The early warning system literature on banking crises has often relied on linear classifiers such as the logit model, which are usually estimated with large datasets of multiple regions of countries. We construct an EWS based on an artificial neural network model with monthly data from the...

We analyze the problem of a policy authority (PA) that must decide when to resolve a troubled bank whose underlying solvency is uncertain. Delaying resolution increases the chance that information arrives that reveals the bank's true solvency state. However, delaying resolution also gives...

This paper introduces a new transmission channel of banking crises where sizable cross-border bank claims on foreign countries with high domestic crisis risk enable contagion to the home economy. This asset-side channel opposes traditional views that see banking crises originating from either...

Using new quarterly U.S. data for the past 120 years, I show that sudden reversals in equity and credit market sentiment approximated by several measures of corporate securities issuance are highly predictive of banking crises and recessions. Deviations in equity issuance from historical...