Ton van Raan and co-researchers investigated the socioeconomic scaling behavior of all cities with more than 50,000 inhabitants in the Netherlands and found significant superlinear scaling of the gross urban product with population size.
Earlier US research showed that socioeconomic performance increases more than proportional, i.e., superlinearly, with the size of a city in terms of number of inhabitants. This urban scaling with a scaling exponent of about 1.15 arises because cities and their direct agglomerations behave as a complex, adaptive system in which by enlargement the underlying network and its physical and social clusters undergo a positive non-linear (superlinear) reinforcement. Twice as much inhabitants means an additional 15% of socioeconomic performance measured in gross urban product. Similar scaling behavior was found by Ton van Raan for the impact of universities.
Together with his coauthors Gerwin van der Meulen and Willem Goedhart (Decisio Economic Consultancy, Amsterdam) Ton van Raan investigated the socioeconomic scaling behavior of all cities with more than 50,000 inhabitants in the Netherlands and found significant superlinear scaling of the gross urban product with population size. Of these cities, 22 major cities have urban agglomerations and urban areas defined by the Netherlands Central Bureau of Statistics. A novel and politically important new element in the study is that for these major cities the superlinear scaling was investigated for three separate modalities: the cities defined as municipalities, their urban agglomerations and their urban areas. The results are recently published in PLoS ONE.
The researchers find superlinearity with power-law exponents of around 1.15 for all three modalities but remarkably both types of agglomerations underperform if we compare for the same size of population an agglomeration with a city as a municipality. In other words, an urban system as one formal municipality performs better as compared to an urban agglomeration with the same population size. This effect is larger for the second type of agglomerations, the urban areas. Even if the benefit would be only 10% of the expected value, then we are still talking in terms of 100 million Euros per city resulting in thousands of jobs. The Leiden urban area is presented as an example.
The authors think that this finding has important implications for urban policy, in particular municipal reorganizations. A residual analysis suggests that cities with a municipal reorganization recently and in the past decades have a higher probability to perform better than cities without municipal restructuring.