Study on the Impact of Logistics Standardisation Construction on Urban Carbon Emission Intensity: Causal Inference Based on Dual Machine Learning

As a key component of the modern circulation system, logistics standardisation (LS) has helped remove critical bottlenecks in the circulation sector and has provided strong support for achieving the “dual-carbon” goals. This study treats China’s national pilot policy for logistics standardisation as a quasi-natural experiment. Using a balanced panel of 282 prefecture-level cities from 2006 to 2022 and a dual machine learning framework, it examines the effect of LS on urban carbon emission intensity (UCEI) and the underlying mechanisms. The results indicate that LS significantly reduces UCEI in pilot cities. This finding remains robust across a series of tests, including controlling for interference from contemporaneous policies, applying instrumental variable estimation, and re-specifying the dual machine learning model. Further mechanism analysis suggests that economic agglomeration, industrial structure upgrading, and green technological progress are plausible channels through which LS reduces UCEI. Heterogeneity analysis shows that the carbon-reduction effect of LS is more pronounced in eastern cities, cities with higher administrative status, non-resource-based cities, cities with better transport infrastructure, and cities where governments pay greater attention to carbon reduction. In addition, the extension analysis shows that LS also helps reduce the emission intensity of major pollutants, thereby contributing to the coordinated governance of pollution reduction and carbon mitigation.