Empirical Relevance of Ambiguity in First-Price Auctions

Abstract      

We study the identification and estimation of the first-price auction model when bidders face ambiguity about the valuation distribution and have maxmin expected utility. We exploit variation in the number of bidders to nonparametrically identify the ambiguity structure (the most pessimistic distribution), the valuation distribution, and the risk-aversion (CRRA) coefficient. We propose a flexible Bayesian estimation method based on Bernstein polynomials. Monte Carlo experiments show that our method estimates parameters precisely and chooses the reserve prices with (nearly) optimal revenues whether there is ambiguity or not. Incorrectly assuming no ambiguity may, however, induce estimation bias with a substantial revenue loss. We apply our method to a sample of U.S. timber auctions and find evidence of ambiguity.