This paper provides an innovative perspective on the role of gold as a hedge and safe haven. We use a quantile-on-quantile regression approach to capture the dependence structure between gold returns and changes in uncertainty under different gold market conditions, while considering the nuances of uncertainty levels. To capture the core uncertainty effects on gold returns, a dynamic factor model is used. This technique allows summarizing the impact of six different indexes (namely economic, macroeconomic, microeconomic, monetary policy, financial and political uncertainties) within one aggregate measure of uncertainty. In doing so, we show that the gold's role as a hedge and safe haven cannot be assumed to hold at all times. This ability seems to be sensitive to the gold's various market states (bearish, normal or bullish) and to whether the uncertainty is low, middle or high. Interestingly, we find a positive and strong relationship between gold returns and the uncertainty composite indicator when the uncertainty attains its highest level and under normal gold market scenario. This suggests that holding a diversified portfolio composed of gold could help protecting against exposure to uncertain risks. Acknowledgement: The authors would like to thank the editor Sushanta Mallick and the two anonymous Reviewers for helpful and insightful comments and suggestions on an earlier version of this article.
Highlights We examine the response of gold returns to different uncertainty indicators. We use a quantile-on-quantile regression model. We develop an uncertainty composite indicator based on six uncertainty factors. The hedge and safe haven ability of gold is conditional on gold market states. The gold's role as a hedge and safe haven depends on the nuances of uncertainty.