IDASA - Budget Information Service
RESPONSE TO THE MTBPS 2001
(Revised version)

Introduction
Government’s spending plans for the next three years are being presented against the backdrop of the continued existence of high levels of poverty and vulnerability in South Africa.
· A very recent estimate of the extent of income poverty in South Africa[1] tells us that 22.8 million people or 51% of the population live below the poverty line (estimated by the Economic Policy Research Institute to be R401 per person).
· The poverty situation of children is a particular concern in light of their vulnerability. The most recent measure of child poverty in South Africa tells us that at least 59.2% of all children age 0-17 are poor and at least 59.3% of children age 0-6 are poor. This translates into 10.5 million poor children age 0-17 and 3.8 million poor children age 0-6. It also shows that half of these children are desperately poor in the sense that they often go to bed hungry[2].

Our specific interest in the MTBPS 2001 is in the extent to which we can expect it to reduce and alleviate poverty and vulnerability.

The MTBPS acknowledges the urgent need for government to spend more and better so that poverty and vulnerability can be reduced. Also, the plans outlined in it hold the promise that government will begin to make more rapid progress in reducing and alleviating poverty. The MTBPS promises to do this by:
· First, presenting a fiscal stance and employment incentive initiatives that promise to assist markets in stimulating more growth and employment creation in the medium to long run.
· Second, stepping up budget allocations to the social services in aggregate, but particularly to social development and health services. The size of this increase is sufficient to allow for a reversal of declining trend in real per capita spending on social services since 1996, but is still fiscally sustainable.
· Third, promising substantial investment in infrastructure and HIV/AIDS, two areas that are particularly important for ensuring reduction and alleviation of poverty and vulnerability.
· Fourth, promising further progress in improving the effectiveness and efficiency of spending particularly by dealing with under spending and improving targeting of social service spending at the poor.

These intentions of government as expressed in the MTBPS are to be applauded.

However, we have the following four-fold concerns about the MTBPS in relation to reduction and alleviation of poverty.

· First, we are concerned that the growth scenario presented in the MTBPS for the MTEF period may be a little optimistic and that markets still hold little prospect for reducing poverty through employment creation.
· Second, we are concerned about the extent to which effective spending on infrastructure and HIV/AIDS – so critical for growth and poverty reduction – will be undermined by under-spending.
· Third, we are concerned about the extent to which progress is being made at targeting funds at the poor in social service programmes.
· Fourth, we are concerned that the size of the proposed allocation to social development may not be sufficient to keep up with demand for social security and welfare services.

These concerns tone down our enthusiasm about the MTBPS 2001. This is because they reduce the size of the impact we can expect it to have on poverty and vulnerability.

Our concerns about the impact of the MTBPS on poverty and vulnerability raise the following as important to ensuring that government’s budgets have a larger impact on the poor:

· First, the recommendations of the Taylor Commission - established to investigate how the social security net can be changed to provide support for growing numbers of unemployed people;
· Second, attending to the institutional and capacity problems that undermine spending on infrastructure and HIV/AIDS;
· Third, making rapid progress in targeting funds in core poverty relief and alleviation programmes.

This brief sets our four concerns about the MTBPS. It is structured as follows.

The first section explains why government’s growth forecast may be a little too optimistic and our concern that the MTBPS still does little to raise prospects for markets to reduce poverty via employment. This further underlines the importance of effective social service spending and making progress on the infrastructure and HIV/AIDS programmes if poverty and vulnerability are to be reduced over the medium term.

The second section looks at the under-spending problems in two areas that we see as key to the strategy to reduce poverty in the MTBPS - infrastructure and HIV/AIDS.

Section three raises the concern about targeting problems in social service spending. We confine ourselves to the norms and standards policy as an example.

Section four explains why we believe the size of the allocations to social development may not be sufficient to meet demand.

1. The optimism in the MTBPS growth forecast and why we expect market growth to do little to reduce poverty via employment

The MTBPS projects that real GDP growth will rise from 2.4% in 2001/02 to 3.1% in 2002/03 and then to 3.6% in 2003/04 and 3.7% in 2004/05. Table 1 presents the MTBPS growth forecast in annual years. It also presents the projections given in Budget 2001 and the real growth performance of the economy between 1997-2000.

Table 1: Growth projections for 2001-2004 compared to growth for 1997-2000

% real	Growth outcome 1997 1998 1999 2000				Growth projection in Budget 2001 and MTBPS 2001
					2001 Bud MTBPS	2002 Bud MTBPS	2003 Bud MTBPS	2004 MTBPS
Growth rate	2.5	0.7	1.9	3.1	3.5 2.6 3.7 2.8 3.3 3.5 3.7

Source: MTBPS, 2001:26 and Budget Review 2001:38

The downward adjustment in growth for 2001 and 2002 compared to what was projected in Budget 2001 and the high growth forecasts for 2003 and 2004 relative to 1997-2001[3] beg the question why we should expect growth to take-off toward the end of 2002 and be 3.5% by 2003.

