Analysis and validation of energy-conservation and emission-reduction effects of economic agglomeration

Economic agglomeration, as a key factor of China’s economic transformation and development, offers great potential for both global and Chinese carbon reduction. This study explores the pathways and principles of economic agglomeration on CO₂ emissions by applying the framework of spatial economic theory. The proposed theoretical model considers the stage of economic development, categorizes carbon emissions into productive and nonproductive activities, extends the production function to construct a carbon emissions function and reasoning, and analyzes the influence of the mechanism of economic resource aggregation on CO₂ emissions at various phases of development. For the empirical analysis, a spatial panel model of 30 provincial districts in China is proposed and combined with a machine learning method to examine the nonlinear relationships between economic resource gathering and carbon release intensity. The results show a typical “N” curve relationship between economic agglomeration and CO₂ emissions. When the thresholds exceed 175.573 and 333.333, the impact of economic agglomeration on emission-reduction changes from inhibition to promotion. In addition, energy consumption can act as a mediating variable for economic agglomeration to affect CO₂ emissions, and there is a typical “inverted U” curve relationship. The “N” curve between economic aggregation and CO₂ emissions with multiple inflection points is validated using machine learning. Accordingly, this study proposes that significant attention should be paid to the energy-conservation and emission-reduction mechanism of economic agglomeration. Furthermore, the linkage mechanism of energy-saving and emissions reduction between regions with complementary advantages is essential.