Equilibrium in risk-sharing games (Q2364537)

The authors model how economic agents choose the beliefs on future uncertain events that they are going to declare to their counterparties and study whether such strategic behaviour yields an equilibrium in a two-period financial economy. Presented results demonstrate how the game leads to risk-sharing inefficiency and security mispricing, both of which are quantitatively characterised in analytic form. It is shown that equilibrium securities have endogenous limited liability, a typically suboptimal feature that is observed in practice. Even though agents' set of strategic choices is infinite-dimensional, the authors, among other things, establish that a Nash equilibrium admits a finite-dimensional characterisation; in fact, the dimension is one less than the number of participating agents. This characterisation in particular provides a concrete algorithm to compute the equilibrium transaction, as well as to prove the existence of a Nash equilibrium for an arbitrary (finite) number of players. In an important case of two players, Nash equilibrium is shown to be unique. Except for the assumed CARA preferences, no extra assumptions on the probability space or random payoffs are made. Further, the authors also show that agents with sufficiently low risk-aversion will prefer the risk-sharing game to the outcome of an Arrow-Debreu equilibrium that would have resulted from the absence of strategic behaviour, leading to an aggregate loss of risk-sharing welfare; this result is valid irrespective of the actual risky positions and subjective beliefs of the agents.