On the Predictive Value of the (Shadow) Real Interest Rate for the Realized Volatility of Gold-Price Returns

Abstract

We use a quasi-out-of-sample forecasting experiment to study the predictive value of a short-term real interest rate for the volatility of gold-price returns. To this end, we use monthly U.S. data for the sample period from 1990/1 to 2022/2, and we study a standard effective-federal-funds-based real interest rate as well as a shadow real interest rate, which accounts for the recent extended zero-lower-bound period. We find that the real interest rate has predictive value for the subsequent realized volatility, and this predictive value turns out to be stronger in several specifications of our forecasting experiment for the shadow real interest rate than for the standard real interest rate. We evaluate the predictive value of forecasts in terms of an asymmetric loss function. Because gold is considered as a safe-haven asset, our results provide some important implications for portfolio decisions of investors.