The Absolute Gini is a More Reliable Measure of Inequality for Time Dependent Analyses (compared with the Relative Gini)

Abstract

I compare time series properties of relative and absolute Ginis to examine which one is better suited for time dependent analyses. In particular, I model Gini trends as a fractionally integrated process and find that there are more mean-reverting absolute Ginis than relative Ginis suggesting that absolute Ginis may be better suited than relative Ginis for time-dependent analyses. I then undertake an estimation of the inequality-growth relationship using popular panel regression methods and find that the absolute Gini is negatively and significantly associated with growth for most models estimated, but none for the relative Gini. I deduce that from an empirical point of view, the absolute Gini may be a better choice when undertaking time dependent analyses.