Decomposing the Difference Between Well-Being Inequality and Income Inequality: Method and Application

Abstract

The authors study the difference between multidimensional well-being inequality and income inequality and propose a method to decompose the difference between the Gini coefficients of income and equivalent income (EI), a multidimensional well-being measure that respects individual preferences towards what constitutes a good life. The authors propose a method to decompose the inequality difference into two parts: the vertical and reranking effects. The vertical effect arises from the correlation between income and non-income dimensions, and between income and preferences. The reranking effect arises from the fact that some persons occupy a different position in the EI distribution compared to the income distribution. The authors also propose a detailed decomposition method based on the Shapley value to decompose each of the two effects by non-income dimensions. The authors apply the decompositions using data for 27 countries, considering five non-income dimensions: unemployment, health, housing, crime and environment. The results show that inequality is much higher for EI that the reranking effect accounts for a large part of the inequality difference, and that health is the non-income dimension contributing most to both effects.