Analyzing the mediating role of industrial structure in the spatial spillover effects of financial agglomeration on carbon emission and regional heterogeneity

In China, achieving financial agglomeration necessitates both the gathering of financial resources and the reduction of carbon emissions and the interrelation between these two goals is significant. In this research, sophisticated econometric models are applied to examine the correlation between financial agglomeration and per capita carbon emissions in China, such as spatial econometric, mixed OLS, and stationary panel models. The research sample is composed of data from 30 provinces and cities in China from 2010 to 2020 and examines both temporal and spatial distributions of these two factors and how they influence each other. The direct effect of financial agglomeration on carbon emission is analyzed through spatial panel model, while indirect effect is analyzed by examining the role of industrial structure upgrading as a mediating variable through mediating effect model. This study also looks at how these effects vary regionally, both directly and indirectly. Generally, the study discovered that financial agglomeration and per capita carbon emissions have significantly positive spatial autocorrelation coefficient in all provinces and cities in China, indicating path dependence and spatial spillover. In terms of distribution trends, financial agglomeration shows an upward trend over time, while per capita carbon emissions grew faster in the early stage, but gradually achieves a steady decrease in recent years. In terms of the impact of financial agglomeration on carbon emissions, the relationship between financial agglomeration and per capita carbon emissions is in the form of "inverted U-shaped." Financial agglomeration indirectly influences the intensity of per capita carbon emissions through the advanced industrial structure, which acts as a mediator. In the test of industrial structure serving as a mediating variable, different regions exhibit different impacts due to regional heterogeneity, with a clear distinction between the central region and the eastern and western regions.