INTRODUCTION

Recent cross-country studies by Honohan and Klingebiel (2003) and Claessens, Klingebiel, and Laeven (Chapter 6) examine to what extent the fiscal costs incurred in a banking crisis can be attributed to specific measures adopted by the governments during the early stages of the crisis, and conclude that “blanket deposit guarantees, open-ended liquidity support, repeated recapitalizations, debtor bailouts and regulatory forbearance all tend to add significantly and sizably to (fiscal) costs.”

The purpose of this chapter is to provide a theoretical framework that can help to understand the different effects of these policies on ex ante risk-shifting incentives and ex post fiscal costs. Specifically, we set up a model of information-based bank runs in which there is a profit-maximizing bank that is funded with insured and uninsured deposits that require an expected return that is normalized to zero. There is a moral hazard problem in that after raising these funds, the bank privately chooses the risk of its loan portfolio. Subsequently, the uninsured depositors observe a signal that contains information on the future return of this portfolio, and withdraw their funds if the signal is bad. In such case, the bank is liquidated unless the government provides the required emergency liquidity or extends the insurance to all depositors. Assuming that the adoption of any of these measures is correctly anticipated by the depositors and the bank, we characterize the equilibrium interest rate of the uninsured deposits and the equilibrium choice of risk by the bank.