Collaborative Research in Experimental Studies of Auctions and Markets with Asymmetric Information

点击翻译按钮获取中文摘要

This project extends and completes experimental investigations of English common value auctions, common value auctions with asymmetric (insider) information, and learning processes in experimental auction markets. The project also finishes a case study of the bidding process in the Houston construction industry. Under the previous NSF grant the investigators found conclusive evidence of a "winner's curse" (winning bidders suffer losses) in English auctions. This is an important result methodologically because it contradicts existing theories of bidding and it is important substantively since the English common value auction is a widely used mechanism for auctions. The experiments employed an English clock procedure whereby prices start at the lowest possible value for the item and increase steadily. Bidders are active until they signal their intention to drop out of the auction, which decision is irrevocable for the sale of that item. The last active bidder wins the item, paying the price at which the next to last bidder dropped out. The experiments demonstrated that the dropout prices of losing bidders reveal information to the other bidders and reduce but do not eliminate the winner's curse. Further research should provide new and useful insights into institutions and practices that mitigate or eliminate entirely the winner's curse in auctions. The case study of the construction industry in Houston is especially interesting. In order to determine the generalizability of the auction market findings, the investigators conducted a series of low price, common value offer auctions in comparing the performance of student subjects with experienced professionals from the Houston construction industry. Qualitatively similar behavior was found across subject pools as in both cases bidders suffered from the winner's curse, losing money as a consequence of winning the item, and increasing the number of bidders exacerbated the losses. These results raised an important question: Why are business executives successful in real world auctions and unsuccessful in laboratory auctions? The case study identifies a number of real world institutional features and practices that reduce the effects of the winner's curse.