The MTBPS view of growth rising to 3.1% in 2002/03, 3.6% in 2003/04 and 3.7% in 2004/05 from only 2.4% in 2001/02 is based on the argument that:
· Lower interest rates and tax rates will further raise consumer demand and investment by private firms;
· Export growth will continue to expand in 2002 and 2003 on the back of the lower production costs now established and real currency depreciation;
· The expansionary fiscal policy framework envisaged for the MTEF period in the form of i) lowering taxes; ii) injecting public spending into the economy (mainly in the form of social service and infrastructure spending) will raise domestic demand.

Fiscal management over the recent past has created an internal macroeconomic environment that is conducive to economic growth being driven through these channels. The following indicators seem to support Treasury’s argument that higher growth is imminent.
· Government consumption expenditure has been brought down from over 20% of GDP to around 18% and government dissaving is expected to be zero by 2004;
· Interest rates are relatively low in real terms;
· Fiscal policy is more expansionary but fiscal discipline is still being maintained;
The decline in capital expenditure – by both government and the private sector - experienced in the 1990s may have begun to be reversed[4] (even if slowly);
· Significant tax relief and investment incentives are to be given in 2001/02
· The injection of money into the economy proposed in the new fiscal framework (most of which is destined for spending on social development services, health and capital) is large.

However, the following considerations lead us to question the MTEF growth forecast.
· First, the high probability that the positive effect on consumers and firms to reduced interest rates and tax cuts may be dampened by uncertainty and lack of confidence related to the falling value of the rand, developments in Zimbabwe and the September 11 attacks. The President of SACOB, Kevin Wakeford, has recently highlighted the impact the negative sentiment is having on growth (Business Day, 7 November). He points out that the Business Confidence Index is already reflecting a negative effect on investment sentiment and the rate of export growth. He predicts that the full negative effect on South African exports will be felt only a year down the line as most export contracts are long term.
· Second, the long lag expected before the part of the fiscal stimulus designated for spending on infrastructure feeds through into high growth due to the need to build institutional capacity around infrastructure spending. We comment further on this below.

Our argument about why prospects for markets to reduce poverty are probably still limited over the MTEF, but particularly in the first two years, in part flows from our story that growth will be less rapid than anticipated. But it is also built around:

· The fact that the reasons for growth not creating a sufficient demand for labour – particularly unskilled labour - are such that we can only expect it to take a long time before the pattern of growth with limited demand for unskilled labour can be reversed[5].

· Second, evidence of how markets have been failing over the recent past to create enough and the type of jobs need to reduce poverty.

· Third, the argument that it is still going to take some time for the measures put forward in Budget 2001 to raise employment (particularly of the less skilled variety) to begin to work.

We say more on the evidence of failure of markets to generate sufficient demand or the right type of demand to reduce poverty and why we expect the key measures introduced to raise demand for labour in Budget 2001 to have limited impact over the next two years, below.

As argued in our submission in February 2001, analysis of the OHS 1995 and OHS 1999 data set by Statistics South Africa revealed that in the late 1990s - more specifically between 1996 and 1999 - markets did not do much to reduce poverty through job creation[6]. This was because:
· The increases in the demand for labour were insufficient to match the growth rate in labour supply;
· The increases in the demand for labour that did occur were concentrated in the skilled category;
· The increases in employment that did occur for unskilled workers were concentrated in the informal sector that offers lower wages and less income security on average than the formal sector;
· The demand for unskilled labour was either stagnant or decreasing over the latter part of the 1990s (Bhorat 2001).

It is difficult to gather data to see whether employment prospects for the poor have improved since then. But recent SARB Quarterly Bulletins lead us to the conclusion that markets continued to do little for poverty reduction in 2000 and the first two quarters of 2001[7].

In Budget 2001 government formally acknowledged the inability of growth to help reduce poverty via employment creation. And it introduced a set of measures to raise the prospects for growth to reduce poverty through employment creation. The most important of these included:
· - A wage incentive mechanism, and
· - A strategic investment incentive.

The following two considerations lead us to the conclusion that these two initiatives to raise the demand for labour (particularly unskilled labour) will have limited impact over the MTEF (particularly the first two years). .
· First, even though some progress has been made on the wage subsidy programme introduced in Budget 2001, it is still in the planning stage. More specifically, the MTBPS (2001:53) says that: `Following the Budget, an inter-departmental task team was established to consider various possible wage incentives …SARS, National Treasury and the Department of Labour will release draft legislation regarding the wage incentive before the end of the year for broad consultation with stakeholders’.
· Second, recent investment incentive schemes such as Coega and the other Spatial Development Initiatives have been undermined by negative sentiment. As argued above, the extent to which we can expect a reversal in negative investor sentiment over the first year or two of the MTEF is not clear.

2 Under-spending in key programmes important for growth and poverty reduction and what is being done to address it

2.1 The new emphasis on increasing capital and infrastructure budgets and spending

Capital investment is one of the main drivers of growth. Investment in infrastructure is also crucial to enhance the efficiency and effectiveness of spending on social services. The planned increase in capital and infrastructure spending in the MTBPS, presented in Table 2 below, is therefore very positive.

