Decomposition of Inequality Measures

Abstract

It is common practice to summarize income or wealth inequality in a population by computing statistical inequality measures like the Gini Coefficient or the Coefficient of Variation. This should be complemented, however, by an analysis of the structure of inequality and an attempt to identify its sources. For this purpose it may be useful to partition the basic population into subpopulations by some socio-economic characteristics (such as employment type, education etc.) and to decompose overall inequality into a part which is due to inequality ``within'' subgroups and a part which is due to inequality ``between'' subgroups. This approach was successfully applied to the analysis of income inequality in England, the U.S.A. and France by Mookherjee and Shorrocks (1982), Cowell (1984), and Bourguignon and Morrisson (1985), respectively.