Why VAT?
1.1 Introduction
In the past 30 years, 45 out of the 54 countries on the African continent1 have introduced the value-added tax (hereinafter ‘VAT’)2 as their main broad-based consumption tax. In doing so, they followed a worldwide trend that started in the late 1960s in the European Union (EU) and South America and spread from there across the rest of the world (Cnossen, 1998a; Ebrill et al., 2001).3 Most African countries introduced a VAT around the turn of the century, but some have done so more recently. The DR Congo, the Seychelles, and Swaziland converted their sales taxes into a VAT in 2012. The Gambia followed their example in 2013, making it the latest member of the African VAT club. Ghana is the only African country that ever removed the VAT (two months after its introduction in 1995), only to reinstate it three years later (Terkper, 1996, 2000). Nine countries in Africa, one in six of all countries on the continent, do not (yet) have a VAT.
In theory, a well-designed VAT is the consumption tax of choice, because it is highly successful in freeing exports from tax and treating imports on a par with domestically produced goods and services through appropriate border tax adjustments (imports taxed, exports free of tax). This neutrality feature is important for the proper functioning of trade within regional economic development communities, which purport to abolish (implicit) discriminatory taxes on imports and subsidies on exports. All countries in Africa belong to one of eight regional economic communities on the continent, distinguished in this book, which attempt to facilitate trade among their members. More generally, the border tax adjustments under VAT are compatible with the tax provisions of the World Trade
1 Not counting Western Sahara, which does not have an institutionalized tax system and whose sovereignty is disputed. The United Nations does not recognize Morocco’s claim on the country or the self-proclaimed Sahrawi Arab Democratic Republic, although the latter is a member of the African Union.
2 In some countries, the VAT is called goods and services tax (GST) but, as is well known, in principle the GST and VAT are identical taxes. In references, this book follows the particular country’s usage, whether VAT or GST.
3 The VAT was pioneered by Siemens (1920) in Germany who called it the ‘improved turnover tax’ (as reflected in its multistage nature) and Adams (1921) in the USA who referred to it as a tax on ‘modified gross income’ (in other words, gross profits, the difference between sales and purchases), but was improved by Lauré (1953, 1957), called ‘le père de la TVA’ (taxe à la valeur ajoutée), in France. An early account of the history of the VAT can be found in Sullivan (1965). Recently, James (2015) provided a detailed account of the worldwide rise of the VAT. There is some confusion about the exact year of Siemens’s proposal, but Pohmer (1983) and Terra and Kajus (2007) point towards 1919.
Organization (WTO), which seeks to replace protective import duties by general consumption taxes, such as the VAT, in an effort to promote regional and global trade (Ebrill, Stotsky, and Gropp, 1999).
In addition to being neutral with respect to foreign trade, a well-designed and well-administered VAT does not distort domestic production and distribution. Thus, under the VAT, it makes no difference to the tax liability how often a product is traded before it reaches the consumer, whether its value is added early rather than late in the production–distribution process, whether the product is manufactured with capital- or labour-intensive technology, or whether a taxable business is incorporated or not. Obviously, these features are important attributes of a ‘good’ consumption tax in economies that leave the optimal allocation of resources to the free play of market forces. Beyond that, the VAT is a productive, stable, and flexible source of government revenue. Since the VAT is collected on a current basis—say, monthly—its revenue-generating capacity is not negatively affected by inflation and the effect of rate changes on revenue is immediately visible.
This is the theory. In practice, a badly designed and/or badly administered VAT may turn into a highly non-neutral sales tax, possibly worse than the taxes it replaced. If refunds with respect to exports are not made or are not timely, the VAT becomes a tax on exporters, hampering investment. Similarly, if the VAT on imports cannot be credited forthwith against the VAT on domestic sales (if necessary, involving a refund if the VAT on imports exceeds the VAT on domestic supplies), the VAT becomes a tax on imports. More generally, a less than full and immediate tax credit with respect to inputs (including capital goods) turns the VAT into a cumulative turnover tax, which distorts consumer and producer choices and provides an incentive for the non-economic integration of business activities. Moreover, a VAT that effectively functions as a turnover tax tends to make the tax burden distribution more regressive, because the value of essential products, disproportionately consumed by the poor, is usually added early in the production–distribution process. Hence, the VAT on them is subject to more cascading (tax-on-tax effects) than the VAT on luxury goods and services.
