The Santa Fe Bar Problem Revisited:
Theoretical and Practical Implications

Amy Greenwald, Bud Mishra, and Rohit Parikh

Abstract

This paper investigates the Santa Fe bar problem in detail from both a theoretical and a practical perspective. Theoretically, it is shown that traditional assumptions of economics, such as rationality, do not give rise to desirable behavior in this problem. Specifically, rationality and predictivity, two conditions sufficient for convergence to Nash equilibrium, are inherently incompatible. On the practical side, it is demonstrated via simulations that computational learning algorithms in which agents are occasionally irrational do yield near-equilibrium behavior.