National Offender Management Service: Maintenance of the prison estate in England and Wales

Executive Summary

National Audit Office Report

1. An increasing prison population, frequent overcrowding and a high turnover of prisoners combine to create substantial pressures on the prison estate. There is a high demand for services such as water and heating, and a high level of wear and tear on fixtures, as well as on the building fabric. Maintaining the estate in a secure and well-ordered condition under these circumstances requires effective planning and delivery of large scale maintenance projects, and responsive local maintenance teams to keep services and accommodation operational at all times.
   
   
2. The estate is accommodating an increasing number of prisoners: excluding privately-run prisons, over 73,000 people were held in custody in 2007-08 in England and Wales, up from 69,000 in 2005-06; and prisons are subject to constant high levels of prisoner occupancy. Accordingly the total estate is continuing to grow and a major capacity building programme is underway. Some old prison wings have been demolished with new wings constructed on the same site, while old wings have received complete or partial refurbishment to extend their life and to meet new standards.
   
   
3. In May 2007, the Government created the Ministry of Justice. A subsequent review resulted in organisational changes effective from 1 April 2008, including the formation of the National Offender Management Service Executive Agency (the Agency). The Agency combined large parts of the former National Offender Management Service headquarters, HM Prison Service and the National Probation Service into one body. It aims to deliver more effective offender management and to strengthen and streamline commissioning of services for offenders from the public, private and third tier sectors, with the goal of improving efficiency and effectiveness. The Agency, through HM Prison Service (the Prison Service), operates and maintains a large and complex estate of 129 prisons in England and Wales. Nearly 50 per cent of current prisons were originally opened in or prior to the 19th century and the oldest building still in use is at HM Prison Lancaster Castle which dates back to 1200. The newest buildings opened in the current decade. Prisons vary greatly in their form and functions as well as in age and size.
   
   

   Maintenance of the estate

   
   
4. Maintenance tasks range from day to day repairs arising from wear and tear or vandalism, planned inspections, and preventive and corrective work, through to the complete refurbishment or the renewal of major assets, such as heating and sewage systems. In 2007 08, the Agency spent around £320 million maintaining the prison estate, down from an estimated £330 million in 2005-06 (in 2007-08 prices), despite an increasing prisoner population. This expenditure also covers maintenance staff, small-scale routine maintenance undertaken by local teams, and major refurbishments of whole prison wings and building services infrastructure.
   
   

   Scope of our study

   
   
5. In this report, we consider whether the Agency has clear aims for prison maintenance, whether it is using the estate maintenance budget efficiently and effectively, and whether it is maintaining the prison estate adequately. This examination covers maintenance of the existing prison estate only. It does not examine the procurement and construction of new prisons or wings to hold the increasing population as part of the Ministry of Justice’s capacity building programme. The examination covers the physical aspects of the current prison estate only. It does not assess overall prison regimes or decency standards, as these are affected by a wider range of factors such as prison operations, staffing and services for prisoners.
   
   

   Key findings

   
   
   

   On the physical condition of the estate

   
   
6. We employed professional chartered building surveyors, who had extensive experience of the prison estate, to accompany us on each of our eight prison visits to help us assess these buildings. In these prisons, the buildings were generally: weather-tight; structurally sound with no evidence of any substantial subsidence or structural movement; and reasonably well decorated. The Governors, prison officers, maintenance staff and the prisoners also generally reinforced this view during our eight prison visits.
   
   

   On the management of prison maintenance

   
   
7. Overall, the Agency has a strong management system for prison maintenance; has clear quality standards for prison buildings and plant; a system for prioritising maintenance programmes for each of the Prison Service’s 12 Areas (11 geographically-based and a single High Security Area with prisons distributed across England and Wales); and devolved management structures for maintenance to Areas and individual prisons.
   
