Low probability, high impact: Policy making and extreme events

The objective of this paper is to derive key determinants for the optimal design of early warning system (EWS) models to anticipate extreme events, such as currency crises in emerging markets. We show how the design of an “optimal” model for policy makers focuses on the choice of three parameters: the degree of risk aversion of failing to anticipate an event, the forecast horizon of the model, and the probability threshold for extracting warning signals. Based on a representative EWS model, we find that, for a given degree of risk aversion, there is a unique combination of the forecast horizon and of the probability threshold that maximizes the policy maker's preferences, yielding the best possible model from a policy perspective. Thus the analysis of the paper allows deriving clear policy implications concerning the choice of action, and in particular the timing of measures to be taken by policy-makers if a financial crisis may be looming in the future.