《E cient Bargaining over Bets》.pdfVIP

E¢ cient Bargaining over Bets KÖr Eliazy and Ran Spieglerz December 1, 2006 Abstract When two agents hold di§erent priors over an unveriÖable state of nature, which a§ects the outcome of a game they are about to play, they have an incen- tive to bet on the gameís outcome. We pose the following question: what are the limits to the agentsíability to realize gains from such speculative bets when their priors are private information? We apply a ìmechanism designîapproach to this question. We characterize interim-e¢ cient bets and discuss their imple- mentability in terms of the underlying gameís payo§ structure. In particular, we show that as the costs of unilaterally manipulating the betís outcome become more symmetric across states and agents, implementation becomes easier. 1 Introduction In many situations people often hold di§erent opinions about how the future will un- fold. While these di§erences may be partially explained by asymmetric information, some di§erences may still persist even if the individuals were to share all their infor- mation. This could be the result of inherent biases that people have in forming their beliefs, such as optimism and pessimism, or overconÖdence in oneís ability to process information. For example, a die-hard Mets fan, who watches or reads the same sports commentary as a die-hard Yankees fan, would still disagree with the latter on the likelihood that his team would win the World Series. Another example includes entre- preneurs, who often do not share the same views as venture capitalists with regards We thank Barton Lipman, Eric Maskin, Ady Pauzner, Wolfgang Pesendorfer, Ariel Rubinstein and especially Eddi