Testing the Ricardian Equivalence Hypothesis in Zimbabwe: An ARDL Bound Testing Approach

Abstract

Government deficit financing, particularly debt for tax swap, is increasingly dominating debate over the role of fiscal policy in macro-economic stabilisation. The Ricardian Equivalence Hypothesis (REH) suggesting fiscal policy impotence has sparked empirical tests in many countries. However, no empirical effort has been devoted to Zimbabwe. The major objective of this study was to econometrically test whether the REH holds in Zimbabwe. Using Bound Testing approach to Cointegration and Error Correction Model within the context of the Auto-Regressive-Distributed-Lag (ARDL) framework developed by Pesaran and Shin (1995, 1999) , we investigated whether a long-run equilibrium relationship exists between Private Consumption and Gross Domestic Product, Government expenditure, Tax Revenue, Total Public Debt and Interest Payments. We analysed the strength of the long run association using Impulse Response Functions (IRFs).  We then run the OLS regression on the reduced form consumption function, derived from Kormendi’s Consolidated Consumption function to test the REH in Zimbabwe. The results show that there is long run association running from Gross Domestic Product (GDP), Government Expenditure, Tax Revenue, Total Public Debt and Interest Payments to Private Consumption. More importantly we found strong evidence against the REH in Zimbabwe and support for Keynesian debt non-neutrality. The findings therefore imply that fiscal policy has a role in macroeconomic stabilisation in Zimbabwe.

Keywords: Ricardian Equivalence, Debt for tax swap, Bound Test of Cointegration, ARDL Model, IRF.