Clustering and regime dynamics for economic growth and income inequality

Abstract

This study explores the dynamic relationship between income inequality and economic growth by using a non-parametric approach and numerical taxonomy as a research method based on data symbolization and clustering methods. The study uses annual data of the GINI index (considering two databases, i.e. the Standardized World Income Inequality Database (SWIID) and the Estimated Household Income Inequality Data Set (EHII)) and the Per Capita GDP Growth Rates (economic growth variable) for two samples, i.e. 38 countries between 1980 and 2015, and 23 countries during the period 1980–2010. This novel methodology is used to detect the existence of clusters of countries sharing similar economic performance that are characterized by the income inequality variable. Once the homogeneous clusters are fixed, using a dynamic econometric approach, we study the Granger causal relationship between economic growth and income inequality, and the positive or negative long-run effects. Our results show that in advanced economies there is an economic growth policy supporting income distribution, while in poor or developing economies economic growth is enhanced by income concentration.