Table 2: Consolidated capital Expenditure

Rbn	2001/02	2002/03	2003/04	2004/05	Percentage growth between 2001 /02 and 2002/03	Average annual percentage growth between 2001/02 and 2004/05
Nominal total budgeted capital	22.6	27.4	30.7	33.5	21.2%	14.1%
Real Total budgeted Capital	22.6	26.0	27.8	29.0	15.1%	8.8%

Source: MTBPS 2001:72.

Table 2 shows that capital spending has been budgeted to increase substantially in real terms, especially between this year and next year. These increases come on the back of increases that were made to capital by the Budget 2001 earlier this year (Budget Review 2001:138). The proposed increases will raise capital spending to 12% as a share of total non interest expenditure in 2004/05 compared to 10% in 2001/02. The increases occur both at the national and provincial levels.

For capital budgets to impact on growth and poverty they have to be spent. In the past this proved to be a problem. The analyses below show that there has been improvement in spending at a national level this year but that more capacity and institution building has to be done at the provincial level[8].

An analysis of the figures given in the Statements of National Expenditure for the second quarter of 2001/02 reveals that at the mid-year point, on aggregate national departments had spent just under half of their aggregate capital budget. This spending seems to be an improvement on last year’s figures for the same period, where national departments had spent 31% of their budget. By the end of last year, departments had spent 80% of the aggregate budget (Statement of National Expenditure for 2000/01).

We look at provincial capital spending in this financial year in Table 3 below. The figures for provincial capital spending in 2000/01 are not yet available.

Table 3: Provincial Capital Expenditure (unaudited)– Second Quarter, 30 September 2001/02 (unaudited)

Provincial budgets and actual spending R’000	Annual budget	Actual spending, second quarter (September)	Actual spending, year to date	% of annual utilised, year to date
Eastern Cape	1,327,341	301,835	385,400	29.04%
Free State	545,112	101,851	172,587	31.66%
Gauteng	1,569,949	(13,404)[9]	137,768	8.78%
KZN	2,383,179	377,415	638,028	26.77%
Mpumalanga	649,208	160,613	322,398	49.66%
Northern Cape	207,299	67,378	129,288	62.37%
Northern Province	1,107,785	176,003	357,152	32.24%
North West	894,856	154,980	259,122	28.96%
Western Cape	971,347	223,516	390,211	40.17%
Total	9,656,076	1,550,187	2,791,954	28.91%

Source: Statement of Provincial Government Expenditure, National Treasury, September 2001; own calculations

Table 3 reveals that in contrast to national government, provincial governments are generally finding it difficult to spend their capital budgets.
· By mid-year (end of September), on aggregate provinces had only managed to spend about 29% of their capital budgets. In contrast, aggregate current spending by mid-year seems to be on track at about 47% of aggregate budget.
· There are provincial variations in the ability to spend. Those provincial governments with large capital budgets seem to be finding it difficult to spend their budgets. Gauteng and KZN seem to be particularly weak[10]. Spending to date in the Western Cape, Mpumalanga and Northern Cape seems to be satisfactory, all being above 40% of their total annual budgets.

The Intergovernmental Fiscal Review 2001, shows that government is aware of the need to improve spending in capital budgets. It is in the process of developing an analysis of the problems underlying poor spending and ways of addressing these problems. For instance, IGFR observes that planning and designing capital projects, as well as implementing them, is a complex and time-consuming activity. Planning and managing authorities do not have sufficient skills, experience and lead-time to plan and manage projects adequately[11]. The IGFR puts forward a number of possible solutions to address this problem, some of which are listed below:
· Provincial government can outsource capital planning and project management to competent consultants
· Provincial governments can create specialised capital planning and management capacity (outside of Public Works Departments, if necessary)
· National line function departments can provide norms and standards for infrastructure provision. These will allow provinces to identify backlogs and prioritise how they are to addressed more readily

2.2 HIV/AIDS allocations in MTBPS: concerns with effective spending in the National Integrated Plan

The National Integrated Plan (NIP), introduced in 2000/01, is one of government’s key programmes for introducing a range of new service delivery approaches to people infected or affected by HIV/AIDS. The NIP is also key to the promise that the MTBPS will reduce poverty and vulnerability. These new approaches are meant to complement existing services and make the entire delivery system for HIV/AIDS more cost effective.

The service programmes covered in the NIP are:
· The Life Skills programme in schools,
· Home Based Care and Support and,
· Voluntary Counselling and Testing (VCT).
It is envisaged that NGOs and CBOs will ultimately deliver many of the new services, and that existing provincial delivery agencies will support them.

The NIP makes resources for implementing these programmes available to the three core social services departments in provinces: Health, Welfare and Education. These resource flows take the form of three conditional grants, one for each of the health, welfare and education departments. At a provincial level, the NIP requires that three departments plan and manage the implementation of the NIP service programme in an “integrated” manner.

The allocation to the total NIP programme (which also covers the planning, technical support and administration costs of the national departments that manage the programme, in addition to the conditional grants) has been substantially increased in the 2001 MTBPS. The increase on previous allocations, as well as how the allocation grows over the medium period is analysed in Table 4 below.

