WHAT can be done to reform the tax system in a useful way? The Minister of Finance, Nhlanhla Nene, will step into the limelight tomorrow to provide an update on the state of government finances and reveal the Treasury’s plans for the direction of national government expenditure and revenues over the next three years.
Of particular interest will be to learn how government revenues are holding up in the face of slower economic growth, and what this may mean for the funding requirements of the government.
Most important is whether or not higher tax rates will be called for – a move that will damage the growth prospects for the economy.
The economy is suffering from both a lack of demand as well as a long-term lack of supply. Government austerity is not an obvious remedy for an economy growing at rates well below even its modest growth potential. Planning for a smaller fiscal deficit, however, might be regarded as a necessary even if an unfortunate concession to the debt rating agencies, though for the sake of economic growth a shallower path for government expenditure would be much better than any further attempt to squeeze higher tax revenues by way of higher tax rates.
It is the unwillingness or inability of the household sector to spend more that is holding the economy back. Without some recovery in household spending propensities, it is unlikely that private spending on plant and machinery or on payrolls can pick up momentum – or that government revenues will grow faster.
Higher tax rates on top of higher electricity and other charges made by local governments and public enterprises could even prove counter-productive, resulting in still slower growth and so less government revenue.
Monthly government expenditure and revenue are highly variable. They require seasonal adjustments and smoothing to establish the latest trends. We undertake this exercise (as shown in the diagrams) with data available to the end of June and extrapolated to June 2015.
The Treasury has had to cope with a sharp slowdown in the growth in revenues that followed the recession of 2009. Correctly in the circumstances, the fiscal deficit was allowed to widen significantly in 2009. But the deficit has remained at these elevated levels since then and, if past trends are sustained, will continue to run at a monthly rate of almost R16 billion unless there are significant changes in the growth in government spending and revenue collections.
It should be noticed that both government spending and revenue collections have been growing consistently in real terms in recent years. Revenue growth recovered strongly with the economy after the recession of 2009, as may be seen.
It may also be seen that revenues have been growing at a faster rate than expenditure in recent years. A policy that is less demanding of revenue from taxpayers, combined with a still slower growth in government spending, would have been more helpful in encouraging household spending and economic growth and possibly have led to higher revenue collections.
Without a pick-up in gross domestic product (GDP) growth, current trends will mean an increase in the ratio of government debt to GDP. The government debt-to-GDP ratio (currently around 42 percent and expected to rise to 44.3 percent by 2016/17, with debt service costs estimated to increase from the 3 percent of GDP to 3.1 percent in three years) is still well below what is regarded as the critical 60 percent level.
There is still room for a temporarily higher ratio, but without faster growth it is unlikely that these debt ratios can trend lower. Total public sector debt, including state-owned companies and municipalities, was 57.3 percent in 2012/13 and will rise further as Eskom and SAA are recapitalised with contributions from Treasury.
The advantage of a low government debt-to-GDP ratio is that it gives scope for counter-cyclical budgets when a recession restricts revenues, as it did after 2009. Some reserve for bad times, in the form of a stronger fiscal balance sheet, needs to be rebuilt. Whether these slow growth times are appropriate for such a purpose is highly moot.
The interest rate bill for the South African taxpayer will rise in response to the larger budget deficits and larger borrowing requirements of the government since 2009. More spending on interest means less spending on other goods and services the government could provide.
Debt service costs in the 2014 Budget were estimated as R114.901bn, or 9.2 percent of consolidated government expenditure. Global capital market trends in long-term interest rates (currently very favourable) will, however, be a much more powerful influence on government debt borrowing costs than the views of the rating agencies. This is especially since the largest proportion of newly issued government paper will be in rands rather than foreign currencies.
The strong foreign appetite for rand-denominated debt provides a strong strategic advantage for the government in difficult economic times that hopefully will be put to good use. Hopefully, government austerity (especially in the form of higher tax rates) will not make the times more difficult.
There is, however, the key issue on which Judge Dennis Davis and his tax committee have been charged to advise the Treasury on tax reforms: how should taxes best be collected in South Africa to promote economic growth and equity?
As Judge Davis has made eloquently clear on the two occasions I have heard him in public discussion, the economic choices for spreading the burden of taxation and so encouraging growth may not be politically acceptable.
Raising the VAT rate and/or eliminating zero rating is the most obvious alternative to higher income tax. Zero VAT rating means a large sacrifice of potential revenues from which the poor could benefit in much more accurately targeted ways. Zero rating fillet steak does not help the poor, though they would protest at having to pay an extra 14 percent for their food.
Food stamps (administratively complex) or direct, well-targeted subsidies for a very few staples (chicken, bread, maize and canned pilchards) designed to hold down their prices by more than a VAT rate increase would raise them could leave the fiscus much stronger after eliminating zero rating (R30bn stronger) and serve to calm the political storm.
The scope for raising income tax collections from individuals by way of higher income tax would seem limited, unless the rate increases applied across all the brackets – hardly a popular step. The higher-income earners already pay a disproportionate share of income tax collected and the scope for raising much more from them, enough to make a meaningful difference to the fiscal deficit, would seem limited and disruptive of economic growth.
We illustrate the Treasury’s dependence on high-income earners. As may be seen, the current income tax structure is highly redistributive, with high-income earners paying a much larger share of income taxes than their shares of income.
Those earning above R750 000 in 2014 will assume a larger share of the income tax burden than their share of income. Those earning above R1 million in 2014 will have earned about 17 percent of all incomes earned and paid about 30 percent of all income tax, or R100bn. We show the share of income by bracket and the share of all income tax paid.
There is one place to look for extra taxes that Judge Davis, at least in the talks I have attended, did not appear to consider. That would be in the form of additional payroll taxes. The great difference between the South African tax structure and those of the developed world is our very limited reliance on payroll taxes.
Companies in South Africa pay a relatively large proportion of all taxes collected by comparison with the US, despite all the belly aching about legitimate company tax saving. By contrast, the US federal government relies very heavily on payroll, that is social security taxes, as we show.
These are taxes on employees that employers collect and may be asked to contribute to. They are paid for by employees as a sacrifice of salaries, especially if the taxes paid have little relationship to the benefits subsequently received.
South Africa has become something of a welfare state and will become more of one if tax means permit it. Funding a welfare state may well make reliance on payroll taxes, levied as a constant percent of total remuneration, unavoidable as a least bad way to raise additional revenues.
The political advantage of the payroll tax is that employers rather than employees can appear to be paying the tax while in the longer run such impositions will lead to a slower than otherwise increase in wages and salaries or other employee benefits provided by firms. What appears as a tax paid by the firms in fact becomes a salary sacrifice.
Inflation will also complicate the perceptions of the incidence of a payroll tax, making it politically more acceptable. Furthermore, starting from a very low base there is great scope for slowly and steadily increasing the rate at which the payroll tax is levied.
The economic case for relying much less on income taxes of one kind or another that have strong disincentive effects on income generation and more on proportional payroll taxes is a strong one.
Higher payroll taxes combined with lower income tax rates and company tax rates would be highly stimulating of economic growth.
Given the difficulty of gaining access to formal employment and the advantages those formally employed enjoy in the form of benefits denied to the informally employed or unemployed, there would be a rough justice in asking the formally employed to assume a larger tax burden in the form of a salary sacrifice, a sacrifice that may well be overtaken by the additional demands for skilled labour that a faster-growing economy would generate.
Brian Kantor is chief economist and strategist at Investec Wealth & Investment.