IDASA: Budget Information Service
SUBMISSION ON NATIONAL BUDGET 1999 / 2000

IDASA: Budget Information Service
19 February, 1999

1. Introduction
The 1999 / 2000 budget maintains the Government’s commitment to tight fiscal discipline. Unfortunately, the minimal resources available for distribution, are disproportionately skewed to wooing the private sector instead of alleviating poverty. This is not a poverty budget - it is, in the first instance, an investment budget. We are concerned about the insufficient resources allocated to poverty alleviation given the trend in provinces towards the crowding out of social services and the fact the most severe cuts in social expenditure are still to occur in the next two years’ budgets. There are sufficient resources within this constrained budget to continue government’s commitment to private sector led growth, but also to provide a greater measure of short-term poverty support.

2. Growth
The government has shown the greatest political will in dealing with growth. Even though the budget is not expansionary in fiscal terms, economic growth is likely to receive a mild stimulus.
• There are significant increases in the Department of Trade and Industry budget towards SMMEs, technological development and Spatial Development Corridors. The Arts, Science, Culture and Technology budget strengthens the resources allocated to technological research.
• Overall capital expenditure is projected to increase slightly in real terms (6.9% per year between 1998-2002) and by a larger proportion than current expenditure (5.3% over 1998-2002) or total expenditure (6.5% over 1998-2002).
• Commitment to macro-economic targets will build investor confidence.

3. Personnel
Over the period of the MTEF, consolidated personnel spending across national and provincial government is set to decline in real terms, by approximately 2% per year. This will presumably allow for reprioritisation of expenditure towards other types of current expenditure (textbooks, health supplies, welfare services) and capital expenditure. Such a saving can only be realised if either civil servants receive real decreases in remuneration or the government affects radical cuts in personnel numbers. There are three reasons why we are concerned with the medium consolidated personnel cost projections:

• At present there is no accepted rightsizing tool or programme. In briefing week, Minister Zola Skweyiya said that a rightsizing programme would be finalised as soon as personnel audits have been completed in all the provinces. This, he predicted, will be by year-end. Even in the most optimistic scenario, this will not impact on the size of the wage bill before mid 2000/01. In the more likely scenario of protracted negotiations with unions, the positive impact of rightsizing will only be felt in 2001/02.
• The chances of a serious real decrease in public sector wages are unlikely given pressure from public service unions to keep pay in the public service in line with inflation.
• The MTEF projections are based on the consolidated projections of national and provincial departments. In the past, many of these departments have underestimated their personnel expenditure considerably. In the key cost-drivers of health and education for example, personnel expenditure in the 1997/98 fiscal year was underestimated by on average 25% (Table 5).
• The projected personnel expenditure does not account for salary drift due to automatic pay progression and actual pay promotions. These factors have in the past added up to 3% of personnel expenditure.

For these reasons, we are concerned that actual personnel expenditure is likely once again to substantially exceed budgeted amounts. This will place considerable pressure on provincial social expenditure, as we argue below.

4. Poverty
The 1999/2000 budget makes provision for R1.1 billion in poverty relief, an increase over the R850 billion allocated last year. The fact that this increase has been targeted at poverty and employment sensitive allocations, such as public works and water affairs, will assist with short-term poverty relief. Furthermore, the significant nominal increase in health spending is pro-poor to the extent that there are increases allocated for immunisation and HIV / AIDS programmes. An important programme for the rehabilitation of hospitals was launched and allocated R200m.

However, for several reasons we feel that the poor will not benefit disproportionately from the budget as was claimed by the Minister. We are aware and accept the tight fiscal constraints, but this requires that every effort is made to mine all opportunities for poverty alleviation and to maximise the effectiveness of available spending on the poor. The emphasis in the budget speech obscures the lack of long-term commitments to poverty alleviation. The crowding out of social expenditures will become even more acute in the next two years. This is because the brunt of lower GDP growth in 1997 and 1998 has been distributed predominantly in the outer years of the MTEF. If we maintain the deficit targets there will be R5bn less available for non-interest expenditure in the outer years than was anticipated at the launch of the MTEF in 1997/8. Reduced growth but retained deficit targets, imply reduced social expenditures.

