This International Monetary Fund (IMF) working paper conducts a Poverty and Social Impact Analysis (PSIA) of replacing Ethiopia’s sales tax with a value-added tax (VAT) in 2003. This reform has not had a major adverse effect on the poorest 40 percent of the population. The VAT is progressive in its incidence, and the higher revenues it brings can provide additional funds for poverty-reducing spending, e.g. primary education. There is significant scope for making education spending more pro-poor by increasing the access of low-income households to schools.
The IMF has been supporting a three-year Poverty Reduction and Growth Facility (PRGF) in Ethiopia since March 2001. The PRGF requires an increase in tax revenue. The sales tax base was limited to imports, manufactured goods, and a few selected services. The new VAT taxes services as well as production, with a uniform rate of 15 percent on most goods and services, zero rates exports, and exempts fewer goods and services than previously. The new VAT is intended to increase revenue, economic efficiency, exports, and growth, but will have differential effects on groups of the population. Some exemptions improve the distributional impact of the tax, but others exemptions may occur for administrative or political reasons. Are the current exemptions in Ethiopia justified?
Using the Ethiopian Central Statistical Authority’s 1999/2000 Household Income, Consumption and Expenditure Survey, the paper finds that:
- VAT is progressive for total expenditure at national level. Most exempt goods and services are disproportionately consumed by the relatively well-off, so the exemptions cannot be justified on equity grounds.
- The progressivity of the VAT is owed mainly to the high ratio of in-kind transactions for poorer households – as the economy achieves more stable growth and becomes more market-based, this ratio should decrease and VAT will become less progressive.
- VAT is regressive or neutral in urban areas, and given growing urbanization mainly of poor people, could seriously affect the poor – authorities should find ways, perhaps restructuring exemptions, to adjust VAT to yield sufficient revenue and increase its progressivity.
Tax incidence should be examined in combination with government spending incidence: benefit incidence analysis shows the distribution of government benefits financed in part from tax receipts. Given a lack of data availability, a recent benefit incidence study on spending on primary education and health in Ethiopia in 2002 – these are crucial areas for poverty-targeted public expenditure because they invest in future human capital as well as the welfare of the current population is described.
- The analysis suggests that average and marginal benefits of public expenditure on primary education accrue mostly to the rich, with particularly inequitable distribution of benefits in rural areas.
- Health expenditure is well targeted, so the poor benefit in absolute terms.
- Recent and projected increases in poverty-reduction budgets promise for yet greater benefits for the poor. If the additional VAT revenues were deployed in primary education and health, the poorest 40 percent of the population would be net beneficiaries.
