The effect of exchange rate volatility on the volume of South African exports

The purpose of this paper is to investigate whether a relationship between export volumes and exchange rate volatility exists as suggested in the ASGISA document. It goes about this by first investigating the theoretical channels that predict the relationship between export volumes and exchange rate volatility. The theoretical prediction though, is ambiguous depending on the justification, model and factors taken into account in reaching the result. Furthermore, the paper provides evidence that the empirical results are ambiguous as well, as some countries tend to exhibit a negative relationship and others a positive relationship. Thus the paper goes on to estimate two measures of exchange rate volatility (a moving average standard deviation and the conditional volatility extracted from GARCH modelling) using the real effective exchange rate. These measures provide varied results in testing for stationarity. A cointegrated model for export volume using the Johansen estimation technique is then estimated with the volatility variables included in either the short or long run model, depending on the sensibility of the results produced.

NOTE: This I am currently revamping this paper, it was my mini-dissertation done toward the completion of my Masters in Economics at UCT. I am still estimating the cointegrated model, and so am not sure if any of my volatility variables will be able to enter in the long run model (in the previous paper this was not possible). I have not summarised the results yet for this reason. The model will be estimated with data ranging from the first quarter of 1961 to the first quarter of 2007 (the reason for the revamp is that this paper previously only considered data up until the first quarter of 2005). The model will also be estimated for the period starting 1972 (after the collapse of Bretton Woods) until the present, or a structural break at this period will be investigated (to see if the behaviour changed as a result of the introduction of the floating exchange rate in spite of the variability of the Real effective exchange rate before this time). The policy implication will be suggested at the end. Originally there was a negative effect of exchange rate volatility on the GROWTH of export volumes (so a significant negative short run effect), but not on the level of export volumes.

Finally, I may (time permitting) include a third estimation of a volatility variable that has been suggested in some of the international literature (a linear moment model/instrumental variable type approach) and the effects of this volatility measure on export volumes.