   
8. The Agency includes its highest priority, large-scale projects in its planned maintenance programme for each year with an intended start date for each project. It may defer the actual start dates for some projects, however, should no funding be available. Other high priority approved maintenance projects within the five year forward plan, particularly smaller-scale ones, may be brought forward to draw on any remaining funds towards the end of the financial year. In our visits to prisons, Governors, Area Estate Coordinators (responsible for the overall maintenance of prisons) and works teams considered that projects brought forward are not always those ranked the highest in the overall list of priorities, although these projects are from the five year maintenance plan which includes only priority projects.
   
   
9. Changes to the start dates of major projects cause uncertainty to the prisons concerned, and to Area Estate Coordinators, over when major projects will begin. This uncertainty creates difficulties in planning the amount and cost of ongoing maintenance work they must still carry out on assets until the major project begins. The Agency is responsible for advising the prisons and Area Estate Coordinators on the actual start dates of planned work. The primary reason for delaying major maintenance projects is population pressures, which limit the space available to move prisoners out of wings requiring refurbishment to alternative prison accommodation.
   
   
10. The Agency considers the operational life, and additional maintenance resources required, of specific refurbishment projects. It does not, however, have long-term plans for managing the economic life of individual plant machinery and other individual assets. It therefore has a limited understanding of the most cost-effective times to switch from servicing and repairs of particular assets to complete refurbishment or replacement.
    
    
11. The Agency recommends planned visual checks in line with manufacturers’ recommended maintenance requirements. The Prison Service is implementing a Service Focussed Maintenance regime which aims to give greater flexibility in visual checks, a greater emphasis on the effective use of resources, and is designed to encourage maintenance teams to use their discretion and assess the opportunities and risks from deferring visual checks that add little value. At March 2009, Service Focussed Maintenance had been introduced in half the prisons in the estate, and in four of the eight prisons we visited. The Prison Service expects to complete the roll-out to the remainder of the estate by May 2009. But even in some prisons in which Service Focussed Maintenance was in place, maintenance teams are still receiving orders from their computer-aided maintenance package, which records maintenance work and repairs (Planet FM), to carry out visual checks on assets which are in continuous use and where any failure would be immediately reported by those affected. Greater adoption of the flexibility in the application of visual checks offered by Service Focussed Maintenance should enable maintenance teams to carry out more corrective or preventive work and servicing, and improve value for money.
    
    
12. We used the Agency’s major maintenance project database to review how contractors were employed nationally and across the 12 Areas. In each Area, more than one company was carrying out major maintenance contracts, to avoid monopolising provision.
    
    

    On the links between refurbishing the estate and maintaining it

    
    
13. During our eight visits to prisons, we found there had been past instances of limited handover arrangements between external contractors and prison maintenance teams following the completion of refurbishment or maintenance work. Poor handovers had resulted in difficulties in ongoing maintenance by the local teams, who have also needed to correct defects in place at the time of handover. In 2008, the Agency introduced a procedure which aimed to remedy this problem, by highlighting the importance of effective handovers. The new procedure aims to strengthen handover arrangements between external contractors and prison works teams for all new capacity and major maintenance projects. The external contractors are required to remedy any reported defects before the handover is signed off.
    
    
14. Prior to the introduction of the new handover procedures, prison maintenance teams had not been routinely consulted in the design and construction of refurbished parts of the estate. Although it is too early to report on the extent to which the new procedures have bedded in, consultation does now take place between senior Prison Service officials, prison maintenance managers and contractors. While the Prison Service assesses the funding required to cover future maintenance, closer joint working between contractors and local maintenance teams gives scope to include proposals for low cost ongoing maintenance and to improve value for money.
    
    
15. The Agency has developed a range of technical specifications for common adoption in refurbishing the estate. These feature in its technical manuals and are raised in discussions with external contractors. Within prisons, however, we found little standardisation of the parts, materials, fixtures and fittings used, even on recently refurbished wings. There was considerable evidence of different specifications and parts used by contractors within and between prisons, a view confirmed by Area Estate Coordinators and site maintenance teams.
    
