The South African Tax System: A Nation in Microcosm

Henry J. Aaron and
Henry J. Aaron The Bruce and Virginia MacLaury Chair, Senior Fellow Emeritus - Economic Studies

Joel Slemrod

December 6, 1999

We were two of nine foreigners invited by the Department of Finance of South Africa to participate in a weeklong workshop to advance a lengthy tax reform process initiated by South Africa’s first democratically elected government shortly after it took power. We believe it would take unusual insensitivity—even callousness—not to become emotionally engaged as well as intellectually committed to the success of this effort to redress, without rancor or recrimination, the consequences of decades of systematic injustice.

All developing countries face difficult problems in trying to marshal limited resources to promote economic growth. Few face a combination of problems as challenging as those confronting South Africa. Economic constraints are conjoined with the political legacy of apartheid. The democratically elected governments that took office in 1994 and 1999 have managed, with quite limited resources, to preserve a remarkable measure of political cohesion and to bring South Africa relatively unscathed through the Asian economic turmoil that threatened to spread to other developing nations. While it will take more than a little good luck to surmount the challenges South Africa faces, current auguries are hopeful.

Income inequality in South Africa is extreme. The standard index of inequality, the Gini coefficient, is 0.58, higher than that of any other country (with the possible exception of Brazil and Colombia) and is exceeded only by the Gini coefficient of the world as a whole. The reality of inequality is palpable. Squalid housing—corrugated iron shacks with neither water nor electricity and single-room dormitories, built for solitary miners separated from their families and now crowded by families of 6 or 8 or 10—exists in townships that border cities and suburbs as affluent as any in the United States. Various townships in the Cape flats, teeming with hundreds of thousands of squatters, border Cape Town, a city distinguished by wealth and beauty. Soweto (SOuthWEst TOwnship), which houses four million people and rivals its neighbor Johannesburg in size, contains a mixture of shacks, simple concrete houses, and spacious dwellings luxurious by any standard.

Crime shadows the lives of everyone—white, black, or colored (the South African term for anyone who is of mixed race or is neither white nor black). It inhibits personal movement and business locational decisions. An outsider is first shocked, then numbed, by the pervasive normality of barred first-floor windows and doors and razor-wire topped walls.

The South African economy is just beginning to recover from its episode of the Asian flu. That economic contagion, which swept through what once were known as the Asian Tigers, created concern that South Africa would suffer a similar fate. To prevent currency collapse, the South African central bank temporarily boosted real interest rates to 15 percent or more. The sustained economic stagnation caused by the interest rate spike has lasted since mid-1997. Protecting the currency doubtless spared capital owners sizeable losses, but the toll on labor has been enormous—unemployment is reported to run as high as 30 percent.

The gradual relaxation of exchange controls will move South Africa into the ranks of small, open economies. The risk of financial capital flight is serious. In 1976, the Soweto riots intensified doubts regarding the sustainability of the old apartheid regime. Since 1994, some investors have been concerned that a democratically elected government might prove inhospitable to capital. South Africa has long suffered from a brain drain, at first because opponents of apartheid fled, then because whites feared loss of position under the popularly elected government, and more generally because the thin layer of highly trained professionals can often find higher wages abroad. The risk of both financial and human capital flight seriously constrains economic policy to alleviate inequality, in general, and tax policy, in particular. Immigration of predominantly poor people from African countries with less economic opportunity is also becoming a major issue, adding to the burden of uplifting the previously disadvantaged communities.

Tax Reform: Background

Under apartheid there were five different tax administrations, including one for each of the homeland “states” of Transkei, Bophuthatswana, Venda, and Ciskei. The post-1994 government abolished the homelands and divided the nation into nine provinces. Most revenue-raising authority resides with the national government, but the provinces have considerable authority over public expenditures. Because South Africa is a large and economically, socially, and ethnically diverse nation, the management of fiscal federalism will pose difficult challenges in the future. Currently, however, the principal tax policy issues concern whether and how to redesign the national tax system.

