Why the proposed Wealth Taxes won't work - IRR

Anthea Jeffery says that what South Africa most needs is a much higher rate of economic growth

South African Institute of Race Relations NPC (IRR) Submission to the Davis Tax Committee regarding the desirability and feasibility of three possible forms of wealth tax, Johannesburg, 31st May 2017


1 Introduction

The Davis Tax Committee (the committee) has invited interested people and stakeholders to submit written comments, by 31st May 2017, on the desirability and feasibility of three possible forms of wealth tax: a land tax, a national tax on property, and an annual wealth tax.

This submission is made by the South African Institute of Race Relations NPC (IRR), a non-profit organisation formed in 1929 to oppose racial discrimination and promote racial goodwill. Its current objects are to promote democracy, human rights, development, and reconciliation between the peoples of South Africa.

The committee has provided no details of what the tax base would be in relation to any of these possible taxes. Nor has it explained what the tax rate would be, or what administrative and enforcement processes might be used in determining and collecting these possible taxes. In the absence of such details, this submission focuses simply on the general points that can be made, within the short time allowed, regarding the desirability and feasibility of these possible taxes in the current South African context.

2 A land tax

A land tax is generally defined as a tax on the unimproved or site value of land, which is paid by the owner. It differs from a property tax, which taxes both the value of the land and the improvements on it.

2.1 Desirability of a land tax in South Africa

A land tax would encourage the more productive use of under-utilised land, bring down land prices, and make more land available for redistribution.

However, most of the under-utilised land in the country is communal or state land. The latter includes much of the land already acquired for land reform purposes, some 70% of which has since fallen out of production. The further redistribution of land which then ceases to produce will not provide effective redress. A new land tax could also undermine food security, push up food prices, hobble the agro-processing sector, and reduce agricultural exports, so harming the trade balance and the value of the rand.

2.2 Feasibility of a land tax in South Africa 

The Local Government: Municipal Property Rates Act of 2004 (the Rates Act) has already extended the obligation to pay rates to all land across the country. However, rates set by municipalities are still relatively low, leaving room to introduce a land tax as well. South Africa’s rates of personal income tax, corporate income tax, and value-added tax (VAT) are also not that high compared to other countries. In addition, since all land owners are already obliged to pay municipal rates, the administrative burden of introducing a land tax would be manageable.

However, the revenue collected by the government, as a proportion of gross domestic product, already stands at 30%, including indirect and other taxes. The tax base is also small, with some 560 000 individuals and 600 companies paying about 60% of all the personal and corporate income tax collected.

Much of the burden of the tax would fall on South Africa’s 35 000 white commercial farmers, many of whom have annual turnover of less than R1m and net earnings of around R300 000 a year. A case study of five commercial farms in KwaZulu-Natal also shows that an additional land tax would be unaffordable for most of them in most years.

A land tax would also add to the challenges confronting black commercial farmers. It would be costly to administer, while its yield is likely to be low. Overall, the tax would make it harder to ensure, as the minister of agriculture, forestry, and fisheries, Senzeni Zokwana, has recently urged, that ‘farmers are encouraged to produce’ and that ‘those coming up are assisted at all costs’.

3 A national tax on the value of property (over and above municipal rates)

A national tax on property would go beyond the taxation of land, for it would tax not only land but also the buildings and other immoveable improvements on it, as the Rates Act already does at the municipal level.

3.1 Desirability of a national tax on property

Property taxes can have ‘significant untapped revenue potential’, as the International Monetary Fund (IMF) suggested in 2013. Since they focus on immovable property, they are more difficult to avoid. They are also assumed to be fair and progressive.

In South Africa, however, the government’s overall tax take is already very high, while the tax base is very small. Moreover, while property itself is immovable, a tax on it may drive away the investment, technology, and skills that give it much of its value. The tax is not necessarily progressive as it can be passed on to others, while fairness requires the regular and accurate valuation of property – which is difficult in practice to achieve.

