The impact of digital transformation on urban green and low-carbon transformation efficiency: evidence from a double machine learning approach

Abstract

This study selected 279 Chinese cities as research samples and integrated double machine learning models, Gini coefficients, and standard deviation ellipses to systematically analyze the dynamic evolution of urban green and low-carbon transformation efficiency amid digital transformation. The main objective was to elucidate the spatial differentiation, temporal evolution, and the impact mechanisms and transmission pathways through which digital transformation affects this efficiency, thereby providing a comprehensive understanding of the coordinated development of China’s green and low-carbon transition. The analysis produced the following key findings: First, the efficiency of urban green and low-carbon transformation shows distinct spatiotemporal evolution characteristics. Although it maintains an overall steady upward trajectory, notable regional disparities persist, with inter-regional differences remaining the main source of imbalance. Second, digital transformation exerts a significant positive effect on urban green and low-carbon transformation efficiency. Benchmark regression results show that a one-unit increase in the level of digital transformation leads to a 0.033-unit rise in urban green and low-carbon transformation efficiency. Third, the impact of digital transformation displays significant regional heterogeneity. Compared with other city types, its positive effect is more pronounced in cities with established industrial bases, key environmental protection functions, and major transport hubs. Finally, green technological innovation serves as a crucial mediator in the process through which digital transformation enhances urban green and low-carbon transformation efficiency.