Option-Implied Volatility Measures and Stock Return Predictability

Do changes in implied volatilities (IVs) or differences among options at different spots on the volatility surface contain predictive information for future returns? The question has been asked repeatedly—and often answered in the affirmative for specific measures—but questions remain. In this article, the authors perform a comparison test on six return predictors that are all computed as differences among implied and realized volatilities and that have been proposed in the literature. The authors consider more measures, more maturities, and longer samples than earlier articles, and they include a variety of appropriate control variables. They find that the difference between at-the-money (ATM) call and put IV (CPIV) and out-of-the-money (OTM) put IV minus ATM call IV are both highly significant at the individual stock level, while a long–short portfolio strategy based on realized minus implied volatility does very well. The alternative volatility combinations are substantially less effective as return predictors