The Committee of Public Accounts has agreed to the following Report:
MANAGING THE RELATIONSHIP TO SECURE A SUCCESSFUL PARTNERSHIP IN PFI PROJECTS
INTRODUCTION AND LIST OF CONCLUSIONS AND RECOMMENDATIONS
- The Private Finance Initiative (PFI) is now established as a major form of Government procurement. There are over 400 PFI contracts currently in force committing departments to future expenditure of around £100 billion.
- These contracts are generally long term arrangements involving public expenditure over extended periods, often for 30 years or more. To achieve value for money over the life of these contracts, the public sector clients (often referred to in this context as "authorities") will need to have a strong contractual framework allied with good relationship skills which will help them to approach projects in a spirit of partnership with their private sector partners.
- On the basis of a Report by the Comptroller and Auditor General the Committee took evidence from the Office of Government Commerce (OGC) and the Major Contractors Group. The Comptroller and Auditor General's Report is based on a survey by the National Audit Office (NAO) of 121 PFI projects where contracts had been let prior to 2000. The survey results include perceptions of authorities and contractors about the early experience of how the contracts they have entered into are working out. We examined the evaluation of PFI projects in progress, how value for money can be maintained in the long term and the skills and guidance that are needed by the public sector to manage these contracts successfully.
- This Report focuses on the early experience of PFI contracts. It is, therefore, a transitional report highlighting issues which will be important to the successful future management of these long term contracts.
- Our key conclusions are:
- Better evaluation is needed of PFI projects in progress. There are 400 PFI contracts now in progress with many more contracts being negotiated. Departments need to analyse rigorously whether their PFI projects are delivering the quality of customer service and value for money expected when the contracts were let. The results of these evaluations should also be monitored by the OGC to enable it to refine the government's approach to the development of future PFI projects.
- Value for money needs to be maintained over the life of these long term contracts. We are very concerned that over one in five authorities consider that value for money from their PFI contracts has diminished, with high prices for additional services an area of concern. As many as 23 per cent of authorities surveyed considered that there had been a decline in value for money in PFI projects after contract letting. It is essential that PFI contracts have appropriate mechanisms in place to ensure that value for money is maintained over the lifetime of a project. Yet only around half of the contracts surveyed had such mechanisms in place. In future all contracts should have appropriate mechanisms, such as benchmarking, market testing, and open book accounting.
- Contractors should expect to lose their investment in PFI projects when things go wrong and to be rewarded reasonably when things go well. If contractors successfully manage the risks that have been allocated to them and deliver the required services then they will expect to earn rewards commensurate with the level of risk they have borne. But if they fail to manage the risks they have taken on then they should expect that part or all of their equity investment in the project may be lost. It will undermine an essential commercial discipline if contractors generally are given the impression that the Government will always bail them out, as has occurred in some individual cases, such as the Royal Armouries Museum or the Channel Tunnel Rail Link.
- Staff responsible for managing PFI projects must be equipped with the appropriate skills. Over £100 billion of public funds is committed to PFI projects, yet the OGC recognises that there are still gaps in the guidance and training on how to manage PFI projects, as distinct from how to negotiate them at the outset.
- Our detailed conclusions and recommendations are as follows:
On the evaluation of PFI projects in progress
- Post-implementation reviews are particularly important for projects where perceived value for money has declined since contract letting. Departments need to identify whether such a decline reflects errors of judgement by the authority when letting the contract, the contractor failing to deliver the service as promised, short term problems during the early period of the service delivery, or other factors such as high charges for additional services.
- There is very little information available on the returns which private sector contractors earn on PFI projects. It is impossible to tell whether these are reasonable in relation to the quality of the service delivery and the risks which the contractors bear. The OGC should complete its current review of contractors' returns as a matter of priority.