Table 4: Comparison of allocations to the NIP in the 2001 Budget and 2001 MTBPS

R million	2000/01	2001/02	2002/03	2003/04	2004/05	% growth 2001/02 - 2002/03	% annual average growth for period reflected
2001 Budget (nominal)	75.0	125.0	300.0	304.5	not given	140.0%	69.4%
2001 Budget (real 2001 Rands)	79.8	125.0	284.9	275.9	not given	127.9%	60.5%
2001 MTBPS (nominal)	75.0	125.0	320.0	422.0	546.0	156.0%	71.0%
2001 MTBPS (real 2001 rands)	79.8	125.0	303.9	382.4	473.5	143.1%	62.4%

Sources: 2001 Budget Review; 2001 MTBPS

Table 4 shows that the MTBPS additions to NIP allocations give a considerable boost to the NIP, especially between this and next financial year where real growth is at 143% as opposed to 128% in Budget 2001. In absolute real terms, the allocation increases by R19m in 2002/03 and R106.5m in 2003/04. For the Budget 2001 allocation, average annual real growth between 2000/01 and 2003/04 is 60.5%. For the MTBPS 2001, the average real growth between 2000/01 and 2004/05 is 62.4%.

Those implementing the NIP in national and provincial government confront large, complex challenges related to the management, funding mechanisms, budgeting, and service delivery involved in a nation-wide intervention which runs across three sectors and two levels of government. In its first two years, the NIP has done a good job of setting up management infrastructure and funding mechanisms, filling co-ordinator posts, and training the staff, counsellors, and care-givers involved. But building the foundations for effective delivery of the NIP has been slow. This slow delivery is evident in the provinces’ under-spending of conditional grants, which is examined more closely below.

In the first year, conditional grants were delivered to the provinces late in the financial year, leaving little time for them to spend the money. In the second year, the process improved, but provincial capacity—in relation to financial and project management and business plan development—appears to be a problem. These observations suggest that to improve the effectiveness of spending, we should enhance the mechanisms by which provinces are funded for NIP activities and strengthen their capacity to spend on NIP programmes. We now look briefly at the nature and extent of conditional grant under-spending.

In the first year of the NIP, according to the 2001 Intergovernmental Fiscal Review, of the R 49 369 829 total allocated for HIV/AIDS conditional grants in Budget 2001/2002, R 18 million or only 36.5% was actually spent by the provinces. Table 5 examines spending of conditional grants against budget by sector in 2000/01. It shows provincial spending was best on the health grant.

Table 5: Spending on NIP conditional grants 2000/01

R million	Allocated for 2000/2001	Spending	Amount spent as percent of amount allocated
Health sector	16.819	10	59.5%
Education sector	26.93	6	22.3%
Social Development sector	5.62	2	35.6%
Total	49.369	18	36.5%

Sources: 2001 IGFR, pgs. 32, 48 & 68. 2001 Budget Review, pgs. 265, 268, & 276.

By mid-year of this year, the provinces have spent only 16.8% of the NIP conditional grants available to them. Table 6 examines NIP conditional grant spending by sector in 2001/02 as of 30 September 2001. In six months, provinces together have only managed to spend 11% of the education grant for the life skills programme and 19% of the social development grant for community and home based care and support. Provinces performed best on the health grant, on aggregate spending 27.5% of the amount available.

Table 6: Percentage actually spent of NIP conditional grants available (at 30 September 2001)

	Education	Health	Welfare	Total
Free State	59.0%	53.7%	20.6%	53.7%
Northern Cape	54.4%	12%	27.1%	23.3%
Eastern Cape	5.1%	71.2%	25.5%	21.8%
Northern Province	13.4%	36.5%	29.6%	20.8%
Western Cape	9.9%	27.9%	0.0%	15.5%
KwaZulu-Natal	7.5%	10.9%	77.1%	10.6%
Mpumalanga	2.7%	9%	0.8%	4.8%
Gauteng	2.4%	3.3%	0%	2.5%
North West Province	0%	22%	0%	8%
Total	10.7%	27.5%	18.8%	16.8%

Source: Statement of the National and Provincial Governments’ Expenditure at 30 September 2001 and own calculations.

The higher rate of spending on NIP health grants revealed in Tables 5 and 6, suggests that processes and capacity in HIV/AIDS units—at the national level and in provincial health departments—are more developed and better positioned to access and transfer the funds. Part of the reason for this is that the NIP is practically and somewhat physically still centred in health—although theoretically all three departments jointly implement the NIP. Thus it is likely that business plan development skills, financial control and reporting skills, and strategic management have been more present in health’s HIV/AIDS units, and are taking root more slowly in the HIV/AIDS areas of the provincial welfare and education departments.

Under-spending on NIP conditional grants is not an argument against allocating additional funds to mitigate the impact of HIV/AIDS. The scope and severity of the epidemic still overwhelmingly demand more than the current allocations. As mentioned above, it suggests we need to focus our attention on improving funding mechanisms and increasing the capacity of provinces to spend effectively.