1.2 VAT’s Role in Africa’s Tax Systems
Africa is often called the most promising continent of the twenty-first century. Living standards are expected to rise considerably in tandem with sustained economic growth.4 But this is unlikely to come about without additional publicly financed investment in education, healthcare, public utilities, and transportation systems. Unfortunately, most government budgets are too small to make this
4 In a cautious assessment, Rodrik (2014) expects moderate and steady growth of perhaps as much as 2% per capita. He believes that the traditional engines behind rapid growth—structural change and industrialization—still seem to be operating at less than full power.
possible. In about half of all countries on the continent, tax revenue as a percentage of GDP (called the tax ratio) is around 15 per cent or less, which is not enough to finance more human and economic development. Until recently, financial aid from developed countries could be used to finance public investment and shortfalls in current government budgets, but this source has largely dried up after the 2007–08 recession in the industrialized world.
This means that domestic tax revenue mobilization is called for if economic development is to proceed (also Bird and Gendron (2007) and International Monetary Fund (2017)). Whether or not tax ratios can be increased depends on each country’s tax capacity—i.e. the tax revenue (expressed as a percentage of GDP) that can be raised in light of the country’s economic and institutional environment. To the extent that a country is not fully realizing its tax capacity, its tax effort—i.e. the actual tax ratio as a fraction of tax capacity—is less than 1. The difference between 1 and the tax effort therefore indicates how much more a country can do to mobilize domestic tax revenue.
The tax capacities and tax efforts of several African (and other) countries have been calculated by Fenochietto and Pessino (2013), who use an econometric model to build a ‘stochastic tax frontier’ for panel data from 25 African (and 88 other) countries covering the period 1991–2012. Country-specific demographic, economic, and institutional characteristics that may change over time are taken into account. Table 1.1 shows the results for 25 African countries that have a VAT. Guesstimates have been added for the 20 countries not covered by Fenochietto and Pessino, derived on the basis of the same variables they use, as to whether each country’s tax effort is likely to be low, medium, or high (see Cnossen (2015)).
In all, 17 out of 45 countries with a VAT can be classified as low-tax-effort countries, meaning that they collect two-thirds or less of the revenue that they can be expected to raise in view of their institutional and economic circumstances. In 20 other countries, the tax effort is labelled medium: the gap between tax effort and tax capacity is one-third to one-fifth of tax capacity. Eight countries exhibit a high tax effort.5 Malawi, one of the poorest countries in Africa, with a per-capita income of US$1,100 in 2016 (purchasing power parity, PPP), has a high tax effort. Egypt, a middle-income country by African standards, with a percapita income of US$11,100 in the same year, has a low tax effort.
The difference between the tax ratio and tax capacity represents the tax policy gap (exemptions and concessional rates), among other factors, rather than the compliance gap (inadequate enforcement and evasion), which is reflected in each country’s proxies for the economic and institutional environment. This book focuses mainly on the policy gap of the VAT, which is often the most important source of tax revenue in African countries. As a revenue-raising instrument, the
5 The high tax effort in Lesotho, Namibia, and Swaziland can partly be explained by the fact that the South African Revenue Service (SARS) collects the import and excise duties for these countries under the umbrella of the Southern African Customs Union (SACU).
Table 1.1 Africa: tax ratios, tax capacity, and tax effort in VAT countries
Country Tax ratio (%) Tax capacity (%) Tax effort Other countries whose tax effort is likely to be low/medium/high
Low tax effort (0.67 or lower)
Guinea–Bissau9.027.40.33
Egypt
Algeria
16.735.90.46
16.836.10.47
Cameroon 12.824.40.52
Ghana 16.932.10.52
Gambia 12.321.10.58
Tanzania 15.326.00.59
Madagascar10.817.00.63
Uganda 12.419.10.65
Ethiopia 11.317.00.66
Burkina Faso 14.121.00.67
Medium tax effort (0.68–0.80)
Congo
27.238.50.71
Niger 13.518.80.72
South Africa
27.838.20.73
Tunisia 25.534.60.74
Senegal 19.426.00.75
Kenya 20.727.10.76
Mali 14.418.80.77
Guinea 14.818.90.78
Togo 15.920.10.79
Morocco 24.330.40.80
High tax effort (0.81 or higher)
Mozambique18.221.40.85
Central African Republic, Chad, DR Congo, Rwanda, Sierra Leone, Sudan
Benin, Botswana, Burundi, Cabo Verde, Côte d’Ivoire, Djibouti, Equatorial Guinea, Gabon, Mauritania, Mauritius
Lesotho, Seychelles, Swaziland, Zimbabwe Namibia 25.327.80.91
Malawi 23.323.80.98
Zambia 16.617.00.98
Note: Countries are ranked in ascending order of tax effort.
