Average Conditional Volatility: A Measure Of Systemic Risk For Commercial Banks

Abstract

We propose using the cross-sectional (daily) average conditional volatility of commercial bank stock returns as a measure of systemic risk for the U.S. banking industry. The performance of this measure is tested using data from the 2008 pre-crisis period. The measure is shown to incorporate individual bank risk as well as the cumulative riskiness of a cross-section of banks. Cross-sectional regressions indicate that individual bank’s probability of default is unrelated to the bank’s conditional volatility during times of low, industry wide risk (as measured by average conditional volatility). However, the bank’s conditional volatility significantly affects its probability of default when the industry is experiencing a high level risk. Regardless of the industry level risk, a bank’s probability of default has a significant negative relation with its capital adequacy (as measured by the proportion of equity capital). Additionally, at an aggregate level, Granger causality tests indicate that the conditional volatility of ‘big’ banks causes the riskiness of medium and small banks to increase.