Far from the Madding Crowd

The open innovation paradigm has become central for understanding firm-level innovation, and is being refined and questioned as researchers explore its limits. However, it has been rather uncritically adopted by many economic geographers since it provides backing for a variety of theories and observations that point towards innovation occurring in geographically concentrated clusters and cities and towards the role that proximity plays. Evidence that innovation can and does occur outside of clusters and cities, whilst widespread, has not displaced the prevailing idea that innovation and geographic concentration are intimately linked. In this paper, I first go over these arguments, focussing on the reasons why innovation is thought to be associated with cities and clusters. I then propose a framework that accounts for the clustering of innovative firms, for the role of cities, and for the fact that empirical work consistently shows that firms introduce first-to-market innovations in a wide variety of geographic contexts, including isolated and peripheral ones. The framework (re)introduces time and place as key factors, which determine not the diffusion, but the value to innovators, of different types of information.