Although the democratically elected government has modestly changed the tax laws, the current tax regime is not fundamentally different from the system inherited from the apartheid years. Revenues generated by the principal taxes are shown in Table 1. Total tax collections in fiscal year 1999 equaled just over 27 percent of GDP, but the tax share is trending down and the government has pledged to reduce it to 25 percent. The personal and corporation income taxes and a broad-based value added tax are the principal revenue raisers. Mining industry taxes, once among the most important sources of government revenue, have declined to less than 1 percent of revenue, as falling prices for gold and other minerals and very generous tax preferences for the mining industry have reduced taxable profits.

Table 1: South African Revenues, 1999-2000

Revenue Source Revenue as Percent of
  Gross Domestic Product Total Revenue



Personal Income Tax 11.7 43.1
Company Tax 3.0 11.2
Value Added Tax 6.7 24.6
Excises 3.4 12.5
Mining Taxes 0.3 1.0
Other Revenue 2.1 7.6

* detail does not add to total because of rounding
Source: Republic of South Africa, Department of Finance, Budget Review, 1999, Pretoria, February 17, 1999.

South Africa’s tax challenges arise from forces familiar to tax policymakers around the world. The South African Revenue Service’s (SARS) list of the six main threats to the personal income tax included fringe benefits, independent contractors, and the complexity of tax legislation. Local conditions—notably, extreme economic inequality, the political imperative to do something about it, and highly mobile capital and labor—complicate efforts to deal with these and other problems. In the early post-apartheid years, the principal problem for the tax authorities was loss of staff, which raised doubts regarding the capacity to maintain revenues. Many tax officials resigned and others preferred severance packages to continued service under the new government. Under apartheid, few blacks or coloreds paid income taxes, and none had any significant role in governing. Most blacks and coloreds worked in the informal economy, and those working in the formal economy typically had incomes below levels at which personal income tax was due. Although the racial balance of political power is no longer as skewed as it once was, economic power remains about as concentrated in 1999 as it was before 1994.

Faced with the desperate situation of the revenue service, with populist calls for redistribution or even expropriation, and with the evident fact that the separate income tax schedules for men and women violated the new Constitution’s prohibition against discrimination based on sex, the government elected in 1994 announced the appointment of a seven-member commission to study the South African tax system and to make recommendations for reform. Officially known as the Commission of Inquiry into Certain Aspects of the Tax Structure of South Africa, it soon became known as the Katz Commission, after its chairman, Michael Katz, a private corporate lawyer. The commission issued nine interim reports from November 1994 through February 1999. Two more reports are expected.

The tax workshop convened in Irene, a quiet town in a farm region between Pretoria and Johannesburg that is being relentlessly absorbed into an emerging megalopolis. The texts for the symposium consisted of the nine interim reports of the Katz Commission; background papers prepared by the Tax Policy Chief Directorate in the Budget Office of the Department of Finance, SARS, and Samuel Thompson, a University of Miami professor of law, on leave in South Africa as part of the U.S. Treasury Department’s Tax Advisory Program, and papers prepared by each of the nine invited foreigners.

In the audience were several members of the Katz Commission, other staff of the Tax Policy Chief Directorate and SARS, professors of economics, accountancy, and law from various South African universities, and representatives from law and accounting firms drawn largely from previously disadvantaged groups. The commissioner of SARS and the deputy director general of the Budget Office attended much of the workshop and the newly appointed deputy minister of finance attended nearly all of it. Minister of Finance Trevor Manuel came for closing speeches. Compared to similar meetings in the United States, attendance was much more diverse, as it included both high government officials, for whom tax policy is a matter of great political sensitivity, and private sector executives and professionals, who will be directly affected by any policy changes.

The Current Tax System

Because income inequality is extreme, all personal income tax and most revenue is collected from a small proportion of the population. The yield of the current system—more than one-fourth of GDP—means that tax policymakers must reform a vigorously functioning tax system and not design a new system, starting from a blank slate. As a result, they must take careful account of existing tax rules, contractual obligations, and transition problems.

The initial conditions are stark, indeed, and sometimes unexpected. The exodus of trained staff posed the initial challenge, and measures to deal with the resulting problems were prominent in the Katz Commission’s first interim report. Because the black and colored population saw the value added tax, introduced in apartheid’s dying days, as an attempt by whites to shift the tax burden onto them, raising the VAT rate or reducing the number of zero-rated commodities has a high political cost. Smuggling is an especially large problem, in part because it was a matter of survival both for the African National Congress in its fight against apartheid and critical for the apartheid government itself in its efforts to evade international sanctions, and in part because South African excise taxes on alcohol and tobacco products are high compared to those of neighboring countries.

