Saturday, May 26, 2007
Posted by D. Daniel Sokol
John Simpson and David Schmidt of the Federal Trade Commission have a new paper that asks the question Difference in Differences Analysis in Antitrust: What Does it Really Measure? on retrospective merger analysis.
ABSTRACT: Merger retrospectives often use a difference in differences (DID) approach to measure the price effects of mergers. As used in these studies, this approach implicitly assumes that the price in the control market fully controls for supply and demand shocks in the treatment market if the two markets experience the same demand and supply shocks. In this paper, we first show that this is only true if the parameters that determine how supply and demand shocks affect price are the same in the two markets. We then show that in many plausible circumstances the DID approach could either overestimate or underestimate the price effects of a merger.