Property taxes are also highly unpopular and generally have to be accompanied by many rebates and reliefs. These add to complexity and reduce yields, which rarely exceed 1% of GDP. To reduce resistance, a visible property tax must be accompanied by visibly increased efficiency and accountability in governance, which will be difficult to achieve.

3.2 The feasibility of a national property tax

Since all property is already subject to municipal rates under the Rates Act, the necessary information on properties and owners across the country has already been gathered. Physical inspection of properties is not needed, as computer-assisted mass appraisal systems and other analytical techniques can be used.

However, many municipalities are short of technical and other skills and are likely to have only limited information on the properties within their jurisdictions. In addition, even if municipal valuation rolls were complete and accurate, they could still not be used in implementing a national tax as municipalities employ different valuation methodologies.

Much of the necessary information will still to have to be gathered, while the tax rates to be applied nationally for different categories of property will have to be determined. A national process to deal with rebates, reliefs, objections, and appeals will also be needed. Accurate billing systems will have to be established and maintained, and effective enforcement mechanisms put in place.

A major obstacle is the culture of non-payment which the African National Congress (ANC) earlier fostered and which has proved difficult to end. Partly for this reason, the amount owing to municipalities, for both rates and service charges, has steadily increased and now stands at some R118bn. Of this, some R78bn is owed by households, R25bn by businesses, and R6bn by government entities (and the remainder by diverse others).

If payment of a national property tax can indeed be comprehensively enforced, this will have important ramifications for the capacity of municipalities to collect on municipal rates. Resistance to paying local authorities could well increase in response to what may be perceived as ‘double taxation’. This could greatly undermine what is a critical source of revenue for many municipalities – and one which the Constitution also expressly allocates to the third tier of government.

Resistance could also increase if municipalities are asked to add a surcharge to their existing rates. Outstanding debt to municipalities would then rise further, putting them under further financial pressure while failing to generate much additional revenue for the national fiscus.

4 An annual wealth tax

An annual wealth tax is a tax which is charged each year on the holding of wealth. It is different from the taxes on the transfer of wealth (estate duty, donations tax, and capital gains tax) which South Africa already imposes. 

Annual wealth taxes are generally applied to the whole range of assets held by an individual, household, or business. These include housing, cash, non-owner occupied property or real estate, jewellery, furniture, cars and boats, the capitalised value of future pension rights, listed shares, shares in private (unlisted) companies and partnerships, and business assets. The tax is levied on net wealth, after the deduction of debt and other liabilities.

Relatively few countries levy annual wealth taxes, as yields are generally low while administrative costs are high. Some countries have nevertheless been investigating the introduction of wealth taxes since 2014, when Professor Thomas Piketty of the Paris School of Economics published Capital in the Twenty-First Century.

In his book, Professor Piketty identifies inequality as the most pressing of all economic problems, ranking well above challenges such as low growth and rising unemployment. He also proposes the introduction of wealth taxes to reduce inequality, saying: ‘The ideal policy for avoiding an endless inegalitarian spiral and regaining control over the dynamics of accumulation would be a progressive global tax on capital.’

Professor Piketty claims to have discovered a basic economic law: that the rate of return on capital (r) generally exceeds the rate of economic growth (g). This proposition can be expressed symbolically as r > g. This means, he says, that ‘all large fortunes grow at extremely high rates’, making the gap between the rich and the poor ever wider. This, he says, is ‘one of the most striking lessons’ of the Forbes list of the 400 wealthiest people.

However, that r will always exceed g is not an immutable principle. Returns on investment can in fact be negative, and the holding of wealth is often fleeting. In addition, 68% of those on the Forbes 400 list in 2013 were ‘self-made’ billionaires, who had earned their wealth, not inherited it. In addition, few of the people on the list remain on it for ten or more years. Analysis also shows that returns on investment for those on the list are much the same as for everyone else. Piketty thus fails accurately to identify the key causes of rising inequality, which lie rather in new ideas and technology, the huge markets that globalisation has made possible, and high incomes for those with scarce skills.

