects all public, material information—including misrepresentations—and that investors rely on the integrity of the market price when they choose to buy or sell stock. Because of the Basic presumption—aptly named the “fraud-on-the-market” presumption—courts may presume that investors relied upon all public, material misrepresentations through their “reliance on the integrity of the price set by the market.” To invoke the Basic presumption, plaintiffs must show that:

  1. The defendant’s stock traded in a generally efficient market.
  2. The alleged misrepresentations were publicly known.
  3. The relevant stock purchase occurred between the time the misrepresentations were made and the truth was revealed.

Although the Supreme Court in Basic also recognized that defendants can rebut the fraud-on-the-market presumption by showing that the alleged misrepresentation did not actually distort the stock price, after Basic, few lower courts actually permitted defendants to make such a showing. Most often, courts found that the question of whether an alleged misrepresentation had impacted the stock price was too closely tied to “merits” issues (such as materiality or loss causation) to be addressed at the class certification stage. As a result, following Basic, class certification became almost a foregone conclusion in securities fraud actions, particularly for cases involving stocks traded on major exchanges such as the NYSE or Nasdaq. (Stocks traded on OTC markets were a different story.)

The altogether unsurprising result of making class certification a fait accompli in most securities fraud suits? An incredible proliferation of securities fraud class actions—and with it, an ever-increasing number of high-dollar settlements following class certification.

In Halliburton II, SCOTUS reaffirmed the “fraud on the market” presumption, but made clear that the presumption can be rebutted.

Over the years, many economists and scholars have disputed the theoretical underpinnings of the presumption—that the stock price in an “efficient” market reflects all public material information or that all (or even most) investors rely upon the integrity of the price. The defense bar took issue with how Basic seemed so out of place with the rest of the Supreme Court’s class action jurisprudence, which generally rejected presumptions. Public companies (and insurers) bemoaned the explosion of often-dubious securities fraud claims and the (unfair) settlement pressure that resulted from rubber-stamp class certifications. But for more than 25 years, the Supreme Court did not revisit Basic.

That changed in 2014 with Halliburton Co. v. Erica P. John Fund, Inc. In that case (commonly called “Halliburton II”), the Supreme Court took up two issues:

  1. Whether to abandon (or modify) Basic’s fraud-on-the-market presumption in securities fraud cases.
  2. Whether (and how) securities fraud defendants can rebut the Basic presumption at the class certification stage.

In taking up the case, the Court was presumably motivated by its decisions just a few years earlier in Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338 (2011), and Comcast Corp. v. Behrend, 569 U.S. ___ (2013), which clarified that class certification requires a “rigorous analysis” of Rule 23’s requirements, and that this rigorous analysis may overlap with merits issues – arguably at odds with Basic’s presumption of reliance.

On the first issue—whether to jettison the fraud-on-the-market presumption—the Supreme Court decided that Halliburton had not shown “special justification” needed to overcome stare decisis. But on the second issue, the court reaffirmed that defendants may rebut the Basic presumption by showing that the alleged misrepresentations did not actually distort the market price of the company’s stock and clarified that this rebuttal may take place at the class certification stage (thereby defeating class certification). Even though examining a misrepresentation’s price impact may dovetail with questions of materiality and loss causation, the court saw no reason to defer it to a later stage of the case. The court thus rejected numerous decisions—including the Fifth Circuit’s decision below—that had refused to consider price impact evidence at the class certification stage.

Since Halliburton II, lower courts have struggled to determine precisely how a defendant can rebut price impact.

Although making clear that defendants could defeat class certification in 10b-5 cases by showing an absence of price impact, Halliburton II unfortunately did not explain how defendants can make that showing. Among the many issues left unaddressed were the defendants’ burden of proof in rebutting the fraud-on-the-market presumption, as well as the types of evidence that would be admissible (and sufficient) to meet that burden, whatever it may be. Does the defendant bear the burden of persuasion in rebutting the presumption, or does the plaintiff bear that burden in obtaining the presumption?  Would evidence of no statistically significant price movement following the alleged misrepresentation be sufficient, or would a defendant also need evidence of no price movement following the alleged corrective disclosure? Should courts ever consider evidence about the absence of price movements on days other than the alleged misrepresentation and corrective disclosure days, such as on days where the purported truth was first revealed to the market? Should courts consider alternative explanations for a price movement following an alleged misrepresentation or corrective disclosure?

Lower courts have struggled to reach clear, consistent, and coherent answers to these questions; unfortunately, in most cases addressing price impact since Halliburton II, courts have restricted the defendants’ evidence in meaningful ways. For example, in Aranaz v. Catalyst Pharm. Partners, Inc., 302 F.R.D. 657 (S.D. Fla. 2014), the court refused to consider evidence that showed that the market already knew the information that was allegedly misrepresented, even though if the market already knew the “truth,” then the alleged misrepresentation or corrective disclosure could not possibly have added to the total mix of information available to the market and thus could not possibly have impacted the stock price. And in Goldman, the district court rejected evidence about the absence of a price movement when the purported truth was revealed, reasoning that such evidence went to materiality, which, under Amgen Inc. v. Connecticut Retirement Plans & Trust Funds, 133 S. Ct. 1184 (2013), is not supposed to be addressed at the class certification stage. With these sorts of evidentiary limitations, it is no wonder that the vast majority of courts that have addressed price impact since Halliburton II have found that the defendants failed to rebut the fraud-on-the-market presumption.

In Goldman, the Second Circuit made clear that courts should consider all evidence that could bear on the price impact of an alleged misrepresentation, even if that evidence is also probative of materiality.