Source: Fenochietto and Pessino (2013, appendix 2) and the author’s assessment based on the economic and institutional indicators used by Fenochietto and Pessino.
VAT is generally regarded as an easier tax handle and less detrimental to economic growth than the income tax.6 VAT revenue enhancement should also make it possible to replace the loss of revenue from import duties, which have to be phased out if closer economic integration and outward orientation, which promote growth, are to be pursued in line with WTO guidelines.
In many African countries, the VAT does not live up to its revenue-raising potential. When introducing the VAT, most countries exempted or zero-rated a wide range of consumer goods, so much so that the so-called standard rate may
6 Empirical evidence collected by Arnold et al. (2011), for instance, suggests that income taxes reduce the rate of economic growth more than consumption taxes do.
be labelled a luxury goods rate, because it is imposed on goods that are mostly consumed by higher-income groups.7 What should be the ‘actual’ standard rate is in fact a set of varying effective rates on exempt consumer goods that differ depending upon the ratio of the tax on inputs, which are used to produce these goods, to price. These rates are capricious and indeterminate; their incidence cannot be determined and the tax on inputs may enter into the price of goods traded across borders.
The wide-ranging non-standard exemptions (so-called because they go beyond the standard exemptions found in the EU—which are quite comprehensive themselves) and zero rates on domestically consumed goods greatly erode the VAT base and hence revenue, and contribute little to a more equitable VAT burden distribution. These exemptions imply that many VATs can be characterized as a combination of import duties (since some two-thirds of most VATs is collected at borders) and excise taxes (since most of the remaining VAT is collected on excisable products).
1.3 Major VAT Policy Issues
This situation calls for reform, principally by proceeding from the notion that the VAT is primarily intended to raise revenue, predictably and efficiently. It produces revenue that grows as its base—consumption expenditures—expands with economic development. It differs from excise taxes in that it does not and should not change people’s behaviour (relative prices should remain unchanged); it differs from the import duties in that it should not be used to support trade policy; and it differs from the income tax in that it cannot and should not be used to redistribute income or stimulate industry through investment incentives. To repeat, under VAT, the goal should be revenue, which can be used, of course, to finance programmes, such as education and basic healthcare, that benefit those who pay the VAT.8
The road towards a best-practice VAT calls for second-generation reforms of African VAT systems aimed at modernizing the tax by broadening the base. Current exemptions for most foodstuffs, public utilities, pharmaceutical products, building materials, agricultural and industrial inputs, and various other goods and services should be removed. Only unprocessed foodstuffs, restrictively defined, might be exempted, although this should not be presumed to be
7 As is well known, if goods and services are exempted, registered businesses buying these goods or services cannot take credit for the VAT charged on their inputs, nor charge VAT on their sales. By contrast, zero-rating means that no VAT is charged on sales while input VAT is creditable (deductible) and refundable if the amount exceeds the tax on output. In various African countries, exempt goods and services are called exemptions without credit, and zero-rated goods and services are called exemptions with credit.
8 Of course, this assumes that the revenue is indeed used to pursue these goals and not used for unproductive purposes. In many instances, budget expenditure controls and audits appear to fall short of minimum standards of probity, but this issue is not the topic of this book.
necessary in all cases. At the same time, transfer duty, registration duty, and stamp duty regimes should be re-examined with a view to abolition or incorporating them in the VAT system, because they resemble distortionary cascade types of turnover taxes, which the VAT seeks to replace. Generally, these regimes yield very little revenue, with the possible exception of transfer or registration duties on immovable property. Traditional excise goods—tobacco products, alcoholic beverages, petroleum products—should continue to be taxed for revenue and, above all, externality-correcting reasons. The case for retaining the excise taxes on goods considered as items of luxury consumption is weak, with the notable exception of passenger cars.
