Measuring Systemic Risk: Common Factor Exposures and Tail Dependence Effects

We model systemic risk by including a common factor exposure to market-wide shocks and an exposure to tail dependence effects arising from linkages among extreme stock returns. Specifically our model allows for the firm-specific impact of infrequent and extreme events. When a jump occurs, its impact is in the same direction for all firms (either positive or negative), but its size and volatility are firm-specific. Based on the model we compute three measures of systemic risk: DD, NoD and ESR. Empirical results using data on the four sectors of the U.S. financial industry from 1996 to 2011 suggest that simultaneous extreme negative movements across large financial institutions are stronger in bear markets than in bull markets. Disregarding the impact of the tail dependence element implies a downward bias in the measurement of systemic risk especially during weak economic times. Two measures based on the Broker-Dealers sector (DD, NoD) and one measure (ESR) based on the Insurance sector lead the St. Louis Fed Financial Stress Index (STLFSI).