Abstract
The main objective of this study is to test Okun’s law in 16 Southern African countries between 1995 and 2019. The study adopts the panel autoregressive distributed lag (ARDL) method and the ARDL error-correction model (ARDL-ECM) and makes use of the pooled mean group (PMG) and the mean group (MG) estimators, based on Pesaran et al. (1999), to arrive at estimates. The study also makes use of the panel Granger non-causality test of Juodis et al. (2021) to investigate the direction of causality between unemployment rate and economic growth.
The study found that although the results of the error correction term are significantly negative for the full sample, the results do not indicate the existence of a long run cointegrating relationship between unemployment rate and real GDP growth rate as the long run effect of latter on unemployment is not statistically significant. Thus, cointegration exists between unemployment rate and some of the statistically significant explanatory variables but not real GDP growth rate. Regression results showed that the long run coefficient of real GDP growth rate is negative but statistically insignificant for all samples except the low-income (LIC) sub-sample. Although the negative estimate of real GDP growth is consistent with theory for all samples, the insignificance of the coefficients refutes the relevance of Okun’s law in Southern African countries. The short-run regression results by country and income group samples show that Okun’s law is mostly invalid and, where it is applicable, the results are insignificant. These findings generally imply that unemployment rate is weakly responsive to real GDP growth rate in Southern African countries. Further, the panel Granger non-causality results show that the direction of causality runs from unemployment rate to GDP growth rate. This finding provides insight that may help to explain the jobless growth in Southern African countries. Furthermore, bidirectional causality was found only for upper middle-income and high-income sub-samples. In light of these results, the study’s policy implications are twofold: (1) that southern African countries deal with the structural constraints that affect labour demand and supply; and (2) improve the macroeconomic policy environment to improve the potential for growth.
Keywords: Okun’s law, economic growth, unemployment rate, pooled mean group, mean group.