Although less urgent, a review of the VAT systems should also be used to take a hard look at ‘standard’ exemptions. Many government and quasi-governmental operations should be made taxable, if only to confront the users of government services with the full social cost of their production and to enhance the integrity and stability of the VAT. Further, immovable property transactions, except sales of used residential dwellings, can be made more fully taxable. The separate taxes on insurance and banking activities, found in several francophone countries, should be reviewed on account of their distortionary impact. Property and casualty insurance can be taxed under the VAT, as can fee-based banking services and gambling. To trim the tax roll of revenue-unproductive VAT payers, the registration threshold should be raised in some countries. The case for a presumptive turnover tax on small VAT-exempt business establishments is not strong.
Design changes should make it easier to administer the VAT. The abolition of exemptions for goods, services, and entities and, more generally, unnecessary design differentiations would reduce the weight and sway of legal opinions and increase the administrative resources available for monitoring compliance with the VAT. Revenue should benefit. Indeed, the VAT design reforms should go hand-in-hand with VAT administration reforms. Even without the design changes, it would be highly worthwhile to make an in-depth evaluation of VAT administrations in selected African countries to see whether the VAT on the statutes is actually the same VAT in practice, i.e. a tax on value added rather than on turnover (if no credit for the VAT on inputs is allowed), exports (because refunds of prior-stage tax are not forthcoming), or imports (again, because no credit is given for the VAT collected at the import stage). Changes in VAT design cannot achieve much if inefficient and wasteful, indeed sometimes counterproductive, administrative processes are left in place.
1.4 Organization of This Book
The book is organized as follows. Chapter 2 dwells briefly on the workings and the legal and economic nature of the VAT and enumerates the characteristics of
a best-practice VAT, against which existing regimes are evaluated. Chapter 3 reviews other broad-based consumption taxes with which the VAT can be compared, especially the retail sales tax (RST). After these introductory chapters, Chapter 4 provides an overview of VATs in Africa, noting their broad characteristics and listing the general issues that arise. Logically, this is followed by Chapter 5, which highlights the shortcomings of the sales tax regimes in African countries that do not (yet) have a VAT and which can likely be resolved under a best-practice VAT.
Since the book is about mobilizing VAT revenues, Chapter 6 examines the revenue performance of the various regimes by calculating their collection efficiencies, which make a distinction between the policy gap (exemptions and reduced rates) and the compliance gap (evasion and lack of enforcement). Possible explanatory factors, such as the level of economic development, the VAT structure, and the effectiveness of the VAT administration, are reviewed.
Poor revenue performance can be attributed to, among other things, the many non-standard exemptions introduced to mitigate the (perceived) regressivity of the VAT. It is widely believed that lower-income groups actually pay more tax as a percentage of income than higher-income groups. Chapter 7 examines the validity of this claim regarding the exemption or zero-rating of food products and agricultural inputs on the basis of empirical evidence, mainly from South Africa. It also discusses the pros and cons of higher-than-standard rates or, alternatively, excise taxes on goods and services considered items of luxury consumption, in order to promote the progressivity of VAT regimes.
VATs have been introduced, among others, to reduce or eliminate the reliance on taxes on international trade. Most African countries have established regional economic communities to promote trade and investment between the participating member states. Chapter 8 reviews the various stages of economic integration that can be distinguished and discusses VAT (and excise tax) coordination in the presence and absence of border controls.
The analysis of the revenue performance of the African VATs and their actual burden distribution establishes the case for taking a closer look at the various tax base concessions, which can truly be called the Achilles heel of African VATs. Chapter 9 opens the discussion by drawing up a taxonomy of standard and nonstandard exclusions, exemptions, and zero rates—highlighting the distortions caused by them and the administrative complexities to which they give rise. Broadly, it compares the treatment of various exemptions in the EU, which have also been adopted by most African countries, with their treatment under New Zealand’s GST and South Africa’s VAT. This is followed, in Chapter 10, by a detailed review and analysis of the manifold non-standard exemptions and zero rates found around the continent. This chapter also makes the case for the comprehensive inclusion of services in the VAT base. Appropriately, it describes the reforms that Kenya and Tanzania, and before them Benin and Senegal, have
undertaken to do away with most non-standard exemptions in order to bring their VATs closer to a best-practice type of tax.