The Personal Income Tax

The personal tax is levied at steeply progressive rates on domestic-source income of individual filing units. For the 1999-2000 tax year, taxable incomes face steeply graduated rates, starting at 19 percent on the first ZAR 33,000 of taxable income [ZAR 6 = US $1], then 30 percent on income up to ZAR 50,000, and topping out at 45 percent beginning at ZAR 120,000 ($20,000). A nonrefundable rebate of ZAR 3,710 (ZAR 6,485 for those 65 or older) lowers liabilities. The tax liability of someone under 65 with ZAR 50,000 of taxable income, for example, is ZAR 7,660 (19 percent of ZAR 33,000 plus 30 percent of ZAR 17,000 minus ZAR 3,710), or an average rate of 15.32 percent. A variety of deductions and exclusions, which are of particular value to middle- and high-earners who may negotiate structured compensation packages, somewhat ameliorates these rates. Despite the low levels of exempt income, only 5.8 million people, or about 23 percent of the adult population, filed returns in 1998, and fewer than one million had incomes of ZAR 80,000 (US $13,333) or higher. The most affluent 20 percent of the population receive 74.3 percent of household income and pay 94.3 percent of personal income tax.

The tax base includes all earnings, minus most contributions to retirement funds, plus interest in excess of ZAR 2,000. Dividends and capital gains are exempt at the household level. No deductions are allowed for mortgage interest, property taxes, or charitable contributions, other than limited gifts to educational institutions. People whose income is substantially from earnings and whose income is less than ZAR 60,000 (SARS has proposed to increase the threshold to ZAR 80,000) are subject to SITE—the Standard Income Tax on Earnings—a system of exact and final withholding collected at the workplace and transferred to SARS by employers. Individuals need not file returns. Two-thirds of income taxpayers are on this system.

Retirement Funds

South Africa has no system of contributory social insurance. Retirees receive support from an income-tested flat benefit, one or more of a variety of tax-sheltered retirement funds to which individuals or their employers may contribute, or ordinary saving. Deposits (except to provident funds) are deductible and there is a 25 percent tax on gross interest and net rental income accruing to retirement funds. Annuities are taxed as ordinary income, but people may also take lump-sum withdrawals, part of which are entirely exempt (ZAR 4,500 times the number of years of fund membership or ZAR 120,000, whichever is lower) and most of which is taxed at reduced rates. The flat benefit, which is nationally uniform, is reported to be generous for rural residents, but meager for urban residents.

The Corporation Income Tax

The corporation income tax rate was lowered from 35 to 30 percent in 1999. Dividends face an additional tax of 12.5 percent, the secondary tax on corporations. In combination with the exemption of dividends and capital gains from personal income tax, this system comprises a partially integrated system. As a general rule, South Africa taxes both inbound and outbound transactions on a source basis. Tax holidays from an earlier era have now been abolished, and there is no investment tax credit. Although inflation has been fairly high, averaging about 8 percent in the mid-1990s, there is no statutory indexing in the tax system.

The Value Added Tax

The South African VAT resembles the New Zealand model. It is a broad-based levy of 14 percent on goods and services, including government and financial services; 19 basic foodstuffs that account for about 10 percent of the base are zero-rated, as are exports. Exempt commodities include items consumed disproportionately by the poor, including corn (mealie) meal, rice, milk, eggs, vegetables, and fruit, but the difference across income classes in the proportion of consumption on zero-rated goods is small. The gains in administrative simplicity from eliminating zero-rating would be considerable. In addition, if the added revenues were spent roughly in proportion to other public spending, the poor collectively would gain far more from the added public services than they would lose from added tax burdens. Nonetheless, groups representing low-income earners and others fiercely resist curtailing or eliminating zero-rating.

