Evaluating the Performance of Merger Simulation: Evidence from the U.S. Airline Industry

This paper uses merger simulations to predict postmerger prices for five airline mergers from the 1980s and compares these predictions with observed postmerger prices. I find that standard simulation methods, which measure the effect of the change in ownership on unilateral pricing incentives, do not generally provide an accurate forecast. By incorporating postmerger information into the model, I measure the relative importance of other factors that also contributed to the observed price changes. The results indicate that the unexplained component of the price change is largely accounted for by supply-side effects. I conclude that deviations from the assumed model of firm conduct play an important role in accounting for the differences between the predicted and observed price changes. This conclusion suggests that the predictive performance of merger simulation would benefit if more flexible models of firm conduct were incorporated into the methodology.