City positionality in supply chain network and urban carbon emission intensity: Evidence from listed companies in China

This study explores the relationship between intercity supply chain networks and urban carbon emission intensity (CEI) in China. While most research investigates the environmental implications of geographical proximity among city clusters, the environmental effects of functional intercity relationships remain insufficiently explored. This study examines how the network topology of economic relations among cities mediates CEI, analysing 29,580 firm-level supplier-buyer transactions from listed companies (2008–2017), aggregated at the prefecture-city level. Using data from the CSMAR database, we construct directed, weighted intercity networks and calculate key network indicators (in-degree, out-degree, closeness, betweenness, eigenvector centrality, and clustering coefficient). These network indicators are integrated into an XGBoost model, with SHapley Additive exPlanations (SHAP) employed to measure the non-linear effects of city positionality on CEI. Results reveal scale-free intercity supply chain networks dominated by key hubs such as Beijing, Shanghai, and Shenzhen. Cities with higher in-degree, out-degree, and closeness centrality in the networks exhibit lower CEI. Moreover, cities with initially low in-degree centrality may exhibit high CEI but improve once their network authority exceeds a certain threshold, revealing a non-linear pattern. Overall, this study advances understanding of urban CEI. via networked geographies and offers critical insights for green supply chain governance in China.