4.1 Desirability of an annual wealth tax

An annual wealth tax would help reduce inequality in South Africa, where asset inequality stands at 0.95 on the Gini coefficient and income inequality at 0.63. As Professor Piketty argues, a wealth tax would compel the more productive use of assets, adding to growth and jobs, while also generating more revenue for increased social spending.

However, assets will often have to be sold simply to pay the tax, which will not stimulate production. A wealth tax will also deter investment, reduce growth, and limit employment. The economic damage is likely to be profound, as shown by Dr Michael Schuyler of the Tax Foundation in Washington DC in modelling the likely impact of Professor Piketty’s proposed wealth taxes on the US economy in 2014.

Dr Schuyler’s model shows that Professor Piketty’s proposed wealth taxes – if the threshold for liability was to start at net wealth of $260 000 – would reduce GDP in the US, then amounting to some $17 trillion, by 6.1% per annum or roughly $1 trillion a year.

His modelling also shows that Piketty’s proposed wealth taxes – at rates of 0.5% for those with net wealth of between $260 000 and $1.3 million, 1% for net wealth between $1.3m and $6.5m, and 2% thereafter – would be even more damaging than Professor Piketty’s proposal to raise top income tax rates to 80% and 55%.

According to Dr Schuyler’s research, if US income tax rates were to be raised to 80% on incomes starting at $750 000 a year and to 55% for those earning around $220 000 a year, then GDP would be reduced by 3.5% or roughly $595bn a year. This is considerably less than the $1trillion reduction in GDP that the wealth taxes would usher in.

Why is this so? As Dr Schuyler explains, a 1% tax on wealth might not sound high, but ‘three factors magnify the potential harm to the economy’. He goes on: ‘First, a wealth tax of a given percent is equivalent to an income tax of a much higher percent. For example, if the pre-tax return on an asset is 8 percent, a 1 percent wealth tax on the asset would take away one-eighth of the income. That is the same tax bite as a 12.5 per cent income tax rate.’ Second, much of the wealth to be taxed would be the productive capital that helps sustain employment and economic activity. Third, Piketty’s wealth tax would be levied every year and on top of all existing taxes.

With South Africa’s annual growth rate now hovering at around 1% of GDP, the country Africa simply cannot afford GDP reductions of the kind that Dr Schuyler’s modelling suggests. Moreover, the impact of the wealth tax would not be confined to the most wealthy. Instead, it would percolate down to the entire population, leaving all South Africans the poorer.

A wealth tax is also likely to be particularly damaging to the asset-rich but income-poor. It also encourages avoidance and evasion, even among those ideologically in favour of it (such as former French president François Hollande). Since the wealthy are better able to avoid and evade, the tax in fact falls primarily on the middle classes. It also encourages a flight of capital and skills.

The costs of implementing such a tax are high, while yields are low. In addition, an annual wealth tax cannot begin to tackle the reasons for inequality and is thus ineffective in overcoming it.

4.2 Feasibility of an annual wealth tax

Banks and other financial institutions already have comprehensive records which can be used to value the assets of the wealthy. In addition, though the rich may try to conceal or export their assets, the sharing of tax information across countries is growing. Moreover, the South African Revenue Service (SARS) already has well-established mechanisms in place to enforce the payment of taxes, which further increases the feasibility of the tax.

However, the valuation of many assets is intrinsically difficult – and especially so where no sale or other transaction has taken place to provide an independent market value. It is particularly difficult to value assets held in private (unlisted) companies, partnerships, trusts, and the like. It is also hard to decide how the capitalised value of future pension rights is to be treated. Moreover, though banks and other institutions may have data on the value of many assets, they are unlikely to have complete information on relevant liabilities. Such information must thus also be collected and cross-checked. All these factors add to the complexity of valuation. So too does the fact that the valuation task has to be repeated at regular intervals, if not every year.

