Important doctrines in diverse areas of law employ structured decision procedures requiring, in rough terms, that the plaintiff first make some demonstration of harm; if but only if that is done, the defendant must make some showing of benefit; and if but only if that occurs, balancing is performed. This Article compares such protocols to unconstrained balancing and finds them to be inferior with respect to the quality of final decisions: they sometimes fail to impose liability even though the harm is greater than the benefit, and they sometimes impose liability even though the benefit exceeds the harm. The Article also develops the principles of optimal information (evidence) collection and shows how structured decision procedures violate every core lesson and presuppose distinctions that often are incoherent or impractical to implement. The analysis addresses concerns about balancing that may motivate structured protocols, how less restrictive alternatives should be assessed, and the extent to which legal proceedings are conducted in conformity with either approach, as well as how they might be reformed.
The full text of this Article may be found by clicking on the PDF link to the left.
* Finn M.W. Caspersen and Household International Professor of Law and Economics, Harvard Law School, and Research Associate, National Bureau of Economic Research. I thank Erwin Chemerinsky, Steve Churchill, John Donohue, Richard Fallon, James Greiner, Vicki Jackson, Benjamin Sachs, Holger Spamann, Matthew Stephenson, Adrian Vermeule, and workshop participants at Berkeley, Chicago, Harvard, Northwestern, and Stanford for helpful discussions and comments; Tracy Cui, Carys Johnson, Kevin Keller, Carsten Koenig, Iacopo Lash, Andrea Lowe, Asher Lowenstein, Clarissa Lu, Junesoo Seong, and Aluma Zernik for research assistance; and Harvard’s John M. Olin Center for Law, Economics, and Business for financial support. Disclaimer: I occasionally consult on antitrust cases, and my spouse is in the legal department of a financial services firm.