The twin deficits hypothesis in east African countries
Abstract
The twin deficits hypothesis asserts a positive relationship between current account balances and fiscal balances in an economy. Previous tests of the hypothesis have generated mixed findings. Some support the hypothesis while others do not. The mixed results could be because most studies failed to account for structural breaks and volatility clustering. This study aimed at testing the twin deficits hypothesis in East African Countries Kenya, Uganda and Tanzania on time series data for a period of thirty seven years from 1980 to 2016 while allowing for structural breaks and conditional heteroskedasticity. The study used the Bai and Perron Global Optimization method to identify the number and location of structural breaks in each country’s data. The VAR-GARCH technique was then used to test the relationship between fiscal balances and current account balances. This technique improves the validity of the tests and hence reliability of the results. The results of the Bai and Perron test showed that structural breaks existed in all the countries with Uganda having more number of breaks. The results of the VAR GARCH analysis revealed that the Kenya and Tanzania had a positive and significant relationship between fiscal balances and current account balances. The results also revealed that Uganda had a positive but insignificant relationship between fiscal balances and current account balances. The results emphasize on the need for governments to maintain favorable fiscal balances in order to improve their current accounts. Allowing for structural breaks in the model makes the use of VARs appropriate since failure of unit root tests is minimized. Allowing for conditional heteroskedasticity in the model results to efficient coefficients, which can be relied upon for policy implementation.
Publisher
University of Nairobi
Rights
Attribution-NonCommercial-NoDerivs 3.0 United StatesUsage Rights
http://creativecommons.org/licenses/by-nc-nd/3.0/us/Collections
The following license files are associated with this item: