Average tail risk and aggregate stock returns

Abstract

We investigate the role of the average risk across stocks in predicting subsequent market returns using measures of risk that capture the higher moments of the return distribution including variance, skewness and kurtosis, as well as measures of tail risk that combine these. We find that average tail risk has statistically and economically significant predictive ability for market returns, even after controlling for market tail risk, suggesting that average idiosyncratic tail risk contains information about future returns. Average tail risk dominates other measures of average risk that have been documented in the literature, such as variance and skewness. Our results are robust to the inclusion of control variables that capture business cycle effects, and to the use of different measures of tail risk.