In regard to funding mechanisms, this year has already seen substantial improvements. In contrast to 2000/01, provincial departments received notification of the amount they would receive in 2001/02 before the beginning of that year and the transfers have flowed timeously to the departments. Provinces also receive written feedback from the national NIP management committee about which aspects of business plans are being funded, and notification of the size and timing of funding tranches. Next year more improvements are on the way. Provincial departments will receive notification of their allocations for the entire MTEF period (not just for the year ahead) and will receive them four months before the start of 2002/03, allowing more planning and preparation before spending has to start. We suggest that national funding authorities also look at simplifying and streamlining funding application procedures and assist provinces to develop monitoring and reporting systems and formats for these grants (and other HIV/AIDS monies which provinces control and spend from centralized HIV/AIDS units).

With regard to provincial capacity to spend funds, recent research by Idasa reveals that HIV/AIDS units, which manage much of the resources specifically earmarked for HIV/AIDS, are struggling to transfer these resources to regions and districts and to induce regions/districts to spend these funds effectively and timeously. These difficulties seem to point to at least two areas of weakness:

· Firstly, units and the regions/district lack project management skills and experience.
The NIP National Working Group is already working to enhance management skills in units; its efforts should be stepped up, and if necessary more resources should be made available for this.

· Secondly, there is an insufficient number of suitable NGOs and CBOs to which units and district/region can transfer funds.
National departments and provincial units are in the process of trying to strengthen existing HIV/AIDS NGOs and CBOs, many of them new and small, but this is a time consuming process. Furthermore, many provinces do not seem to have allocated adequate resources to this end. More resources should be made available to strengthen NGOs and CBOs at national and provincial levels. Innovative NGO support schemes to open up NGO service delivery and spending need to be emphasized. In this regard, the NGO mentorship programme initiated by the National NGO Funding Unit in the HIV/AIDS Chief Directorate holds promise.

3. The inclusiveness of targeting using the norms and standards in public ordinary schools

The Norms and Standards for School Funding was introduced in 2000 as one of government’s key initiatives to improve the effectiveness of spending in school education. It aims to distribute non-personnel and non-capital spending between public schools. It has a re-distributive thrust in that the poorest of the poor schools and learners are earmarked to receive the largest chunk of non-personnel and non-capital spending. We want not only to highlight the problems of under-spending of funds allocated to the norms and standards policy but also raise concerns about the impact in practice of the targeting and redistribution associated with it.

Table 7: Under-spending on the budgeted norms and standards allocations in 2000

Province R million	Ecape	Fstate	KZN	Mpuma	Ncape	Nwest
Budgeted 2000	279.923 [12]	113.9	174.6	99.574 [13]	57.7 [14]	87.3
Total spent	3.467	67.872	131.6	63.613	R49.5	53.694
Under-spending as a % of total budgeted allocation	-	40.40%	24.60%	36.1%	0.1%	38.50%[15]

Source: Provincial education departments, 2001

The bulk of under-spending reported in Table 7 is attributed to poor public schools. Discussions with provincial education departments reveal that the main reasons for lack of spending are insufficient infrastructure facilities such as telephones, electricity, water and key teaching and communications equipment facilities. Under-spending problems were compounded by lack of capacity in Districts and School Governing Bodies.

Table 8 provides a breakdown of proposed per learner allocations on the school funding norms in 2001.

Table 8: Proposed per learner allocations for the school funding norms 2001

Quintile 1

Quintile 2

Quintile 3

Quintile 4

Quintile 5

Eastern Cape

R223

R159

R127

R96

R32

Free State

R241

R199

R163

R152

R131

R120

R100

R88

R44

R19

Gauteng

R364[16]

R258

R207

R156

R52

KZN

R216

R192

R168

R144

R120

R102

R90

R78

R54

R36

Mpumalanga

R51[17]

R32

R25

R22

R8

Northern Cape

R430

R371

R300

R230

R170

North West

R176

R154

R143

R107

R11

Western Cape

R193

R179

R158

R133

R65

Source: Provincial education departments 2001

If we look at the per learner allocations for the different quintiles in table 8 and compare them with the allocations last year (see table C in appendix), we see success in extending benefits of redress to more poor learners in 2001.

We welcome the targeting for its success in extending the benefits of redress to more poor learners in 2001. This goes a long way to improve the effectiveness of spending in school education. Our concerns are as follows: Schools in the middle of the redress table (in quintile 3) continue to bear the brunt of targeted spending. Parent communities and learners served by these schools are not noticeably different from learners and schools in the two poorest categories. Thus, in practice, income-poor learners in the third quintile are excluded from the core redress benefits of this policy. This observation suggests targeting strategies for these groups of learners need to be refined. There is ample flexibility in the policy to allow the monetary benefits of redress to be more evenly distributed to these schools. Provincial education departments indicate that greater redress benefits are likely for poor learners over the medium term. This would require targeting and distribution strategies to remain sensitive to the plight of the broadest majority of poor schools.

The problems mentioned above do not make targeting less important. Instead, they imply that provincial education departments should carefully align their funding strategies with the broader aims of educational improvement. This requires them to contextualise targeting in a way that recognizes the complexity of poverty in schooling communities.

