Measuring and Testing Systemic Risk from the Cross-Section of Stock Returns

Abstract

Abstract This study proposes a novel measure of systemic risk that is obtained by aggregating downside risk information from the cross section of assets. In contrast to existing studies, we expand the analysis of systemic risk to many assets and focus on marginal measures of tail risk that are aggregated using a Fisher-type test to detect the risk of systemic events. The presence of downside risk for each asset of the cross section is examined through a bootstrap test of first-order stochastic dominance between the underlying tail distribution and the tail distribution of the residuals of a multivariate DCC-GARCH model. The application of these methods to the cross section of the FTSE-100 stock returns provides overwhelming evidence on the presence of financial instability during the period 2006–2009. Interestingly, we also find compelling evidence of systemic risk during the 2012–2015 period coinciding with the European debt crisis and after the outbreak of the coronavirus disease 2019 pandemic.