The class action plaintiffs’ bar celebrated yesterday’s Supreme Court’s decision in Tyson Foods, Inc. v. Bouaphakeo (pdf), rejecting Tyson’s challenge to class certification. One lawyer called it “a huge David v. Goliath victory.”

But when plaintiffs’ lawyers wake up this morning and focus on the details of the Court’s opinion, they are in for a serious post-celebration hangover.

The Court’s reasoning for the first time maps a clear route for defendants to use in challenging plaintiffs’ use of statistical evidence in class actions. It also provides important guidance for defendants about preserving the ability to challenge plaintiffs’ reliance on statistics.


Tyson involved claims under the overtime pay requirements of the Fair Labor Standards Act (FLSA). The named plaintiffs, hourly workers at a Tyson pork processing plant, claimed that the company failed to compensate them or the other 3,344 employees for time spent donning and doffing protective equipment and walking to and from their work stations. Prior to trial, the district court certified the FLSA claims as an opt-in collective action and the state-law claims as a Federal Rule of Civil Procedure 23(b)(3) class action. (The Court assumed that the standards for both types of certification are similar.)

Because Tyson did not keep records of each employee’s donning and doffing time, the plaintiffs relied on what they characterized as “representative evidence,” which consisted principally of two studies. One, conducted by their expert Dr. Mericle, used 744 “videotaped observations” to conclude that employees in the “cut and retrim departments” averaged 18 minutes a day, and those in the “kill department” averaged 21 minutes per day.

The second study, conducted by Dr. Fox, used each employee’s time records—combined with the Mericle time estimates—to determine which employees were entitled to overtime pay based on the addition of the donning and doffing time to their work weeks. The study found that 212 class members were not entitled to recover, and that the remaining class members were entitled to an aggregate award of $6.7 million.

The jury returned a special verdict finding that donning and doffing time at the beginning and end of the work day was compensable, but that donning and doffing for meal breaks was not. It awarded $2.9 million.

Tyson moved to set aside the verdict. It argued that the individual variations in donning and doffing times—reflected in the Fox study—meant that the class should not have been certified, because it did not satisfy Rule 23(b)(3)’s “predominance” requirement: that “questions of law or fact common to class members predominate over” any questions affecting only individual class members. Both the district court and the court of appeals rejected Tyson’s arguments.

The Supreme Court’s decision

The Supreme Court addressed two issues.

Predominance and statistical analysis

The first issue is whether the class satisfied the predominance requirement—and the parties agreed that the predominance question turned on whether the Fox study could properly be used to prove the donning and doffing time. Tyson argued that these were “necessarily person-specific inquiries” while the plaintiffs asserted that—based on the Mericle study—it could “be assumed each employee donned and doffed for the same average time observed” in the study. Thus, the Supreme Court stated, “[w]hether this inference is permissible becomes the central dispute in the case.”

The Court rejected the argument, advanced by Tyson and some of its amici, for a categorical exclusion of representative or statistical evidence in class actions, and refused to “establish general rules governing the use of statistical evidence . . . in all class-action cases.” That is because the use of such evidence to establish classwide liability “will depend on the purpose for which the evidence is being introduced and on ‘the elements of the underlying causes of action.’”

A key question, the Court said, is whether “each class member could have relied on [the statistical evidence] to establish liability if he or she had brought an individual action”; if so, “that [evidence] is a permissible means of establishing the employees’ hours worked in a class action.”

The Court held that this test was met here, because of a legal rule applicable to FLSA claims: Anderson v. Mt. Clemens Pottery Co., 328 U.S. 680, 687-88 (1946), held that when an employer has not complied with its statutory obligation to keep proper records, an employee “if he produces sufficient evidence to show the amount and extent of [uncompensated] work as a matter of just and reasonable inference”; the employer then must “come forward with evidence of the precise amount of work performed or with evidence to negative the reasonableness of the inference to be drawn from the employee’s evidence.”

Under Mt. Clemens, the Court stated, the Mericle study would have been a valid method of proof in individual cases. It therefore was a valid means of proof in the class action.

Tyson does not categorically approve the use of statistical evidence in class actions generally. Indeed, the Court reiterated at the end of its analysis that the question requires a case-specific inquiry: “[w]hether a representative sample may be used to establish classwide liability will depend on the purpose for which the sample is being introduced and on the underlying cause of action. . . . The fairness and utility of statistical methods in contexts other than [FLSA actions] will depend on facts and circumstances particular to those cases.”

Here are some of the case-specific inquiries identified by the Tyson Court:

  • Whether the statistical evidence is probative of the factual issue that controls the application of the governing substantive rule. The Tyson Court’s distinction of Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338 (2011), which had rejected a “Trial By Formula” approach for a Title VII discrimination class action, is illuminating:

The plaintiffs in Wal-Mart did not provide significant proof of a common policy of discrimination to which each employee was subject. “The only corporate policy that the plaintiffs’ evidence convincingly establishe[d was] Wal-Mart’s ‘policy’ of allowing discretion by local supervisors over employment matters”; and even then, the plaintiffs could not identify “a common mode of exercising discretion that pervade[d] the entire company.” . . . .

