The Risk Profile of Gold Mining Stocks

Abstract

The purpose of this paper is to assess the economic determinants of gold mining stocks. We are especially interested in the relationship between gold stocks and the gold price. To examine the risk profile of the gold stocks, we regress the excess return of a portfolio consisting of 182 gold mining stocks, on seven risk factors: the market return, the gold price, the USD exchange rate, the long-term interest rate, the oil price, the small-minusbig portfolio and the high-minus-low portfolio. The regressions are run both at the mean, using ordinary least squares, and at different quantiles, using the quantile regression approach. To study cross-sectional differences, we apply the Fama-MacBeth regression on the most relevant risk factors. The results from the quantile regression show that only a few factor coefficients significantly differ from the ordinary least squares estimates. However, both the interest rate and the market factor show signs of upper tail dependency with the gold stock portfolio. The gold price is the only factor that is significant over the entire return distribution, even though we find that several factors significantly explain gold stock returns both at the mean and at different quantiles. The gold price factor constitutes most of the explanatory power, consequently being the main driver of the return of gold stocks. Interestingly, we find that big market capitalisation gold stocks exhibit stronger gold risk exposures, and lower average returns, than small capitalisation gold mining companies. This is accompanied by the Fama-MacBeth regression, where we find that both the gold price exposure, market capitalisation and book-to-market equity explain the cross-section of gold stocks. The same analysis shows no significant explanatory power of the market beta on gold stocks. This is of relevance to investors, as it suggests that the market prices gold risk, but not market risk, for gold mining stocks. As such, an individual who invests in a gold mining company should expect higher average returns for a stock with a high gold beta, than for a stock with a low gold beta.