Public expenditure and economic growth in Sub-Saharan Africa

Abstract

This study had three objectives; to explore causality between aggregate public expenditure and economic growth, to assess whether convergence exists in public expenditure and economic growth and to estimate the effect of functional expenditure on economic growth. A Sub-Sahara African data set covering the period 1980-2007 was employed. Unidirectional Panel Granger causality between public expenditure and economic growth all in per capita terms, basing on Wagner’s law, was found to run from per capita GDP to per capita expenditure at 1% significance level. On the other hand, aggregate expenditure was found to Granger cause GDP at 5% level of significance and the results were also unidirectional. Since per capita GDP is a preferred measure of economic growth in a number of studies, real per capita GDP granger causes per capita expenditure in a panel of 43 countries. Results of convergence showed that high income countries converged faster in terms of per capita expenditure pointing to the fact that there is no tendency for the rich and poor countries to reduce differences in their public expenditure over time. On the other hand, although beta convergence in per capita income was found to exist in the panel as a whole, the robust test revealed divergence. Simultaneous analysis of sub-panels revealed that low income-coastal-resource scarce countries converged faster than the low income-landlocked- resource poor countries at 33% and 9% respectively. This indicates that poor countries exposed to trade in goods and services are likely to grow faster. The results also indicate that natural resources have a positive effect on the catch-up process of both middle and low income countries. Therefore labor, capital, trade openness and natural resource base are significant in explaining economic convergence process in SSA for the period under review. In a panel of fourteen countries with available data, defense expenditure was found to significantly reduce economic growth and the results were consistent in all models except the dynamic panel data. However, education, health, general public services and social welfare expenditures were only significant in enhancing economic growth measured in terms of GDP. There is need to reallocate resources to functional expenditures with a bias against defense spending. Monitoring outputs and outcomes from functional expenditures is implied. The negative coefficient on economic affairs expenditure needs further investigation.