DP953 | Knowledge Spillovers and the Geography of Innovation and Production

Publication Date


JEL Code(s)


Programme Area(s)



This paper examines the extent to which innovative activity clusters spatially and links the geographic concentration of innovative activity to the degree to which knowledge externalities exist. Because one obvious explanation of why innovative activity tends to cluster geographically more in some industries than in others is that the location of production is more spatially concentrated, we first examine and then control for the geographic concentration of the location of production. Based on a data base that identifies innovative activity for individual states and specific industries for the United States, as well as the geographic distribution of production, we find that the propensity for innovative activity to cluster spatially is greater in industries where the creation of new knowledge and knowledge spillovers is more important. Even though such industries also tend to exhibit a greater geographic concentration of production, the empirical evidence suggests that the propensity for innovative activity to cluster is more attributable to the role of knowledge spillovers and not merely the geographic concentration of production.