Does Private Information from Options Markets Forecast Aggregate Stock Returns?

Abstract

We show that the average difference between the implied volatilities of call and put options on individual equities, which we term the implied volatility spread (IVS), has strong predictive power for stock market returns at horizons between one and six months, with monthly in-sample and out-of-sample R-squares around 5%. Controlling for other common predictive variables increases the significance of IVS and lengthens the horizons at which it is significant. We further show that IVS forecasts future surprises in aggregate earnings and growth rates in GDP and aggregate dividends. We conclude that this predictability is likely driven by expectations of stock lending fees embedded in IVS and is inconsistent with explanations based on informed option trading, time-varying risk premia, or illiquidity.