Auctions with competing sellers and behavioral bidders

Abstract

With the rise of the Internet, the use of auctions has become increasingly prevalent. Nowadays, consumers can buy a myriad of goods by means of online auctions – cars, holidays, clothing, sports items, electronics and even lab equipment are auctioned on the Internet. Economists have studied auctions for several decades, but have traditionally focused on a monopolistic auctioneer selling a single good in a standard auction to a fixed number of fully rational bidders, who are solely concerned with maximizing monetary payoffs. The recent flourishing of online auctions, however, requires that the traditional literature is modified in at least three ways. That is, auction theorists studying online auctions need to consider that bidders may have non-standard preferences and endogenously enter competing auctions, where these auctions may include non-standard ones. The contributions in this dissertation revolve around these three departures from the traditional literature. The dissertation begins by studying entry into competing auctions. On the Internet, goods are sold in numerous simultaneous auctions, allowing bidders to choose which auction to enter. Hence, auctioneers compete against one another to attract bidders, and should be strategic when deciding which auction to offer. In the first study, I theoretically investigate which auctions–first price or second-price–are selected by competing auctioneers when risk averse bidders endogenously enter these auctions. I show that when choosing between entering the first-price and second-price auction, bidders enter each auction with equal probability if they are risk neutral or exhibit constant absolute risk aversion. However, bidders enter the second-price auction with greater probability if they exhibit decreasing absolute risk aversion, and enter the first-price auction with greater probability if they exhibit increasing absolute risk aversion. As risk averse bidders overbid in first-price but not in second-price auctions, this implies that competing sellers have a dominant strategy to select first-price auctions when bidders exhibit nondecreasing absolute risk aversion. If bidders exhibit decreasing absolute risk aversion, sellers may also select second-price auctions. The second study continues to investigate entry into auctions, but considers non-standard auctions and non-standard preferences. By means of an exploratory experiment, I analyze bidders' decisions between participating in an auction with or without a Buy-It-Now option and buying at a posted price. Which selling mechanism is preferred by consumers? Can these preferences be explained by differences in expected payoffs across mechanisms, or do consumer characteristics also play a role? I find subjects enter the posted price less often than the auctions. These preferences can be explained by expected payoffs—subjects use a cut-off based on the posted or Buy-It-Now price when making entry decisions—and by impatience and risk aversion. Furthermore, I find strong evidence for the existence of a gender effect, as males are more likely to enter mechanisms involving bidding, and females are more likely to enter mechanisms involving buying. Finally, in the third study, I theoretically investigate how bidding behavior in first-price and second-price auctions is affected by the presence of social competition and, in particular, social comparison concerns. Social psychologists have argued that the tendency to compare ourselves to others may generate competitive behavior. Auctioneers and auction theorists alike have argued that the presence of social competition is one of the main drivers of the success of auctions. Adopting a model of interdependent preferences, where bidders experience envy and/or pride when comparing their payoffs to those of others, I show that social comparison concerns result in more competitive bidding behavior in both first-price and second-price auctions. This effect is driven by the anticipation of envy; anticipating pride does not generate overbidding in isolation.