On the Decomposition of Polarization Indices: Illustrations with Chinese and Nigerian Household Surveys

Abstract

This paper explores the link between polarization and inequality and proposes some analytical methods to decompose the Duclos, Esteban, and Ray (2004) polarization index by population groups or income sources. In some cases, the decomposition methods were extended to the Esteban and Ray (1994) one. The main aim of these decomposition methods is to extend the interpretation derived from polarization indices to that of contribution components. Results drawn from Chinese data conclude that even if inequality has increased sharply during the last two decades, the pure polarization component was remained constant or even decreased on average. On the other hand, results from the 2004 Nigerian survey conclude that the population is spatially polarized, and this, based on geo-ecological zones. Furthermore, the two income sources, namely, Employment income and Non farm business income, significantly contribute to total polarization.