Ethiopia introduced value added tax (VAT) as a replacement of sales tax in January 2003. The Ethiopian VAT has annual turnover threshold of ETB 500,000 and is chargeable with zero rate on exports and 15 percent on all others except exempted supplies. One of the purposes of the Ethiopian government in replacing the previous cascading type sales tax with VAT is to improve the revenue to Gross Domestic Product (GDP) ratio and enhance economic growth. Examination of the revenue performance of VAT since its introduction indicates that even though there has been an increase in government revenue after the introduction of VAT, it appears that the ratio of tax revenue to GDP has been fluctuating.

The government’s commitment to use the VAT as an instrument of enhancing its revenue position and fostering economic growth is noted in its efforts of strengthening the administration such as the establishment of the Ministry of Revenue and computerization of the VAT system although there remains a lot to be done to make the VAT achieve its intended objectives. To support the government’s efforts of enhancing its revenue position and enable the VAT system achieve its intended objectives, the design needs to be based on the principles of fairness, efficiency and administrative simplicity, among others.

In the light of the above the purpose of this paper is to examine the design of VAT in Ethiopia focusing on such features as tax coverage, exemptions, small traders and threshold level, rate structure, VAT accounting and reporting, collection and refund. In order to assess the Ethiopian VAT design in terms of these features, comparative analysis is made with the VAT system in Kenya and the model VAT design in the world – GST in New Zealand. The study reveals the areas which currently have their bearing on the operation and the revenue generation capacity of the VAT that could be improved without significantly affecting the government’s other objectives.

1. Introduction

While the principal source of a government’s revenue should be taxation, in many Sub-Saharan African nations this is often not the case. Many of the Sub-Saharan African countries rely on foreign sources of finance namely foreign loan and aid. For instance taking foreign aid only, in Guinea Bissau, aid represents about 37.3 percent of GDP; similarly in Malawi and Sierra Leone the aid to GDP ratio is 26.2 percent and 28.7 percent respectively. In Kenya the ratio is 4.9 percent (Cheeseman and Griffiths 2005, p. 3).

Expanded domestic revenue base (especially taxation) offers a promise of greater autonomy in the future and a break from restrictive aid and loan conditionalities. With this vision, recently, many poor countries have become preoccupied with improving tax systems. For instance, in the case of Ethiopia in an effort to increase the government’s domestic revenue the government replaced the sales tax with Value Added Tax (VAT), the Ministry of Revenue was established giving autonomy to the government’s revenue organ, and the tax system is computerized. Similarly, in Kenya the introduction of VAT and the establishment of the Kenyan Revenue Authority are claimed to have contributed to the improvement in the revenue position of the government.

In light of the above, the purpose of this paper is to assess the design of VAT in Ethiopia focusing on such features as tax coverage, exemptions, small traders and threshold level, rate structure, VAT accounting and reporting, VAT credits and refunds. In order to assess the Ethiopian VAT design in terms of these features, comparative analysis is made with the VAT system in Kenya and the model VAT design in the world – GST in New Zealand. Although both Kenya and New Zealand have had longer experience than Ethiopia in administering VAT, the choice of New Zealand is because New Zealand’s GST/VAT is cited by many tax experts as a model towards which VAT systems in the world need to strive. In the case of VAT in Kenya, with the intention of looking at Ethiopia’s VAT in relation to that of other developing countries, the choice is primarily based on the availability of information. Thus, the paper intends to identify the areas on the Ethiopian VAT design, which are currently having adverse impact on the operation and revenue generation capacity of the tax and also on the fairness and administrative simplicity of the VAT system. Accordingly, section two presents a review of the financial structure of the Ethiopian government. Section three discusses the design of Ethiopian VAT focusing mainly on issues such as tax base, rate structure, exemptions and refund. Section four presents an overview of VAT design in Kenya. The fifth section presents the main design features of goods and services tax (GST) in New Zealand. Section six presents a comparison of the main design features of VAT/GST in Ethiopia, Kenya and New Zealand. Finally, lessons from Kenya and New Zealand and concluding remarks are presented in section seven.

2. A review of the financial structure of the Ethiopian Government

An examination of the government's expenditure program over the last decade or so reveals fluctuation. For instance, in the year 1992/93 government expenditure as a per centage of GDP was 19.57 per cent, which increased to 33.7 per cent in 2002/03 and then declined to 29.4 per cent[1] in 2003/04 (NBE, 2003/04). The structure of government expenditure is dominated by recurrent expenditures that absorbed about 66 per cent of the total in 1992/93, 68 per cent in 2002/03 and 59.5 per cent in 2003/04 (NBE, 2003/04). The trend in the two components of expenditure (recurrent and capital) is similar to that of the total expenditure. As a share of GDP, recurrent expenditure increased from 14.7 per cent in 1995/96 to 24.8 per cent in 2002/03 while capital expenditure showed a slight surge from 9.4 per cent in 1995/96 to 11.6 per cent in the 2002/03 fiscal year (AfDB/OECD 2004).

In spite of such a surge[2] in expenditure, government revenue showed a small rise. For example, total government revenue as a share of GDP increased from 18.4 per cent in 1995/96 to an estimated 20.5 per cent in 2002/03 (AfDB/OECD 2004). Government revenue that falls short of financing the entire expenditure program is raised mainly from taxation that contributed about 78 per cent (table 1 in the appendix) of total revenue in 2003/04; this was an increase from 69 per cent in 1992/93, evidencing the significance of taxation in the government's finance (G/Egzihabher 2005).

Examination of tax revenue of the Ethiopian government as a share of GDP indicates that the ratio increased from 12.5 per cent in 1995/96 to 15.3 per cent in 2001/02. The tax to GDP ratio[3] declined to 14.4 per cent in 2002/03 and then it showed a rise to 15.2 per cent in 2003/04 fiscal year (see table 2 in the appendix). In 2004/05 fiscal year, the tax to GDP ratio again dropped to 12.7 per cent (NBE 2004/05). This reveals that the tax to GDP ratio has been fluctuating. Furthermore, it is far below the resources needed to finance even the recurrent expenditures, which were estimated to be 24.8 per cent of GDP in 2002/03 (AfDB/OECD 2004; NBE 2003/04) evidencing the gap between the revenue and expenditure requirements of the government.