A non-stationary panel data approach for examining convergence in South Africa

Economic convergence has received much attention since the 1980s when researchers tried to ascertain whether low-income countries would stay that way in the long run, or they would gain ‘developmental traction’ and become the affluent nations of the future. This article gives fresh insight on this topic from an African perspective by comparing 39 countries—South Africa, 32 Organisation for Economic Cooperation and Development (OECD) members and 6 Latin American countries. The author investigated their average steady-state equilibria and tested convergence trends from 1980 to 2019. The Solow–Swan model was tested. Furthermore, this study applies panel econometric modelling to determine the relationship between the variables analysed in the convergence analysis. This commenced with the Levin–Lin–Chu and Im–Pesaran–Shin panel unit root tests. Then, the Kao test and the vector error correction model were used to evaluate the cointegration and relationships between variables. The findings revealed that South Africa’s economic performance is significantly lower than the OECD average gross domestic product per capita with an annual growth rate of 0.54%, which falls below the ‘iron law of convergence’ hypothesis. JEL classifications: C01, C32, C33, E13, F62, F63