Up- and Downside Variance Risk Premia in Global Equity Markets

This paper studies the variance risk premium from a new perspective by disaggregating the total premium into upper and lower semivariance premia. To this end, we provide novel tools for computing conditional expectations using traded options as well as moment generating functions. Across a dataset of global stock market indices, we find that the variance premium is almost exclusively driven by downside risk. Our results are robust with respect to the sample period. These findings substantiate the hypothesis found in the literature that the variance premium is largely driven by the left tail of the index return distribution.