Estimation of cost synergies from mergers without cost data : Application to U . S . radio ∗

This paper develops a new way to estimate fix cost synergies from mergers without using cost data. The estimator uses a structural model in which companies play a dynamic game with endogenous mergers and product repositioning decisions. This formulation has several benefits over the standard static merger analysis. In particular, it corrects for a sample selection of more profitable mergers and captures follow-up mergers as well as a post-merger product repositioning. The basic idea behind the estimator is treating mergers as endogenous. This allows for a comparison of mergers observed in the data with counterfactual ones based on simulated long-run gains for different levels of cost synergies. A bounds estimator is produced which, depending on variation in the data, provides partial or pointwise identification of the parameters. The framework is applied to estimate cost efficiencies after the de-regulation of U.S. radio in 1996. I find that average cost savings (across all owners) over 1996-2006 amount to $0.5M (22%) per owned station. Total cost savings only from mergers after 1996 amount to about $1B per year. This outweighs the loss of advertiser surplus caused by increased market power,which has been estimated to be $223M by Jeziorski (2011). ∗I would like to thank: Lanier Benkard and Peter Reiss, Joseph Harrington, Elena Krasnokutskaya, Ilya Segal, Alan Sorensen, Andrew Sweeting, Benjamin Van Roy and Ali Yurukoglu, as well as participants of numerous seminars. †Department of Economics, Johns Hopkins University