| Joshua P. Davis, Randy Stutz, Berk Bahceci
Cartels, Private Enforcement, Section 1 of the Sherman Act
AAI submitted an amicus brief on May 19 urging the Ninth Circuit Court of Appeals to grant en banc rehearing to address a flaw in a recent panel opinion vacating class certification in a closely watched price-fixing case.
In Olean Wholesale Grocery Cooperative, Inc. v. Bumble Bee Foods, LLC, a district court certified three classes of purchasers, including direct purchasers and two groups of indirect purchasers, seeking to recover for the confessed price fixing of the three leading producers of packaged tuna, Bumble Bee Foods LLC, Starkist, and Chicken of the Sea. The defendants had either sought leniency or pled guilty after a Department of Justice Investigation, and several of their executives have been sentenced to prison.
Unable to contest liability given their admissions, the defendants focused extensive resources and attention on defeating class certification. In district court proceedings, they introduced rebuttal experts seeking to counter plaintiffs’ economic experts, which had introduced statistical analysis attempting to show that the price fixing caused widespread injury across the respective classes. The district court, after a three-day evidentiary hearing, found plaintiffs’ experts more persuasive and held that the plaintiffs’ common statistical evidence of impact was sufficient to help satisfy Rule 23’s predominance requirement, though it allowed that defendants could still challenge the admissibility and probative value of the common statistical evidence at trial.
On interlocutory appeal, the defendants, supported by the U.S. Chamber of Commerce and the Washington Legal Foundation, argued that the district court erred by refusing to definitively resolve the battle of the experts at class certification, and that plaintiffs’ expert statistical analysis was inherently problematic because it relied on the average overcharges to the classes, thereby masking the possibility that some of the class members were uninjured by the price fixing. The defendants maintained that, because plaintiffs’ expert evidence could not necessarily sustain a jury finding for every class member, it should not be a permissible means of establishing that common questions would predominate at a class trial.
In an amicus brief submitted last August, AAI argued that such evidence need only be relevant and reliable to be admissible; it does not have to assure that each plaintiff would prevail on the merits of the impact element in an individual action. Rule 23 requires only that common “questions” must predominate over individual questions at trial; it cannot be read to suggest that the questions’ answers must be determined to permit class certification. Moreover, any uninjured class members may be identified after trial, and longstanding case law prevents defendants from capitalizing on the uncertainty created by their own illegal conduct, including uncertain damages calculations.
The AAI brief also argued that the court should unequivocally reject the defendants’ effort to cast categorical doubt on statistical analysis, and specifically regression modelling, in antitrust cases. Regression models frequently rely on averaging techniques, but that is not where they begin and end. Such models are routinely accepted as reliable methods of proving widespread injury to antitrust classes because econometric techniques can control for price changes caused by supply and demand factors and then focus on the uniformity of differences across class members to reliably show common impact.
All three judges on the merits panel adopted the position advocated by AAI in rejecting defendants’ categorical arguments on the use of statistical analysis and regression modeling to prove class-wide impact in antitrust cases. However, the three judges sided with defendants in holding that the district court erred by refusing to resolve the disagreement among the parties’ experts over the number of potentially uninjured members in the class. And the panel then split over the standard for determining whether the presence of uninjured class members may defeat predominance. The panel majority concluded that the district court, before certifying a class, must find that only a “de minimis” number of class members are uninjured. Judge Hurwitz, partially dissenting, maintained that neither the text of Rule 23 nor Ninth Circuit precedent permit the court to implement such a requirement.
In the aftermath of the panel opinion and partial dissent, neither party petitioned for panel or en banc rehearing, instead agreeing to accept remand. But on April 28, the Court sua sponteordered briefing on whether en banc hearing is warranted and directed the parties to focus on the “de miminis” issue that divided the panel.
AAI’s brief argues that en banc rehearing is warranted because the panel majority’s ruling is unduly rigid and will undermine the efficacy of private antitrust class actions. The brief points out that in this case and many other antitrust cases, the presence of uninjured class members does not create a risk that individual questions will predominate over common questions at trial because the plaintiffs rely solely on common statistical evidence of aggregate damages, meaning the claims rise or fall as one, both as a legal and practical matter. Moreover, common issues need only predominate in a case as a whole, not as to each element of a claim. Impact need not be a common question for common questions to predominate as a whole in an antitrust case.
The brief was written by Professor and AAI Board Member Joshua Davis of the University of San Francisco Law School, with assistance from AAI Vice President of Legal Advocacy Randy Stutz and AAI Research Fellow Berk Bahceci.
