Abstract:
International reserves of West African countries rose sharply by 84.3% between 1991 and 2001 and 287.1% between 2001 and 2011. However, due to high macroeconomic instability in the form of persistent asymmetric shocks, output variability and fiscal policy distortions, the reserves were inadequate and the countries in the sub-region are continually faced with the problem of balancing the costs and benefits from reserves holdings. Previous studies have paid little attention to the factors determining reserves and the prospect of reserves pooling to reduce the prevalent macroeconomic instability. This study, therefore, examined reserves determinants and the effects of a prospective reserves pooling on macroeconomic stability in the Economic Community of West African States (ECOWAS) subregion between 1981 and 2011, a period within which there was an availability of uniform time series data on the macroeconomic variables.
The buffer stock model was employed to capture the determinants of reserves (imports, external debts, government spendings, exchange rates, foreign direct investments, exports and investment) and the Optimum Currency Area (OCA) asymmetric shock framework was utilized to evaluate the impact of reserves pooling on macroeconomic stability. Data were collected from World Bank and International Monetary Fund databases. The fixed and random effects models in the panel data framework were estimated to capture the demand and supply determinants of reserves. The impulse response and variance decomposition in the domain of Panel Vector Autoregression (PVAR) were employed to estimate the impact of reserves pooling on macroeconomic stability. The PVAR estimations were undertaken in an ex ante counterfactual analysis and considered reserves held in an autonomous state as well as under a 60.0% pooling scheme as recommended by the West African Economic and Monetary Union (WAEMU) as the optimal pooling arrangement. Diagnostics and robustness tests (reserves optimality, PVAR stability, residual normality and cointegration tests) were carried out to ascertain the reliability of the estimates. All estimates were validated at p=0.05.
The statistically significant demand determinants of reserves in the region were imports (-0.43), external debts (-0.83), government spendings (-0.03) and exchange rates (0.94) while the supply determinants were foreign direct investments (0.09), exports (0.05) and investment (0.02) for the Fixed Effect Model. The unbiasedness of the Random Effect Model was rejected using the Hausman test. Generalised one standard deviation innovation significantly reduced macroeconomic instability under the pooling scheme when compared to autonomous state of the individual countries as expressed in the behavioural models estimated. Percentage instability reduction from reserves pooling were external debts (22.9%), government spendings (26.8%), investments (54.8%), trade flows (56.9%), gross domestic product (58.3%) and exchange rates (60.7%). Overall, macroeconomic instability in the region was reduced by 46.7% when the countries entered into the 60.0% reserves pooling arrangement.
Adequate reserves holdings through reserves pooling arrangement insulated the economies of countries in West Africa. Therefore, monetary authorities in the region should intensify efforts towards the successful establishment of international reserves pooling.
Keywords: International reserves pooling, Fixed and random effects models, Panel vector autoregression, Macroeconomic stability, ECOWAS.
Word count: 474
