Subgroup and Shapley Value Decompositions of Multidimensional Inequality – An Application to Southeast European Countries

Inequality is a multidimensional phenomenon though it is often discussed along a single dimension like income. This is also the case for the various decomposition approaches of inequality indices. In this paper we study one- and multidimensional indices on inequality on data for three large Southeast European countries, Bulgaria, Romania and Serbia. We include four dimensions in our measure of multidimensional inequality income, health, education and housing. We apply various decomposition methods to these one- and multidimensional indices. In doing so, we apply standard decomposition techniques of the mean logarithmic deviation index (I0) and decompositions based on regression analysis in conjunction with the Shapley value approach.