Flushable Wipes, Take Three: The Second Circuit Gets Injunctive Standing Right, But Classwide Damages Models Wrong
As our readers know, we’ve kept a close eye on the “flushable wipes” litigation—known variously as Kurtz v. Costco and Belfiore v. Procter & Gamble—as it has bounced between Judge Weinstein’s courtroom in the Eastern District of New York and the Second Circuit. The cases raise several issues important to class-action defendants, including the necessity of a rigorous damages model at the class-certification stage; the availability of injunctive relief to customers who are already wise to the alleged deception; and the appropriateness of massively multiplied “statutory damages” in the class context. We (and others) had hoped that the Second Circuit would use the case to provide clear answers to these questions and to remedy the New York federal courts’ status as a hotbed for questionable class-action complaints. But with that court’s latest ruling—fortunately, an unpublished and non-precedential one—those hopes may have gone down the tubes.
We’ve covered the background of this case in our previous posts, so we’ll keep it brief here. A handful of manufacturers sold pre-moistened sanitary wipes, which they advertised as “flushable.” Some plaintiffs disagreed with that characterization and sued, alleging that they had paid a “price premium” for the defendants’ wipes because of the purportedly false “flushable” description. Back in 2017, the district court certified the case for class treatment under both Rule 23(b)(2) (with regard to the plaintiffs’ claim for injunctive relief) and Rule 23(b)(3) (with regard to the plaintiffs’ claim for statutory damages). As for the latter claim, the district court found that the plaintiffs’ mere proposal to employ a modeling technique called “hedonic regression” satisfied their burden to show that issues common to the class predominated over individualized ones.
In 2019, the Second Circuit granted interlocutory review and sent the case back for the district court to try again, noting that “Plaintiffs’ expert ha[d] not demonstrated, but ha[d] merely alleged, that he [could] apply hedonic regression analysis to establish on a classwide basis whether the class members paid a price premium for Defendants’ products attributable to the ‘flushable’ representation.”
On remand, the plaintiffs’ expert actually performed the regression analysis that he had previously proposed, and—shockingly!—concluded that consumers had paid, on average, a 6–8% “price premium” attributable to the “flushable” claim. The defendants raised a number of serious issues with this regression analysis. Among other things, the plaintiffs’ model relied on incomplete and improper data. The model’s finding of a “price premium” attributable to the “flushable” claim was also at odds with reality, as the ultimate removal of that claim from the product packaging did not lead to any drop in the products’ actual price. Moreover, slight changes to the plaintiff’s model—e.g., with regard to time period studied and control variables used—resulted in major swings in the reported “price premium,” and at times, its disappearance. Such an analysis, the defendants argued, did not satisfy the Supreme Court’s edict that a plaintiff’s damages model must withstand “rigorous analysis” to comply with Rule 23. The district court, however, concluded that these critiques went “to [the model’s] weight rather than [its] admissibility” and once again certified both injunctive and damages classes. A second interlocutory appeal followed, leading to the Second Circuit’s recent decision.
Second Circuit’s Decision
That court disposed of that second appeal in an unpublished summary order. As a threshold matter, the Second Circuit rejected the defendants’ arguments that the remaining named plaintiff was an atypical and inadequate class representative because he had “sacrificed potentially higher-value plumbing damages claims in order to advance lower-value, but more easily certifiable, claims based on a price premium theory,” and because he had “continued to buy the wipes [even] after he learned that they were not flushable.”
The Second Circuit then turned to the district court’s certification of the injunctive-relief claims. Here, the news was better. To have standing to seek injunctive relief, the court explained, the named plaintiff must be at risk of suffering a future injury that is “actual and imminent, not conjectural or hypothetical.” At a bare minimum, such a plaintiff must allege that she “intends to purchase” the challenged product again “in the future.” We always doubted that a stated intent to purchase a product again is truly enough to satisfy the Supreme Court’s “actual and imminent injury” standard—but the plaintiff in Kurtz did not even allege that much. Absent any “indication that [the named plaintiff would] buy Defendants’ flushable wipes products in the future,” the Second Circuit easily found “no likelihood of future injury,” and thus, no standing to represent an injunctive class. (As it happens, just days later, the Second Circuit issued a published opinion in a different case holding unequivocally that injunctive relief is unavailable in ordinary consumer false advertising cases. We just covered that decision here.)
Arriving at the main event—the plaintiffs’ damages model—the Second Circuit held that Judge Weinstein had not “abuse[d] [his] discretion” in certifying a damages class, now that the plaintiffs’ expert had actually conducted the analysis that he had previously proposed. Importantly, the panel recognized that, in an appropriate case, several of the defendants’ critiques could justify rejecting a plaintiff’s damages model and denying class certification. For example, a regression model’s “fail[ure] to account for major variables” may render it “so incomplete as to be inadmissible as irrelevant.” And “cherry-picking data to artificially generate a particular result” may do the same. Here, however—citing the “extensive” proceedings before the district court and the deferential standard of review—the Second Circuit declined to disturb Judge Weinstein’s conclusion that the flaws in the plaintiffs’ model did not rise to that level.
