Collaborative Networks as Determinants of Knowledge Diffusion Patterns

This paper examines if interpersonal networks help explain two widely documented empirical patterns of knowledge diffusion: (1) geographic localization of knowledge flows, and (2) easier transmission of knowledge within firms than between firms. A regression framework based on choice-based sampling is used to estimate the probability of micro-level knowledge flow between inventors of U.S. patents. Knowledge flows are measured using citations made by patents originating between 1986 and 1995 from around 3,300 firms and organizations. Interpersonal ties are inferred using “social proximity graphs” constructed using collaboration data for over one million inventors of U.S. patents since 1975. The existence of a direct or indirect interpersonal tie is found to be associated with a greater probability of knowledge flow, with the probability increasing with the directness of the tie. Even more interestingly, the estimated effect of geographic co-location and firm boundaries on the probability of knowledge flow decreases once interpersonal ties have been accounted for. In fact, geographical co-location and firm boundaries have very little additional effect on the probability of knowledge flow between inventors that have close direct or indirect ties.