This paper develops a new Early Warning System (EWS) model for predicting financial crises, based on a multinomial logit model. It is shown that EWS approaches based on binomial discrete-dependent-variable models can be subject to what we call a post-crisis bias. This bias arises when no...

The objective of this paper is to implement a prototype of a currency crisis model as part of an early warning system framework for Uganda. The financial systems of developing countries like Uganda are especially vulnerable and therefore robust instruments to predict crises are needed. Our model...

In recent years, Fund staff has prepared cross-country analyses of macroeconomic vulnerabilities in low-income countries, focusing on the risk of sharp declines in economic growth and of debt distress. We discuss routes to broadening this focus by adding several macroeconomic and macrofinancial...

Early warning systems (EWSs) are subject to restrictions that apply to exchange rates in general: fundamentals matter but their influence is small and unstable. Despite this limitation four major lessons emerge: First, EWSs have robust forecasting power and thus help policy-makers to prevent...

To complement the early warning signals literature, we study the determinants of banking and currency crises for small states and currency boards. Building on the crisis dataset by Laeven and Valencia (2020), we estimate a binominal logit model to identify the determinants of crises, and as a...