BETWEEN LOGIC AND EXPERIENCE: ERROR COSTS AND UNITED STATES v. MICROSOFT CORP.

摘要

This Article uses United States v. Microsoft Corp. to assess the ongoing debate over how antitrust doctrine should deal with the risk of mistakes and the consequent problem of error costs. I discuss the case in the context of the Justice Department's longstanding involvement in general-purpose computing, with an eye to the ways free and open-source software development practices ("F/OSS") may affect the structure of those markets in the future. In particular, I focus on the D.C. Circuit's two-tiered approach to causation with respect to the monopolization cause of action. The Circuit's approach is best understood as a doctrinal tool designed to minimize error costs. There are two main schools of thought regarding relative error costs. One school, for which Judge Frank Easterbrook has been a consistent and eloquent advocate, holds that society suffers more when courts wrongly find a defendant liable than when they wrongly find no liability. On this view, monopolization cases should be restricted to targeting the few business practices that we understand thoroughly enough to say with confidence that such practices undermine competition. Outside this set of practices, judges should stay the heavy hand of antitrust. A key claim of this school is that markets correct mistakes better than courts do, so I refer to it as the market correction approach. I agree with this view because experience teaches that this approach causes less harm than extensive judicial management of large firms. As I discuss in Part I, however, I concede that, as a strictly logical matter, the arguments for this view are surprisingly weak.