Estimating Discrete Games with Many Firms and Many Decisions : An Application to Merger and Product Variety

This paper presents a new method for estimating discrete games based on bounds of conditional choice probabilities. The method does not require solving the game and is scalable to models with many firms and many discrete decisions. We apply the method to study merger effects on firm entry and product variety in the retail craft beer market in California. We simulate an acquisition of multiple craft breweries by a large brewery and find that the acquisition would induce firm entry and product entry by non-merging firms. However, these changes are insufficient to offset the negative welfare effects resulting from the higher prices and decreased product offerings by the merging firms