The Geography of Innovative Firms

Organised by

• Department of Economics

Abstract:

A large body of literature has documented that knowledge spillovers between firms are local and decay with distance. In a world with multi-plant firms, such local spillovers imply that a firm’s spatial expansion can impact both its own exposure to existing ideas and the spillovers it generates for others, shaping aggregate innovation and growth. We investigate this idea empirically and theoretically. We first provide evidence on the importance of geographical firm expansion for cross-firm spillovers by linking data on firms’ geography and patent citations. Using an event-study design, we show that a firm’s expansion into a local market increases citations to its existing patents by inventors in that market. Motivated by this evidence, we develop a model in which multi-plant firms can improve their innovation process by accessing ideas across multiple markets. The distribution of these ideas across space depends on the location decisions of all firms in the economy. A key takeaway is that when knowledge spillovers are local and ideas are sufficiently differentiated across firms, innovative firms are too geographically concentrated relative to the social optimum.

Co-authors: Craig Chikis and Benny Kleinman