Thursday, May 29, 2008
Posted by D. Daniel Sokol
Lawrence Frankel of DOJ Antitrust has an interesting piece on The Flawed Institutional Design of U.S. Merger Review: Stacking the Deck Against Enforcement.
ABSTRACT: The institutional design of federal merger review in the United States leads to systematic underenforcement of merger law. This is so even though substantive merger law is relatively well settled, most mergers are not anticompetitive, and the review process properly permits them to proceed without challenge. However, with regard to the small number of mergers that raise significant competitive issues, and where the analysis of the merger's likely consequences turns on highly fact-specific, complex, information-intensive economic assessments based on imperfect data, two fundamental aspects of institutional design create systematic underenforcement. First is the asymmetric availability of judicial review: A determination by the antitrust agency that a merger is anticompetitive is subject to court challenge and judicial review, but a contrary determination is not. Thus, an agency's false positive (erroneous condemnation of a merger as anticompetitive) may be corrected in court, but its false negative (erroneous clearance of a merger as not anticompetitive) is almost always final. Second is the particular nature of judicial review: Because a court's evaluation of an agency determination that a merger is anticompetitive is non-deferential even though the court has less information, less expertise, and fewer resources than the agency and because the agency bears the overall burden of proof in a situation that involves considerable inherent uncertainty, courts will reject the agency's determination more often than they should. Thus, judicial review may correct most false positives, but it will also create additional false negatives. Together, these structural design elements likely lead to significantly more false negatives than false positives and a higher overall error rate than might be produced by various alternative systems. This institutional design might be appropriate if false positives were inherently more harmful than false negatives, but that is unlikely in the context of modern merger control. Thus, the system probably leads to approval of (or suboptimal relief for) an inefficiently large number of anticompetitive mergers. Examination of other advanced merger review systems may yield important insights including ways to correct this problem.