Bosker, E. Maarten Brakman, Steven Garretsen, Harry Schramm, Marc
Year of Publication:
CESifo working paper 1728
The empirical literature on city size distributions has mainly focused on the USA. The first major contribution of this paper is to provide empirical evidence on the evolution and structure of the West-German city size distribution. Using a unique annual data set that covers most of the 20th century for 62 of West-Germany's largest cities, we look at the evolution of both the city size distribution as a whole and each city separately. The West-German case is of particular interest as it has undergone major shocks, most notably WWII. Our data set allows us to identify these shocks and provide evidence on the effects of these `quasi-natural experiments' on the city size distribution. The second major contribution of this paper is that we perform unit-root tests on individual German city sizes using a substantial number of observations to analyze the evolution of the individual cities that make up the German city size distribution. Our main findings are twofold. First, WWII has had a major and lasting impact on the city size distribution. Second, the overall city size distribution does not adhere to Zipf's Law. This second finding is largely based on the results of unit root tests for individual cities to test for Gibrat's Law, the latter being a requirement for Zipf's Law to hold for the overall city-size distribution. Together these two findings are consistent with theories emphasizing increasing returns to scale in city growth.