The Impact of Ego-Network Stability on Firm Innovation Efficiency From the Perspective of Multilevel Network

Firm innovation is typically embedded within multilevel networks, where dynamic changes at various levels of the network can significantly influence firm innovation. Previous research has often overlooked the dynamics of these networks and failed to consider the differential impacts that distinct network levels have on firm innovation. Drawing on social network theory, this study develops a theoretical model that explores the relationship between ego-network stability and innovation efficiency from a multilevel network perspective. We analyze patent data in the domains of artificial intelligence and intelligent manufacturing equipment in China, covering the period from 2012 to 2022, and test our hypotheses using a Data Envelopment Analysis-Tobit Regression (DEA-Tobit) model. The results reveal that organizational ego-network stability exhibits an inverted U-shaped effect on innovation efficiency. Additionally, urban ego-network stability and knowledge ego-network stability enhance the influence of organizational ego-network stability on innovation efficiency. This paper contributes to the theory of multilevel network embeddedness by introducing organizational collaboration networks, urban collaboration networks, and knowledge networks as external contexts that affect firm innovation and by examining their relationships with innovation efficiency from a dynamic perspective.