This article reports the results of a first-price sealed-bid auction experiment, which has been designed to test the Nash equilibrium predictions of individual bidding behavior. Subjects faced in 100 auctions always the same resale value and competed with computerized bids. Three treatments were considered which varied with the conditions of information feedback. In earlier experimental work an overbidding above the risk neutral Nash equilibrium has been frequently reported. Our data provide evidence that this overbidding regularity can be a consequence of the standard information feedback in auction experiments of revealing only the winning bid after each auction. By means of learning direction theory we explain the individual bidding dynamics. Finally we apply impulse balance theory and make long run predictions of the individual bidding behavior.
Experimental economics first-price sealed-bid auctions independent private value model computerized competitors bidding theory risk aversion