Centre for Scientific and Industrial Research (CSIR)
The upgrading, maintenance and rehabilitation of public sector assets
Andrew Merrifield (amerrifi@csir.co.za) Division of Building and Construction Technology, CSIR

Introduction
The urgency to upgrade, maintain or even rehabilitate the public sector capital assets has been identified as a potential risk to the fiscus. During the MTEF Review, the Review Team requested information from national departments on infrastructure backlogs with particular emphasis on trying to identify state of the country’s asset base and estimated costs to upgrade, maintain and/or rehabilitate their assets.

The asset status of the Departments of Public Works (DPW), Transport (DoT), Health (DoH), and Education (DoE) has been analysed because these Departments were able to supply information on their assets. It is assumed that other state infrastructure is a similar state of disrepair, but we were unable to get adequate data from the responsible agencies during the period of the MTEF Review.

Infrastructure Backlogs (upgrade, maintain and rehabilitate)
The general extents of backlog and potential cost in nominal terms to the fiscus are summarised in the table below. The number of years required to remove the backlogs is estimated by the departments themselves or the consultants employed to calculate these for them.

 

Total backlog and estimated years of spending required

Annual amount not currently budgeted for

Public Works

R8.8 billion over five years

R1.5 billion

Health

R13 billion over ten years

R1.3 billion (Ave.) R2.4 (max)

Education

R14-20 billion over nine years

R1.6 billion

Road Transport

R11 billion over ten years

R1.1- 3.4 billion

Total

R 46.8 –52.8 billion

R5.5-7.8 billion per annum


It can be argued that if the upgrading, maintenance and rehabilitation of public sector assets is not addressed with some urgency, the nation risks not only losing its resources, but also continued neglect could seriously impair the nation’s ability to meet its social and economic policy objectives. This argument is generally accepted by political and administrative decision-makers in the public sector. Yet, historically budget constraints have meant that insufficient resources were allocated to addressing the problem.

To understand the budgetary implications for this neglect, it is worth noting that historical evidence and research conducted in the Provinces for the MTEF Review confirmed that the pressures of current expenditure have traditionally crowded out capital expenditure. In South Africa, as elsewhere, capital budgets are only allocated once personnel and statutory obligations have already been provided for. This has meant that infrastructure delivery targets have been sacrificed or delayed in an attempt to keep public spending within sustainable fiscal limits. Such practices have given rise to the growth of the current infrastructure backlogs that exacerbate other historical inequalities.

The unseen element of the problem, is that while new capital expenditure is subordinated to current spending pressures, there is even less allocated to maintenance and rehabilitation of existing assets. The traditional budgeting practice of accounting for maintenance as part of current expenditure has meant that it has been even easier to shift these allocations to cover overspending in other areas of current expenditure. Fortunately, the shift to Generally Recognised Accounting Practices (GRAP), and especially to the resource costing of capital expenditure and the inclusion of maintenance under capital expenditure, should ensure that this problem becomes more visible in the future. In terms of GRAP, future budgets will have to account for these costs, so that lack of maintenance and rehabilitation will be reflected in terms of depreciated assets.

A final point to be noted about the rehabilitation and maintenance problem is that most physical assets (buildings, roads, structures) can operate with minor maintenance expenditure (less than 5% of value) for long periods of time. Most physical assets can even survive for many years (ten to twenty years) without any maintenance. However, such neglect means that the very fabric of the structure is eventually threatened by this lack of maintenance, and within a very short time afterwards, the asset needs complete replacement.

Roads are a good example of this relationship. Roads in good condition need resealing every five years to maintain their structural integrity (cost R50 000 per lane/km). If resealing is not done in time their condition falls from fair to poor, requiring rehabilitation to restore them (cost R200 000 per lane/km). If no rehabilitation occurs, the poor condition deteriorates to very poor, requiring reconstruction (cost R800 000 per lane/km). This example suggests that maintenance is sixteen times cheaper than reconstruction.

