New economic geography reloaded : localized knowledge spillovers and the geography of innovation

Abstract

Despite the increasing and newly inspired interests in geographical economics and industry location theory, the majority of existing New Economic Geography models ignores the interdependence between spatial concentration, knowledge diffusion, invention and growth. For this reason, the paper exclusively surveys the emergence and development of New Economic Geography Growth models in the context of the existing geography of innovation literature. The first part of the paper contributes with a classification of first- and second-nature causes of agglomeration and clustering. This part will also discriminate between static and dynamic externalities. Therefore, the chapter particularly compiles the differences between urbanization and localization externalities, and MAR, Jacobian and Porter externalities. A second concern of the paper is to highlight the modeling peculiarities of New Economic Geography Growth models. Besides approaching the main differences and similarities between first- and second-generation NEG models, the paper additionally reviews and discloses complemental contributions to the geography of innovation literature in the course of time. For this purpose, the paper examines in a meta-study 61 empirical contributions, which are related to the knowledge production function, the concept of spatial dependence and knowledge spillovers. The meta-study is complemented by bibliometric research. The paper ultimately concludes that the empirical studies that are related to the concept of (localized) knowledge spillovers and spatial association have caused a fundamental upgrading of the New Economic Geography literature towards non-pecuniary externalities. Consequently, the paper shows that recently developed second-generation NEG models offer alternative backward and forward linkages, which similarly determine centripetal and centrifugal forces, circular causality and finally the geography of innovation.