• In his budget speech, the Minister claimed that social expenditures have been maintained despite fiscal constraints. This may be true in global terms, but it ignores the current trend of decreasing spending school text-books, stationary, medical supplies and capital expenditure (Tables 1 – 3). Even though provinces have largely met their deficit requirements (and even posted a surplus), the real per-capita transfers to provinces have declined (Table 4), while personnel expenditures will consume opportunities for reprioritisation. Growing personnel expenditure in largely stagnant provincial budgets have been financed by decreases in spending on school text-books, stationary, medical supplies and capital spending in social services. It may be claimed that efficiency gains will offset real declines in provincial allocations. However, if current trends continue, efficiency gains are likely to be absorbed by personnel expenditure.
• The Budget does provide for greater equalisation of revenue allocations between provinces. However, this equalisation is unlikely to lead to greater equalisation in per capita social service provision between provinces. This is because the overall allocation to the provinces is decreasing in per capita terms (Table 4). For example, the Eastern Cape will receive the second highest per capita transfer by 2001/02, but this transfer will still be lower in real terms than it received in 1998/99. The scale of backlogs in the weakest provinces and commitments to personnel expenditure will further crowd out intra-provincial reprioritisation. It remains to be seen whether the backlog component in the horizontal split formula will significantly address backlogs. This is unlikely as the amount is not earmarked and will be subject to the same pressures as other capital expenditure.
• The decrease in allocations to the recently formed Policy Co-ordination Unit in the Deputy President’s office is alarming. The likely impact is to erode the opportunity created to co-ordinate poverty and employment expenditures for maximum impact (It is worth noting that the Youth Commission, for example, receives double the allocation to the Policy Co-ordination Unit). It also decreases the Deputy-Presidents capacity to monitor the impact of government budget on the poor, preventing future refinements in targeting.
• The decrease in the housing budget (5.8% in nominal terms) is unfortunate given government indications that this is precisely the moment when labour-intensive house building is coming on stream.

5. Taxation
We must commend SARS for the astounding and unanticipated increase in revenue collection. The success this government has enjoyed in collecting taxes is surely a sign of good governance, and we believe that the poor will ultimately benefit from a stable fiscus that continues to collect from those with the ability to pay.

Taxation benefits were distributed carefully to all income earning segments of the population, but the private sector benefited disproportionately. This implies that, as was the case in last year’s proposals, the tax proposals for 1999 / 2000 will have a mildly regressive impact on an otherwise highly progressive tax system.

While taxpayers will surely welcome relief in the effective rates of income tax they will have to pay, the continued reduction of taxation from this progressive source means that the overall tax system is less progressive. If we think about the reductions as targeted transfers to taxpaying households, one has to agree that it is very strange to transfer 0 Rands to households earning R10,000 in taxable income; while giving R95 to those earning R19,000 in taxable income; R415 to those earning R45,000 in taxable income; and R1,475 to those earning half a million in taxable income! In a country with income distributed as unequally as in South Africa, it would have been appropriate to distribute the transfers in the opposite direction. This would imply distributing R1,475 to citizens without income, or in low-paying employment, and not bothering with what amounts to a pittance distributed to wealthy citizens.

The impact of reductions in company taxes is difficult to measure, because analysts differ in terms of their assessments of who ultimately pays the tax. In all likelihood, the reductions will benefit the rich more than the poor, but if – as entrepreneurs promise – this will lead to more job creation, the poor could benefit in the long-term. Unfortunately, the track record of supply-side economics under Reagan and Thatcher were not particularly effective in generating new jobs in the U.S. or the U.K., and our own recent growth record does not indicate strong employment effects. It is also important to note that the cost of the reduction in company taxation amounts to approximately R2.5bn. This is sufficient to increase pensions by at least a further 50% of the granted increase, or to expand the housing budget by 80%. The potential positive spin-offs for employment and poverty relief are considerably more tangible. At the least, we feel a more equitable share of these resources could have satisfied the Government’s commitment to investment led growth, as well as brought a guaranteed greater measure of relief to the poor.

While the minister resisted pressures from wealthier segments of society to push the burden on to the Value Added Tax, excise taxes on tobacco and alcohol products – while justifiable in health terms – were increased yet again. These are highly regressive taxes, as the poor consume these products, and represent a more significant portion of total income than for the wealthy.

6. Budget reform
• Documentation
The Finance Department has made commendable progress on improving budget documentation. It is important to single out the preparation of the National Expenditure Survey and improved Budget Review for special mention. These documents lay the basis for departmental review in a manner that has never before been possible in this country.

• Budget Reform white paper
It is most unfortunate that the publication of the Budget Reform White Paper is once again delayed. It is now four years late. As we have argued before, the absence of a map for this ambitious programme erodes Parliament and civil society’s ability to monitor and participate in the process. It is worth mentioning that South Africa is the first developing country to undertake a shift to outcome budgeting on this scale. Change on this scale will benefit from deeper legislative and civil society participation. For example, negotiations over the Treasury Control Bill would not have been as protracted, if a clear map of budget reform had been provided. This would have clarified, amongst other points, whether it is the appropriate document to address virement and whether a split bill was necessary.

