Draft notes for a presentation to the Portfolio Committee on Finance
and the Select Committee on Finance
by Professor Rick Krever,
Director, Taxation Law and Policy Research Institute,
Deakin University, Australia
1. Should we be guided by overseas examples or South Africa's needs?
Much of the debate about the taxation of capital gains in South Africa has, to date, drawn on overseas experience. Over the past decades, industrialised countries have steadily been moving in the direction of more comprehensive taxation of capital gains, though in recent years some have lowered the effective rates on capital gains. Some developing countries have moved to tax capital gains as ordinary income and others have moved to relax the taxation of capital gains. Some say that South Africa wishes to be an industrialised power and it should look for the tax system that allows it to operate most effectively in the context of the norms in industrialised states. Others say that South Africa's competitors for foreign direct investment capital are the newly industrialised and developing countries and they should provide its models.
I begin this presentation with a simple caution - the answer to the question "what is the optimal tax treatment of capital gains in South Africa" is not to be found in the tax system of another country. It will be found here, in South Africa. South African conditions are unique and no one else's rules will provide the answer for South Africa.
One key reason why foreign rules will not provide the answer is because all the foreign rules were developed in the context of overall tax systems and the rules only make sense in the context of the total package. Over the course of these hearings, the Committee will no doubt be treated to countless examples of "cherry picking" - we shouldn't have a capital gains tax because New Zealand doesn't tax capital gains, we shouldn't recognise gains and losses when property is transferred on death because the Netherlands doesn't tax at that point, we shouldn't recognise gains and losses by companies on the sale of shares because Germany is going to exempt these, and so on and so forth. We can be sure that in the case of every foreign example submitted in the arguments, we won't be told the rest of the story - there may not be a formal capital gains tax in New Zealand but gains on land in New Zealand are subject to ordinary income taxation with no concessional treatment, the absence of recognition on transfer of property on death in the Netherlands is offset by a comprehensive wealth tax, non-recognition of gains and losses on the sale of shares by a company is part of a comprehensive and very different company and shareholder tax regime used in Germany.
2. What are capital gains in South Africa?
I've suggested that the design of an appropriate capital gains tax regime in South Africa is to be found in South Africa, not overseas, though general economic impacts of broader tax design experiences will be helpful in answering very general tax questions. If we're to look for the answers to capital gains taxation in South Africa, an appropriate place to start is by asking what exactly are capital gains?
Here, South Africa's experience is almost unique, as capital gains in South Africa have a meaning that is almost unique - a legacy of English common law precedents and Dutch-Roman civil law concepts.
The story of capital gains in South Africa can be traced back to to the late middle ages, when England had two separate legal systems, the common law legal system of the king and the equity system of the ecclesiastic courts. The two systems ran in parallel, with separate courts and separate powers. The king s courts had the power to right wrongs by making a person who had done something wrong to another person pay damages to the wronged person. The ecclesiastic courts enjoyed no similar power. However, they were able to achieve justice in cases brought before them by developing the notion of a "trust". They would recognise the legal ownership of property but say the legal owner had a moral obligation to hold the title in the property for another person, the person entitled as a matter of justice to the benefits of the property. This separation of legal title and beneficial interest in property was known originally as a "use", which later became known as a "trust".
As time went on, trusts lawyers developed more sophisticated notions. The most important of these was the separation of beneficial interests into two periods of time - life interests and remainder interests. A person could leave his property to a trust, with the income from the trust going to his wife for her life and then the property itself going to his eldest son. The life interests became known in trust law terminology as "income interests" (because the life beneficiary was entitled to income of the trust during her lifetime) and reminder interests became known in trust law terminology as "capital interests" (because the remainder beneficiary would eventually receive the capital invested in the trust.