Although buildings do not deteriorate in precisely same manner as roads, there comes a stage when lack of maintenance means that the fabric of such structures is destroyed. Recent examples of the need to replace buildings due to neglect, is the replacement of the Hendrik Verwoerd Hospital by the new academic hospital in Pretoria (R200 million), and the replacement of the King George Hospital by an new academic hospital in Cato Manor in Durban (R500 million).
Estimating the Extent of the Problem

Department of Public Works
The Department of Public Works’ (DPW) budget and actual spending for the MTEF review period is given below.

Maintenance versus New Construction Expenditure (Current R’000)

Budget

1994/95

1995/96

1996/97

1997/98

1998/99

1. New Construction

347 678

389 057

465 257

602 180

1 087 929

2. Maintenance

346 626

308 768

466 849

490 939

526 900

2/1 as %

99

79

101

82

48

Actua

         

3. New Construction

316 188

424 922

453 898

625 047

159 230

4. Maintenance

357 069

337 842

440 635

517 727

81 616

4/3 as %

113

80

97

83

51


The table suggests that the proportion of the budget allocated to maintenance has declined significantly in recent years, and this trend is mirrored in actual spending. The DPW noted, in a submission accompanying the above data, that it "serves no purpose to add to the country’s infrastructure if funds are not made available to maintain existing infrastructure". Noting that provision was being made in the MTEF for a R1 billion budget for new capital expenditure, the DPW went on to say that "[i]t therefore appears that a higher priority is given to adding to the state’s property portfolio than preventing the existing portfolio from deterioration and eventual disintegration".

The DPW commissioned a study to determine the condition of the public sector portfolio under its control (including general government accommodation, military and criminal justice facilities). It received the following estimates,

Public Works Maintenance Backlog (Current Prices)

Type of Structure

Maintenance Backlog

Military Bases

R5 560 million

Prisons

R725 million

Magistrates Courts

R386 million

Administration offices

R109 million

Police stations

R944 million

Total (nominal cost)

R7 725 million

Total (inflated costs over 5 years)

R8 787 million


The study indicates that an average budget allocation of R1.5 billion over and above the current DPW allocation will be needed for the next five years to eliminate the identified backlog. However, it is possible to assume that this may be an under-estimate, since their study only focused on the maintenance backlog and did not look at refurbishment or upgrading. It also did not account for the need to maintain new structures erected during the above period and it did not include an estimate for future maintenance of the public sector estate.

Department of Health
The Department of Health (DoH) undertook a comprehensive facilities audit in 1996, enabling it to make a fully informed assessment of its requirements. The audit showed that at least R8 billion was required to rehabilitate existing facilities and, if this money was not spent, that possibly 30% of the existing health infrastructure will soon be beyond repair. The DoH proposed a ten-year rehabilitation programme, which would require their budget allocation to gradually increase to about R1.6 to R 1.8 billion before declining again.

However, on top of the rehabilitation requirement, the DoH estimated they need another R5 billion over the next ten years to maintain their facilities. During this period, the budget would need to allocate an additional R500-R600 million per annum to pay for maintenance and ensure that the existing and new health facilities remained fully operational. Since Health is a Provincial responsibility, these requirements would ultimately need to come out of existing Provincial budgets. Unfortunately the Provincial visits undertaken by the Review Team suggest that, with the possible exception of Gauteng, there is no indication that Provinces are able to make adequate provision for such capital expenditure.

Department of Education
The Department of Education (DoE) controls a large proportion of public sector estate in the form of classrooms and other educational facilities. Their ‘Schools -Register of Needs’ provides a basis for estimating that R14.6 billion is required to address the maintenance and rehabilitation backlog. Using a time horizon of nine years (2005), an annual budget of R1.6 billion will be required to address the backlog. Similar to the case of Health, this expenditure should come from the Provincial budgets, but indications suggest that it is unlikely that most Provinces have the resources to allocate sufficient funds to address these problems.

The Register of Needs does however provide a clear picture of the state of the nation’s schools.

