Raising Urban Productivity or Attracting People? Different Causes, Different Consequences
34 Pages Posted: 28 Nov 2006
Date Written: May 2006
This paper investigates growth differences in the urban system of the EU12 over the last decades of the 20th Century. Models in which growth of real GDP p.c. and rates of population growth are the dependent variables are compared. This suggests that it makes sense to model GDP growth in a European context. The analysis supports the conclusion that systems of urban governance are significantly related to economic growth, as is the distribution of highly skilled human capital and R&D activity. In addition, evidence is found supporting the conclusion that integration shocks in the EU favour core areas but when all else is controlled for peripheral regions experienced a systematic positive growth differential. Careful testing for spatial dependence reveals that national borders are significant barriers to adjustment but we can resolve such problems by including a set of variables designed to reflect spatial economic adjustment mechanisms where cities are densely packed so their economies interact. Models of population growth show some similar results but interesting and revealing differences. Strong evidence is found that there are substantial national border effects impeding the emergence of a full spatial equilibrium across the EU's urban system. Better climate is the single most significant variable but only when expressed relative to the national (not EU) mean. As with economic growth, there are significant national border effects in patterns of spatial dependence. Concentrations of human capital and R&D, however are if anything negatively associated with attracting population - a finding which parallels the finding that a better climate relative to the national mean is associated with slower rather than faster growth of real GDP per capita.
Keywords: growth; cities, local public goods, human capital, convergence, territorial competition
JEL Classification: H41, H73, O18, R11, R50
Suggested Citation: Suggested Citation