A Causality Test of the Revenue-Expenditure Nexus in Ghana

Abstract The paper investigates the revenue-expenditure nexus for Ghana. The study covers the period 1980-2013. It examines whether increases in government revenue cause increases in government expenditure or increases in government expenditure cause increases in government revenue. It also examines if changes in government expenditure and revenue have feedback effects on each other. The stationarity test indicates that both variables are stationary at the levels when the test is done with a constant and a trend, and are first difference stationary when the test is done with a constant but no trend. The paper analyses the long-run relationship between government expenditure and government revenue using the Ordinary Least Squares (OLS) method. The short-run relationships between the two variables are tested in a Vector Autoregressive (VAR) framework. The results show a very strong long-and short-run relationship between the variables. The second period lag of the revenue variable shows a negative relationship between government revenue and government expenditure. This indicates the possibility of the absence of Fiscal illusion in every two years of increased government expenditure. Granger causality test is done to determine the direction of the causal relationship between government expenditure and government revenue. The test gives a unidirectional causality running from revenue to expenditure. This implies government revenue causes government expenditure. Therefore, evidence of Tax-spend hypothesis is found. The implication is that, government must improve its revenue generation efforts in order for it to fund its ever increasing expenditure and to control the frequent fiscal slippages.