Path Dependence and Path Breakthrough: Evolutionary Dynamics of Innovation Networks in China's Semiconductor Industry

ABSTRACT This study uses co‐applied patent data from China's semiconductor industry, covering the period from 2001 to 2023, to construct networks of innovation collaboration between cities for each year. It applies social network analysis and a Temporal Exponential Random Graph Model (TERGM) to examine the evolution and formation mechanisms of these networks from the perspective of path dependence and path breakthrough. The results show that the network evolved from a structure dominated by a few core cities to a polycentric configuration with stronger cross‐regional collaboration. Traditional core cities retained substantial influence, while several emerging cities increasingly acted as bridges between communities, reflecting growing functional differentiation across these communities. The TERGM estimates reveal that prior ties persist significantly and that endogenous structures reinforce collaboration. The positive coefficients for Memory and GWESP indicate that established ties and triadic closure strongly shape subsequent collaboration, supporting the presence of path dependence. Even after controlling for relational history and endogenous network structure, urban resource endowments and dyadic covariates remain important. Economic development, IT human capital, R&D investment, higher education resources, and semiconductor industrial bases significantly promote collaboration. Geographical distance discourages collaboration, whereas similarity in technological knowledge portfolios facilitates it. Differences in administrative hierarchy, however, do not independently influence tie formation. Overall, China's intercity semiconductor innovation network does not simply reproduce established ties. Rather, urban resources and technological conditions create scope for reconfiguration, even as historical ties continue to exert influence. These findings reveal the coexistence of path dependence and conditional path breakthrough, providing a dynamic relational perspective for understanding how innovation networks evolve in strategic technology industries.