A large-scale evaluation of merger simulations

Grant description:

This project asks whether standard merger simulation techniques in industrial organization effectively predict price changes in observed mergers, and if not, if predictions depart from reality systematically and in a way consistent with efficiencies or coordinated effects. Using scanner data, the authors will run a standard merger simulation on a large set of completed mergers and compare predictions to outcomes, creating a comprehensive retrospective of the effects of mergers on prices, which will inform us of whether typical approved mergers in the United States tend to increase prices. They will also study the sources of the prediction error.

Regulation of merger policy is a primary tool of competition policy in the United States. Merger simulations are used to decide whether mergers are anti-competitive or whether they should be permitted. This ambitious project could provide a wealth of information about consummated mergers and the predictive power of merger simulation techniques, contributing to the infrastructure used to regulate competition.