How Digital Infrastructure Drives Urban Green Innovation Efficiency: Evidence from Chinese Cities

A balanced panel of 284 Chinese prefecture-level cities covering 2010–2021 is used to examine the relationship between digital infrastructure and urban green innovation efficiency (EUGI). We construct an entropy-weighted index to measure digital infrastructure, while EUGI is estimated through a super-efficiency SBM model that incorporates undesirable outputs. The empirical results indicate that better digital infrastructure is associated with higher EUGI. This positive relationship remains statistically significant after using alternative measures, introducing more restrictive fixed effects, applying instrumental-variable specifications to address endogeneity, and accounting for spatial dependence. Similar evidence is obtained from the Broadband China pilot. The estimates from both the conventional DID model and the Callaway–Sant’Anna approach, which accounts for different policy adoption times, are positive. We also investigate three possible channels underlying this relationship. The estimates are consistent with the roles of green technological innovation, entrepreneurial activity, and industrial upgrading. Formal interaction tests further show that the relationship between digital infrastructure and EUGI differs across regions, between resource-based and non-resource-based cities, and according to the level of local policy support. Overall, the results indicate that digital infrastructure development does not necessarily produce equal green innovation gains across cities. The extent to which digital capacity is translated into green innovation outcomes is also shaped by local innovation resources, industrial conditions, and policy support. Digital infrastructure investment should thus be combined with measures that reflect local innovation and industrial conditions rather than following a uniform policy approach.