Network externalities of the innovation network in China’s five urban agglomerations: based on “buzz-and-pipeline” theory

In the era of the knowledge economy, innovation networks have emerged as crucial drivers in shaping urban economic development patterns. Network externalities represent a key mechanism for understanding the economic impact dynamics of innovation networks. This paper constructs a three-dimensional innovation network integrating knowledge, technology, and innovation. By applying social network analysis and GIS-based spatial analysis, it explores the spatiotemporal evolution of innovation networks in China’s five major urban agglomerations from 2011 to 2020. The dual-weight spatial econometric model is further employed to empirically examine the network externalities of intra- and inter-agglomeration innovation networks. The main findings are as follows: (1) From 2011 to 2020, intra-agglomeration networks displayed an uneven spatial distribution and a clear hierarchical structure. Inter-agglomeration networks evolved from a hub-and-spoke pattern to a diamond-shaped spatial structure. (2) Innovation networks generate significant externalities for urban economic development. The intra-agglomeration “buzz” effect produces stronger positive externalities, complementing the inter-agglomeration “pipeline” effect. Knowledge networks and technology networks interact, reinforcing these externalities. (3) Industrial upgrading moderates network externalities, but this effect remains confined to intra-agglomeration networks and has not yet overcome the inter-agglomeration threshold. (4) Network externalities vary across urban agglomerations and depend on the degree of network embeddedness, exhibiting clear heterogeneity. At the theoretical level, this paper constructs a multidimensional spatial model of urban innovation networks, integrating knowledge and technology flows. Drawing on “buzz-and-pipeline” theory, this paper proposes a framework that combines intra- and inter-agglomeration perspectives, offering new insights for the study of innovation network and network externalities.