The fiscal challenge

The 2009 Medium-term Budget (MTB) to be released on 27 October is being awaited with more interest than usual. It is probably the most important MTB to date, as the sustainability of the prudent fiscal policy that South Africa has followed for the past 10 years is in question for the first time.

Since the introduction of the MTB the budget deficit in the relevant fiscal years has, with one exception, been smaller than initially expected owing to better-than-expected tax collection. However, the challenge over the next few years will be quite different, namely to curb spending until the economy, and therefore tax collection, has recovered sufficiently.

The 2009/10 budget

By now it is common knowledge that this year’s budget deficit could be up to double that budgeted (viz. between 7% and 8% of GDP), mainly as a result of lower-than-expected tax collection.
However, the expenditure side will also not go according to plan, inter alia owing to the higher-than-budgeted wage and salary increases granted in the public sector. The statement in the 2009 Budget Review that “if government is to maintain its commitment to increasing public employment, some moderation of general salary increases will be appropriate”, seems to have been forgotten.

The cause of the problem with this year’s budget is partly an over-optimistic set of macro-economic assumptions, especially as far as economic growth is concerned. The expected growth rate of -2% for 2009 is considerably lower than the +1,2% assumed in the budget. However, the National Treasury is definitely not the only institution that had not foreseen such a deep recession.

The MTB will contain the first official revised estimate of tax revenue for the current fiscal year, although statements by various members of the cabinet have already indicated that it will be reduced by at least R60 billion to R70 billion. In other words, government tax revenue is expected to be about 10% less than budgeted, and 11% less than in 2008/09.

2009/10 vs. 1992/93

The reduction appears even more drastic when compared to the experience during the recession in the early nineties. For example, in 1992/93, when the economy contracted by 2,1% (in other words similar to the contraction expected this year), tax revenue nevertheless increased by 6% compared to the previous year. This was mainly thanks to a rise of 13% in personal income tax against the backdrop of an increase of 21,3% in the remuneration per worker in the non-agricultural sector. On the other hand, company tax and VAT declined by 6,7% and 6,8% respectively.

In the five months to August 2009 state revenue from personal income tax amounted to 37,1% of the amount budgeted for the year compared to 38,6% in the corresponding period last year. It is therefore unlikely that personal income tax will come to the state’s rescue as in 1992/93, despite higher-than-expected wage and salary increases. The decline in the contribution of personal income tax to total tax revenue from approximately 40% in 1992/93 to approximately 30% in 2008/09 has in any case resulted in government revenue becoming more cyclical by nature.

Company tax (including secondary tax on companies) currently amounts to 29,6% of the budgeted amount compared to 32,7% at the same time last year. However, the bad news is still to come, as the tax paid by companies in the past few months was based to a large extent on last year’s gains. On the other hand, the earnings base of the JSE All Share Index is currently 22,4% lower than a year ago, which is a good indication of what is happening to company profits. The increase in the effective company tax rate in spite of a lower nominal rate has probably caused an increase in the sensitivity of company tax to the business cycle.

However, VAT will probably show the largest under-recovery. At present it amounts to 28,9% of the budgeted amount compared to 38% at the same time last year. The sharp decline is still puzzling in the light of the fact that nominal GDP is still rising. For example, in the five months to August 2009 VAT amounted to 8,2% of final consumer spending by households, whereas it was 10,2% in the corresponding period in 2008.

It therefore seems as if there has been a drastic increase in non-compliance – could companies be reluctant to pay VAT because their cash flow is under pressure? If this trend continues, VAT could show a deficit of up to R40 billion against budget, which would mean a massive drop of 17% compared to 2008/09. This is nearly three times the decrease in 1992/93.

However, the expenditure side of the budget will also contribute to a higher-than-expected deficit. Government spending in the five months to July 2009 amounted to 40,7% of the amount budgeted for the year, which is significantly higher than the 38,8% of last year. If this trend is mechanistically projected further, spending in the current fiscal year could exceed the budget by R36 billion.

But spending is not consistent, and the National Treasury has indicated that the outcome for the year should be better than implied by these figures. Nevertheless, it will be interesting to see which upward adjustments will be made on the spending side to reflect inter alia the inordinate wage and salary increases for public servants and the rising debt service burden.

A budget deficit of around R180 million or 7,5% of GDP in the current fiscal year will therefore not be surprising. It represents a deterioration of 6,5 percentage points on the 2008/09 fiscal year compared to a deterioration of 3,6 percentage points in 1992/93.