Over time, many dispute came before the ecclesiastic courts over the issue of which beneficiary of a trust had the right to gains derived by the trust - did they belong to the life beneficiary or the capital beneficiary? In the case of some gains such as interest, the matter was fairly easy to resolve. But who had the right to gains from the sale of trust property? The income beneficiary would argue he or she should keep all the gains derived by the trust in the beneficiary s lifetime. The capital beneficiaries would say the gain related to the property that would come to them so they were entitled to the gains.
To resolve the disputes, the ecclesiastic courts had to develop fairly sophisticated tests to characterise gains as either income gains or capital gains. It's important to note that the distinction had no tax consequences - the tests merely resolved which of two competing parties would be able to keep the gains.
In the late 1800s, the ecclesiastic courts of England were combined with the ordinary courts and the trust law concepts of the ecclesiastic courts were fully incorporated into the common law. The U.K. income tax schedules at the time referred to various types of income and U.K. judges started to use trust law concepts to define what "income" in the income tax meant. They presumed the term should have the same meaning in tax law as in trust law so gains that would have gone to a capital beneficiary had the gain been derived by a trust were said to be outside the scope of the income tax law.
When the U.S. adopted income tax law for the third time in 1913, representatives for taxpayers made similar arguments - they said the income tax law only caught gains that would be given to income beneficiaries of a trust. In perhaps the most famous tax case in American law, Eisner v. Macomber, the taxpayer argued there was a fundamental difference between income an capital. Capital, the taxpayer said, was like a tree and income was the fruit borne annually by the tree. The U.S. Supreme Court agreed, saying income was like fruit and capital was like a tree, but when you sell the tree for more than you paid for it, the gain is "income" for income tax purposes, whoever it might go to as a matter of trust law.
The U.S. proved to be the exception in the English-speaking world. All other English-speaking countries, drawing directly from U.K. trust law principles, distinguished between income gains and capital gains and said that without specific statutory inclusion measures, only gains that would be income gains for trust law purposes were caught by income tax legislation. Interestingly, often the judges quoted from the U.S. decision in Eisner v. Macomber, citing the decision as authority for the proposition that the difference between capital and income was like the difference between trees and fruit, carefully omitting the second limb of the quote, namely that if you sold a tree for a profit, the gain should be income for income tax purposes.
The English common law tests for distinguishing capital gains from income gains had absolutely nothing to do with any rational economic distinction or any notions of ability to pay taxes. They were designed not to affect a tax burden but rather to decide which of two competing claimants should be entitled to a gain. The tests turned largely on the subjective intention of the trustee - did the trustee act to derive gains that would better the income beneficiary or the capital beneficiary. This was twisted for tax purposes into a subjective intention by the taxpayer to derive immediate or longer-term gains. Other tests were derived from the U.K. common law with some South African variations that seem to have been incorporated from various Roman-Dutch property law concepts. In addition to the subjective intent of the taxpayer, the tests include the frequency with which gains are derived, whether they are periodic in nature or irregular, whether they arise in the context of ordinary business activities or as a separate investment activity, whether the gains are used or applied by the taxpayer to fund ordinary expenses that one would normally apply a regular income towards or whether they were used to pay for extraordinary acquisitions or consumption, and so on.
The South African income tests have proved in many ways to be far more narrow than their counterparts in other jurisdictions that adopted the U.K. trust law income gain/capital gain distinction. Particularly in recent years, the income notion has been greatly expanded in many of those jurisdictions to include gains that are ancillary to income gains or derived from secondary income-producing activities. Thus, for example, in many English-speaking jurisdictions, gains derived by an insurer from the sale of shares would be ordinary income gains on the assumption that the purchase and sale of shares is a necessary part of an insurance business. In South Africa, by way of contrast, taxpayers continue to argue that these gains are capital gains, currently excluded from taxation entirely.
Another important feature of the South African capital gains concept is its application to companies as well as natural persons. The notion that companies, legal persons, could have subjective intentions similar to natural persons and that the subjective tests used to distinguish income and capital gains could be applied equally to companies and individuals was not unique to South Africa - the same theory was applied in almost all English-speaking countries and in the U.S., where the capital gains concept was the product of the legislature, the capital gains rules applied to both companies and natural persons.
