Knowledge polycentricity and technological diversification: evidence from U.S. Metropolitan Statistical Areas

Existing literature on regional technological diversification has primarily viewed regions as points, while little attention has been paid to their inner spatial configurations. To address this gap, this paper investigates the impact of knowledge polycentricity on regional technological diversification. Using the USPTO patent data (1976–2020), we conduct econometric analyses on 354 U.S. Metropolitan Statistical Areas and 558 four-digit CPC technology domains. Integrating the attribute-function polycentricity framework from knowledge polycentricity literature and relatedness-complexity diversification framework from Evolutionary Economic Geography studies, we find that attribute polycentricity leads to unrelated and less complex diversification, whereas functional polycentricity triggers related and more complex diversification. The primary results still hold under instrumental variable estimation and a series of robustness checks. We contribute to Evolutionary Economic Geography literature by incorporating intra-urban knowledge polycentricity into the study of technological diversification.