In the meantime the public sector’s borrowing requirement has increased drastically, resulting in upward pressure on long-term interest rates. How detrimental this will be to the private sector’s access to capital will only become clear once the economy has recovered to the extent that companies are starting to think of expanding.

Fiscal policy in a time of recession

The global recession has resulted in expansionary fiscal policy being the order of the day and the sharp rise in South Africa’s budget deficit is therefore justifiable. However, the increasing signs that the recession is bottoming have already caused the focus to shift to an orderly reversal of the fiscal stimulus, as inter alia discussed at the recent G20 summit.

Policy-makers have repeatedly indicated that fiscal policy will not be tightened until such time as the economy recovery is well entrenched and that the reversal will have to be well coordinated internationally. However, South Africa is merely a drop in the ocean as far as the global economy is concerned and will not really be part of the coordination exercise, setting it free to pursue its own course.

The relaxation of fiscal discipline will probably still be acceptable until, but not beyond, 2010. Therefore, if South Africa has not taken effective action to put its government finances on a sound footing again by 2011, markets will not put up with it. Although the International Monetary Fund (IMF) expects the public debt to GDP ratio in the developed countries to continue rising until at least 2014, the opposite applies to the emerging-market countries that are members of the G20 group, and this is the benchmark against which South Africa will be measured.

Compared with other G20 countries, South Africa’s fiscal stimulus initially appeared relatively large if the depth of the recession in the various countries is taken into account. In a speech to the Helen Suzman Foundation and the Gordon Institute of Business Science on 6 April this year, the then Minister of Finance, Trevor Manuel, said: “Our fiscal response as a ratio of the slowing in our gross domestic product has been larger than nearly all other countries, except for the United States”, measured by the ratio between the change in the budget balance (-3,9 percentage points) from 2007 to 2010 and the change in the output gap over the same period (-4,1 percentage points). In the meantime both these figures are considerably worse.

This has been confirmed by a set of notes on the macro-economic aspects of the crisis prepared by IMF staff as background to G20 meetings. In February, before the national budget was tabled, they estimated South Africa’s discretionary stimulus for 2009 to be 1,3% of GDP. In March, that is after the budget, they raised their estimate to 1,8% of GDP in 2009 (compared to the average of 1,2% for the G20 countries) and -0,6% in 2010. At that stage South Africa and the United Kingdom were the only countries that were expected to implement a restrictive fiscal policy as early as 2010.

However, in June the IMF forecast that South Africa’s discretionary fiscal stimulus would amount to 3% of GDP in 2009 and 2,1% in 2010, which would put it in the top quartile of the G20 countries.

At that time they predicted that the overall budget deficit would increase to 4,1% of GDP in 2009 compared to the budgeted figure of 3,9% in the February budget and 4,3% in 2010, instead of 3,1% as budgeted. In the IMF’s Article IV report on the South African economy released in September, the budget deficit is estimated at 5,7% of GDP in 2009/10 and 4,7% in 2010/11. And notwithstanding the repeated upward adjustments this is still too optimistic.

If one considers that the change in South Africa’s economic growth since 2007 has in the interim increased to an expected 7,5 percentage points compared to 6,9 percentage points for the G20 countries, the local stimulatory package suddenly no longer seems quite so large, especially if one takes into account that the increased budget deficit can be attributed mainly to a sharp non-discretionary decline in tax collection and not higher discretionary spending.

The medium-term budget outlook

The question on everyone’s lips is fiscal sustainability. The key variable to watch is the so-called primary balance – whereas the ordinary budget balance is simply the difference between income and expenditure, the primary balance is calculated as income less expenditure, excluding the cost of public debt. A negative primary balance (or primary deficit) implies that the government is borrowing money to pay its interest and that its total debt and the concomitant interest burden will therefore continue increasing as long as this is the case. If the nominal interest rate on the government debt furthermore exceeds the nominal growth rate in GDP, the ratio of government debt to GDP will also continue to increase.
 
In terms of the February budget South Africa would have a primary deficit equal to 1,6% of GDP this year and 0,9% in 2010/11, with a return to a nil balance by 2011/12. The lower-than-budgeted tax revenue will obviously have an extremely negative impact on these figures. The IMF expects inter alia a primary deficit of 3,4% of GDP this year, 2,2% next year and 1% in 2011/12. The public debt to GDP ratio is therefore expected to continue rising in the medium term to reach 33,2% of GDP by 2011/12.

