Naturally-occurring auction markets are characterized by endogenous determination of bidder entry and exit. In contrast, most previous theoretical and experimental research has assumed that the number of bidders in an auction is exogenously determined. The proposed research involves developing the theory of and undertaking experiments for auctions with endogenous entry and exit decisions. The first application will be to paired private value auctions, where the value of an auctioned item may differ for different bidders, and to common value auctions, where the value of an auctioned item is the same for all bidders. This application is motivated by earlier experiments with common value auctions in which it was observed that: (1) in small groups of 3 or 4 people, subjects' earning were only about 65% of predicted profits; and (2) in larger groups of 6 to 9 people, subjects incurred losses, followed by bankruptcies, causing attrition in the number of bidders. Theory and experiments that are designed to accommodate endogenous entry and exit of bidders will be developed. The new auction framework will be based on independent draws of individual agents' opportunity costs of bidding. These will be used in both the common value and private value environments. Paired experiments for common and private value auctions that will test the implications of the new models will be compared, including their ability to predict individual exit and entry behavior. The endogenous-bidders extension of the economics of auction markets will lead to various insights, and provide a more fundamental conception of the winner's curse.
