Market Variance Risk Premiums in Japan as
Predictor Variables and Indicators of Risk Aversion
Abstract
This article evaluates the predictive performance of the market variance risk premium (VRP) in Japan on the Nikkei 225 returns, credit spreads, and the composite index of coincident indicators. Different measures such as expected and ex-post VRPs, which are constructed from model-free implied and realized variances, are used to verify the predictability. Moreover, the VRP is estimated by the Bollerslev, Gibson and Zhou (2011) method using Japanese macroeconomic variables to approximate the dynamics of the representative investor's relative risk aversion. The main empirical findings are: (i) the ex-post VRP, which is defined as the difference between implied and ex-post realized variances, is useful in predicting the Nikkei 225 returns, whereas the expected VRPs, which are the differences between implied and current or model-based realized variances, lose their predictive ability, (ii) the expected and ex-post VRPs provide significant predictability of credit spreads and the composite index of coincident indicators, (iii) the VRP involving Japanese macroeconomic variables contains plausible business cycle dynamics of the Japanese economy.
