The Macroeconomics of Fiscal Policy in Sierra Leone: An Empirical Investigation of Fiscal Sustainability, Fiscal Policy Shocks, and the Tax Smoothing Hypothesis

Samuel Bonzu

Research output: ThesisDoctoral Thesis

286 Downloads (Pure)

Abstract

Abstract
This thesis comprises three empirical chapters that examine the on-going issues in fiscal policy debates in Sierra Leone. The first empirical chapter, Chapter Two, assesses the sustainability of public finance in Sierra Leone. We examined sustainability by employing unit root tests on primary balance, cointegration between government revenues and expenditures, and finally, the fiscal reaction function showing the relationship between lagged debt ratio and primary balance ratio. A battery of unit rest tests was performed on primary balance ratio and the results confirm sustainability. Likewise, different tests for cointegration between government revenues and expenditures were applied and the results show that fiscal policy over the years is weakly sustainable. Controlling for endogenous structural breaks, the results show that the sustainability became weaker post 1984. Finally, we estimated a policy rule that further confirms that fiscal policy in Sierra Leone under the review period was sustainable.

Chapter Three examines the short-run effects of fiscal policy shocks on key macroeconomic variables within the framework of structural vector autoregression. The key results show that output and private consumption are persistently crowded-in by positive innovation in government spending. Government spending moderately increases private investment and government revenue. Shocks to government revenue temporarily reduce output and investment. Both spending and tax shocks are inflationary and interest rate rise to spending increases in the short term. Additionally, government investment expenditure strongly stimulates the economy in contrast to government consumption expenditure.

Chapter Four investigates the existence of optimal taxation in Sierra Leone. ‘Optimality’ here means if the government smoothed tax over time. Following Barro (1979), we employed different unit roots tests on tax rate to examine its stationarity properties. Moreover, we employed univariate autoregression and vector autoregression to examine the unpredictability of tax rate. Both approaches confirm that tax rate changes are unpredictable which implies that tax smoothing hypothesis hold under the review period.
Original languageEnglish
QualificationPh.D.
Awarding Institution
  • Royal Holloway, University of London
Supervisors/Advisors
  • Spagat, M, Supervisor, External person
Award date1 Dec 2018
Publication statusUnpublished - 19 Nov 2018

Cite this