On maintaining value for money in the long term
- As many as 23 per cent of authorities surveyed considered that there had been a decline in value for money in PFI projects after contract letting. Yet only around half of the contracts surveyed had mechanisms for ensuring continued value for money over the lifetime of the project such as benchmarking and open book accounting. All contracts should have appropriate mechanisms in future.
- The survey showed a very low proportion of authorities, just 15 per cent of those surveyed, with arrangements to share in refinancing gains. Refinancing can give rise to excessive returns to the private sector from PFI deals, beyond even the private sector's reasonable expectations. The OGC should introduce its new guidance on refinancing as quickly as possible to ensure that authorities have contractual rights to share in refinancing gains.
- PFI contracts often contain specific procedures for the parties to vary the deal, and 55 per cent of authorities with such change procedures had already used them in the early years of their PFI contracts. Long term contracts must provide room for flexibility in the face of changing circumstances. But concern has arisen over high charges for additional services, suggesting that authorities need to watch that change procedures are not abused as a covert means for increasing the profit margins of the contractors.
- The fact that 58 per cent of authorities with a performance review process had made performance deductions from payments due to PFI contractors suggests many authorities are not getting the service they require. If bids are priced on the assumption that actual performance will fall short of the required level, then contractors may not have a strong incentive to perform well. It is up to departments to ensure that their PFI contracts do not accommodate persistent under-performance.
- The transfer of risk inherent in a PFI deal cannot protect the authority from the risk that the private sector simply fails to deliver what may be a key public service. It is essential that the authority actively manages this ultimately untransferable business risk.
- The essence of PFI deals is that the private sector contractor should take appropriate risks in return for appropriate rewards (such as the risk on volume of users in return for a share of user revenue). It is not for the public sector authority to insulate the contractor from the consequences of the risks it has been paid to take on. The public sector should certainly not reward private sector failure by agreeing to reduce the risk of the contractors losing their equity investment when the private sector has not delivered, as occurred in the case of the Royal Armouries and the Channel Tunnel Rail Link.
On skills and guidance
- Some authorities provide little or no training on contract management and there appear to be significant shortcomings in authorities' current approach to managing PFI contracts. The OGC should introduce further guidance and training on the key principles of good contract management, on evaluating the value for money of PFI projects in progress, and on mechanisms for maintaining value for money. Drawing on illustrations from successful PFI projects, the OGC could show how to combine insistence by the authority on getting what it paid for with a partnership-based approach throughout a long-term deal.
- Familiarity with the project and how the contract is intended to operate are essential requirements for any staff engaged in managing PFI contracts. Staff continuity between the procurement and the subsequent management of the contract is desirable. Where this is not possible, there should be a gradual hand-over between the staff who negotiated the deal and those who will be responsible for post-contract management to ensure that there is continuity in the authority's knowledge and understanding of the project.
- Public sector officials involved in developing and managing PFI projects need commercial awareness, and the Civil Service has taken steps to recruit staff from the private sector. Departments should consider whether such staff would benefit from training in the proper conduct of public business.
EVALUATION OF PFI PROJECTS IN PROGRESS
Perceived value for money
- Responses to the NAO's survey of 121 PFI projects where contracts had been let prior to 2000 showed that 81 per cent of authorities said that the value for money of their PFI projects was currently satisfactory or better. But 15 per cent said value for money was marginal and 4 per cent said it was poor (Figure 1). And 23 per cent of authorities who gave a perception of value for money considered that it was not as good as at contract letting. We noted that the positive perceptions given by many authorities were based on a subjective impressions of what was happening on their PFI projects. Having entered into the contracts authorities were not likely to admit that they had got it wrong.
Figure 1: Authorities' perceptions of their projects' value for money
Based on 98 authorities who gave their perception of value for money at the time the contract was let, as well as at the time the survey was completed.
