A re-examination of long-run Purchasing Power Parity (PPP) hypothesis: the case of two Southern African countries

Abstract

In this paper, we have examined the validity of the PPP hypothesis for two Southern African countries, namely: Lesotho and Zambia. We have utilized four econometric tests to examine the existence of the PPP hypothesis in Lesotho and Zambia. These tests include two unit root tests without structural breaks—the DF-GLS test and the Ng-Perron test; and two unit root tests with structural breaks—the Perron test and the Zivot-Andrews test. We extracted the data on the real exchange rate for Lesotho, and for Zambia, from the Penn World Tables, version 7.1, which is compiled by Heston et al. (2012). We found that the PPP hypothesis was supported in the case of Lesotho, but rejected in the case of Zambia. The implication of this finding is that Lesotho is unlikely to profit immensely from trade and investment arbitrages, whereas Zambia is more likely to profit immensely from trade and investment arbitrages, by trading with the US.