Untangling the relationships between proximity dimensions – an in-depth study of collaboration in the Danish cleantech industry
Last modified: 2011-02-11
The increasingly interactive nature of innovation processes has been widely studied in economic geography lately, especially the effect of geographical proximity on innovation. While some authors stress that localised learning is central to innovative activity, others argue that this particular focus has led to a neglect of non-regional knowledge linkages.
However, it is theoretically rewarding to move away from a discussion of proximity in purely physical terms. In a widely quoted paper, Boschma (2005) introduces a five dimensional model of proximity consisting of geographical, cognitive, organisational, social and institutional proximity. He emphasises that some cognitive proximity is a prerequisite for interactive learning contrary to the other dimensions, where one form of proximity can substitute another.
However, geographical proximity is different from the other dimensions, as it does not directly impact collective learning, but rather facilitates other types of proximity. As Malmberg & Maskell (2006) note, Boschma’s conclusion is actually supporting the localised learning theory. Yet, the relationships between geographical proximity and the other four proximities have not been empirically tested. This paper aims to contribute to this field of research by analysing collaborative development projects in the Danish cleantech industry. An ordered logit model will be applied to a unique database of 180 inter-firm collaborations created through in-depth interviews with cleantech firms. The analysis will in this way reveal to what extent the other proximity dimensions are consequences of geographical proximity.
Additionally, the paper will provide a detailed picture of the use of knowledge networks in the cleantech industry. This is in itself a novel contribution, as the industry is rarely conceptualised as a whole in academic work due to the lack of a cleantech-code in industrial classifications. Thus, a further aim is to identify similarities and differences in innovation patterns between firms from diverse backgrounds such as wind energy, biomass energy and green construction.