Other Revenue Sources

Aside from the personal and corporate income taxes and the value added tax, South Africa raises money from a variety of customs and excise taxes. These taxes yield a bit over 10 percent of total revenue. South Africa also imposes an estate tax at a flat rate of 25 percent on estates valued at more than ZAR 1,000,000, with relief for assets transferred twice within a 10-year period, and a donations tax on inter vivos transfers of wealth. Transfers of ZAR 25,000 per year per donor per donee are exempt. The tax appears to be easily avoided through mechanisms U.S. estate planners would easily recognize, and avoidance seems to be unencumbered by restrictions enacted in the United States in the past two decades. The fact that the estate tax yields only one-eighth of a percent of total tax revenue attests to the ease with which it is avoided.

As a country rich in diamonds, gold, and other natural resources, South Africa has at various times derived considerable revenue—as much as 10 percent of total tax revenue in the early 1980s—from taxation of extractive industries, despite extremely generous concessions. Revenue from these taxes has collapsed to less than 1 percent of total tax revenue as the price of gold and, to a lesser extent, other natural resources has fallen.

The Interim Reports of the Katz Commission

The Katz Commission reports reflect a tremendous amount of work and embody sophisticated analyses of the legal and economic aspects of taxation. These reports compare favorably with the classic comprehensive reviews of taxation such as the Meade Report in the United Kingdom, the Carter Commission Report in Canada, the U.S. Treasury’s Blueprints for Basic Tax Reform, and the three-volume 1984 report that led up to the Tax Reform Act of 1986.

The most striking shortcoming of the Katz Commission reports, compared to those prepared abroad, is the relative lack of quantitative underpinning. Compared to reports of the U.S. Treasury Department, Congressional Budget Office, or Joint Committee on Taxation, the table-to-text ratio in the Katz Commission reports is minuscule and the infrequency of quantitative evidence to buttress arguments or even to illustrate magnitudes is striking. The government has limited capacity in revenue estimation and modeling— it lacks the staff to do microdata-based analyses of either the corporation income or the value added tax—but it is currently in the process of significantly upgrading its capabilities. This state of affairs exists partly because resources are limited and other priorities have crowded out analysis. In addition, the relationship between the Tax Policy Chief Directorate in the Department of Finance and SARS is not well designed to encourage the collection and exchange of data. This relationship is bureaucratically sensitive but is still evolving.

The Katz Commission has avoided calls for fundamental reform, such as replacement of the income tax with a consumption tax or dramatic shifts in tax rates. Nonetheless, it has developed hundreds of recommendations, ranging from technical fixes to proposals for fairly major structural changes. Some recommendations have already been implemented. Others have been shelved for a variety of reasons. Scores of proposals remained on the table as Thabo Mbeki became president of the second democratically elected government in June, 1999.

The Workshop

As a prelude to the final reports of the Katz Commission, the Tax Policy Chief Directorate (with support from the U.S. Agency for International Development) sought the views of tax analysts from abroad on the major tax policy issues facing South Africa. The panel of experts was not asked to make recommendations for policy reform, although each member made closing comments containing more or less specific suggestions about tax policy. Several common themes emerged, although nothing of what follows was formally approved by the group or endorsed by the South African tax authorities.

We began where the Katz Commission also started—with administration. The overriding principle in the short run, we agreed, should be to keep tax rules as simple as possible, given the demands of modern commerce. In the medium and long term, a major effort should be made to build SARS’ quite limited administrative capacity. The problem runs deeper than legislative drafting and bureaucratic capacity, however. Much of the economy operates without formal accounts. A sizeable fraction of the population is illiterate. These conditions currently preclude effective and equitable operation of complicated tax rules. Increasing literacy and the spread of modern accounting will come in tandem with economic development. Meanwhile, the advisers emphasized the importance of introducing a system of taxpayer identification numbers and of increased funding for the South African Revenue Service.

There was less consensus about structural reform of the tax system, although most advisers counseled South Africa to keep the single-rate character of the VAT, hold the line on or pare back the number of zero-rated items, and subject to the very sensitive political constraints, consider raising the rate slightly. Some advisers urged consideration of replacing the multiple-rate (six brackets) personal income tax with a two-bracket individual income tax schedule, in which the top rate is close to the single corporation income tax rate and possibly also to the rate of tax on pension buildup. Because of the threat of flight by capital and skilled labor and the pressures of international and especially regional tax competition, no one advocated increasing the top corporate rate or the top personal rate. Professor Sijbren Cnossen demurred, favoring instead the dual income tax now used in Scandinavia. Under this system, labor income is subject to a graduated tax and capital and business income are subject to a separate flat tax. Agreement was unanimous that South Africa should avoid special tax preferences for favored sectors and tax holidays.

