You are not currently logged in.
Access JSTOR through your library or other institution:
Don't Bet on It: Contingent Agreements with Asymmetric Information
James K. Sebenius and John Geanakoplos
Journal of the American Statistical Association
Vol. 78, No. 382 (Jun., 1983), pp. 424-426
Stable URL: http://www.jstor.org/stable/2288651
Page Count: 3
Preview not available
If two people have different probability assessments about the realization of an uncertain event, they can design a contingent agreement such as a bet or gamble that offers each of them positive expected value. Yet, in the process of formulating this kind of agreement, information about the basis for each person's probabilities may be indirectly revealed to the other. The very willingness to accept a proposed bet conveys information. This article models a process by which private, asymmetrically-held information is progressively unveiled as a possible contingent agreement is discussed. If the two parties share priors and their information partitions are common knowledge, simple discussion of the acceptability of any proposed bet is shown to reveal enough about the parties' private information to render the bet unacceptable.
Journal of the American Statistical Association © 1983 American Statistical Association