Kurtz’s conclusion as to typicality and adequacy strikes us as problematic. If the named plaintiff literally did not care whether the wipes he purchased were “flushable,” as his behavior suggests, then it’s hard to understand how he could have been harmed by the alleged falsity of the “flushable” claim—even if its presence on the package did entail a “price premium.” As the Seventh Circuit has , the payment of a higher price for a product than one might have paid, standing alone, is not an Article III injury-in-fact, let alone an actionable consumer-protection violation.
Kurtz’s treatment of the damages issues is similarly troubling. In particular, it is hard to reconcile with the Second Circuit’s landmark decision in In re Initial Public Offering Securities Litigation (“IPO”), 471 F.3d 24 (2d Cir. 2006). There, the Second Circuit squarely “disavow[ed]” the notion “that an expert’s testimony may [satisfy a plaintiff’s Rule 23 burden] simply by not being fatally flawed”—i.e., by not being “so flawed that it would be inadmissible.” As IPO explained, even if the plaintiff has adduced an admissible damages model, the district court must “weigh [the] conflicting evidence” presented by both sides, id., and actually determine which side’s expert is more persuasive as to whether common issues predominate. See also In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305 (3d Cir. 2008) (holding that an expert’s damages model “should not be uncritically accepted as establishing [Rule 23 predominance] merely because the court holds [it] should not be excluded”). The Kurtz panel’s refusal to engage with the plaintiff’s damages model, beyond deferring to the district court’s finding that it was not “so incomplete as to be inadmissible,” appears to conflict with IPO—not to mention precedent from other circuits, like Hydrogen Peroxide.
The Kurtz panel believed that its hands-off approach was in accord with the Supreme Court’s decision in Tyson Foods, Inc. v. Bouaphakeo, 136 S. Ct. 1036 (2016). Apparently, in the panel’s view, Tyson Foods suggests that flaws in a plaintiff’s damages model that do not rise to the level of rendering it inadmissible cannot defeat class certification; they just mean that the certified class may eventually lose on the merits. But, as the Third Circuit recently explained, that is not what Tyson Foods actually held. Tyson Foods involved an unusual procedural posture (a motion to decertify a class following a jury verdict in the class’s favor) and a unique substantive-law context (a claim under the Fair Labor Standards Act, which employs a distinctive burden-shifting framework). See In re Lamictal Direct Purchaser Antitrust Litig., 957 F.3d 184, 191-92 (3d Cir. 2020). Moreover, in Tyson Foods, the defendant “did not [challenge] the statistical validity of [the plaintiff’s] studies …, nor did it attempt to discredit [the plaintiff’s] expert with testimony from a rebuttal expert.” Tyson Foods, 136 S. Ct. at 1044. Thus, as the Third Circuit noted, Tyson Foods “does not control” in other procedural or substantive contexts, and it “certainly does not overturn [the] longstanding rule” that, at the class-certification stage, courts must conduct a rigorous analysis of dueling expert opinions and “resolve … by a preponderance of the evidence” which one is more persuasive. Lamictal, 957 F.3d at 191-92.
That is not the only questionable aspect of the damages discussion in Kurtz. Although both parties had assumed that Daubert’s standards for admissibility of expert testimony applied at the class-cert stage, the panel remarked in a footnote that this proposition is “far from well-established.” Perhaps there is no binding precedent from the Supreme Court or Second Circuit expressly so holding. But the Kurtz footnote overlooked that the Supreme Court has expressed “doubt” about the argument that Daubert doesn’t “apply to expert testimony at the certification stage of class-action proceedings.” And other circuits have held, in persuasive published opinions, that Daubert does so apply. The Second Circuit’s omission of these facts from its discussion is unfortunate. However, the panel does not actually purport to decide the question of Daubert’s applicability, so this footnote is pure dictum.
In sum, the latest Kurtz decision is a mixed bag. It reaches the right result on standing and injunctive relief, but its discussion of typicality and adequacy—and, especially, of damages models and predominance—leaves much to be desired. The one bright spot is that Kurtz teaches that a plaintiff seeking class cert must actually create a working damages model to demonstrate predominance, and cannot merely propose to do so down the road. However, once a plaintiff submits an actual damages model, the decision suggests a degree of deference that is difficult to square with binding Supreme Court and Second Circuit precedent.
Fortunately, because the Kurtz decision is unpublished, future Second Circuit panels and in-circuit district courts are not bound by it. And even on its own terms, the Kurtz decision makes clear that its result was premised on the unique and “extensive” record before the district court, rather than a wholesale rejection of the defendants’ critiques. In fact, all that Kurtz actually held vis-à-vis the damages class is that the district court did not abuse its discretion by granting certification based on the evidence before it. Kurtz did not hold that a district judge should certify a class (or, still less, that it must do so) whenever a plaintiff presents the type of “hedonic regression” submitted in this case. Future courts, therefore, remain free to subject such models to the “rigorous analysis” that they so richly deserve, and that precedent demands. We hope that they will do so.