From "predominance" to "resolvability": A new approach to regulating class actions

This Article develops normative and doctrinal innovations to cope with a pivotal yet undertheorized question in most proposed class actions: assuming that a class has adequate representatives, how much variance among class members' circumstances should courts tolerate? Class actions seeking monetary damages would be much less controversial if the factual circumstances of every class member were exactly alike. In an imagined world of perfect homogeneity, shifting from an individualized to an aggregative mode of adjudication would promote efficiency, mitigate collective action problems, and counterbalance defendants' inherent economies-of-scale without sacrificing accuracy or redistributing entitlements among class members. In the real world, however, most classes encompass at least partially heterogeneous claims spread across a spectrum of merit and economic value. This diversity creates opportunities for strategic behavior that can distort the outcome of trials or of settlements negotiated in the shadow of trial. The Article defines and explores three phenomena that create such distortions: "cherry-picking," "claim fusion," and "ad-hoc lawmaking." The potentially mischievous consequences of heterogeneity in class actions suggest that courts should have a normative theory and doctrinal mechanism to distinguish between acceptable and excessive diversity among proposed class members. The Article addresses the normative question by introducing and justifying three principles to structure certification criteria: "finality," "fidelity," and "feasibility." The Article then applies these principles to assess the forty-year-old "predominance" standard that currently governs how courts decide whether to certify diverse damages classes. This analysis reveals that the predominance standard is conceptually incoherent and that widely-cited doctrine applying it is normatively unsound. To fill the gatekeeping function that the predominance standard ineffectively attempts, the Article proposes a new "resolvability" test for courts to apply when deciding if a class action would be an appropriate procedural vehicle for adjudicating diverse claims and defenses. The proposed test would permit certification of class actions seeking money damages only when: "The court has a feasible plan to answer all disputed questions of law and fact that must be resolved before entering judgment for or against class members under the law governing each class member's claim and applicable defenses." The Article then discusses broader implications of its proposal that highlight a dynamic relationship between substantive and procedural constraints on regulation of mass risks.