4. Why we believe the size of the allocation to social development may not be sufficient

Poverty and vulnerability reduction are the two primary goals of the Department of Social Development (DSD) through the “provision of social grants, welfare services and development programmes in a global and human rights context.”

The following indicators show the achievements of the social development departments:
· More than 4 million individuals now benefit from the non-contributory social security system. (MTBPS 2001: )
· The more than 655 poverty relief programmes (PRP) are providing individuals, families and communities with income, skills and food security. (National Department of Social Development, Submission to Parliament 2001).
· Approximately 70% of both the non-contributory social security system and Poverty Relief Project beneficiaries are women, the majority located in rural areas, indicating the strong re-distributive thrust in the delivery of services from urban to rural areas and from men to women. (National Department of Social Development Annual Report, 2001).

The MTBPS promises more progress in alleviating poverty and vulnerability through more spending by social development departments. Social development services have been prioritized for the MTEF 2002/03-2004/05. Table 9 illustrates nominal allocations and real rates of growth in the consolidated social development budgets over the medium term. Expenditure on social development is expected to increase from R34 billion in 2001/02 to R43.6 billion in 2004/05 – an increase of R9.6 billion in nominal terms.

Table 9: The social development consolidated national and provincial budget 2002/03-2004/05

R billion	2002/03	2003/04	2004/05
Social development (Nominal)	38	41.1	43.6
Social development (Real)	36	37	39
Real Growth rates	3%	3%	6%
Welfare as % of the Social services Budget (Real)	27%	27%	27%

Source: MTBPS 2001:64.

We argue that even without allowing for any restructuring of the social security net that may be called for by the outcome of the Taylor Commission and failure of markets to reduce poverty through employment, the size of the increase to social development is insufficient. This view is based on 3 considerations.

· First, the promise to link grants to inflation on cost.
· Second, the impact that the recent Grahamstown High Court decision on back pay will have on costs. The court decided that back pay must be given from the date of application and not just for three months[18]. (Section 11 of the Social Assistance Act) by amendments to the Social Assistance Act of 1992.
· Third a range of factors that indicate a rapid increase in demand for social security and social development services.

We list these factors that indicate a rise in demand below:

First, the recent data on trends in up-take rates suggests rapid growth in demand for grants over the MTEF period. Table 10 and 11 illustrate the rapid increase in the demand for grants between March 1997 and July 2001. The annual growth of the care dependency grants increased by 80.4%, the foster care grant dependency grant (22.2%) and the old age grant (2.2%).

Table 10: Trends in beneficiary numbers

	April 1997	March 1998	March 1999	March 2000	July 2000	Average annual growth
Old age	1737682	1702647	1806493	1859197	1893077	2.2%
Disability	732322	660198	631372	611325	633777	-3.5%
Foster care	41865	43906	70650	79437	93357	22.2%
Care dependency	2895	10126	15234	24073	30628	80.4%
CSG			27577	321906	1125277

Source: Intergovernmental Fiscal Review 2001:65

Table 11 shows strong annual average growth rates of social security grants from 1997 to 2001 in poorer provinces i.e. Mpumalanga (14.70%), North West (11.50%) and the Northern Province (13.80%).

Table 11: Grant beneficiaries per province

R million	1997	1998	1999	2000	2001	Average annual growth
Eastern Cape	630514	583705	589010	661432	791211	5.80%
Free State	174727	181994	174562	176879	228644	7.00%
Gauteng	335721	323947	324758	374785	455090	7.90%
KZN	594916	617568	633627	677898	875021	10.10%
Mpumalanga	157091	164656	158780	199342	272096	14.70%
Northern Cape	110301	113592	124080	121186	110443	0.00%
Northern Province	339555	276465	343875	406858	525306	11.50%
North West	189980	194579	207211	243931	318432	13.80%
Western Cape	366719	384220	388956	384342	366208	0.00%
Total	2899524	2840726	2944859	3246634	3942451	8.00%

Source: Intergovernmental Fiscal Review 2001:65

Second, the number of child support grant beneficiaries increased from 27 577 in March 1999 to 1.5 million beneficiaries in October 2001 and it is expected that the 3 million target for the CSG could be reached by 2002/03 and not 2003/04 as planned (MTBS, 2001).

Third, the impact of HIV/AIDS suggests an increase in the demand for the child support grant and foster grant due to an increase in the number of orphans. It is estimated that South Africa will have 1 million orphans by 2005 and 2 million children younger than 15 years of age by 2010 (UNAIDS, 2001). We expect that, provided the system is changed to make it easier for children to access the foster care grant, most of the increased demand associated with HIV/AIDS will fall on this grant which has a high value

Fourth, the introduction of disability panels in rural areas should increase the take-up rate of the disability and care dependency grants over the medium term.

Fifth the recently introduced norms and standards for social security stipulate that qualifying beneficiaries must receive grants within 3 months of qualifying. In the past people had to wait up to 2 years in some provinces.

