6533b7d4fe1ef96bd1261afa

RESEARCH PRODUCT

Disparités géographiques et convergence des régions européennes : une approche par l'économétrie spatiale

subject

description

The aim of this dissertation is to analyze the convergence process between European regions by taking into account spatial patterns of economic activities. First, with the help of literature reviews on theoretical and empirical analyses of growth and convergence and on economic geography models, we show in chapter 1 why geographic disparities should be introduced in the study of convergence between economies. Next, the spatial statistic and econometric tools allowing to reach this goal are detailed in chapter 2. Finally, three empirical studies of the convergence process in presence of geographic disparities are carried out. They are based on a sample for per capita GDP of 138 regions, belonging to 11 countries for the period 1980-1995. The data are extracted from the Eurostat-Regio databank. In chapter 3, an exploratory spatial data analysis shows that the GDP distribution is characterized for the whole period by a strong geographic concentration (positive spatial autocorrélation) and by a North/South polarization pattern (spatial heterogeneity). In chapter 4, regressions linking the average annual growth rate between 1980 and 1995 to the per capita GDP in 1980 are carried out. The estimation of a beta-convergence model with spatial error autocorrelation and structural instability allows detecting positive and significant geographic spillover effects between neighboring regions and differentiating the convergence process between northern regions from the convergence process between southern regions. In chapter 5, the estimation of Markov chains indicates that the mobility of a region within the GDP distribution is very slow and that it is strongly conditioned by its neighbors, highlighting a geographic lock-in effect.