From industrial district to innovation system – the first century of the geography of innovation

Abstract

 One of the defining factors of regional resilience is local innovation potential and activity. The territorial, regional analysis of this innovation potential is closely linked to geography and to the geography of innovation. At the same time, innovation geography is one of the youngest disciplines of our time that has produced competing theories about the production of innovation in space, across time and in different institutional settings. We demonstrate that the scientific field of the geography of innovation was ’born’ only in the eighties, after decades of analysis about economic growth and regional development that paid little attention to endogenous factors of innovation production. Our research objective is not only to explore but also to quantify the theoretical-historical dynamics of the theories of innovation geography. We apply big data tools in order to point out paradigm shifts in approaches such as the neoclassical school, location theories, industrial district and learning regions, networks and innovation systems or the growth pole theories of post-war Keynesian spatial planning regarding innovation potential. Large-scale, digitized datasets concerning the literature has only been available since the 21st century. Our method, based on a big data approach is a novel, viable and reliable way to gain more insight into the inherent shifts in scientific trends, compared to traditional literature reviews which tend to build on more subjective methodological decisions. This may provide support for economic policy to find theoretical frameworks that help to identify barriers to innovation and strengthen regional comparative and competitive advantages as a basis of regional economic resilience. The approach used in this paper shows considerable shifts in basic assumptions of theoretical models, with the increasingly frequent use of the terms ’economic instability’, ’asymmetric information’ or ’knowledge networks’. These mark a deviation from traditional neoclassical regional economics. The study argues that innovation system models took the position of earlier industrial district models and presents a quantified mental map of the most important tendencies in the geography of innovation.