Fishy Class Certification: A Packaged Tuna Antitrust Case and a Shift in Class Certification Standards.

• Author: Newton, Mac

Olean Wholesale Grocery Coop., Inc. v. Bumble Bee Foods LLC, 31 F.4th 651 (9th Cir. 2022), cert. denied sub nom., Starkist Co. v. Olean Wholesale Grocery Coop., Inc., 143 S. Ct. 424 (2022).

1. INTRODUCTION

   Class certification confers a fearsome power on plaintiffs. Organizing a group of otherwise disparate potential claimants through a class action produces a mighty concentration of power that plaintiffs can wield against defendants. This power is exerted on defendants in the form of "hydraulic pressure" to settle, (1) shifting focus from litigation to settlement. (2) The risk involved with one jury standing between a defendant and "potentially ruinous liability" often proves to be intolerable for corporate defendants. (3) The in terrorem nature of class actions means that just by obtaining class certification, plaintiffs achieve a major victory. (4) This notion is borne out by empirical results, with one study that spanned nine years finding that, once certified, classes "almost always settled." (5)

   In light of the awesome power that is class certification, a number of criteria, chiefly in Rule 23 of the Federal Rules of Civil Procedure, aim to certify only classes that are both fair and efficient. (6) To obtain certification, plaintiffs must satisfy all four prerequisites in Rule 23(a)--numerosity, commonality, typicality, and adequate protection of interests--and conform to one of three types of actions listed in Rule 23(b). (7) The most controversial of these three types is 23(b)(3), under which plaintiffs seek monetary damages. (8) Rule 23(b)(3) requires that issues "common" to the class "predominate" over those that affect individual class members. (9)

   The rule states, in a single sentence, the bar that plaintiffs must clear in order to certify a class. (10) When the rubber of the rule meets the road of the real world, however, the rule provides but minimal guidance. Classes are often drawn in broad strokes. By nature, they consist of individuals with some common experience, such as every student subject to a challenged admissions rule, every employee working under the same compensation policy, or every consumer purchasing the same product. (11) The nature of this line drawing gives rise to a problem: what if some individuals who share the common experience were nonetheless completely uninjured by the defendant's conduct? Can those uninjured plaintiffs still be part of a certified class? If so, how many uninjured plaintiffs does Rule 23 tolerate? That is the central issue this Note addresses. Possible solutions run the gamut: a strict requirement that there be no uninjured class members, tolerance of a negligible or "de minimis" number of uninjured plaintiffs, or an acceptance of a significant number of uninjured plaintiffs at the certification stage.

   Regardless of the approach that a court uses, satisfying the predominance requirement mandates that plaintiffs come forward with common proof on behalf of the class. In the quest to provide such proof, plaintiffs routinely turn to expert witnesses to establish a model that demonstrates injury. (12)

   Prospective class members advanced such a model in Olean Wholesale Grocery Coop., Inc. v. Bumble Bee Foods LLC, an antitrust action against several of the world's largest manufacturers of tuna. (13) The plaintiff model purported to demonstrate injury in the form of higher prices the class paid. (14) The model was fiercely contested by the defendants, who, through their own expert, claimed that antitrust injury could not be established for nearly a third of the class and accused the plaintiff model of relying on faulty methodology. (15) The district court did not decide between the dueling experts and credited some claims made by both, ultimately certifying the class. (16) Following class certification, the Court of Appeals for the Ninth Circuit affirmed the decision not to pick between the battling experts' conclusions. (17) Its opinion broadened the path to class certification and created a circuit split when it rejected the rule that all or substantially all class members must have been injured to obtain certification. (18)

   Part ii of this Note sets forth the factual background and procedural posture of this dispute. Part III lays out both the doctrine of predominance and a court's obligations at the class certification stage. Part IV details the majority's rationale for affirming the refusal to find in favor of either expert's model and catalogues the dissent's qualms with that decision. Part V examines the holding of the case against the backdrop of Rule 23's text, precedent, and the requirement of standing. It also highlights key areas in which the holding is in tension with those doctrines and discusses the need for the Supreme Court's input on this issue. Part VI concludes by discussing the implications of the decision on future class certification disputes.

