Related variety, recombinant knowledge and regional innovation. Evidence for Sweden, 1991-2010

This study investigates how related variety in the regional employment mix affects the innovation output of a region. Departing from the idea of recombinant innovation, previous research has arguedthat related variety enhances regional innovation as inter-industry knowledge spillovers occur more easily between different but cognitively similar industries. This study combines a novel dataset and related variety measures based on network theory, which allows a more nuanced perspective on the relationship between related variety and regional innovation.The principal novelty of the paper lies in employing new data on product innovations commercialised by Swedish manufacturing firms between 1970 and 2013. In this respect, it allows a direct measure of regionalinnovation output as compared to patent measures, usually employed in similar studies. The second contribution of this paper is that we employ network-topologybased measuresof related variety that allowus to measure relatedness as the recombination rather than direct flow of knowledge. We argue that this measure comes closer to the notion of innovation as spurred by recombination and show that this measure is a superior predictor of innovation activity.