    

    On performance management of prison maintenance

    
    
16. The Agency has the ability to monitor and manage performance if Planet FM is used consistently, key maintenance fields are completed and performance targets are effectively designed. The Agency does not, however, routinely analyse the type, number or location of prison maintenance tasks over time, or how much is being, and has been, spent in total and by each prison on maintenance. The lack of such analysis substantially restricts the Agency’s ability to assess the maintenance funding required in future years or the cost-effectiveness of the various maintenance delivery structures. Such analysis would enable the Prison Service to produce robust financial and performance management information, allowing Area and senior managers to understand and manage better the overall maintenance of the prison estate.
    
    
17. The Agency has limited ability to monitor and manage the performance of maintenance work because key maintenance performance targets are not effectively designed. A target for 100 per cent completion of all planned maintenance work, for example, is automatically met as the system only counts planned maintenance tasks issued to staff, and not all tasks logged. Prison works managers only issue maintenance tasks which they are sure maintenance teams can complete.
    
    

    Conclusion on value for money

    
    
18. The Agency has obtained good value for money from its expenditure on prison maintenance and was maintaining the prisons we visited sufficiently well to preserve physical security, prisoner capacity, prisoner and staff safety, and their own and legal standards. The Agency has introduced procedures aimed at improving the handover of major maintenance projects from external contractors and to assess the future maintenance costs of refurbished or replacement assets.
    
    
19. Our findings from benchmarking estate management externally, and our review of internal standards, structures and funding arrangements, indicate that the Agency does not plan the maintenance of assets over their whole economic life, or prioritise and schedule major maintenance work as robustly as it might. Nor does it monitor information on asset management and manage risks fully which would aid the more effective operation of assets. Long term maintenance is complicated by a lack of consultation between those designing and constructing large scale refurbishments, and the prison maintenance teams charged with maintaining them. There is also little standardisation of parts and fittings. The effective rollout of Service Focussed Maintenance would bring flexibility in scheduling planned inspections and allow maintenance teams to give greater emphasis to corrective or preventive maintenance, helping the Agency meet the maintenance priorities for the estate.
    
    

    Recommendations

    
    
20. We recommend as follows:
    
    
    1. Flexing the start times of approved major projects to avoid expenditure over-runs can affect maintenance priorities across the financial year. As a result, some projects may commence that do not always rank highest in the overall list of priorities. The Agency should have a robust and consistent system for deciding which high priority major approved projects to start on and for ranking projects relative to each other. Decisions to defer any of these projects when the total estimated costs of all approved projects exceed maintenance budgets in any year should reflect relative priorities and an assessment of the relative costs of holding over projects.
       
       
    2. The Agency does not have long-term maintenance plans for individual assets over their economic life and does not have a full understanding of the optimal times to switch from servicing and repair of an asset to its complete refurbishment or replacement. The Agency should develop long term plans for maintaining plant, equipment and other assets, over their economic life, in line with its planning for the maintenance needs of major refurbishments over their economic life. Developing long-term plans for individual assets would help the Agency make explicit decisions between funding a maintenance task now, delaying it until a later date, or other options such as funding the asset’s refurbishment or replacement. Better informed decisions would contribute to reduced total maintenance costs in the long term.
       
       
    3. The Agency recommends planned visual checks of assets in line with manufacturers’ recommended maintenance requirements, but many assets do not give an early indication of future failure, and the failure of some assets is obvious when it occurs and are effectively self reporting. The checks divert resources and offer less value than corrective or preventive servicing maintenance, or carrying out reactive maintenance tasks if the asset fails to work. In completing the roll-out of Service Focussed Maintenance across the estate, local maintenance teams should use the flexibility it provides to defer visual checks of plant or equipment unless these are part of taking corrective or preventive actions.
       
       
    4. The Agency does not make the most of joint working and consultation at all levels between those designing and building new or refurbished parts of the estate and those responsible for operating and maintaining the estate. Within the Agency’s developing integrated estate function, it should:
       • reduce the costs of assets over their whole working life, through including low-cost maintenance options into the initial design, building plans and contracts for refurbished and replacement assets; and
       • adopt in full its new procedures for handover between external contractors and prison maintenance teams to improve the transfer of knowledge to prison maintenance teams on the infrastructure and services in the buildings and the materials, fixtures and fittings used, and to confirm the effective working of assets and the correction of any defects prior to handover.
         
