Welfare-poverty measurement

This paper provides an approach to poverty measurement that relies on the interpretation of poverty as a welfare loss. Our contribution is twofold. On the one hand, we analyse the relationship between individual and aggregate indicators, by introducing the notion of “distributive impact of poverty” (a measure of the poverty loss due to the inequality among the poor). We show that a welfare inequality measure can be expressed as the sum of the average individual welfare poverty plus the distributive impact of poverty. On the other hand, we extend the analysis to the case of a society made of several population subgroups, by using a decomposability principle consistent with this approach. An empirical application, regarding educational poverty in the OECD, out of the data in PISA 2009, illustrates the extent of our method.