The evolution of income inequality in the European Union during the period 1993–1996

Abstract

The objective of this work is to analyse the income inequality in the 15 EU countries during the convergence process to the Monetary Union, using the information contained in the European Community Household Panel, corresponding to the four first waves. Using the inverse second order stochastic dominance concept, an ordering of these countries has been carried out. Furthermore, this ranking allows one to determine if the differences among EU country members have increased or decreased during this particular period. Whether the inequality of income has diminished within and between countries over time was studied. Gini's generalized family indices proposed by Donaldson and Weymark (Journal of Economic Theory 22: 67–86, 1980 and 29: 353–8, 1983) and Yitzhaki (International Economic Review 24: 617–28, 1983) have been used. This allows one to test the sensitivity of the results obtained to different degrees of inequality aversion and to different equivalence scales, taking into account household sizes.

Acknowledgements

This paper has benefited from support of the European Communities TMR network Living Standards, Inequality and Taxation [Contract #ERBFMRXCT980248], the Spanish Ministry of Education [Project #PB98-0546-C0202], the Spanish Ministry of Science and Technology [Project #SEC2003-08397] and Fundación BBVA. The usual disclaimer applies.

Notes

There is a previous research on poverty and social exclusion (EUROSTAT, 2000) that compares inequality across the EU members using this data set. However, they concentrate on only one wave of this panel (1996) and exclude Sweden and Finland. They focus on one inequality index (the standard Gini coefficient) and on one equivalence scale (the corresponding OECD one).

Inverse n-degree stochastic dominance introduced by Muliere and Scarsini (1989) has appealing properties in terms of consistency with very general class of the Yaari rank-dependent Social Welfare Functions. See for instance Zoli (1999) and Aaberge (2001) theorems on n-degree inverse stochastic (welfare) dominance. Given two distributions F and G defined over a non-negative random variable X with finite expectations. The distribution F n-degree indirect stochastic dominates G, denoted as F ≽ 

G, for any integer n ≥ 1, if and only if F

(p) ≤ G

(p) for all p ∈ [0, 1]. Define

as the nth degree inverse distribution function, for n = 1, as:

and for n ∈ {2, 3, …}as:

It can be noticed that, in the extreme case, as v tends to infinity, the S-Gini converges to

. Obviously, this index is very much dependent on the minimum level of income, which maybe statistically unreliable.

Hence, Finland and Sweden are not included since they are observed only twice and once respectively.