Spatial knowledge dynamics of innovation processes: local and non-local aspects of buzz and collective learning

Most studies in economic geography think of innovation as an outcome measured by patents or innovation statistics. This simplistic view on innovation causes scholars to draw conclusions supporting a spatial determinism that favours local ties over international. This paper aims to understand if geographically proximate partners are more important at some phases of the innovation process than at other phases. First the paper takes a critical look at the fuzzy conceptualization in economic geography, including local buzz, face-to-face, collective learning and global pipelines. Second, the paper applies a novel, biographical methodology, namely innovation biographies, which places the innovation event as its analytical focus and study the process as it unfolds over time (from idea-generation ? to problem-solving ? to implementation). The paper presents three innovation biographies from suppliers to the wind turbine industry. Results suggest that local linkages (buzz) are crucial in the early stages of generating new ideas, whereas in other phases (problem-solving and implementation) geographical co-location does not play an important role, although these processes are highly dominated by collective learning processes and require face-to-face contact. In sum, the innovation biography method contributes in uncovering innovation processes and how these rely on many different configurations of spatial knowledge dynamics, including buzz, local ties and global pipelines. The findings imply that policy should be designed to support companies working across regional and national borders and not only favour local networking.