Current account sustainability: evidence from South Africa

• Main Author: North, Andrew
• Other Authors: Kahn, Brian
• Format: Dissertation
• Language: English
• Published: University of Cape Town 2015
• Subjects: Economics
• Online Access: http://hdl.handle.net/11427/12509

Bibliography : p. 83-88. === This paper investigates a set of "leading indicators" behind large and persistent reductions ('reversals') in South Africa's current account. It begins by providing a theoretical analysis of current account sustainability based on the intertemporal optimisation of consumption, saving and investment decisions. Drawing on these theoretical concepts, a non-structural approach is then adopted in considering a broad range of indicators that are instructive in evaluating the sustainability of current account deficits. In particular, a country's macroeconomic structure and policy, its external position, and other factors such as expectations, socio-political stability and credibility, are identified as important. Based on event study methodology (Eichertgreen et al, 1995) and work by Milesi-Ferretti and Razin (1997, 1998), a multivariate probit model for South Africa's current account 'reversals' is estimated. Using quarterly data from 1965 to 1996, a binary dependent variable is constructed and used to analyse the impact of various theoretically important variables. The results show that South Africa's current account reversals have been influenced mainly by changes in the country's external macroeconomic environment. These include increases in the OECD economic growth rate, improvements in the country's terms of trade (especially through rises in the dollar gold price), and increasing levels of foreign debt. Reversal periods are typically also associated with rising levels of exports and falling levels of imports. While developments on the capital account were expected, prime facie, to have had a significant impact on the current account, this is not borne out well in the regression results. FDI flows, however, are found to be a statistically significant indicator of current account reversals over the sample period.