This paper examines the spatial dyanmics of Europena unemployment by applying non-parametric methods. The use of different levels of administrative data, the study of female and youth unemployment, and the analysis of regressions trees are the main innovations with respect to the related work by Overman and Puga (2002). Stochastic kernels analysis confirms that the emerging feature of the dynamics of European unemployment rate is polarization. Moreover regression trees shows that unemployment growth rates are more similar across surroundings areas (particularly within the same Member State), than across regions with similar specialization or educational attainment. Our results suggest that polarization is primary demand-driven.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.