Collusion Theories in Merger Analysis: Still Alive and Kicking

In a 2008 paper published in this journal, I described the continuing success of coordinated interaction (collusion) theories in maintaining their role as an alternative analytical technique to the unilateral effects theories used in Federal Trade Commission merger reviews. While recent Agency commentary and guidelines have suggested a further shift in policy towards unilateral effects analysis, collusion analyses remain entrenched in the internal files. Staff appears to apply the theory most compatible with the available facts.

After an overview of developments in merger policy, this paper compares the level and outcome of collusion and unilateral effects analyses over the 1993-2010 time period. Remarkably little change in relative activity is observed and challenge rates remain relatively constant once the samples are standardized for entry impediments. Focusing on the counts of significant rivals, the data shows that both four-to-three and three-to-two mergers tend to end in challenges when entry evidence is strong. Moreover, a reasonable case can be made that the challenge rates are higher in homogeneous goods markets when five or more premerger rivals exist, although the small sample size limits this conclusion.

Thus overall, facts, not theory, appear to affect FTC evaluations.