[T]he [Wal-Mart] Court held that the employees were not similarly situated, [and therefore] none of them could have prevailed in an individual suit by relying on depositions detailing the ways in which other employees were discriminated against by their particular store managers.

In Tyson, the Court reached the opposite conclusion, because

the study here could have been sufficient to sustain a jury finding as to hours worked if it were introduced in each employee’s individual action. While the experiences of the employees in Wal-Mart bore little relationship to one another, in this case each employee worked in the same facility, did similar work, and was paid under the same policy. As Mt. Clemens confirms, under these circumstances the experiences of a subset of employees can be probative as to the experiences of all of them.

That explanation shows the importance of case-specific analysis, and the limited scope of Tyson’s holding.

  • Whether reliance on statistical evidence to certify a class will “deprive [the defendant] of its ability to litigate individual defenses.” Defendants in class actions frequently rely on the provision of the Rules Enabling Act stating that a rule may not “abridge . . . any substantive right.” 28 U.S.C. § 2072(b). The Tyson Court emphasized this provision’s relevance to class actions, referring to “the Rules Enabling Act’s pellucid instruction that use of the class device cannot ‘abridge . . . any substantive right.’” It concluded that Tyson was not deprived of its ability to litigate individual defenses because “there were no alternative means for the employees to establish their hours worked” and no defense for Tyson other than challenging the study. In many other types of class actions, however, the defendant will be able to defend by presenting class member-specific evidence, and in such situations the defendant should have a strong basis for defeating class certification. Indeed, the Tyson Court stated that would have been the case in Wal-mart.
  • Whether the statistical evidence would be admissible in an individual action under the balancing tests established by Federal Rules of Evidence 401 and 403—which the Court cited repeatedly in its opinion. That, in turn, will often depend on the other evidence that is available; the unavailability of actual employee-specific information was key to the Tyson Court’s analysis. In other contexts, of course, actual plaintiff-specific information will be available.
  • Whether the statistical study is admissible under Federal Rule of Evidence 702. The Supreme Court noted that Tyson “did not raise a challenge to [the plaintiffs’] experts’ methodology under Daubert [v. Merrell Dow Pharmaceuticals, Inc.]; and, as a result, there is no basis in the record to conclude it was legal error to admit that evidence.”

Excluding uninjured class members

The second issue that the Tyson Court addressed was whether class certification was improper because there was no mechanism for ensuring that only injured class members would receive compensation.

Again, some factual background is useful. The jury’s award of $2.9 million was considerably less than the $6.7 million award indicated by the Fox study. Tyson argued that the lower damages amount meant that the jury had rejected the Mericle study’s time estimates, and that it therefore was not possible to know which class members were entitled to damages. Awarding damages to each class member would mean that some individuals found by the jury not to be injured would nonetheless receive compensation.

The Court stated that “the question whether uninjured class members may recover is one of great importance,” but concluded it was not properly before the Court in this case “because the damages award has not yet been disbursed, nor does the record indicate how it will be disbursed.” It held that Tyson “may raise a challenge to the proposed method of allocation when the case returns to the District Court.”

But the Court also noted that “this problem appears to be one of [Tyson’s] own making,” because Tyson had rejected the plaintiffs’ suggestion to bifurcate the liability and damages phases of the trial, which was made “for the precise reason that it may be difficult to remove uninjured individuals from the class after an award is rendered.” Whether “any error should be deemed invited” by Tyson was an issue left to the district court.

Chief Justice Roberts, joined by Justice Alito, addressed this issue in detail in a concurring opinion, concluding that the plaintiffs’ proposal for allocating the award would not weed out uninjured plaintiffs—and expressed skepticism about reliance on invited error because “Tyson’s insistence on a lump-sum jury award cannot overcome the limitations placed on the federal courts by the Constitution” (referring to Article III’s injury-in-fact requirement).

The Court’s emphasis on the “great importance” of the issue of uninjured class members, combined with the Chief Justice’s focus on Article III, means that defendants should continue to focus on this issue in litigating class actions. To the extent that a class is defined broadly—and the liability determinations in the merits phase of the case are not certain to weed out all uninjured parties—there is a strong argument that the class certification decision cannot stand.

However, the Court’s reference to invited error means that defendants should take care to propose, or at least not oppose, procedures that would ensure that uninjured class members are identified and excluded from any eventual damages award.

In sum, the Tyson decision is far from a home run for plaintiffs. The Court supplemented Wal-Mart by defining in greater detail the standards that must be satisfied when plaintiffs seek to rely on statistical evidence to transform individualized issues into common ones for purposes of Rule 23(b)(3)’s predominance inquiry.