A panel vector AutoRegression analysis of income inequality dynamics in each of the 50 states of USA

Abstract

Purpose

The purpose of this paper is to investigate the income inequality dynamics in each of the 50 states of USA over the period 1981-2011.

Design/methodology/approach

The paper estimates an augmented Kuznets curve panel Vector AutoRegression in per capita income, economic freedom, educational attainment, unemployment, and population ageing along with evaluating generalized impulse responses functions (GIRF) and generalized forecast-error variance decompositions (GFEVD).

Findings

All the variables are integrated of order one and are panel cointegrated. Kuznets’ hypothesized inverted U-shaped relationship between inequality and growth is not supported by the data. Unemployment and population ageing have statistically significant positive effects on inequality in the long-run; education has statistically significant negative impact; economic freedom has statistically insignificant positive effect. Long-run bidirectional causality exists among the variables. GFEVD show that excluding income inequality itself, variation in income inequality is more influenced by perturbations in per capita income, educational attainment, and unemployment. GIRF corroborate the results of the GFEVD.

Originality/value

This paper fulfills an identified need to study the causal relationship between inequality and its determining factors without assuming the a priori exogeneity or endogeneity of the underlying variables.

Citation

Onafowora, O. and Owoye, O. (2017), "A panel vector AutoRegression analysis of income inequality dynamics in each of the 50 states of USA", International Journal of Social Economics, Vol. 44 No. 6, pp. 797-815. https://doi.org/10.1108/IJSE-06-2015-0154