Exchange rate co-movement and volatility spill over in Africa

Abstract

Although research interest in why and how economies are related to each other continues to accumulate (Joyce, 2013, Todd, 2008), our understanding of the mechanisms through which economic turmoil in one region transmits to another is still unclear. The motivations for this study were to examine the volatility transmission in the African foreign exchange market and the global world through financial interdependence and factors explaining exchange rate co-movement. The thesis set out to find answers to the questions of volatility spill over within, from and to the Africa region and also to identify factors of exchange rate co-movement changes.

The study employed the multivariate GARCH with the unrestricted full BEKK version to estimate the volatility transmission. The BEKK GARCH and the panel data approach were used to analysis the determinants of co-movement. Monthly data of GHC/USD, Ksh/USD, ZAR/USD and GBP/USD for volatility transmission analyses and quarterly data for the influence of exchange rate co-movement were analysed. Data were sourced from the International Monetary Fund’s direction of trade statistics and international financial statistics. The selection of countries for the study was influenced by common colonial ties, flexible exchange rate regimes, high depreciation of the currencies and competition in the same exports markets.

The results from the VAR(1)- multivariate GARCH (1,1) analysis, suggest that, there is a statistically significant mean spill over at 5% level. Regionally, significant mean transmission is found from the South Africa rand to the Ghana cedi only. Significant mean transmission is identified mainly from the global world to Africa and bidirectional transmission between South Africa and the global world.

Exchange rate volatility transmission results show that there are high positive correlations between currencies and significant interaction between second moments of these currencies. First, global variance of exchange rates significantly influenced a volatility spill over in Africa supporting the idea that foreign exchange markets in Africa are more prone to the cross-over effect of developed economies. Regional volatility spill over is only significant from South Africa to Ghana. The BEKK estimation revealed significant correlation between African countries and the global world.

The panel data analysis of the bilateral exchange rate correlation showed that foreign exchange markets in Africa are highly positively influenced by the shock from the global world. This relates to an average world interest rate and capital account openness. Similarly, from the demand side, the trade linkage appeared to have a long run positive effect on any co-movement. Regional differences in interest rate and financial development negatively influenced the exchange rate co-movement.

The major finding implies that the foreign exchange market in Africa is more prone the global market than intra regional volatility spill over in both volatility and mean spill over. Thus, the African foreign exchange market is under meteor shower hypothesis. Exchange rate correlation is explained by a shift in the global economies and a low level of financial market development in the region.

Policy makers can achieve financial and exchange rate stability if they are able to control macroeconomic factors and cope with changing circumstances in the region and from outside. In addition, authorities should be able to moderate exchange rate fluctuations through strong institutional set-up since economic governance is now a force to be reckoned with in macroeconomic management and regional leaders should also conduct an exchange rate policy on a regional basis in order to cope with any shocks and exchange rate volatility.