Risk spillover effects of climate uncertainty on commodity markets: From the dual perspective of physical risk and transition risk

The issue of climate change is deeply intertwined with human activity and has a significant impact on human existence. To examine the spillover effects of climate change and climate policy on China's commodity market, this study adopts a dual-model approach, combining the time-varying parameter vector autoregressive (TVP-VAR) model and the dynamic conditional correlation generalized autoregressive conditional heteroskedasticity (DCC-GARCH) framework for empirical analysis. Impulse response graphs and dynamic conditional correlation coefficients, we analyze the spillover intensity of climate uncertainty and climate policy on the commodity market from both aggregate and pairwise perspectives via a spillover index. We find that the risk spillovers from climate uncertainty (CU) and climate policy uncertainty (CPU) to commodity markets are dynamic and time-varying, and three distinct peaks in climate uncertainty risk spillovers are identified during 2010–2012, 2016–2018, and 2021–2022. Compared with other industries, the energy and nonferrous sectors are more significantly affected by both physical and transition risks. Based on the TVP-VAR framework, physical climate risks are found to have a stronger spillover effect on China's commodity market. From the perspective of the CU index, the impact on the commodity market is generally negative, whereas the impact of the CPU index is generally positive. While, according to the DCC-GARCH framework, CPU dominates risk spillover, indicating that transition risks are substantially greater than physical risks. The CU index remains relatively stable and positively correlated with various industries, whereas the CPU index shows no such correlation.