Does Digital Economy Drive New Quality Productivity? Evidence From Dual Machine Learning Models

ABSTRACT Advancing new quality productivity (NQP) is essential to deviate from the traditional paths of productive forces and economic growth. However, further research is needed to uncover effective development strategies for NQP. By applying dual machine learning (DML) on panel data for 282 Chinese cities from 2009 to 2022, this study examines how digital economy (DE) influences NQP. Several key findings emerge from the analysis. First, China's NQP has considerable potential for improvement, with the development of new factors and technologies being the main barriers. Second, DE positively influences the advancement of NQP as confirmed by various robustness checks. Third, DE promotes NQP by promoting green innovation, facilitating human resource agglomeration, and improving urban informatization levels. Fourth, the effects of DE on NQP are heterogeneous. From a regional heterogeneity perspective, DE aids NQP advancement across the eastern, central, and western regions, while from an economic heterogeneity perspective, the DE in the Yangtze River Delta, central Yangtze River, and Beijing–Tianjin–Hebei economic areas significantly drives NQP progress. Theoretically, this study innovatively applies the DML method to examine how DE influences NQP, thereby effectively addressing endogeneity concerns and deepening the present theoretical understanding of their mechanisms and heterogeneity. Practically, this study provides policy insights for governments to design differentiated DE strategies and promote regionally coordinated development.