Fortunately for securities fraud defendants, the tide may be starting to turn. On appeal in Goldman, the Second Circuit reversed the district court’s grant of class certification based (in part) on the district court’s refusal to consider evidence about the absence of a price impact when the truth was revealed, even though that evidence also “touche[d] on materiality.” As the court explained, price impact “differs from materiality in a crucial respect”—“[i]f a defendant shows that an ‘alleged misrepresentation did not, for whatever reason, actually affect the market price’ of defendant’s stock, ‘there is no grounding for any contention that the investor indirectly relied on that misrepresentation through his reliance on the integrity of the market price.’” Accordingly, the Second Circuit’s opinion makes clear that district courts should consider all evidence that could bear on price impact, even if that evidence is also potentially probative of materiality. The court was also bothered by the fact that the district court failed to hold an evidentiary hearing or hear oral argument—which was at odds with the district court’s duty under Wal-Mart and Comcast to “rigorously” scrutinize class certification.

Not everything in Goldman was helpful for securities fraud defendants, however. The Second Circuit also agreed with the district court that defendants bear the burden of persuasion to rebut the fraud-on-the-market presumption—which would mean that plaintiffs do not bear the burden of persuasion to invoke the presumption. In so holding, the Second Circuit split from the Eighth, which in IBEW Local 98 Pension Fund v. Best Buy Co., 818 F.3d 775 (8th Cir. 2016), held that under Federal Rule of Evidence 301, defendants merely bear the burden of production in rebutting the Basic presumption—plaintiffs retain the burden of persuasion. The Second Circuit did, however, disagree with the district court’s conclusion that defendants had to meet their burden with “conclusive” proof of the “complete absence of price impact.” Showing the absence of price impact by a mere “preponderance of the evidence” is sufficient, according to the court.

Despite the tension between Goldman and Best Buy, the Second Circuit’s decision is a helpful tool for defendants seeking to defeat class certification in dubious securities fraud cases. The principal takeaway from Goldman is that in opposing class certification in 10b-5 cases, defendants should be prepared to offer any evidence that could potentially sever the link between the alleged misrepresentations and an impact on the stock price—and to use the Goldman decision to convince district courts to consider that evidence. This includes evidence about price movements (or the lack thereof) on days not at issue in the complaint, but that could be probative of the market’s reaction to the revelation of the purported truth that was misrepresented or withheld. This also includes evidence showing plausible alternative explanations for statistically significant price movements following an alleged misrepresentation or corrective disclosure.

Goldman is now back before the district court on remand, where the court is once again considering whether to certify a class. We will continue to monitor the case to see how the district court applies the Second Circuit’s price impact guidance.

The Supreme Court’s decision last summer in Bristol-Myers Squibb Co. v. Superior Court of California, 137 S. Ct. 1773 (2017), is my pick for “2017 Class Action Practitioners’ Case of the Year”––and it’s not even a class case.

One of the most exciting areas in the class action arena is personal jurisdiction. Stifle that yawn. The Supreme Court’s recent decisions in this area could have a major effect on the scope of class actions where the claims sound in state law. Plaintiffs’ lawyers have long sought to bring class actions in the most plaintiff-friendly venues. As a general proposition, a plaintiff would prefer to cram a big nationwide class of claims into the most favorable venue instead of having to bring either several single-state class actions in several venues or a nationwide class in a less-favorable venue. Where plaintiffs are successful, this kind of forum shopping makes class cases all the more difficult for defendants: Not only do defendants have to face the high stakes of a large number of aggregated claims in one action, they also have to fight all those claims in a hostile venue.

The Supreme Court’s recent decision in Bristol-Myers Squibb, has given defendants a powerful new argument against this tactic. Bristol-Myers Squibb involved a mass action in which nearly 600 plaintiffs who did not live in California joined their claims with California residents’ claims. The defendant challenged personal jurisdiction, but the California courts found specific jurisdiction to exist. By an 8-1 vote, the United States Supreme Court reversed, holding that the California courts could not exercise personal jurisdiction over the defendant as to claims of non-resident defendants who could not satisfy the well-settled test for specific jurisdiction. As a result, the plaintiffs face the choice of either bringing their nationwide class in a forum where the defendant is subject to general jurisdiction (i.e., state of incorporation of principal place of business, thanks to the Supreme Court’s recent Daimler decision) or limiting the class to those plaintiffs whose claims involve the kind of minimum contacts necessary to support the exercise of specific jurisdiction.

As mentioned above, Bristol-Myers Squibb is not a class action case––but that is likely of no consequence. The decision rests on the constitutional due process rights of defendants, and, under the Rules Enabling Act, the purely procedural class action rules cannot abrogate those rights. (We’ll address this point further in a future post, along with looking at how lower courts have applied Bristol-Myers Squibb in the class context.)

Practically speaking, Bristol-Myers Squibb should curb the most blatant forum shopping in diversity cases. Defendants should use it to channel more litigation to the jurisdictions where defendants are incorporated or headquartered or where the plaintiffs reside. The case may also lead to an increase in single-state class cases, which is a potential downside—but defendants can waive personal jurisdiction as a defense to permit consolidation where appropriate. The increase in state-specific suits may also make removal under diversity or CAFA easier, as plaintiffs will be less free to use joinder to defeat diversity jurisdiction and will likely run afoul of the no-similar-case rule of 28 U.S.C. § 1332(d)(4)(A)(ii). We also predict an uptick in multidistrict-litigation petitions, which both plaintiffs and defendants can perceive as an advantageous way of consolidating geographically dispersed litigation.