Sixth, the Constitution does not distinguish between an individual’s right to social income security shelter or protection. The White Paper for Welfare stipulates an 80:20 split in resource allocation between social security and social development programmes. Yet, the current split is 90:10 resulting in the under-funding of social development services. This has led to the under-funding of constitutionally mandated social welfare services. In July 2001 more than 9 000 children awaiting trial and sentenced were in prison. Table 12 shows the huge discrepancy between the number of children in prison and the provincial capacity of secure care facilities. Yet, the Constitution clearly states that alternative care should be provided for children guilty or suspected of criminal offences.

Table 12: Capacity of secure care facilities and number of children in prison

Province	Secure Care Facilities Capacity (1999/2000)	Number of children in prison younger than 16 years July 2001	Number of children in prison younger than 18 years July 2001
Free State	35	143	648
Eastern Cape	35	324	1268
Gauteng	470	223	1931
Kwazulu-Natal	80	436	2135
Mpumalanga	55	40	453
Northern Cape	30	56	427
Northern Province	30	19	298
North West	35	96	555
Western Cape	220	376	1613

Source: Department of Correctional Services, 2001

Seventh, the impact of HIV/AIDS will increase the demand for social development services and meeting this demand will be costly. The research paper on social security submitted to MINMEC in June 2000 found that 5-10% of orphaned children would be in need of institutional care. Non-Governmental Organisations are currently funded to a tune of R1000 child/month. The cost of caring for children in state institutions is substantially higher than caring for children within their extended family and communities. The health and education departments are responsible for implementing and delivering home and community based care for people infected and affected by HIV/AIDS as outlined in the National Integrated Plan for Children Infected and Affected by HIV/AIDS. This is in recognition of the fact that the state is not able to accommodate all children in state institutions. Conditional grants to the value of R 96 million, R131 million and R2004/05 have been allocated to both departments to HCBC in 2002/03, 2003/04 and 2004/05. However, costing under-taken and presented by both departments to Parliament in February 2001 estimated that they would need R370 million in 2002/03, R737 million in 2003/04 and R 1 099 million in 2004/05. The costing study excludes the cost of school uniforms, food-parcels, school fees etc. (make the easier to read)

Appendix

Table A: The child poverty situation in South Africa according to OHS 1999 and 1995

Province (%)	OHS 1995 Poverty rate Poverty share		OHS1999 Poverty rate Poverty share		% change in poverty rate
WC EC NC FS KZN NW G M NP	39.2 76.9 64.8 72.0 59.0 69.6 22.4 66.2 64.2	6 23 2 7 22 10 5 8 17	25.3 75.2 50.8 61.5 63.8 60.4 38.3 59.4 68.4	3 23 2 6 24 8 8 7 18	↓ 13.9 ↓ 1.7 ↓ 14 ↓ 10.5 ↑ 4.8 ↓ 9.2 ↑15.9 ↓ 6.8 ↑ 4.2
South Africa	58.8 100		59.2 99		↑ 0.4

Source: October Household Surveys (OHS) 1995 and 1999, analysis conducted by Woolard 2001.
Note: A child was counted as poor if s/he lived in one of the bottom 40% of households when ranked according to income per adult equivalent.

Table B: Employment, unemployment and economically active, 1996-1999

‘000s of people	1996	1997	1998	1999
(a)Total employed ‘000 people Employed in formal sector Employed in formal sector (STEE) Employed in formal sector in activities not in STEE Employed in informal sector Employed in agriculture Employed in domestic service	9 287 6792 5242 1550 996 759 740	9 247 6726 5139 1587 1136 717 668	9 390 6390 4945 1445 1316 935 749	10 369 6564 4840 1724 1907 1099 799
(b)Total unemployed Official definition Expanded definition	2 224 4 566	2 451 5 202	3 163 5 634	3 158 5 882
(a+b)Total economically active Based on official definition of unemployed (without discouraged work seekers) Based on the expanded definition of unemployed	11511 13853	11698 14449	12553 15024	13527 16251

Source: Statistics South Africa results from the 1999 OHS, on website.

Note: The categorisation of the data follows that of statistics South Africa:
· All formal sector activities are not covered by the survey, so the OHS formal sector employment estimates have to be added to the formal STEE estimates to get an estimate of total employment in the formal sector.
· Employment in the informal sector generally generates less income and certainly less insurance and job security, than work in the formal sector, so it is useful to present this as a separate category.
· Due to concern over the reliability of the estimates of employment in agriculture and how to classify domestic workers on the informal/formal sector scale, these types of employment are presented separately.

Table C: Per learner allocations to the norms and standards in 2000

Quintile 1

Quintile 2

Quintile 3

Quintile 4

Quintile 5

Free State

R193

R158

R152

R139

R114

Gauteng

R342

R227

R177

R129

R40

KZN

R71

R63

R55

R47

R39

R33

R29

R25

R18

R12

Mpumalanga

R223

R140

R109

R97

R32

North West

R175

R125

R100

R75

R25

Western Cape

R196

R179

R157

R140

R123

R112

R101

R84

R67

R45

R28

Source: Provincial education departments 2001

---

[1] Done in 2001 by Prof M. Samson from the Economic Policy Research Institute.