Subsequent chapters dig deeper into the treatment of difficult-to-tax sectors, including governments and their agencies, education and healthcare, immovable property, financial services, insurance, and gambling. These ‘standard exemptions’, as they are often called, are generally based on the EU’s Common VAT Directive (2006) (previously called the Sixth VAT Directive). This Directive, originally issued in 1977 and not significantly changed since, does not meet the requirements of a modern best-practice type of VAT adopted by countries such as Australia, Canada, New Zealand, Singapore, and South Africa (along with the other members of the Southern African Customs Union (SACU): Botswana, Lesotho, Namibia, and Swaziland).
Chapter 11 on governments and activities in the public interest includes a discussion of the VAT treatment of public utilities and cultural goods. Chapter 12 on immovable property shows that the application of the VAT to newly created property is, in theory, equivalent to the VAT on rents and rental values generated during the lifetime of the property. The similarities and differences between the VAT and property transfer taxes are also reviewed. As shown in Chapter 13, the proper inclusion of financial services in the VAT base still eludes the grip of VAT experts. However, various substitute treatments found in South Africa and New Zealand, for instance, are worth considering. Chapter 14 shows that insurance can be included in the VAT base by taxing premiums and imputing a VAT credit to indemnifications. The equivalent applies to gambling, briefly discussed in Chapter 15.
Chapter 16 discusses VAT registration thresholds and the treatment of farmers. A high threshold is recommended to keep the number of taxable persons manageable. The VAT on exempt small businesses should be confined to the tax on their inputs. Additional, substitute assessment schemes are not favoured. Farmers are usually categorically excluded from the VAT for political reasons and because they do not keep adequate accounts. Circumstances permitting, they should be taxed subject to a (high) registration threshold. Farm inputs that have no alternative use outside the agricultural sector might be zero-rated to mitigate the effects of the VAT on the prices of unprocessed foodstuffs.
Tax administration is key to tax policy, particularly in developing countries. This book would not have been complete, therefore, without a discussion of basic administrative processes in Chapter 17. These processes comprise registration, filing, payment, collection, enforcement, and audit. The chapter concludes with a note on fraud, a major issue in African countries.
Chapter 18 concludes the book with a summary review of the findings and preferred choices on exclusions, exemptions, and zero rates discussed in the previous chapters. It contrasts the present situation in most African countries with a workable best-practice VAT. The ideal does not have to be attained overnight, but
should not be lost sight of since it tends to indicate the type of reform measures, often second-best, that are in line with the ideal. Throughout, it should be kept in mind that VATs are revenue-generating taxes, as opposed to progressivityenhancing taxes, such as the income and property taxes, or externality-correcting taxes in the form of excise taxes and environmental levies.
Basic information on the VAT regimes in African countries can be found in Appendix A on the VAT treatment of essential goods and services and in Appendix B on the VAT and excise tax treatment of luxury products. Further, Appendix C shows details of the sales tax and excise tax systems in African countries without a VAT. Appendix D provides background information by presenting some basic economic and institutional indicators for African countries with a VAT. Knowledge of all these details is essential in evaluating the VAT regimes and drawing up the comparative tables in the text. Although every effort has been made to collect the latest data and to verify their accuracy, no doubt some information is out-of-date or possibly incorrect. In many instances, moreover, the (non-)taxable status of a product or service had to be inferred from the fact that it was not mentioned in the legal provision or the schedule of the VAT act that lists the exemptions or the goods and services subject to a lower-than-standard or zero rate.
1.5 Acknowledgements
Best practice elsewhere has been the lodestar for the review. Various publications have analysed and described VAT practice around the world, some with particular reference to Africa. These publications, listed in Box 1.1, have been valuable sources of reference, not least because of the literature cited in them. This book differs from these publications in that it digs deeper into the details of VAT design on the African continent, which are relevant for the discussion of various policy issues.
The book proceeds from the premise that VATs (and other taxes) in Africa (and elsewhere), as works in progress, should be scrutinized continuously and improved as experience is gained and economic development proceeds. I am inspired by Richard Bird and Pierre-Pascal Gendron’s motto that proper tax design makes good tax administration easier and by Nora Lustig’s judgement that efficient VATs can result in a net fiscal system that is equalizing. At the same time, I am tempered by Michael Keen’s reality check that mistakes made at introduction are hard to undo. With these well-grounded reasonings in mind, I hope that this book contributes to a growing awareness of the need for more neutral, simpler, and more revenue-productive VATs reformed in line with commonly agreed notions about a best-practice VAT, suitably adapted to the circumstances of each country. Recent regional conferences and workshops on VAT across the African continent indicate that the VAT is starting to receive the limelight it deserves.
