The knowledge economy and Dutch cities

How can cities and metropolitan regions remain prosperous and competitive in a rapidly changing economy? In our paper we argue that ‘the knowledge economy’ offers perspectives for growth and added value creation. The paper clarifies what elements the knowledge economy actually consists of, how it can be measured in statistical indicators, in which regions and cities in the Netherlands the knowledge economy has its most significant imprints and what statistical association there is between these regions and cities and relatively good economic performance of firms. We test two contrasting hypotheses often heard in the international literature. The current embedding of knowledge externalities in endogenous economic growth theory have led to important contributions that stress the urban character knowledge transmission in particular. The reasoning is that if knowledge spillovers and –externalities are important to growth and innovation, they should be more easily identified in cities where many people are concentrated into a relatively small geographic space so that knowledge can be transmitted between them more easily. Much recent research indeed finds a limited extent of spatial spillovers and a large degree of local clustering. Alternatively, a large body of literature on Western spatial configurations of innovation and high-technology firms predominantly stresses the supposed ‘urban field’ character of firm performance: location and agglomeration aspects do not seem to have a systematic impact on the distribution of innovative and growth inducing activities over space. We test the urban hypothesis using spatial econometric modeling techniques. On the one hand, the fact that a distance squared distance weight matrix in spatial lag estimations fits the performance data best in relation to knowledge economy factors indicates that spatial relations are limited and urban fixed. On the other hand, the significance of several spatial regimes though (especially those of the Randstad core region, the so-called intermediate zone and medium-sized cities) indicates that the urban structure related to the knowledge economy and economic performance is not straightforward hierarchical (largest cities are not the relatively most attached to the knowledge economy). Both hypotheses (urban and non-urban) are too extreme to fit the Dutch situation. We also conclude that the locational attributes of the factor ‘knowledge workers’ are much more significantly related to economic growth and added value (in practically all specifications over regimes and spatial lag estimations) than the R&D-based innovation input factor. This questions Dutch policy initiatives that mainly focus on R&D as stimulator of the ‘knowledge economy’.