Crash risk connectedness in commodity markets

Abstract

ABSTRACT In contrast to the extant literature on returns, volumes, and volatility spillovers, we examine the crash risk connectedness of 13 commodity markets in the energy, metal, and agricultural sectors, spanning January 2019 to December 2020. We use 5-minute high-frequency data to construct two proxies of daily crash risk and apply spillover analysis and network graphs to show that overall connectedness among commodity uncertainties is most significant in the aftermath of COVID-19. High intra-group connectedness and low inter-group connectedness are observed. Since there is an inter-commodity disconnection and intra-commodity connection, the findings suggest that commodity investors diversify crash risk by choosing commodities from different commodity groups rather than picking from within the commodity group. Interestingly, copper and gas exhibit a disconnection from their respective groups, making them suitable diversifiers for other commodity groups and individual commodities in their own groups. Moreover, this property of copper and gas is not much affected during COVID-19. Crash risk connectedness is asymmetric regarding long- and short-run investment horizons, and fear of 2nd wave of COVID-19 signifies pronounced long-term connectedness. Our findings are robust to the use of the alternative specification. Individual investors and commodity export-dependent economies can learn important implications from our analysis.