The contrast with non-English-speaking countries is significant. In Europe and much of the rest of the non-English-speaking world, both developed and developing countries, company income tax is based on profits. The profit notion is synonymous with gain - that is, any gain is profit and thus subject to company income tax in most countries. There is no distinction between income gains and capital gains at the company level.
Several English-speaking countries have been moving in the direction. Thus, in the United States and Australia, for example, companies capital gains are now fully included in taxable income without concession. The South African proposal of different, and higher, inclusion rates for companies may be seen as a move in the same direction, though given the much broader concept of capital gains for companies in South Africa, it will remain very generous compared to examples found in many developed and developing countries.
In conclusion, we currently have a dramatic tax distinction in South Africa depending on whether gains derived by both natural persons and companies are characterised as income gains or capital gains. The distinction is based on no rational economic criteria or tests - rather, it is derived almost entirely from tests developed by ecclesiastic courts centuries ago for an entirely different purpose. Most importantly, in the context of the year 2001 in South Africa, those tests afford taxpayers tremendous scope for tax planning. The most important element of tax planning in South Africa is the structuring of arrangements so gains will be characterised as capital gains rather than income gains under the judicial tests used to distinguish the two types of gains. A large percentage of the accounting and legal intellectual resources in this country are devoted to the profitable exercise of planning for the preferable characterisation of gains and the distinction between fully taxed and fully tax exempt gains is so great that taxpayers are willing to incur significant economic costs in terms of inefficient structures or arrangements only to ensure their gains will be characterised as the tax-free category of gains.
3. The need to allocate domestic investment capital most efficiently
Following on the closing point made above in the context of the South African notions of income gains and capital gains, the tax consequences of how a gain is characterised are dramatic - depending on its characterisation, a gain may either be fully taxed or fully exempt from tax. Relying on astute tax planning advice, taxpayers with access to professional assistance can often rearrange affairs or organise transactions or investments with the object of greatly increasing the odds that their gains will be characterised as tax-free capital gains. But at a more general level, the fundamental tests to distinguish income and capital gains are fairly well-settled and almost all persons with savings have a basic idea which sorts of investments will yield gains likely to be called (tax-free) capital gains and which sort will yield gains likely to be called (fully-taxed) income gains. The distinction in terms of tax consequences in South Africa means that taxpayers can derive much lower rates of return on assets yielding capital gains and still be better off after tax than had they derived much higher levels of income gains.
History has shown in a fairly conclusive way that for all its faults in terms of income distribution and polarization of economic opportunity - faults that can be addressed through appropriate government action - the free market economy is the best system discovered to date for developing an efficient economy. In the context of a global economic system, any nation that wants to compete in the world market must have an economy as efficient as its competitors. That efficiency will be achieved only if the basic tenet of a market economy is followed - the persons holding a nation's investment capital must be seeking the highest rates of return when they invest their capital. So long as capital flows to the assets and activities that yield the highest rate of return, the market will pressure all businesses to achieve maximum efficiency gains so their rates of return will be sufficient to attract further investment capital.
Many commentators have observed that the market economy may work best in the absence of any tax system, as this will ensure all investment decisions are made purely on the basis of economic criteria. The most efficient tax system, in terms of economic considerations, is the tax system that treats all types of gains exactly the same so the relative difference between alternative investment choices remains exactly the same. Thus, for example, if the choice faced by an investor in a no-tax world is a 20% return on a risky investment or a 15% return on a less-risky investment, the difference in a 50% tax world should be the choice between a 10% after-tax return and a 7.5% return. In other words, if the tax is to operate efficiently, it should not affect investment choice at all. The difference between 15% and 20% in a pre-tax world or 7.5% and 10% in a post-tax world is exactly the same in proportional terms - in both cases, the less risky investment offers three-quarters of the return of the more risky investment, so the relative risks and benefits remain the same. Under the neutral or efficient tax, the relative choice remains the same and investment decision will be made solely on rational commercial economic grounds, with no tax bias.
