Abstract
Recent international evidence on the Variance Risk Premium, defined as the difference between implied and realised return variation, shows that this metric has predictive power in forecasting aggregate stock market returns. This evidence seems to be strongest for the US market. In this article we conduct a study on the ability of the Variance Risk Premium to predict aggregate stock returns for both the US market and the South African market, allowing us to contrast our South African results against the US. Our empirical study using US data confirms the predictive power of the Variance Risk Premium in the US market. The evidence from our empirical study in the South African market however reveals little or no predictive power of South African market returns, even when combined with typical economic predictor variables. Thus the use of a South African Variance Risk Premium for predictive purposes in the South African environment is currently not supported by this research. Our analysis does however suggest that the US VRP does have some predictive power for the SA stock market, but that this relationship is period dependent and might not persist into the future.
Notes
1 While most of the published research refers to the VRP as the Variance Risk Premium, in Bollerslev, Tauchen and Zhou (2009) the authors comment that replacing the variance with volatility, leads to relatively similar results.
2 The history for the SAVI only goes back to 2007. For prior periods we use the implied volatility of the 3month ALSI near futures contract from INet BFA.