Pro-poor growth and the lognormal income distribution

A widely accepted criterion for pro-poorness of an income growth pattern is that it should reduce a (chosen) measure of poverty by more than if all incomes were growing equiproportionately. Inequality reduction is not generally seen as either necessary or sufficient for pro-poorness. Because empirical income distributions fit well to the lognormal form, lognormality has sometimes been assumed in order to determine analytically the poverty effects of income growth. We show that in a lognormal world, growth is pro-poor in the above sense if and only if it is inequality-reducing. It follows that lognormality may not be a good paradigm by means of which to examine pro-poorness issues.