Since a wealth tax imposes great hardship on those with little income, tax authorities must also make provision for rebates and deferments. But these undermine the efficiency and equity of the tax, while also reducing its yield. Yield is also generally low, raising questions as to whether the high costs of implementation are justified. Enforcement is also often difficult, especially if questions around the accuracy of valuations give rise to litigation. The growth of cryptocurrencies will also make it easier in time for many to conceal their wealth.

5 No simple cure for inequality

The wealth taxes under consideration by the committee cannot address the underlying reasons for the inordinate asset inequality evident in South Africa. These reasons are deeply rooted in the past, but they also have much to do with what the government has done (or failed to do) since 1994.

For example, while some 1 million whites own their homes, so too now do 7.7 million black Africans. This shows a rapid shift from the profound injustice of the apartheid era, when home ownership by Africans was largely prohibited. But whites generally have title deeds to their houses, whereas most Africans still do not.

In addition, despite the constitutional promise of tenure reform, some 16.5 million Africans living in former homeland areas have yet to be accorded individual title to their customary plots.

Under the land reform programme, moreover, some 8.2 million hectares of land have been transferred from whites to blacks. But beneficiaries have generally been barred from obtaining individual ownership and are often confined to being tenants of the state.

Major redistribution via the budget has brought about a rapid increase in living standards, with 63% of South Africans (up from 37% in 2001) now falling within the middle-ranking measures (LSMs 5 to 8). Income inequality nevertheless persists, with average annual household income among Africans standing at R113 200 in 2015, as opposed to some R631 400 among whites. However, the persistent disparities thus evident are rooted in complex socio-economic factors, which additional wealth taxes cannot overcome.

Key factors include the following:

- the median age of whites is 39 while that of Africans is 24, and older people generally earn more;

- whites have better education and skills, for 74% of whites have completed matric versus 33% of Africans, while 29% of whites have post-school education compared to 6% of Africans;

- some 6% of whites are unemployed, compared to 30% of Africans;

- whites tend to stay longer in their jobs than Africans (71 months as opposed to 51 months), partly because Africans are often headhunted to fill racial targets;

- roughly 8% of whites own their own businesses, compared to 3% of Africans;

- almost all whites live in urban areas, whereas the African population is only around 60% urbanised and urban incomes are generally higher than rural ones; and

- some 78% of whites grow up in two-parent households while only 29% of Africans do so, which has important ramifications not only for household income but also for self-confidence and future achievement.

Also important is the changing structure of the economy and the impact of government policies. As the economy has modernised, so the contributions to GDP of the primary sectors (agriculture and mining) has shrunk. Yet these are the sectors with the greatest capacity to absorb unskilled labour. Both these sectors have also been adversely affected by government policies, which have undermined property rights, eroded business confidence, deterred investment, and contributed to job losses.

At the same time, the government has failed to improve the schooling system, which remains one of the worst in the world, despite the large revenues allocated to it each year. Hence, the great majority of African youths leave school without the skills that would help them gain entrance to the financial sector, for example, which has grown rapidly since 1994 and now contributes some 20% to GDP.

Factors of the kind described above provide the primary reasons for persistent inequality in South Africa. By their very nature, they cannot be overcome by means of additional taxes on land, property, or net annual wealth. On the contrary, such taxes are likely to have many negative consequences for South Africa’s already struggling economy. They will also be expensive to administer and enforce, while their yield is likely to be limited.

6 Better solutions

South African tax practitioners note that the country already has three forms of wealth taxes – estate duty, transfer duty, and donations tax – which bring in about 1% of tax revenue. In addition, the tax burden in South Africa is also already high, while much of it (as earlier noted) falls on a relatively small group of individuals (numbering some 560 000) and companies (about 600) that pay around 60% of all personal and corporate income taxes.

