A review of Zimbabwe’s macro-fiscal forecasts

Abstract

Purpose

The purpose of this paper is to evaluate accuracy of macro fiscal forecasts done by Government of Zimbabwe and the spillover effects of forecasting errors over the period 2010-2015.

Design/methodology/approach

In line with the study objectives, the study employed the root mean square error methodology to measure the accuracy of macro fiscal forecasts, borrowing from the work of Calitz et al. (2013). The spillover effects were assessed through running simple regression in Eviews programme. The data used in the analysis are based on annual national budget forecasts presented to the Parliament by the Minister of Finance. Actual data come from the Ministry of Finance budget outturns and Zimbabwe Statistical Agency published national accounts.

Findings

The results of the root mean square error revealed relatively high levels of macro-fiscal forecasting errors, with revenue recording the highest. The forecasting errors display a tendency of under predicting the strength of economic recovery during boom and over predicting its strength during periods of weakness. The study although found significant evidence of GDP forecasting errors translating into revenue forecasting inaccuracies, the GDP forecasting errors fail to fully account for the revenue errors. Revenue errors were, however, found to be positive and significant in explaining the budget balance errors.

Originality/value

In other jurisdictions, particularly developed countries, they undertake regular evaluation of their forecasts in order to improve their forecasting procedures, which translate into quality public service delivery. The situation is lagging in Zimbabwe. Given the poor performance in public service delivery in Zimbabwe, this study contributes in dissecting the sources of the challenge by providing a comprehensive review of macro fiscal forecasts.

Citation

Mudhunguyo, C. (2017), "A review of Zimbabwe’s macro-fiscal forecasts", African Journal of Economic and Management Studies, Vol. 8 No. 4, pp. 433-440. https://doi.org/10.1108/AJEMS-03-2017-0060

Download as .RIS