Continuum Nash Bargaining Solutions

Abstract

Nash`s classical bargaining solution suggests that n players in a non-cooperative bargaining situation should find a solution that maximizes the product of each player's utility functions. We consider a special case: Suppose that the players are chosen from a continuum distribution $\mu$ and suppose they are to divide up a resource $\nu$ that is also on a continuum. The utility to each player is determined by the exponential of a distance type function. The maximization problem becomes an optimal transport type problem, where the target density is the minimizer to the functional \[ F(\beta)=H_{\nu}(\beta)+W^{2}(\mu,\beta) \] where $H_{\nu}(\beta)$ is the entropy and $W^{2}$ is the 2-Wasserstein distance. This minimization problem is also solved in the Jordan-Kinderlehrer-Otto scheme. Thanks to optimal transport theory, the solution may be described by a potential that solves a fourth order nonlinear elliptic PDE, similar to Abreu's equation. Using the PDE, we prove solutions are smooth when the measures have smooth positive densities.