         
    5. The Agency has developed technical specifications for prison buildings, equipment and fixtures and technical manuals for contractors, but a wide variety of specifications and materials has been used in recent refurbishments. Centrally, the Agency has shifted towards performance specification, leading to a wide variety of solutions, driven by a desire to ensure a greater degree of competition during the procurement of projects. The Agency should use whole lifecycle costing to optimise and reduce the number of different products being used in projects. It should require contractors to adhere to its technical specifications and to use its approved materials and products on refurbishment projects. Consistent use of such supplies would provide greater buying power, economies of scale, commonality of spares and common training of the staff charged with maintenance and repair of the assets.
       
       
    6. The Agency does not implement a systematic process for analysing the type, number or location of prison maintenance tasks over time, or how much is, and has been, spent in total and by each prison on maintenance. The Agency collects most of the required information, but stores it in several separate databases that are not joined up.
       The Agency should:
       • improve the links between the various databases it uses to store information on maintenance tasks, in order to improve the monitoring of maintenance projects, allow analysis of prison workloads and the identification of trends, and to identify examples of good practice or where improvements can be made; and
       • monitor overall maintenance demands over time through more systematic analysis of its Planet FM and Work Package Management System data, so as to improve its understanding of the performance of maintenance projects over time.
         
         
    7. The Agency has the ability to monitor and manage performance if Planet FM is used consistently, key maintenance fields are completed and performance targets are effectively designed. Key maintenance performance targets are not, however, effectively designed. For example, the 100 per cent target for completion of all planned maintenance work is automatically met, as the system only counts planned maintenance tasks issued to staff, and not all tasks logged. The Agency should develop robust Key Performance Targets for planned and reactive maintenance, capable of determining real differences in practice and performance across Areas. Guidance from the National Audit Office, Cabinet Office and HM Treasury [Footnote 1] highlights the criteria for effective key performance targets. Areas should develop consistent reporting of Service Delivery Agreement data.
       
       
    8. The Agency is hampered in assessing maintenance funding required in future years or the cost-effectiveness of the differing structures across Areas for managing prison maintenance by not having as full a picture as it could of current spending. The Agency should examine the Estate Planning Tool and Estate Performance Measurement System which the Ministry of Defence has developed to manage its own estate. Such a system should help track the condition of the prison estate over time and allow the Agency to estimate the performance of contractors [Footnote 2].
       
       
    9. The Agency collects most of the information necessary to monitor the progress of maintenance work, but stores this information in several separate databases that are not joined up. The resulting difficulty in combining and analysing performance data makes it difficult for the Agency to assess how well prisons are performing in carrying out maintenance projects. The Agency should improve the links between the various databases it uses to store information on maintenance tasks, in order to improve the monitoring of maintenance projects, allow analysis of prison workloads and the identification of trends, and to identify examples of good practice or where improvements can be made.
       
       
    10. There are wide differences between central performance data and reports from prison maintenance teams on the degree to which vandalism by prisoners causes reactive maintenance work. The Prison Service is unable to identify centrally the overall amount and cost of maintenance work resulting from vandalism and does not know which prisons have the highest disturbance and vandalism levels. Maintenance staff should always fill in the fields on Planet FM recording the reason for maintenance work to improve the reliability of information. The Prison Service should use this information to see how far maintenance costs reflect issues of prison discipline, so that action can be taken to limit the damage and related maintenance costs.
        
        
21. Further recommendations taken from previous NAO reports which are relevant to the findings of this report can be found in Appendix 3.

Footnotes

1. Setting Key Targets for Executive Agencies: A Guide, HM Treasury, Cabinet Office, NAO November 2003. [Back from Footnote 1]
   
   
2. Managing the Defence Estate: Quality and sustainability (HC 154, 23 March 2007). [Back from Footnote 2]