Spatial spillover effects of the digital economy on green innovation based on time varying networks

Introduction The digital economy reshapes interregional connections beyond geographical proximity, with implications for green technology innovation. Methods Using 2014–2023 panel data for 30 Chinese provinces, this study constructs an asymmetric, time-varying digital-economic network based on a modified gravity model and applies social network analysis and a Spatial Durbin Model. Results The results show that the network exhibits a persistent core–periphery structure alongside changes in provincial positions. Under the baseline network specification, digital economy development is positively associated with local green technology innovation and has a significant negative indirect effect on network-connected provinces, consistent with resource-competition and siphon-related forces outweighing knowledge-diffusion forces in the cross-regional channel. This negative indirect effect remains significant under an alternative geographic matrix, although the negative total effect becomes insignificant, indicating that the overall effect is sensitive to the definition of spatial linkages. Conditional decomposition shows that the negative spillover deepened before 2019, moderated between 2019 and 2021, and remained stable thereafter. Discussion Provincial evidence suggests that stronger connectivity does not necessarily improve the relative innovation performance of late-developing regions, as its benefits depend on local absorptive capacity and resource-retention capabilities. These findings highlight the importance of combining interregional digital connectivity with local innovation-capacity building.