Nonlinear Dynamics in Conditional Volatility

Abstract

Investors pay a substantial premium to hedge against fluctuations in volatility—the variance risk premium (VRP). The asset-pricing literature has presented numerous models with jumps in economic fundamentals to reproduce the properties and the time variation of the VRP. This paper shows that these quantitative results are almost exclusively driven by an inaccurate measure of conditional volatility. Solved accurately, conditional volatility exhibits—counterfactually—a strong procyclical pattern and the models do not deliver a sizeable VRP in response to jumps in state variables. The notion that the VRP is purely a compensation for fluctuations in macroeconomic uncertainty does not hold.