Working Paper: CEPR ID: DP8149
Authors: Geert Bekaert; Robert J. Hodrick; Xiaoyan Zhang
Abstract: We examine aggregate idiosyncratic volatility in 23 developed equity markets, measured using various methodologies, and we find no evidence of upward trends when we extend the sample till 2008. Instead, idiosyncratic volatility appears to be well described by a stationary autoregressive process that occasionally switches into a higher-variance regime that has relatively short duration. We also document that idiosyncratic volatility is highly correlated across countries. Finally, we examine the determinants of the time-variation in idiosyncratic volatility. In most specifications, the bulk of idiosyncratic volatility can be explained by a growth opportunity proxy, total (U.S.) market volatility, and in most but not all specifications, the variance premium, a business cycle sensitive risk indicator. Our results have important implications for studies of portfolio diversification, return volatility and contagion.
Keywords: Contagion; Diversification; Growth Opportunities; Idiosyncratic Volatility; Regime Switching Model; Return Correlation; Trend Test; Variance Premium; Volatility Dynamics
JEL Codes: C52; G11; G12
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Growth opportunity proxy (O40) | Aggregate idiosyncratic volatility (C43) |
| Total US market volatility (N22) | Aggregate idiosyncratic volatility (C43) |
| Variance premium (G19) | Aggregate idiosyncratic volatility (C43) |
| Aggregate idiosyncratic volatility across countries (F39) | Common driving force (C69) |