The Condition of schools

 

Good & Excellent

Minor repairs

Very weak & Weak

Eastern Cape

40%

45%

15%

Free State

38%

46%

16%

Gauteng

58%

36%

6%

KwaZulu Natal

29%

48%

23%

Mpumalanga

41%

50%

8%

North West

48%

42%

9%

Northern Cape

40%

49%

11%

Northern Province

34%

25%

41%

Western Cape

80%

19%

1%

National

42%

40%

18%


The table indicates that only 42% of the nation’s schools are in good to excellent condition (requiring perhaps 5-10% of replacement cost for maintenance). Another 40% are in need of minor repairs (requiring perhaps 25-30% of replacement cost) while the remaining 18% are in poor to very poor conditions (requiring perhaps 50-100% of replacement cost). On the basis of an estimate of a total worth of between R50-60 billion for the Education estate, and using the Health audit’s estimate that one third of the estate needs replacement, a more likely estimate is that the Education rehabilitation and maintenance backlog could be as high as R18-20 billion.

Department of Transport
The Department of Transport (DoT) has just completed an assessment of the rehabilitation and maintenance costs of the national, provincial and local road infrastructure as part of their ‘Moving South Africa’ project. Unlike many other asset categories, roads infrastructure has been closely monitored by the former Provincial administrations, and in most areas, a reliable asset register indicating the condition of the roads infrastructure remains6. The DoT estimated that it has a total backlog of around R38 billion and a maintenance and rehabilitation backlog of R11 billion. Its annual new construction, maintenance and rehabilitation requirements in the following table.

Annual requirements for roads funding (Rmillion, current prices)

 

Upgrading Gap (new capex and M&R)

Average per annum needs to maintain current service levels

 

Current Spend

Back-log (est. 10 years)

New Capex

Maint and Rehab

Support

Interest

Total Need

National

470

237

179

536

147

-

1 099

Provincial

2 531

2 144

1 369

3 606

800

-

7 919

Local

4 143

1 413

1 002

1 161

433

314

4323

Toll

1 100

-

1 020

273

182

805

2 280

Total

8 244

3 794

3 570

5 576

1 562

1 119

15 621


The table reflects that there is possibly a R5.1 billion shortfall in the current funding of public sector roads (R15.6 billion - R2.3 billion (toll roads) - R8.2 billion (current spending)). If neither new construction nor backlogs were addressed, it may be possible that the existing spending could cover the rehabilitation and maintenance, support and interest (R8.3 billion p.a.). However this is unlikely to be desirable or acceptable given the economic policy objectives of the current Government.

If the infrastructure backlogs, support and interest payments are covered by current spending, then only R1.8 billion is available for current maintenance and rehabilitation, leaving and annual maintenance and rehabilitation shortfall of R3.7 billion. If the total maintenance and rehabilitation backlog of R11 billion is averaged over ten years however, at R1.1 billion, then either the annual allocation for new capital expenditure and/or the reduction in infrastructure backlogs would not be adequately funded.

The table also shows that the annual requirement for rehabilitation and maintenance is 56% more than that for new construction, reflecting perhaps the need to preserve an already established infrastructure base. It is also worth noting that Provincial roads require the most rehabilitation and maintenance (65% of the total maintenance and rehabilitation requirement). The experience of Provincial visits suggested that, with the possible exception of Gauteng, this investment is unlikely to be provided for. In Kwazulu alone, the current year’s rehabilitation and maintenance budget is approximately half of what is required and most of the Provinces road infrastructure has not received routine maintenance for up to 17 years.

Conclusion
The above information indicates that there is a grave problem with the state of nation’s infrastructure, and if maintenance and rehabilitation continues to be under-funded, we run the risk of losing a significant proportion of our physical assets. If this were to happen, it would not only be difficult for the Government to ever meet its ambitious service delivery targets, but the efficiency of our economy could be seriously impaired.

It is not only important that Parliament understands the full budgetary impact of the maintenance and rehabilitation backlog, but they begin to do address it. The estimates given above are merely an indication of a much bigger problem, since only a portion of the nation’s upgrading, maintenance and rehabilitation backlog have been accurately assessed. Anecdotal evidence from other sectors suggests that their infrastructure could be in even worse condition. Therefore, if the full extent of the backlog were taken into account in the budget, we would need to increase infrastructure spending to at least double the current budget amount.