VOLATILITY OF CAPITAL FLOWS IN EMERGING AFRICAN ECONOMIES

ABSTRACT

This study examines the empirical nature of capital flows in some selected African countries –i.e. South Africa, Morocco, Egypt and Botswana. In addition it also investigate empirically such issues of sudden stops and reversals. The study has employed the widely used Generalised Autoregressive Conditional Heteroskedasticity (GARCH) model and considers the Exponential Generalised Autoregressive Conditional Heteroskedasticity (EGARCH), which was tailored to empirically examine absolute and relative magnitudes of the flows in these economies. Moreover, it detailed the volatile nature of capital flows in these countries, including the identification of periods of sudden stops and reversals as well as measuring the extent of their historical volatilities. Thus, revealed some interesting results, such as strong evidence of persistence in capital flow volatility for all the countries under investigation and that the volatility does not indicate a tendency of reversal to its previous mean. Furthermore, it identify that, the previous information of each capital component (foreign direct investment and portfolio flows)   demonstrates a strong effect on the behaviour of capital flow volatility across all the countries under review. Additionally, conforming to a priori, the study revealed that portfolio investment is more volatile than the foreign direct investment (FDI) for all countries. Further, it discover inherent asymmetries in the volatility behaviour which is underpinned by the large information asymmetries in most capital market in these countries. This finding is presumed by the consistency of the EGARCH over other ARCH models. Overall, the study indicates that FDI is the most volatile flow in Botswana, South Africa, Egypt and Morocco, in that order. While PINV is most volatile in South Africa, it is followed by Egypt, Botswana and Morocco respectively.