Fundamentals-Based Panics An Analysis of Bank Runs in Environments with Risky Long-Term Assets

Abstract

This thesis analyzes the effects of uncertain long-term fundamentals on bank runs. Its main contributions are, under uncertain fundamentals, (i) establishing a minimum expected return that enforces incentive compatibility for risk-averse agents, and (ii) deriving a unique equilibrium and calculating the probability of a bank run. There are two models. The first is a modified version of the classic Diamond-Dybvig model in which the economy can be in a good or bad state at the final date. I derive the minimum expected return such that the incentive compatibility constraint is satis ed. The second model has only two agents who receive noisy signals about the probability of the good state of the economy. I make assumptions about relative utilities of certain payoffs, which generate a unique threshold equilibrium in which each agent runs the bank if and only if their signal falls below some critical level. There is an explicit formula for the threshold and, most importantly, the probability of a bank run.