Leading Indicators of Crisis Incidence: Evidence from Developed Countries

Abstract

We search for early warning indicators that could indicate important risks in developed economies. We therefore examine which indicators are most useful in explaining costly macroeconomic developments following the occurrence of economic crises in EU and OECD countries between 1970 and 2010. To define our dependent variable, we bring together a (continuous) measure of crisis incidence, which combines the output and employment loss and the fiscal deficit into an index of real costs, with a (discrete) database of crisis occurrence. In contrast to recent studies, we explicitly take into account model uncertainty in two steps. First, for each potential leading indicator, we select the relevant prediction horizon by using panel vector autoregression. Second, we identify the most useful leading indicators with Bayesian model averaging. Our results suggest that domestic housing prices, share prices, and credit growth, and some global variables, such as private credit, are risk factors worth monitoring in developed economies.

Keywords: Early warning indicators, Bayesian model averaging, macro-prudential policies

JEL Classification: C33, E44, E58, F47