If, however, different tax rates apply to different types of gains, investment decision making will be distorted. Consider, for example, a world in which income from a riskier entrepreneurial business is fully taxed at 50% and gains from a less risky investment in appreciating property is tax-free. Assume, once again, that the first investment has a 20% return and the second a 15% return. Before tax, the investor has a choice of a 20% return or a 15% return. After-tax, the investor has a choice of a 10% return or a 15% return. Any rational investor, however inclined he might otherwise be to invest his capital in a higher risk entrepreneurial business, is more likely to reduce exposure in that investment and divert capital to the investment yielding lower pre-tax and higher post-tax returns.
As the Committee is well aware, vested interests have long put forward a whole range of arguments in favour of differential treatment of capital gains. Many experts may argue there may be some merit to some of the arguments as they apply to particular types of assets in very particular circumstances for a variety of compliance cost, administrative cost, and other reasons. It is quite clear, however, that a blanket preference for all gains that are called capital gains in South Africa, particularly when the notion is based on judicial concepts developed for different purposes in a different age, is simply contrary to the principles of a sound market economy. In effect, those calling for blanket exemptions or preferences for capital gains are asking for a tax subsidy for a particular type of gain to raise the after-tax return from investments yielding that type of gain so taxpayers will invest in those assets rather than the ones in which they would invest in a genuine market economy with a completely neutral tax system. For their own reasons, they would repudiate the fundamental tenet of a market economy - taxpayers should invest solely on the basis of expected rates of return - with the inevitable consequence of a less efficient economy overall and one that will find it difficult to compete with enterprises operating in jurisdictions with more neutral tax systems and, thus, more efficient businesses.
A key decision facing South Africa, therefore, is how it wants to place itself economically in the 21st century. If it wants to erect barricades and develop less efficient enterprises in the hopes that they can survive in a closed economy, retention of the current tax distinction is logical. If it wants to compete effectively on the world market, it must adopt a tax system that builds upon the tremendous potential and resources in this country to develop enterprises able to compete on the basis of world's best practice, efficiency resulting from the efficient allocation of investment capital without tax distortion.
4. The need to encourage riskier and more entrepreneurial investment
The dynamics of world commerce are changing rapidly. It is not possible to sustain a viable economy on the basis of being a provider of raw commodities and farming output in the 21st century. Nor, given the unlimited number of potential competitors, is it sustainable to rely on value added based on cheap labour. Economic participation in the 21st century will be based on value added manufacturing based on innovation, quality and low-cost production based on continually increasing efficiency and the provision of new services and traditional services in new and efficient ways. Investment in low-yielding fixed assets, particularly immovable property and buildings, will not build the foundations for national economic success in the 21st century.
It was noted in the previous section that the effect of exempting capital gains from taxation in South Africa is to encourage investment in lower yielding, low risk assets which, after the tax exemption, can yield a higher tax return. This pattern has been repeated in almost all countries that originally had exemptions or severe concessions for capital gains. The reverse was equally true - when capital gains were subject to tax the same as or almost the same as other types of income, investors switched investment for low-yield, low productivity assets, particularly immovable property, into higher risk assets and supported entrepreneurial innovation.