If one includes the contingent liabilities resulting from the guarantees the government has issued inter alia to Eskom, the debt ratio increases to more than 40%. However, this is still below the limit of 50% of GDP set by the National Treasury.

The table below sets out the required primary balance to stabilise the ratio of interest bearing debt to GDP at 35% (no interest is payable on contingent liabilities) for different combinations of nominal interest rates and growth rates in a simple static analysis. Although it shows that the required primary balance is not very high it nevertheless implies a substantial correction from the current situation. In my view South Africa will have to aim for a primary surplus of at least 0,5% of GDP, compared with a deficit of 3,5% in the current year.

	Required primary balance (% of GDP) to stabilise the government debt/GDP ratio at 35% of GDP
	Growth rate →
Interest Rate ↓
	7	8	9	10	11
7	0	-0,35	-0,70	-1,05	-1,40
8	0,35	0	-0,35	-0,70	-1,05
9	0,70	0,35	0	-0,35	-0,70
10	1,05	0,70	0,35	0	-0,35
11	1,40	1,05	0,70	0,35	0

South Africa therefore has time on its side to regain control of the fiscal trends. But although a rash tightening of fiscal policy is unnecessary, formulating a credible plan to curtail the increase in public debt is vital.

What is the outlook for the next three years, starting with the revenue side?

During the recession in the early nineties, government revenue declined from 24% of GDP in 1990/91 to 21,7% in 1992/93, in other words by 2,3 percentage points. It took eight years to recover to the level that prevailed in 1990/91 and averaged 22,3% of GDP in the three years after 1992/93. During those three years the economy achieved an average real growth rate of 2,5% per annum and during the eight years up to 1999 growth averaged 2,7% per annum.
According to the IMF’s forecast the South African economy will grow by 3,3% a year on average over the next three years and by 3,8% a year over the next five years. It can therefore reasonably be expected that government revenue as a percentage of GDP will recover more quickly than after 1992/93, supported by its more cyclical nature, but then the expected decrease as a percentage of GDP is much larger. If the share of wages and salaries in national income was to increase at the expense of profits, as the unions are aiming to achieve, it will create an additional stumbling block as the average tax rate for individuals is lower than that for companies.
 
According to the IMF’s forecast government revenue will amount to 25,3% of GDP by 2014, which will still be lower than the 26,4% in 2008. If government revenue is going to recover so slowly, raising taxes to make new spending plans possible might be a great temptation. Apart from the fact that tax increases (especially as far as company tax is concerned) would have a negative impact on the economy, it is a political hot potato. The only tax proposal that is likely to enjoy support is to increase personal income tax for the “rich”.

If we define the “rich” as those with a taxable income of more than R300 000 in 2006/07, we will find they represented only 4,4% of taxpayers, but accounted for 45,1% of personal income tax. If these ratios still apply, this small group of individuals will pay income tax equal to 9% of GDP in the current fiscal year.

This implies that the average tax rate of this group will have to be increased by 11% in order to deliver 1% of GDP in additional revenue for the government – a drastic step with many unwelcome side effects. In short, it is impossible to significantly relieve the pressure on government revenue over the next few years by means of tax increases.
Strict discipline with regard to the expenditure side of the budget is therefore unavoidable, and government expenditure will simply have to increase at a slower rate than nominal GDP. And this in the context of the spillover effect of this year’s abnormally high wage and salary increases for public servants, a sharp rise in the debt service burden, and a commitment on the part of the ruling party to increase social spending. Capital spending usually bears the brunt of any cut backs in government spending, but this would not be a good idea in view of South Africa’s infrastructure needs.

The best we can hope for is that the budget deficit will decline to between 3% and 4% of GDP over the term of the next MTB.

Conclusion

The consequences of this year’s fiscal catastrophe will be with us for some time to come. The context in which fiscal policy will be applied in the next few years is now radically different from that foreseen in this year’s budget. However, whether all the political role players realise this is an open question. Especially those who envisage a greater role for the state in the economy will have to think again. In fact, the squeeze in which the government finances is going to find itself necessitates a greater role for the private sector, inter alia in the development of infrastructure through public-private partnerships.

The fiscal outlook also creates a very fragile base for radical new initiatives such as the proposed national health service and national social security system. Caution will have to be the operative word and risky initiatives, the outcome of which cannot be predicted with reasonable certainty, should be avoided. Any new initiative will have to be budget neutral, in other words it will have to be financed from new income, bearing in mind that the tax burden is quite high already.