[2] Table A in the appendix presents the results of this study of child poverty which was done by Ingrid Woolard of the Univeristy of Port Elizabeth and commissioned by Idasa in 2000. It also compares the results from the OHS 1999 with those from the OHS 1995 to illustrate that the level of child poverty does not seem to have been falling. This study used the relative concept of income poverty to measure poverty, which counts a child as poor if s/he falls into one of the bottom 40% of households in South Africa’s income distribution when they are ranked according to adult equivalent income. However, we can infer that the results tell us about how many children are poor in the sense of having too little income; a study by Haarmann into child poverty in 1999 indicates that the level of income per capita in households in the bottom two quintiles puts children in these households below the absolute poverty line.

[3] With the exception of 2000.

[4] A recent analysis of real capex growth by Andre Roux of Investec shows the real capex growth is now about 3 % relative to about –10% in June 1995 and –7% in late 1999. The most recent SARB Quarterly Bulletin (September 2001:3) states that: `fixed capital formation, which lays the foundation for future economic growth, has been growing at a solid pace over the past year’.

[5] The SARB Quarterly Bulletin for March 2001 (pp14-15) recently set out the following reasons for the weak labour absorption capacity of the South African economy: i) Firms having greater preference for capital-intensive rather than labour intensive production processes (partly due to higher unit labour costs but also due to industrial action); ii) The introduction of new production technologies tends to raise the demand for a small number of highly skilled workers and at the same time lower the demand for less-skilled workers; iii) Shift in the production structure of the economy from an emphasis on primary and secondary sector activity toward the service sectors which are less reliant on larger numbers of unskilled workers; iv) The decline in the investment ratio; v) The process of right-sizing the public service in order to raise the efficiency of public service delivery.

[6] Table 2 in the Appendix reproduces the unemployment data for 1996-1999 from Stats SA that produces this story and we presented in February.

[7] For example: The SARB Quarterly Bulletin for March 2001:15-16 describes the labour market situation in 2000 as follows: “According to the latest Survey of Total Employment and Earnings (STEE) by Statistics South Africa, (private) employment totals declined again in the third quarter of 2000. Comparing the average employment level in the first three quarters of 2000 with that of the first three quarters of 1999, there was a decline of 2.7% in measured formal-sector employment, following declines of 3.7% in 1998 and 1.9% in 1999. Employment in the public sector declined (even) more than in the private sector during 2000.” And, the most recent Quarterly Bulletin tells us that:‘A brief respite in the longer-term decline in regularly surveyed formal-sector non-agricultural employment in the fourth quarter of 2000 was followed by renewed spate of job losses in the opening months of 2001….the long-term decline in employment in the non-agricultural formal (private) sectors of the economy continued in the first quarter of 2001….The contraction in public sector employment in the early months of 2001 was even more severe than in the private sector’ (September 2001:1-2 and 13-14).

[8] The spending figures discussed here should only be seen as crudely indicative of progress in actual capital spending for the following two reasons:

· Measuring the adequacy of spending progress by comparing the proportion of budget spent to the proportion of time elapsed in a financial year, for capital spending is only extremely crude benchmark. While the technique makes sense for current budgets, which should tend to be used at an even pace across the year, capital spending can be uneven because an individual capital project consist of different sized “lumps” of spending that take different amounts of time to spend. In economics, we say that capital expenditure is “lumpy”.

· In some cases, the spending data does not reflect actual spending of capital that has occurred. This may be the case when the capital budget is retained by a line function department and project amounts are only paid over a public works department when the latter claims them upon completion of the entire project. This is the case in Gauteng province. A thorough assessment of capital payment arrangements across government is required before we can rigorously interpret the numbers in table 3.

[9] This negative amount is to correct for overpayment made to the Public Works Department in the first quarter.

[10] Some of the apparent lack of progress in the Gauteng can be accounted for by the payment system for capital spending it uses, mentioned in footnote 8.

[11] Lead-time is the time between when authorities are notified about their capital budgets for a period and when they have to start spending them. It is the period in which authorities can plan and set up projects.

[12] The Eastern Cape Department of Education only allocated funding to section 21 schools on the Norms and Standards policy. The amount spent refers only to section 21 school spending, and it is thus impossible to determine the percentage under-spending in the department for section 21 or non-section 21 schools.

[13] The global amount that was set aside for Mpumalanga was R108.6 million. The education department top-sliced 10% of these funds to make provision for schools that were not correctly placed on the Resource Targeting Table and for new schools that were built after allocations were made.

[14] The Northern Cape retained a portion of the available funds at Head Office for the administration of the policy, which means that the budgeted amount was actually R49.49 million. This means that the percentage in the table refers to small over-spending instead of under-spending.

[15] The North West Department of Education also had over-expenditure in schools that were not classified as poor and the amount totals R12 864 million.

[16] The per learner allocations for Gauteng represent the average allocation in a quintile. This means that there are learners in all these quintiles that receive more or less than the average indicated in the table. The same goes for Northern and Western Cape.

[17] The figures for Mpumalanga exclude R143 million, which has been centralised and not distributed according to the resource targeting table.

[18] See section of the Social Assistance Act of 1992.