How many uninjured class members is too many for class certification? For decades, this question has generated significant controversy between and within federal judicial circuits and led to widely divergent outcomes. Particularly given upward trends in class litigation and the staggering liability and costs such suits can generate, many businesses, industry groups, government entities, and other constituencies were hopeful that the U.S. Supreme Court would resolve—or at least clarify—the issue in Labcorp v. Davis, where the Court had granted certiorari on the question of whether federal courts may certify a damages class under Federal Rule of Civil Procedure 23(b)(3) when the class includes some members who lack any injury under Article III of the U.S. Constitution. 

But on June 5, 2025, in an 8-1 per curiam order, the Court dismissed the writ as “improvidently granted” without explanation, leaving the question for another day—and leaving lower courts and litigants with an uncertain legal landscape. The decision is unsurprising, however, as oral argument revealed concerns among some Justices that the core certification question had not properly been raised from a procedural standpoint.  

In Labcorp, the defendant, which provides diagnostic testing services, installed self-service kiosks with touchscreens in 2017 to allow patients to check in for appointments. The plaintiffs, who were legally blind, brought a class action in the Central District of California under the Americans with Disabilities Act and California’s Unruh Civil Rights Act, alleging the kiosks were not accessible to legally blind individuals, and seeking certification of a class giving rise to potential damages of $500 million per year. The district court certified a California-specific damages class, and the defendant sought interlocutory review with the Ninth Circuit, arguing, inter alia, that the class was overly broad because it contained uninjured members, including blind patients who were not interested in using kiosks. After the request for review was filed, the district court amended the class definition, but the defendant did not formally amend its appeal to include the later definition. The Ninth Circuit affirmed the district Court’s original certification ruling, and the defendant petitioned the U.S. Supreme Court.

The Court initially granted certiorari on the propriety of certifying a class that included at least some uninjured members. But after an oral argument focusing in substantial part on the changing class definition and related procedural complexities, the Court dismissed the writ as improvidently granted in a one-sentence decision. 

Although the Court did not explain its reasoning, Justice Kavanaugh argued in dissent that the Court presumably did not wish to address a threshold mootness issue raised by the plaintiffs. Specifically, the plaintiffs argued that Labcorp appealed the wrong certification order, citing the district court's revised class definition. Justice Kavanaugh found this argument unpersuasive and would have addressed the core certification question, reasoning that the district court’s revisions to the class definition were immaterial and that Labcorp lacked any procedural options for appealing the revised definition and had therefore acted reasonably. In any event, based on the exchanges at oral argument, Justice Kavanaugh’s assumption that other Justices desired a cleaner procedural record appears plausible. 

Citing the United States as amicus curiae, Justice Kavanaugh went on to state that he would have reversed the Ninth Circuit’s decision and held that Rule 23’s requirement that common issues of fact and law predominate does not allow for certification of damages classes that include both injured and uninjured class members. Justice Kavanaugh cited the “real world consequences” of certifying classes inflated with uninjured members, which can “threaten massive liability,” force businesses to settle rather than bet the company on the uncertainties of a class-wide trial, and indirectly harm consumers:

“Importantly, the coerced settlements substantially raise the costs of doing business. And companies in turn pass on those costs to consumers in the form of higher prices; to retirement account holders in the form of lower returns; and to workers in the form of lower salaries and lesser benefits. So overbroad and incorrectly certified classes can ultimately harm consumers, retirees, and workers, among others. Simply put, the consequences of overbroad and incorrectly certified damages class actions can be widespread and significant.”

Justice Kavanaugh also referenced the more nuanced question of whether Rule 23 allows certification of damages classes that include a de minimis number of uninjured class members. However, he ultimately found this question had not been presented by Labcorp’s appeal because the Ninth Circuit’s standard allows for certification of classes with more than a de minimis number of uninjured members. 

It bears noting that at oral argument, multiple Justices expressed skepticism toward defense arguments that certified classes can include no uninjured class members, which suggests a zero-tolerance stance on the presence of such members at the certification stage may face judicial headwinds. But in at least some cases, advocating a Rule 23 predominance standard that emphasizes quantitative absolutes or strict numerical thresholds at the outset arguably misses the point, since determining precisely how many uninjured class members exist in a population often requires the types of individualized inquiries that render class certification inappropriate. 

At bottom, those hoping for the Supreme Court to strike a material blow against certification of overly broad classes, or to clarify the question of how many uninjured class members is too many, will have to wait. In the interim, the Court's dismissal of the writ demonstrates the importance of maintaining a clear record in the trial court and on appeal, particularly in the face of evolving class definitions. Further, the dissent’s views on Rule 23’s predominance requirement, and the harms associated with certification of overly broad damages classes, provide compelling arguments for defendants who choose to resist certification. Moreover, separate from the policy consequences associated with overly broad classes, courts and litigants must still face the practical challenge of efficiently litigating and trying such classes—and of ferreting out uninjured members—thus providing defendants with superiority and manageability arguments in addition to the predominance problems noted by the dissent.