Source: NAO survey
- The OGC attributed the fact that the majority of projects were perceived as delivering at least satisfactory value for money to competent staff applying good practice and guidance. In these projects the management of value for money was a priority, there was focus on outputs and service levels, and specialist experts such as the former Treasury Taskforce and Partnerships UK helped to keep the projects on track. However, there was no room for complacency in the level of achievement to date. Once any project, whether PFI or non-PFI, became operational the client's perception of value for money might well change as factors in the original evaluation related to the construction phase of the project ceased to be relevant, and other aspects of quality assumed greater importance. The client's view of the relative importance of monetary and non-monetary considerations could also change over the life of the project.
Monitoring of value for money
- There are 400 PFI contracts currently in force committing departments to future expenditure of around £100 billion. But we have seen no rigorous assessment of value for money on PFI contracts in force, or empirical evidence as to whether they are good value for money or not. Every pre-contract appraisal of any substance should indicate how the proposals concerned will be evaluated after completion and how the results of the evaluation will be disseminated. While overall responsibility for looking at individual projects rested with the relevant departmental accounting officers, OGC was obtaining feedback from a number of sources: the post-contract stage of its Gateway Review process, Partnerships UK, reports from the National Audit Office and other parties on individual projects and the OGC's networking with departmental private finance units.
Reductions in perceived value for money
- The 23 per cent of authorities surveyed by the NAO which considered that value for money was not as good as at contract letting gave a variety of reasons including high charges for additional services, user dissatisfaction and expected benefits not being realised. The OGC said that in some cases a common understanding of what was really required and what each party had to deliver only became apparent after the award of the contract. In some cases the authority might have made a mistake and had selected the wrong partner. Projects such as the Armed Forces Pay and Administration Agency PFI project and RAF Mail had experienced initial difficulties, but had subsequently improved as a result of redefining the relationship between the authority and the contractor.
Contractors' rates of return
- The private sector participants in a PFI project may earn returns from providing services as contractors to the project or on their investment in the project. The OGC is reviewing value for money and the levels of return the private sector gets on its investments in PFI projects to ascertain whether it is commensurate with the risks in undertaking government contracts. There is a risk that if a contractor builds high rates of return into the contract, it will be more expensive than if undertaken in the public sector. If its study shows that contractors' rates of return are excessive, the OGC will wish to look at the steps needed to introduce more competition into the market place. High contractors' returns might also reflect other factors such as the private sector putting a very high premium on certain risks which had been transferred by authorities. If this was the case, the authorities would be advised to bring such risks back into the public sector.
- The OGC did not know what level of returns would be highlighted by its current study but believed that shareholders' returns of 8 to 15 per cent in real terms would be reasonable. Our Report on the refinancing of the Fazakerley PFI prison showed that the contractor had made significant extra profit as a result of refinancing, increasing the shareholder returns to 39 per cent. The OGC said this was an early prison where there had been significant risk for the private sector, since in public sector procurement in areas like prisons there had been significant cost and time overruns. It would expect the private sector to make higher returns on successful early contracts. As the PFI market matured and the risks became better understood by the private sector, their rates of return might be expected to diminish towards the average returns earned in other market sectors. The OGC review was assessing whether returns were in fact declining. Even on early PFI contracts some private sector organisations had lost large amounts of money. The Chairman of the Major Contractors Group told us that in his company, Kier Group, returns on PFI contracts were about 2.5 per cent of turnover compared with one per cent on other contracts. However, the risks were significantly greater in PFI projects and there were high bidding costs.
MAINTAINING VALUE FOR MONEY IN THE LONG TERM
Value for money mechanisms
- Although well run competitions will produce competitive initial prices for PFI contracts, authorities also need contractual means for maintaining value for money during the contract period. The percentages of contracts surveyed by the National Audit Office with contractual value for money mechanisms are shown in Figure 2. The OGC expected the percentages to become much higher, as standardisation, advice and guidance affected more of the contracts. Some, such as benchmarking, might require the assistance of third parties with specialist market knowledge. The OGC did not necessarily expect, however, to see all the mechanisms used in every contract. Gate 5 of its Gateway Review process was a formal review after the service was in operation to see whether the benefits in the original business case had been realised and to confirm what authorities were doing about the ongoing monitoring of value for money.