2. FACTS AND HOLDING

   The defendants, Bumble Bee, StarKist, Chicken of the Sea, and their parent companies, are the largest tuna producers in the united States, accounting for about eighty percent of the domestic packaged tuna market. (19) In 2015, an investigation by the Department of Justice sparked a string of indictments against several of the defendants for a criminal price-fixing conspiracy ranging from late 2011 to late 2013. (20) Bumble Bee, StarKist, and three individual executives pleaded guilty. (21) A jury convicted Bumble Bee's former CEO, whereas Chicken of the Sea cooperated and admitted its guilt in exchange for lenience. (22) Following the criminal proceedings, several purchasers of the defendants' products filed putative class actions against the defendants, alleging that they had engaged in a conspiracy from 2010 through at least the end of 2016. (23) These actions were initially consolidated in a multidistrict litigation proceeding in the Southern District of California. (24)

   The purchasers consisted of three putative subclasses: (1) direct purchasers of the defendants' products (e.g., major retailers and small mom-and-pop shops that then sold the tuna to consumers), (2) indirect purchasers that bought tuna in bulk for prepared food or resale (e.g., restaurants), and (3) individual end purchasers for personal consumption. (25) The plaintiffs moved to certify the three subclasses in 2018. (26)

   In an attempt to demonstrate class-wide impact in the form of an overcharge, the plaintiffs proffered economic analyses by three different expert witnesses who deployed substantially identical methodology. (27) The defendants countered with expert evidence of their own, advancing numerous criticisms of the plaintiff model. (28) Both sides deployed regression models that attempted to isolate changes in price attributable to the conspiracy by controlling for other factors. (29) Although the parties clashed over the expert analyses for all three subclasses, it was the direct purchaser class that proved to be the most violent battleground.

   Applying his regression model, the plaintiff expert found that the direct purchasers incurred an average overcharge of 10.28% as a result of the conspiracy. (30) Using the outcome of the regression analysis, the plaintiff expert created a but-for price--the expected price if there was no conspiracy. (31) The plaintiff expert then compared the predicted but-for price to the actual prices the direct purchasers paid and found that 94.5% of the class had at least one purchase above the but-for price. (32) The defense expert challenged the plaintiff expert's conclusions on multiple grounds. (33) The primary critique was that the plaintiff model's reliance on an average overcharge of 10.28% was faulty and masked individualized differences among the subclass members, such as bargaining power. (34) The defense expert performed his own test of the plaintiff model, abandoning the average overcharge figure and instead evaluating the overcharge based on each customer. (35) The results indicated that no positive, statistically significant overcharge occurred for 28% of the direct purchaser class. (36) Thus, the crux of the parties' dispute centered around the propriety of the single uniform overcharge of 10.28%. (37)

   Although the parties differed on the number of uninjured prospective class members by more than 20%--roughly 6% of the class per the plaintiff model and approximately 28% according to the defense--the district court chose not to credit one model over the other. (38) Instead, the court noted that even class members for whom the model failed to produce statistically significant results could use it as proof "as it pertains to similarly situated class members." (39) The district court deemed the criticism of the plaintiff model nonfatal because the plaintiff expert could explain criticized results "using sound econometric principles that are not obviously contrary to the theory of the case." (40) Responding to another claim by the defense expert--that the plaintiff expert improperly used a cost index rather than the defendants' actual accounting data--the district court recounted the arguments and concluded that the plaintiff expert's "use of costs in this case [was] reasonable." (41)

   The district court conceded that the plaintiff methodology contained "potential flaws" (42) and that the defendants' criticism was "serious and could be persuasive to a finder of fact." (43) Nonetheless, the district court certified the class, concluding that the defendants had not sufficiently demonstrated that the plaintiff model was "unreliable or incapable of proving impact on a class-wide basis." (44) In making its decision, the district court noted that econometric evidence should be allowed as common proof "where [it is] plausibly reliable." (45)

   The defendants appealed the order certifying the class to the Ninth Circuit. (46) A panel heard the case and vacated the district court's certification order, remanding the case for further proceedings. (47) The panel...