Many of the advisers expressed concerns about the narrowness of the tax base, both at the top and bottom of the income distribution. The absence of a capital gains tax provides huge opportunities to avoid tax by characterizing labor or business income as capital gains. Without better data than are now available in South Africa, it is impossible to know how much income is converted to untaxed capital gains and to what extent such avoidance reduces effective tax rates on upper-income households. Although most taxes are collected from only a small fraction of upper-income households and total revenues amount to more than one-fourth of GDP, opportunities for tax avoidance, as opposed to outright evasion, are limited for low- and moderate- income earners and are plentiful for recipients of capital income and high-earners who are in a position to arrange compensation consisting of undertaxed fringe benefits. Widespread avoidance not only produces obvious distributional consequences but also threatens taxpayer morale by creating a sense that the affluent can escape the burden of taxation.

Moving more small businesses into the tax net is another critical issue. Businesses with annual turnover of less that ZAR 150,000 need not register as VAT vendors, and most such businesses do not remit any income tax, either. Professor Amaresh Bagchi described the Indian experience with presumptive taxation, and several of us supported consideration of a system under which a presumptive tax liability based on easily observed inputs is assessed in the absence of business records that establish true income. Ideally, this approach provides an administratively feasible way to expand the tax net. Nonetheless, some South Africans complain that such presumptive taxation would be unfair because it would expand the tax net only now, when business opportunities historically denied to blacks are open to everyone.

The pressure of regional and international tax competition lay behind the consensus that taxes on mobile capital should be kept fairly low. Nonetheless, the foreign consultants agreed that South Africa needs to strengthen its defenses against transfer pricing and other mechanisms to shift taxable income out of the Treasury, although they disagreed on how this goal should be accomplished within the territorial system now in place. For this reason and others, most of the advisers suggested that South Africa consider moving toward consolidated taxation of corporations and their subsidiaries; currently, each reports separately.

Closing Perspectives

The tax policy challenges facing South Africa are intellectually fascinating, but finding a way to meet them is much more than interesting: it is vital to the success of South Africa’s efforts to sustain a multiracial democracy. The black and colored majority’s forbearance from extracting vengeance for the feral oppression they endured at the hands of the white minority is impressive and humbling. But forbearance cannot endure if the lot of the once- disenfranchised population fails to improve. Given low average incomes, such improvement cannot come in a politically and economically sustainable way from redistribution, but must flow from sustained and broadly based economic growth.

The investments in human capital and public infrastructure to support such growth and the government social supports to make bearable the wait until these investments bear fruit place heavy and conflicting burdens on the government. Raising revenue is vital for the survival of the nation. In its efforts to raise revenues, the South African government faces the pressures of all small, open economies, as well as the deep scars and unhealed wounds of apartheid.

On the plus side, there is much expertise and sophistication in tax matters and great energy and good will. But the stake in success is high and the price of failure is terrifying. At one point during the workshop, we read a news story downloaded from the Internet about the budget maneuverings back in the United States and chuckled over the brouhaha regarding a seemingly trivial controversy. It called to mind a comment made to an American journalist by a South African: “I don’t want to live in an exciting country. I want to live in a nice boring country like yours.”


Goodman, David, Fault Lines, Berkeley: University of California Press, 1999.

Malan, Rian, My Traitor’s Heart: A South African Exile Returns to Face His Country, His Tribe, and His Conscience, Vintage 1991.

Mandela, Nelson, Long Walk to Freedom: The Autobiography of Nelson Mandela, Little Brown & Co., 1995.

McLure, Charles (1999), “Financing Provincial Government in South Africa: Short-Run Solutions That Maintain Long-Run Options,” unpublished (June).

Thompson, Leonard, A History of South Africa, Revised Edition, New Haven: Yale University Press, 1995.

van der Berg, Servaas, “Redirecting Government Expenditure,” in Peter Moll, Nicoli Natrass, and Lieb Loots, eds. Redistribution: Can It Work in South Africa? Cape Town: David Philip, 1991.