Figure 2: The use of value for money mechanisms
Value for money mechanism
Percentage of projects including such mechanisms
Profit and other gain sharing mechanisms
Open book accounting
Sharing of refinancing benefits
Source: NAO survey
- At the time of the NAO survey only 15 per cent of authorities indicated that they had the right to share in refinancing benefits. The OGC agreed that this was not acceptable. In 1999, it had alerted departments to the question of refinancing and issued further guidance in 2000. It had put out to consultation its revision of the Standardisation of PFI contracts which it proposed to publish in March 2002, and which would explicitly steer authorities towards a 50:50 share of refinancing gains and retaining the right to approve any refinancing. It now expected all contracts being let to include appropriate refinancing provisions.
Dealing with change
- PFI contracts are generally of a long term nature so not all changes that may be required can be foreseen. The NAO survey showed that 55 per cent of the authorities surveyed had already used change procedures to update their contracts. Most of the changes made related to changes in services, the introduction of new services, and additional works and changes to the design of buildings. The OGC considered that most of the changes were fairly minor while some, such as the authority and the contractor agreeing to joint testing of the service, were beneficial and showed that the partnership was working well. The Major Contractors Group also said that there were far fewer changes under PFI deals compared to a conventional building contract.
- One of the reasons authorities gave, however, for a decline in value for money was high charges for additional services. A recurring feature in traditional construction projects is that departments accept lowest price tenders, but contractors then seek to increase their profit margins through variations and claims for additional work. The OGC standardisation guidance emphasises that authorities should configure PFI contracts to satisfy themselves that change would be made on a value for money basis through the life of the contract.
- We were concerned that a contractor facing losses because it had failed to allow sufficient contingencies for risks occurring might seek to cut the level of service. The OGC said that an authority would have remedies against the contractor for failure to deliver the contracted level of service because there would be contractual payment deductions for poor performance. 58 per cent of authorities surveyed by the NAO who had an agreed performance deduction review process had made deductions in accordance with that process. The Major Contractors Group said that these deductions put pressure on contractors, though some cases, such as a contractor taking a hospital ward out of action for maintenance, did not arise from an assessment of the contractor's performance under the contract, and would be expected to arise during the contract period. It would be unreasonable to expect every hospital ward to be available all of the time throughout a 30 year contract. Contractors would probably have taken the likelihood of deductions into account when pricing the contract and the OGC confirmed this was normal business practice.
- The Committee has considered reports on the Passport Agency and National Insurance Recording System 2 deals, where the contractor failed to deliver the service required yet costs were borne by the relevant department, so there was a lack of real risk transfer. The OGC said that effective risk transfer was being achieved in certain PFI contracts and in some the private sector partner, having failed to perform adequately, was suffering significant financial penalty. For example, on the PRIME project the contractor had suffered a performance penalty of £5.6 million which had acted as an incentive for the contractor to improve.
- There was now a much better understanding among public sector clients that certain types of risk that could not be transferred, particularly the ultimate responsibility for delivering a public service. As risk management techniques were developed within the public sector, with an understanding that there were retained risks, focus was being sharpened on what sort of risks authorities could sensibly seek to transfer to the private sector and those that would have to be retained in the public sector. The public sector would always try to extend the envelope of risk that the private sector would take. What would be the optimal allocation today would not be the optimal allocation tomorrow. The Gateway Review process examined risk transfer in individual projects and it was developing enhanced guidance for accounting officers on risk management systems. The Major Contractors' Group said the private sector had struggled significantly with the additional risks they were asked to take on board as a result of the PFI. There was now a debate, which it considered healthy, on which party was best able to manage each individual risk.
