The Winner's Curse: Experiments with Buyers and with Sellers

Abstract

This paper presents a replication and extension of experiments with the "winner's curse" which were initiated in John Kagel and Dan Levin (1984, 1986) and Douglas Dyer et al. (1989). The common-value auction involves firms bidding for an item of unknown common value. Since the value of the item is unknown, the winners can bid more than the value and thereby lose money. The winner's curse occurs if the winners of auctions systematically bid above the actual value of the objects and thereby systematically incur losses. The phenomenon is said to occur possibly in the bidding for such natural resources as mineral rights, where the value of the mineral is unknown but each firm has an estimate of the value. Due to the field nature of the data, doubts have existed as to the actual existence of the curse. The Kagel and Levin (1986) paper tested for the existence of the phenomenon in a laboratory setting. The hope is that, by achieving a thorough understanding of the phenomenon as it might exist in simple laboratory environments, economists will become better equipped to identify and study the phenomenon in more-complex field settings.