• Outstanding legislation
We are concerned that, despite significant progress, the Treasury Control Bill has not yet been passed by Parliament. This Bill is at the core of a transparent and accountable budget process. The resulting truncated legislative period on the Treasury Control Bill is lamentable given the considerable public and civil society participation in the hearings on this Bill. Similarly, the persistent delays in the Money Bills Amendment Procedure Bill are preventing legislatures from exercising effective power over the budget. One of the unfortunate effects is that civil society will continue to focus on questions of the size of the envelope, rather than, in our opinion, the greater challenges of intra-vote reprioritisation.

• Changes to the budget process
We suggest the following budget process changes to maximise the effective involvement of legislatures and civil society.
(i) The Medium Term Expenditure Budget Policy Statement (MTBPS) is a unique opportunity to broaden monitoring and discussion of macro-economic policy. To take advantage, it is necessary to insert an appropriate point in the Legislative agenda for open hearings and debate on this topic.
(ii) The MTBPS does not substantially enable earlier monitoring and analysis of inter or intra-vote reprioritisation. We still face the situation that civil society and Parliament have only couple of days and limited capacity to analyse and debate a document that takes 18 months to prepare. Further, even once the National Budget is presented, it is very difficult to understand the impact on poverty alleviation, given that provincial budgets have not yet been delivered.
(iii) We therefore suggest:
- That the Budget presentation is moved to January, subsequent to the publication of the MTBPS, while national and provincial legislature debates should remain in February. This will allow 2 –3 months for in-depth analysis and debate that will become increasingly necessary as amendment powers are introduced.
- That the national and all nine provincial legislatures present the budget to their respective legislatures on the same day in January. This synchronisation will allow the Department of Finance to prepare accurate and consolidated national and provincial accounts enabling detailed consideration of the social sector expenditures. An earlier synchronised presentation will have the added benefit of allowing markets to adjust more gradually to the budget.
(iv) We wish to re-iterate our call for a revision of the accounting procedures for the
independent monitoring institutions. At present, several of these Commissions receive their funding through one of the departments they are meant to monitor. For example, the Gender Commission receives it funding through the Justice department; the Financial and Fiscal commission through the Finance Department and the Independent Electoral Commission through Home Affairs. As an alternative we suggest that a separate vote is established for these institutions, who should negotiate as a group, together with national departments, for their allocations. This arrangement will increase the bargaining power of the Commissions as a group with the Finance Department, while enabling monitoring of the entire expenditure relating to Commissions.

7. Conclusion
There are several aspects of the 1999/00 budget that deserve praise. Amongst these, the continued commitment to fiscal discipline and the improved documentation are most notable. All sectors receive some benefit from the budget. Committed spending has meant that there are very few resources available. Unfortunately, the allocation of available resources is skewed strongly towards the private sector rather than to alleviate poverty. We do not disagree with the principle of decreasing company taxation in an attempt to stimulate investment-led growth. We do argue that in the present context, a greater proportion of available resources should have gone to poverty alleviation and public sector –led growth and a smaller proportion to private sector tax benefits. Our primary concern arises from evidence that the trend in provincial spending is for declining revenue allocation and personnel expenditure to crowd out social expenditure. Given the structure of the MTEF, provincial social expenditure will come under even greater pressure in the next two financial years. This budget is likely to aggravate these trends.

Idasa feels that the time has arrived for Government to publicly introduce a greater measure of realism into its medium-term fiscal and poverty framework. It is clear, that the Government has been unable to meet most of the poverty targets that were set in the original RDP. There are several reasons for this, including strong exogenous factors, such as the Asian Crisis and inherited debt burden. The Government has also been unable, despite sincere efforts to meet most of the targets established within GEAR. Most notably, the growth and employment targets have proved well beyond our capacity to reach. Even the deficit target, clearly objective number one, has been elusive. It is therefore time to review macro-economic and poverty policy and re-establish a measure of credibility and realism in these estimations. This is also an appropriate moment to unify the fiscal and social side of macro-policy. It is not sufficient to claim that the RDP is endogenous to departmental spending. In reality, deficit targets are crowding out social targets. In order to unify the RDP and GEAR, the government needs to update GEAR targets and insert a parallel set of revised medium-term poverty targets that are congruent with fiscal constraints. A national conference on macro-policy, with broad stakeholder participation, would be an ideal opportunity to bind the nation to a realistic poverty and fiscal strategy.

TABLES

Table 1: Provincial Capital and Personnel Expenditure

Table 2: Medical supplies in the provincial health budgets as % of total provincial budgets (stores and livestock).

Table 3: School books and stationery in the provincial education budgets as % of total provincial budgets (stores and livestock).

Table 4: Real per capita transfers to provinces

Note: These figures are taken from p.107 of the Budget Review. Local government conditional grants for 1998/9 and 1999/00 were excluded since they are phased over this period. These figures were then adjusted for inflation based on figures from the Budget Review p.56. Population figures were projected from Census 96 with population growth rates from the Demographic Information Bureau.

Table 5: Consolidated budgeted and revised personnel expenditure for 1998/99