Andrew Wellsted, director at Norton Rose Fullbright, warns that imposing yet more taxes on this small tax base could cause more harm than good, saying: ‘There is already a proliferation of different taxes that are largely borne by the same tax base. These include transactional taxes such as VAT, existing wealth taxes, such as estate duty, transfer duty and donations tax, as well as income tax, which has just been increased to a relatively high 45% in the [2017] budget’. Also relevant is South Africa’s dividends withholding tax, which was increased from 15% to 20% (a 33% rise) at the same time.

Warns Mr Wellsted: ‘The addition of yet another tax, such as a wealth tax, will be asking the same contributors to apply more funds towards the fiscus… [Yet] there are a number of studies which show that, at some point, asking taxpayers to contribute too much can lead to a reduction in the taxes collected.’ Keith Engel, chief executive of the South African Institute of Tax Professionals, echoes this concern, saying: ‘Many people are now paying more than 50% of their income in tax once VAT and other indirect charges are taken into account. A hefty new tax could be a breaking amount for many.’

At the same time, none of the three wealth taxes being mooted is likely to achieve its stated objectives. A land tax is unlikely to improve the productivity of under-utilised land, most of which is owned by the state or traditional communities. If it results in many commercial farmers being forced to sell some or all of their land, this will erode food security and push up food prices, which will harm the poor in particular.

If forced sales make it easier for the government to acquire land for land reform purposes, this might at first sight appear as if redress for past injustice is being provided. In practice, however, this would be an illusion. The land would remain in state ownership and would thus empower the government, rather than the disadvantaged majority. In addition, much of the land thus acquired by the state would be likely to fall out of commercial production, as has already happened on some 70% of transferred farms.

A national property tax is unlikely to curb the speculative holding of land, especially as municipalities generally already charge significantly higher rates on vacant land within their jurisdictions so as to encourage its more productive use. At the same time, a national property tax will be difficult and costly to administer. It will also be onerous to enforce, while its yield is likely to be limited. That the debt owed to municipalities, for both municipal rates and other charges, has already grown to some R118bn provides an important pointer to the enforcement challenge likely to arise. Introducing a national property tax will also interfere with a major source of revenue for local government – and one which the Constitution expressly assigns to the municipal sphere.

An annual wealth tax is unlikely to reduce inequality, as this has many and complex causes going far beyond the capacity of a wealth tax to address. An annual wealth tax is also likely to exacerbate existing capital flight; encourage a further exodus of scarce skills; reduce incentives to work, invest, and save; hinder the country’s already limited capacity for capital formation; lower the anaemic growth rate (0.3% of GDP in 2016); and worsen the crisis of unemployment, especially among the young and inadequately skilled. The tax will encourage evasion and avoidance. Its burden will fall principally on people in the middle class who, by dint of much hard work and self-discipline, have managed to acquire homes and savings with a value that is high enough (perhaps largely because of inflation) to bring them within the ambit of the tax. This will hurt the established middle class – but it will also greatly harm the emergent middle class, which already faces many daunting obstacles in building up its income and its assets.

What South Africa most needs is a much higher rate of economic growth, at 5% of GDP or more (as recommended by the National Development Plan). It also needs a number of structural reforms to:

- bolster property rights,

- reform labour laws,

- shift away from damaging BEE requirements to a new system of ‘economic empowerment for the disadvantaged’ which would help the many rather than the few,

- boost the efficiency and lower the costs of the public service,

- expand essential infrastructure,

- reform the education system to build up the skills base,

- counter family breakdown,

- reduce the burden of crime, substance abuse, and domestic violence, especially in poor communities, and

- put an end to the fraud and inflated pricing which currently taints some R200bn of the government’s annual procurement spend.

Additional wealth taxes can neither resolve these problems nor help the disadvantaged to get ahead. Comments Dan Foster, tax director at law firm Webber Wentzel: ‘Ultimately, South Africa, like all developing countries, needs more growth and not more taxes. Taxes lead to wealth destruction, low investment, low returns, low growth, and lower tax collections.’

South African Institute of Race Relations NPC (IRR)

 31st May 2017