Heterogeneous impact of climate policy uncertainty on regional economic development in China: Evidence from double machine learning

Climate change presents a significant challenge to the global economy. To achieve the goal of carbon neutrality, China has proposed diverse climate policies. However, the uncertainty surrounding these policies poses substantial risks, especially in regions with significant differences in economic development levels. This study explores the heterogeneous impact of climate policy uncertainty (CPU) on the gross output value per capita in regions with different economic development levels. Firstly, a cluster analysis of provincial economies is conducted, and then a double machine learning (DML) method is employed for causal inference analysis. Finally, the model is verified through a robustness test, further confirming the robustness and accuracy of the conclusions. The results show that the impact of CPU differs considerably among regions with varying economic development levels: a significant positive effect is observed in developed regions, a moderate positive effect in moderately developed regions, a negative effect in less-developed regions, and a slight positive effect in economically underdeveloped regions. In the process of promoting the "dual carbon" goals, formulating differentiated policy responses based on the industrial structure and fiscal situation of different regions is of great significance for mitigating the adverse effects brought about by CPU.