On inter-industry relatedness and regional economic development

Abstract: The dissertation aims to advance our understanding about the role of local industry structure in regional economic development. More specifically, it investigates how relatedness between local industries (i.e. similarities in what kind of knowledge industries use) contributes to and constrains regional economic development. Empirically, the dissertation investigates Swedish regional development in 1991-2010.Inter-industry relatedness is a key concept in evolutionary economic geography and has a prominent role in narratives explaining the evolution of regional economies. The diversity of local industries and relatedness between them serve as sources of feedback and path dependencies that mould regional development trajectories. It is widely accepted that firms benefit from the local presence of other firms in related industries and such local related variety increases regional employment growth. Often these growth benefits are assumed to be ‘in the air’ or in local ‘buzz’ taking the form of pure knowledge spillovers via unintended interactions between individuals. The dissertation argues that the role of market-mediated knowledge flow channels like labour mobility is underestimated.Economic diversification is a major source of dynamics in regions. Researchers and policymakers increasingly recognise that existing industrial structure and local capabilities condition which new activities will be feasible to develop in regions. This is supported by empirical studies showing that regions tend to diversify into activities that are related to their current industry mix. For example, regions manufacturing cars are more likely to start producing motorcycles than tennis rackets. The dissertation demonstrates that regions’ potential for such related diversification has an inverted U-shaped relationship with regional size. The highest potential is in large and medium-sized regions, while core regions have few related diversification opportunities since they already host most industries. Actual diversification events are largely in line with the potential – fewer new related industries enter small and core regions than mediumsized and large regions. The dissertation also proposes a framework that connects various regional development paths (ranging from path shifting to path creation) to the features of local industrial structure that supports these paths (e.g. related variety, related diversification potential) and to the mechanisms associated the paths (e.g. incremental innovation, entrepreneurial cost discovery).Finally, the dissertation challenges the current static view of relatedness and demonstrates that relatedness is a dynamic phenomenon. Furthermore, growth impact of emerging, stable and disappearing relatedness linkages varies by regional type. Relatedness ties have a ‘best before date’ and over time deplete their potential to generate inter-industry knowledge spillovers and stimulate innovation. At a more general level, the dissertation calls for studying the role and evolution of inter-industry relatedness as part of broader structural change processes.