The leading and without doubt most dramatic example of this is the American experience. The Republican White House under Ronald Reagan was committed to a broad income tax base that would restore economic neutrality to the income tax base. To accomplish this, the U.S. raised its inclusion rate for capital gains to 100%. Immediately, there was a flood of money out of real estate and less productive investments into high tech, innovative and entrepreneurial businesses. Much of the diverted investment went into the computer and software business. The result - directly attributable to two factors, the alignment of capital gains and ordinary income rates and relaxation of prudential investment rules for tax-preferred pension funds - was the growth of silicon valley. The sector has never looked back. The Australian story is similar. The introduction of capital gains tax in 1985 led to significant investment in new and entrepreneurial ventures and a corresponding fall off in investment in less productive immovable property. The government unfortunately overcorrected the overheating economy with significant interest rate rises which led to a brief recession, but since there has been a significant shift in the economy, with "new economy" ventures replacing the older traditional industries as the growth engine of the nation.
It appears the dramatic effect of capital gains taxation on encouraging innovative, entrepreneurial and higher risk investment is due to three factors. The first is simply the adoption of a rational market environment with tax neutrality between investment alternatives. When faced with after-tax returns similar to pre-tax returns, investors make sound investment decisions based on commercial considerations only and realign their portfolios with a shift to from low risk to more entrepreneurial activities.
The second aspect of a capital gains tax that encourages investment in higher risk entrepreneurial ventures is the recognition of capital losses. Under current law, if a taxpayer invests capital in a low risk venture, the chances of losing the funds are low. If the funds are invested in a high risk venture, the gains on disposal of the investment will be tax-free but if there's a loss on disposal, the entire investment has been lost. If capital gains are taxed and capital losses recognised for tax purposes, taxpayers are willing to invest more in risky investments since they have a type of government-sponsored insurance policy connected with their investment. In return for taxpayers agreeing to pay part of their gain to the government as tax if they win, the government agrees to subsidise their losses by allowing taxpayers to fully offset the losses against otherwise taxable gains. With the government picking up almost half the losses if an investment proves to be unsuccessful, taxpayers will invest more funds in more risky activities.
The third aspect of capital gains taxation that favours small business and entrepreneurial investment is the reduced cost of capital for such enterprises. When Australia included capital gains in the income tax base in 1985, one of the strongest proponents of the tax was the chair of the largest bank in the country. At the time, business, and in particular small business was faced with crippling interest rates on debt capital. It was the view of many in the finance world that the high rates were in part a consequence of the enormous diversion of domestic investment capital to investment in assets, particularly immovable property, generating lower yields of relatively predictable capital gains. If more of these funds were invested in debt, they said, interest rates would fall and further investment in riskier equity would also reduce rates by reducing demand for loan capital.
The relationship between interest rates and the level of inclusion of capital gains in the income tax base is impossible to ascertain, as so many other factors also affect the rate of interest. It is clear, however, that whatever cost of capital would otherwise apply to small business, the cost will fall with a more rationale allocation of investment capital as the after-tax rates of return from alternative investments more closely tracks the pre-tax rates of return from those investments.
The relative cost of capital and the effect of exempting capital gains from taxation has an interesting subsidiary effect on patterns of ownership. The price of assets reflects in part the effect of tax on gains related to those assets since the tax affects the after-tax rate of return. If business profits are fully taxed but capital gains are exempt from tax, property yielding more if its return in the form of business profits will lower in value (to increase the after-tax rate of return) and property yielding most of its return in the form of capital gains will increase in value as the tax benefits are partially capitalised into the price. What is the effect of these results on foreign investors? Consider, for example, the case of an American company or a European company which is fully taxed on its profits (be they capital gains or income gains under South African concepts). Because business profits and capital gains are subject to the same tax rates in the home country and these taxpayers are not subject to the tax biases of South African investors, these foreign investors will find active businesses in South Africa yielding current profits a bargain and assets yielding capital gains overpriced. As South Africans invest in lower yield properties, particularly immovable property, foreign investors will be acquiring active businesses that yield higher pre-tax (and for them post-tax) rates of return. One effect of an exemption for capital gains in South Africa is to subsidise takeovers of active businesses by foreign investors.
