My latest working paper, which bears the same title as this post, is now available on SSRN. In the paper, I address the challenge created by the Supreme Court’s 2007 Leegin decision, which abrogated the 96 year-old rule declaring resale price maintenance (RPM) to be per se illegal. The Leegin Court held that instances of RPM must instead be evaluated under antitrust’s more lenient rule of reason. It also directed lower courts to craft a structured liability analysis for separating pro- from anticompetitive instances of the practice.
Since Leegin was decided, courts, commentators, and regulators have proposed at least four types of approaches for evaluating instances of RPM. Some of the approaches, like that advocated by the American Antitrust Institute, focus on whether an instance of RPM has raised consumer prices. Others, like that set forth in the pending Toys-R-Us case, focus on the identity of the party initiating the RPM (manufacturer or retailer(s)?). Some, like that proposed by Professor Marina Lao, focus on whether the product subject to RPM is sold with retailer services that are susceptible to free-riding. One approach, that endorsed by the FTC, mechanically applies factors the Leegin Court mentioned as relevant, but with little consideration of the potential for proof failures.
My paper critiques these four approaches from the perspective of decision theory (or what Josh and Geoff might call error cost analysis). Recognizing that antitrust liability rules always involve a risk of imposing social costs — either losses from under-deterrence if the rule wrongly acquits anticompetitive acts or losses from over-deterrence if it wrongly convicts procompetitive practices — decision theory says liability rules should be tailored to minimize the expected total cost of a liability decision. Specifically, the optimal rule will minimize the sum of decision costs (the costs of reaching a decision) and expected error costs (the costs of getting the decision wrong).
To evaluate how the proposed RPM rules fare from a decision-theoretic perspective, I begin by considering the theoretical harms and benefits associated with RPM and the empirical evidence on the incidence of those various effects. This analysis leads me to conclude that most instances of RPM are pro- rather than anticompetitive. I then consider whether wrongful convictions or wrongful acquittals are likely to cause greater social losses, and I conclude that wrongful convictions threaten greater harm. Taken together, these two conclusions call for a liability rule that tends to acquit more instances of RPM than it convicts. The proposed liability approaches, by contrast, are tilted toward conviction. Moreover, several of the proposed approaches would condemn instances of RPM even when the preconditions for anticompetitive harm are not satisfied.
Finding each of the proposed liability analyses to be deficient, I set forth an alternative approach that (1) reflects the economic learning on RPM (with respect to both the theories of competitive effects and the empirical evidence of those various effects), (2) is aimed at minimizing the costs of incorrect judgments, and (3) would be fairly easy for courts and business planners to administer. The proposed approach, in short, aims to minimize the sum of decision and error costs in regulating RPM.
Please download the piece. Comments are most welcome.
