The scope of the hypothesis of Bayesian equilibrium (Q1085080)

The author raises the question of: What behavior, if any, can be explained as the Bayes (Nash) equilibrium of some game? The answer provided by the author is that almost any nondominated behavior can be rationalized as Bayes equilibrium behavior. He shows, in particular, that given any Bayesian (coordination) game with positive priors, and given any vector of nondominated strategies, there is an increasing transformation of each utility function such that the given vector of strategies is a Nash equilibrium of the transformed game. One of the conclusions he arrives at as a result of his analyses is that only if one places some (severe) a priori restrictions on the functional forms of utilities and priors, that one gets meaningful restrictions on behavior. The author also comments that his ''conclusions are similar to those, from Walrasian general equilibrium theory, which apply to aggregate excess demand functions. They do not, however, imply that the hypothesis of Bayesian equilibrium has no content; rather that it is only the combination of assumptions on utilities, priors, and equilibrium which jointly provide meaningful implication''.