Connectedness and risk spillover in China's commodity futures sectors

Abstract

This study employs minimum spanning tree and generalized forecast error variance decomposition methods to investigate the connectedness and risk spillovers across China's commodity sectors from January 2016 to December 2021. The results show that total connectedness within the commodity system is time varying. Chemical is the main risk driver, while other sectors occasionally dominate the system. These two methods achieve consistent results in identifying the systemically important sector and dynamic connectedness. In addition, we find that Chinese economic policy uncertainty and the investor sentiment index have significant impacts on total connectedness. Our findings have implications for preventing systemic risk for policymakers and managing commodity portfolio risk for investors.

High–low volatility spillover network between economic policy uncertainty and commodity futures markets

“…Second, when the net contribution of commodities changes rapidly, the market structure also changes. For instance, during the COVID-19 pandemic, almost all commodities experience significant fluctuations, which is in line with Long et al (2024). Furthermore, the same crisis event would have different impacts on different types of commodity markets, which is largely similar to those of Xiao et al (2020).…”

“…Figure 6 As shown in Figure 6, network centrality of commodity markets changes over time in the high and low volatility spillover network. According to Long et al (2024), due to the presence of bull and bear markets, the dependency structure of the commodity market changes over time, which may result in changes in central commodity markets. First of all, we find that after the global financial crisis, some commodities in the high volatility network layer show an increase in closeness centrality.…”