45th Congress of the European Regional Science Association: "Land Use and Water Management in a Sustainable Network Society", 23-27 August 2005, Amsterdam, The Netherlands
The overall Swedish total fertility rate (TFR) has been shown to fluctuate quite dramatically during the last 40 years, from 2.4 in 1965 reaching down as low as 1.5 in 1999. Although numerous studies in the past have tried to explain this fluctuation, there has been surprisingly little focus on impacts from local or even regional differences. Thus, the purpose of this paper is to analyse, for the first time, whether there exists a significant dispersion in the number of child births normalised by the number of fertile women among Swedish municipalities for the period 1974-2002 and then provide an explanation for occurred differences. Regional differences, as well as the presence of spatial dependence, are first confirmed in an exploratory spatial data analysis. Particularly noticeable is the volatility and how local hotspots emerge and disappear during the study period. Subsequent regression analyses are performed for a selected number of years where we solve for spatial dependence and use economic-, institutional-, sociological-, and geographical characteristics of the municipalities as explanatory variables.