We show that the equal-weighted average stock volatility analyzed by Goyal and Santa-Clara (GS, 2003) forecasts stock returns because of its co-movements with stock market volatility. Moreover, contrary to the positive relation hypothesized by GS and many others, we find that the value-weighted average stock volatility is negatively related to future stock returns when combined with stock market volatility. This puzzling result reflects the fact that the valueweighted average stock volatility is negatively correlated with the consumption-wealth ratio, and its predictive power vanishes if we control for the latter in the forecasting equation. The idiosyncratic volatility proposed by GS thus provides no information beyond the forecasting variables advocated by Guo (2003)
https://doi.org/10.20955/wp.2003.025
follow @stlouisfed