The effect of smart city construction on the green evolution of enterprises under the formation of new-quality Productivity: Based on double machine learning models

Abstract

Smart city construction (SCC) profoundly influences high-pollution industrial enterprises in adopting greener development pathways as the new-quality productivity (NQP) emerges. This study applies a double machine learning (DML) model based on panel data from 1118 high-pollution industrial enterprises in China from 2010 to 2022, utilizing the Lasso regression algorithm for prediction and estimation, and adopting cross-validation with a 1:2 sample split ratio to avoid model overfitting. After measuring the NQP levels for 1118 enterprises, this research uses the "Smart City" strategy as a quasi-natural experiment to examine the impacts and underlying mechanisms of SCC on the green evolution of high-pollution enterprises in China. The results reveal that SCC shows a substantial positive effect on the green evolution of high-pollution enterprises. For high-pollution enterprises with well-developed digital infrastructure, located in central cities, and with higher levels of digitization and green innovation, the positive effect of SCC is more pronounced. Regarding the underlying mechanisms, SCC effectively promotes NQP, new-quality labor (NQL), new-quality means of labor (NQML), and new-quality objects of labor (NQOL) in high-pollution enterprises, thereby promoting their green evolution. This study not only refines the measurement system for NQP of enterprises (NQPE) but also provides policy recommendations for the Chinese government to advance SCC and promote the green evolution of high-pollution enterprises.