Content | We use laboratory experiments to examine the relative performance of the English auction (EA) and the first-price sealed-bid auction (FPA) when procuring a commodity. The mean and variance of prices are lower in the FPA than in the EA. Bids and prices in the EA agree with game-theoretic predictions, but they do not agree in the FPA. To resolve these deviations found in the FPA, we introduce a mixture model with three bidding rules: constant absolute markup, constant percentage markup, and strategic best response. A dynamic specification in which bidders can switch strategies as they gain experience is estimated as a hidden Markov model. Initially, about three quarters of the subjects are strategic bidders, but over time, the number of strategic bidders falls to below 65%. There is a corresponding growth in those who use the constant absolute markup rule. |