The Nature of Global Income Inequality and Its Determinants

The paper examined the nature of global income inequality from 1980 to 2023 through linear and non-linear trend models. It decomposed into cycles through the Hamilton regression filter model and the H.P. Filter model and also framed a forecast model for 2050 using the ARIMA (1,1,1) model using data on the income share of the top 10% and bottom 50% from the World Inequality Database. It was found that global income inequality is inversely U-shaped. The ARIMA (1,1,1) model of global income inequality during 1980‒2023 is significant, convergent and stable, and the forecast model for 2050 is stationary and convergent to equilibrium. During the cyclical path the income inequality and growth are downward while the unemployment rate is upward in the period of global recession in 2008‒2009. Still, income inequality and unemployment rate are upward, while growth is downward in the recession during Covid-2019‒2020. The article also examined the cointegration and vector error correction analysis among global income inequality, global unemployment rate, global GDP growth rate and global inflation rate, respectively during 1991‒2023, using data from the World Bank Database and World Inequality Database by applying Johansen cointegration test (1988). It was found that there is one significant cointegrating equation which implied that the variables are associated in the long run where the cointegrating equation stated that income inequality is negatively related with unemployment rate and positively related with inflation and growth rates where all are significant except inflation rate. There is also a short-run causality from income inequality to the inflation rate. The cointegrating equation converges to equilibrium insignificantly at the speed of adjustment of 1.21% per annum, with its trend coefficient being positive and insignificant. The impulse response of income inequality to unemployment and growth rate is convergent, while the response to inflation rate is divergent. The VECM is stable and nonstationary, showing autocorrelation and multivariate non-normal problems. Target rate of inflation, tight monetary policy, progressive taxation and higher investment in small-scale manufacturing sector and infrastructure might be effective for benefiting good long run association among them.