5. The need to encourage greater foreign portfolio and direct investment
While South Africa is an important economic power in its region, it is not a significant player in terms of world capital markets. To sustain its economic growth, South Africa must rely on foreign capital. This will take the form of portfolio investors who purchase non-controlling interests in South African companies and foreign direct investors who establish subsidiaries in South Africa to conduct businesses in this country. As the proposed law now stands, both groups of investors will be completely exempt from tax on gains from the disposals of their shares in South African companies, whether they hold a small portfolio interest in those companies or are 100% owners of South African subsidiaries.
The proposed complete exemption from taxation of capital gains realised by non-residents on sales of shares in South African companies should thus ensure the tax will have no effect whatsoever on levels of foreign investment. Nevertheless, it has been suggested in some quarters that imposing a capital gains tax in South Africa will deter foreign direct investment. No evidence has been offered to support this assertion and in all logical respects it is counter-intuitive as most foreign direct investors in South Africa seek to derive ordinary business profits through their South African subsidiaries though retailing, manufacturing, exploitation of natural resources, services such as airlines, and so forth. Not only are the foreign owners unaffected by the taxation of capital gains, but so too are their South African subsidiaries as their income will not normally be characterised as capital gains. It might be argued, however, that some South African subsidiaries will be able to characterise some of their gains as capital gains. Will the inclusion of taxable gains derived by their South African subsidiaries into the South African income tax base deter foreign investors? International experience overwhelmingly suggests this will not be the case.
In the case of most countries from which foreign direct investment capital flows to South Africa, profits distributions from South African subsidiaries are either exempt from tax when received by the parent company if they have been taxed in South Africa or they are taxed subject to a credit for any South African tax imposed on the profits when they were first derived by the South African subsidiary. In either case, the only effect of an exemption in this country for South African source capital gains would be to transfer tax revenue from the South African Revenue Service to a foreign revenue service. That is, if capital gains derived by a South African subsidiary are not taxed in South Africa, the parent will be taxed when the profits are repatriated to the foreign shareholders. The exemption would not lower the overall tax burden; it would just affect who collects the tax. If the object of an exemption is for South Africa to subsidise the U.S. or Australian treasuries, for the sake of economic efficiency gains it would make more sense to tax capital gains and mail annual subsidy cheques to the U.S. and Australian governments.
6. Revenue yield
Predictions of yields from capital gains taxes are notoriously unreliable. In virtually every case of jurisdictions starting with the English distinction between income and capital gains, the revenue form inclusion of capital gains in the tax base turned out to be far higher than predicted - in Australia's case, almost tenfold in a short period. And in every case, the revenue greatly exceeded collection costs notwithstanding predictions by some that the broader tax base would be expensive to administer.
Part of the reason for the great underestimation in each case was the failure of tax authorities to realise just how much avoidance was going on as taxpayers entered into arrangements to convert fully taxable income into non-taxable capital gains. In Australia's case, for example, authorities revised initial estimates downwards when tax professionals assured those authorities that there was very little conversion of ordinary income into capital gains. Only after capital gains were included in the tax base were authorities able to discover how enormous was the revenue leakage from arrangements to convert other types of income into capital gains.
The true revenue story lies not in the capital gains tax, however, but in the increased ordinary income tax take following the inclusion of capital gains in the income tax base. Depending on how much the differential between capital gains and income gains is reduced, the incentive to rearrange affairs and convert income gains to capital gains is reduced. The large revenue gains from taxing capital gains will not come from taxing capital gains in the same manner as other types of income, but rather from increased revenue from taxes on other types of income. Moreover, this will rise in the longer term as economic inefficiencies and distortions are minimised and businesses operate more efficiently. Also, significant deadweight losses in the form of fees for tax planning and associated activities will be reduced as these amounts are diverted from tax minimisation planning to productive investment. Equally important are the potential gains from the economy as the enormous intellectual resources of tax lawyers and tax accountants currently spent on developing schemes so gains will be characterised as capital gains rather than income gains are diverted to more productive activities. And finally, from the perspective of the revenue authority, the costs of administering the capital gains provisions will almost certainly be offset in the longer term by reduced costs of chasing taxpayers who have engaged in schemes to convert income gains to capital gains.
