Exploring the drivers of China's carbon price spatial correlation network structure

This paper adopts the modified gravity model, social network analysis and quadratic assignment procedure methods to measure the characteristics and analyze the driving factors of China's carbon price spatial correlation network structure in Beijing, Shanghai, Tianjin, Hubei, Guangdong, Chongqing, and Fujian from 2013 to 2023. The modified gravity model constructs the carbon price spatial correlation network, the social network analysis detects the network structure, and the quadratic assignment procedure model analyzes the driving factors of the carbon price spatial correlation network. The results show that China's carbon price has a complex spatial correlation network structure, with Guangdong, Hubei, Beijing and Shanghai at the center of the network, and Chongqing, Tianjin and Fujian at the periphery of the network. The network density and the network correlation show an increasing trend from 2013 to 2023, while the overall spatial correlation is weak. In 2023, Hubei and Shanghai exhibit high degree centrality and are central in the network. Hubei, Beijing, Tianjin, Fujian and Guangdong have high closeness centrality and play the role of central actors, with Hubei leading in carbon price. Hubei, Shanghai and Guangdong have the highest degree of betweenness centrality, acting as intermediaries and bridges in the network. Beijing and Tianjin are in the bidirectional spillover plate, Fujian and Guangdong are in the agent plate, Hubei and Shanghai are in the main outflow plate, and Chongqing is in the main inflow plate. GDP per capita, diesel price, average temperature and geographic distance are important driving factors, significantly affecting the carbon price spatial correlation.