Experimental evidence of bank runs as pure coordination failures

This paper investigates how the level of coordination requirement affects the occurrence of bank runs as a result of pure coordination failures in controlled laboratory environments. We find that miscoordination-based bank runs can be observed when the coordination parameter, defined as the amount of coordination among depositors choosing to wait that is required for them to receive a higher payoff than depositors who choose to withdraw, is high enough. In particular, we can divide the values of the coordination parameter into three regions: the run region characterized by high values of the parameter, the non-run region characterized by low values of the parameter and the indeterminacy region characterized by intermediate values of the parameter. In the run (non-run) region, all experimental economies stay close or converge to the run (non-run) equilibrium. When the coordination parameter lies in the indeterminacy region, the outcomes of the experimental economies vary widely and are hard to predict. Some learning and path-dependence effects are also detected when the coordination parameter is located in or around the indeterminacy region. Finally, we show that behavior of human subjects observed in the laboratory can be well accounted for by the logit evolutionary algorithm. JEL Categories: D83, G20