It has often been argued that in an inflationary environment, indexation of capital gains is required to prevent the taxation of inflation gains that do not represent increases in real economic power. The argument is difficult to sustain on a theoretical level and almost impossible to sustain on a practical level. It is true that inflation distorts the measurement of real gains. However, this is true of all elements of the income tax system, not just capital gains. Inflation artificially increases the nominal gain of ordinary traders disposing of trading stock in the course of their businesses. Inflation artificially increases the profits of persons using machinery or equipment by failing to recognise the true costs of their equipment through the depreciation rules. Inflation artificially increases the income of lenders who are taxed on real interest returns and the additional return that represents compensation for the loss in real value of their principal. Inflation artificially decreases the profits of borrowers by allowing them a deduction for the inflation and real components of interest payments without any offsetting recognition of gain in respect of repayment of loan principals in devalued rand.
It is conceivable to index the South African income tax system for inflation. Other countries have indexed their tax systems and shown it is a feasible, if horribly administratively complex, possibility. However, indexing only one part of the tax system would be the worst of all worlds - economically inefficient and distorting and administratively complex and costly. And of all the types of capital income for which indexation might be considered, capital gains is arguably the least deserving. Unlike gains such as interest which are taxed annually, capital gains that accrue over many years are taxed only when an appreciated asset is sold. Persons who realise capital gains thus already enjoy a significant tax concession in terms of tax deferral and it makes no sense to compound this concession with a further selective concession in the form of indexation.
One effect of indexing capital gains will be to exacerbate tax avoidance arrangements. Many persons deriving capital gains have some debt. It would be a completely unjustifiable rule that allowed taxpayers to deduct the full nominal cost of interest payments to derive gains while taxing only the nominal inflation-adjusted gains. It would be the ultimate whipsaw against the revenue authority. However, it would be impossible to prevent this system if taxpayers ensure their investments generate a small current income such as rent while most of the gain could be characterised as capital gain.
However weak the theoretical case for selective indexation of capital gains only, the practical issues must be the paramount consideration. Without exception, jurisdictions that have experimented with indexation of capital gains have found it to be a terribly complex and costly concession. Moreover, the highest costs inevitably fall on small investors. Consider the case of a small investor making small monthly deposits into a unit trust or wrap fund, saving for his or her retirement. Every month's deposit will be a different element of the cost base, subject to a separate inflation calculation. By the time the person retires, the inflation calculations could fill many binders of pages, with 12 separate calculations for each year, possibly multiplied by a number of different investments. It's almost certain that the investor will be unable to make the calculations his or herself, meaning that person will have to lose a not insignificant part of the retirement savings just to have a tax adviser calculate the indexation adjustments.
The Australian experience in this respect provides a good illustration of the pitfalls of indexation. Capital gains were introduced to the income tax base in Australia in 1985. By 2000 it was clear to everyone that indexation was no longer sustainable. The corporate sector welcomed its abandonment and full inclusion of capital gains into taxable income as part of a package that saw a lowering of the company tax rate. For individuals, indexation was replaced with a partial exclusion rule similar to that proposed for South Africa, though not nearly so generous.
The inclusion of capital gains in the income tax base will do much to increase the efficiency of the South African tax base. It is unfortunate that it is being proposed to include capital gains on such a concessional basis. The benefits of capital gains taxation in terms of a more efficient investment environment, greater investment in entrepreneurial and riskier projects, and reduced deadweight losses and administrative costs will not be fully realised until capital gains are taxed on the same basis as all other profits, with base broadening offset by a reduction in tax rates if desired as revenues grow. Until a fully rational tax system is politically achievable, partial taxation of capital gains will go some way to achieving these objectives.