This paper builds a model in which clustering emerges as the equilibrium outcome driven by informational imperfections and the importance of knowledge and uncertainty in innovative production even when knowledge is fully exclusive and appropriable. This motive for clustering will be stronger the more important new knowledge in the industry. It thereby helps to account for the empirical finding that economic activity clusters more strongly in industries where new knowledge plays a bigger role. It weakens the case for local knowledge spillovers to explain innovative clustering.
knowledge innovation innovative production vector representation complementarity