Ninth Circuit Holds That Statistical or “Representative” Evidence Can Establish Predominance, But Vacates Order Certifying Class Where District Court Failed to Resolve Factual Issues Concerning Uninjured Class Members

King & Spalding

On April 6, 2021, the Ninth Circuit for the first time addressed a plaintiff’s burden to show predominance at the class certification stage. In Olean Wholesale Grocery Coop. v. Bumble Bee Foods LLC, the court joined the First, Second, Third, Fifth, and Seventh Circuits in holding that a plaintiff must prove predominance by a preponderance of the evidence, concluding that this standard best comports with the U.S. Supreme Court’s directive that district courts perform a “rigorous analysis” to determine whether Rule 23(b)(3) has been satisfied. The Ninth Circuit also clarified that classes may include no more than a de minimis number of uninjured class members, though it declined to adopt a “numerical or bright-line rule” concerning what percentage of uninjured class members would exceed this threshold. [Disclosure: King & Spalding filed an amicus brief in support of Appellants in this appeal.]

  • Various purchasers of tuna products brought this class action suit against defendants StarKist Company, Tri-Union Seafoods d/b/a Chicken of the Sea, and Bumble Bee Foods LLC alleging a price-fixing conspiracy to artificially inflate the prices of tuna products. There was “little dispute over the existence of a price-fixing scheme” given that all three defendants had either pleaded guilty to or admitted to price fixing. The district court therefore focused on whether plaintiffs satisfied the requirements of a Rule 23(b)(3) damages class.
  • In support of their class certification motion, plaintiffs relied on experts who performed statistical analyses of the class members’ damages, concluding, among other things, that nearly 95% of tuna purchasers in one of the classes overpaid for tuna by an average of 10.28%. By contrast, defendants’ experts concluded that 28% of putative class members suffered no injury at all.
  • The district court certified plaintiffs’ three proposed classes, concluding that defendants’ challenges to plaintiffs’ experts’ methodologies were “ripe for use at trial” but “not fatal to a finding of classwide impact,” and that, despite “potential flaws,” the methodologies were “reliable and capable of proving impact.” In so ruling, the district court stressed that “determining which expert is correct is beyond the scope of a class certification motion.”
  • Defendants appealed the district court’s determination that Rule 23(b)(3)’s predominance requirement was satisfied, arguing that (1) representative evidence cannot be used to establish predominance, and (2) even if it could, plaintiffs’ expert analyses failed to do so because a significant percentage of the class may have suffered no injury at all.
  • With respect to defendants’ first argument, the Ninth Circuit approved the use of statistical evidence in establishing predominance in certain circumstances. The Ninth Circuit further determined that plaintiffs’ particular statistical evidence was permissible because: (1) it could have been used to establish liability in a class member’s individual suit, (2) it was consistent with plaintiffs’ underlying price-fixing theory of liability and thus did not run afoul of Comcast, and (3) the use of averaging assumptions in a regression analysis can be appropriate and was appropriate in this particular case.
  • Turning to defendants’ second argument, the Ninth Circuit held that even though plaintiffs’ statistical evidence could be used to satisfy the predominance requirement, district courts “must still rigorously analyze the use of such evidence to test its reliability and to see if the statistical modeling does in fact mask individualized differences.” In so ruling, the Ninth Circuit stated that district courts must resolve any factual disputes bearing on Rule 23 requirements and cannot “outsource” the predominance determination to a jury.
  • The court went on to explain that “a key factual determination courts must make is whether the plaintiffs’ statistical evidence sweeps in uninjured class members,” and that a class containing uninjured members may be certified only where the number of such members is de minimis. Though the Ninth Circuit declined to adopt a numerical or bright-line rule defining de minimis, it concluded that predominance has not been met “under any rubric” if more than one-fourth of the class members were unharmed. The Ninth Circuit remanded the case to the lower court with instructions to determine the number of uninjured class members.
  • A judge on the Ninth Circuit subsequently called for a vote to determine whether the case should be reheard en banc, with the parties directed to submit simultaneous briefing on this issue and specifically address whether Rule 23(b)(3) “requires a district court to find that no more than a ‘de minimis’ number of class members are uninjured before certifying a class.”
    • Notably, the U.S. Supreme Court may soon provide further guidance on this important issue. Last month, the Court heard arguments in Ramirez v. TransUnion LLC—another appeal from the Ninth Circuit—on the issue of “[w]hether either Article III or Rule 23 permits a damages class action where the vast majority of the class suffered no actual injury, let alone an injury anything like what the class representative suffered.” We discussed the Ramirez case in our March and December issues last year.
  • Regardless of how the “uninjured class member” question is ultimately decided, the Olean decision reinforces several important class action themes for defendants. While the court endorsed the use of “representative evidence” in appropriate cases, it underscored that there are “reason[s] to be wary of overreliance on statistical evidence to establish classwide liability.” For instance, because the Rules Enabling Act provides that class actions are merely a procedural tool that cannot “abridge, enlarge or modify [a plaintiff’s] substantive rights,” representative evidence cannot be used to establish classwide liability if a plaintiff could not use the same statistical evidence to establish liability in an individual suit. Moreover, the court reiterated that establishing predominance “goes beyond determining whether the evidence would be admissible in an individual action.” Thus, a district court cannot merely conduct a Daubert analysis to determine whether an expert’s opinion is sufficiently reliable to be presented to jurors at trial. Rather, at the class certification stage, a district court must actually resolve factual disputes between dueling experts if the issues bear on whether the requirements of Rule 23 have been met.
  • Read the Ninth Circuit’s opinion in Olean here.

Written by:

King & Spalding

King & Spalding on:

Reporters on Deadline

"My best business intelligence, in one easy email…"

Your first step to building a free, personalized, morning email brief covering pertinent authors and topics on JD Supra:
*By using the service, you signify your acceptance of JD Supra's Privacy Policy.
Custom Email Digest
- hide
- hide