Counting Rivals or Measuring Share: Modeling Unilateral Effects for Merger Analysis

Posted by D. Daniel Sokol

Malcolm Coate (FTC) has posted Counting Rivals or Measuring Share: Modeling Unilateral Effects for Merger Analysis.

ABSTRACT: This paper explores the FTC’s unilateral effects merger policy using a sample of 184 investigations undertaken between 1993 and 2009. A review of the files suggests that roughly half of the sample is evaluated with a dominant firm/monopoly model, while the rest of the cases require a more complex unilateral effects analysis. Deterministic modeling based on the number of significant rivals suggests that the four-to-three transaction in a market with impediments to entry represents the marginal merger challenge. Case specific facts explain deviations from this rule and suggest that critical diversion ratios fall into the 25-30 percent range. Share based indices (post-merger market share, change in the Herfindahl, or a share-based Gross Upward Pressure on Price variable) can be used, but require the definition of a market and do not predict outcomes as well as the significant rivals’ model. An Appendix details the various reasons why the staff declined to apply a unilateral effects analysis to conclude a merger was likely to substantially lessen competition in a broader sample of differentiated products mergers.