Measuring Skewness Premia

We provide a new methodology to empirically investigate the respective roles of systematic and idiosyncratic skewness in explaining expected stock returns. Forming a risk factor that captures systematic skewness risk and forming idiosyncratic skewness sorted portfolios only require the ordering of stocks with respect to each skewness measure. Accordingly, we use a large number of predictors to forecast the cross-sectional ranks of systematic and idiosyncratic skewness which are considerably easier to predict than their actual values. Compared to other measures of ex ante systematic skewness, our forecasts create a significant spread in ex post systematic skewness. A predicted systematic skewness risk factor carries a significant risk premium that ranges from 7% to 12% per year and is robust to the inclusion of downside beta, size, value, momentum, profitability, and investment factors. In contrast to systematic skewness, the role of idiosyncratic skewness in pricing stocks is less robust. Finally, we document how the determinants of systematic skewness differ from those of idiosyncratic skewness.