Real exchange rate misalignment and growth of tradable sectors in South Africa: a sectoral dynamic panel data approach
Loading...
Date
2017
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Zululand
Abstract
Disappointing growth outcomes in South Africa on the back of a weakening currency have once again underscored the need for examining carefully how undervaluation relates to economic growth. Evidence on this subject has been accumulated from aggregate studies which do not take us far in understanding how undervaluation affects particular sectors. This study contributes to the existing knowledge by using a sectoral approach to establish the link between undervaluation and growth of South Africa’s agriculture, mining, manufacturing, tourism, personal services and financial sector for the period 1985 through 2014. It addresses endogeneity of the real exchange rate by using the system Generalised Methods of Moments (GMM) technique and reverse causation by using initial rather than contemporary values of explanatory variables. Two measures of undervaluation are used. The first baseline measure is adjusted for the Balassa-Samuelson effect by regressing the bilateral exchange rate between the South African rand and the United States dollar on income per capita using the dynamic ordinary least squares technique. The second measure is computed from a model of a small open economy where the exchange rate is undervalued if the prevailing real exchange rate is lower than the equilibrium real exchange rate dictated by its fundamentals. Using these two measures as regressors in a conditional convergence growth specification, a positive and significant relationship between undervaluation and sectoral growth emerges with a percentage increase in undervaluation being estimated to raise average annual growth by 0.17 per cent holding constant sector specific factors. This effect increases with capital accumulation rather than employment creation and is particularly more relevant and sizeable for sectors that start off poor. These findings are robust to using alternative panel estimation techniques that include the Augmented Mean Group (AMG) and the Common Correlated Effects (CCE) estimators. Time series techniques are then employed to track the effect of undervaluation on individual sectors. Using the bounds testing procedure due to a mixture of variable integration, the results indicate that the impact of undervaluation is not uniformly distributed across sectors. In the short-run period, undervaluation promotes growth in the tourism, financial, mining and personal services sectors while agriculture and manufacturing are all hampered by undervaluation both in the short-run and long-run. Only the mining sector appears to benefit from undervaluation in both short-run and long-run periods. The latter result could be reflective of immense global competition that turn South Africa’s agriculture and manufacturing exports into non-tradables at the margin. Only the mining sector appears to benefit from undervaluation and we take this to reflect presence of an already high global demand for South Africa’s minerals. Overall, time series results emphasize the importance of controlling sectoral heterogeneity and that the effect of undervaluation must not be generalised across sectors. For policy issues having found undervaluation promoting growth of a minor sector (mining) and hurting growth of major sectors in terms of contribution on gross domestic product, our analysis suggests that undervaluation is not the solution to South Africa’s current slow growth as claimed in recent literature that support undervaluation-led growth; it is in fact part of the problem.
Description
A dissertation submitted to the Faculty of Commerce, Administration and Law in l fulfillment of the requirements for the Degree of Master of Commerce in the Department of Economics at the University of Zululand, 2017
Keywords
growth --currency --undervaluation --economic growth --South Africa