Bailing out the private sector
- The Committee's Report, The Renegotiation of the PFI-type Deal for the Royal Armouries Museum in Leeds, found that the public sector had bailed out the private sector. The OGC said that cost overruns and risks which had a high probability of occurrence and a low impact to be picked up by the private sector. But the risk of catastrophic failure to provide a public service, which had a low probability but very high impact, could never be transferred to the private sector. The public sector had to keep providing the service, and if the authority had followed standardisation of contracts, it would have rights to step in take over the running of the contract and would be able to seek other providers. Under the standardisation of PFI contracts there were termination provisions that defined the circumstances in which an authority could terminate the contract where the contractor was in default for not delivering a public service.
Procurement times and bidding costs
- The OGC considered that the considerably longer timescale for putting PFI contracts into place, as compared with conventional procurement, was offset by the better performance of the private sector in bringing assets in on or ahead of the schedule demanded by the contract. It also compared favourably with what had been seen in the past when assets had been provided late compared to original plans, and the cost of the overrun had fallen to the public sector.
- The Major Contractors Group was concerned about the time being taken to put contracts in place. It had the impression that a lot of public sector organisations did not understand the importance of time to the private sector. The OGC said it was in dialogue with a number of departments to understand the reasons for the excessive timescales which gave rise to additional cost to the public and private sectors, and to see what could be done to shorten the timescales.
- The Chairman of the Major Contractors Group said that for one contract, valued at £70 million which his company, the Kier Group, had bid for and won, there had been £4 million of bid costs at risk until the deal was signed. His company was not being asked to invest £4 million up front in any other part of his business. His company had also lost other PFI contracts it had bid for after incurring costs of up to £1 million before being eliminated from the competition. In the light of this risk he considered the higher returns for PFI projects were fully justified, and believed that some prospective competitors were actually excluded from the competitive process by the high costs of entry.
Risks to the competitive process
- There is a risk that high bidding costs could encourage contractors to discuss with each other their approach to bidding for PFI contracts. The Major Contractors Group said that, at the level of individual contracts, its members did not discuss such practices as pricing and contingency funds for penalty clauses or any other matters which related to how they would approach individual contracts and tendering. Their members did, however, discuss issues such as risk allocation and general points of contract and bid costs where they felt it appropriate to take these issues forward as an industry for discussion with the OGC.
SKILLS AND GUIDANCE
- PFI is used to procure a wide range of public services, including accommodation, hospitals, prisons, roads, computer systems and Government accommodation. There are over 400 PFI contracts currently in force committing departments to future expenditure of around £100 billion. It is vital, therefore, that PFI contracts are managed well and that staff responsible for managing the contracts have the appropriate skills and guidance.
- The OGC described the C&AG's Report as a useful stimulus as it focused on the importance of contract management and was a reminder that it needed to continue to maintain and strengthen the focus on this area. It regretted that contract management disciplines had not been in place earlier, and that many authorities provided little or no training in contract management skills. It also had concerns about the blend of experience and appropriate skills employed in some contract management teams.
- The OGC said it was taking action to improve authorities' capability to commission and manage projects through advice and guidance, the use of the Gateway Review process and a stronger focus on wider commercial skills. It would introduce new training courses and encourage people to build careers in contract management. At the operational and business level the OGC had inherited much advice and guidance from predecessor organisations. It plans to provide revised, coherent and updated contract management operational guidance by the end of April 2002. For accounting officers, the OGC had provided more focused advice on the critical issues of managing partnering arrangements and why service contracts could go wrong. It also published a document on 15 April for accounting officers on value for money in complex procurements. The need for contract management manuals would feature as part of its updated advice and guidance. As part of the Gateway Review process review teams would test what methods were in place to facilitate the ongoing management of a contract.
The Gateway Review process
- To ensure that authorities have the capability to achieve value for money at contract award and to maintain it, the OGC had introduced the Gateway Review process in February 2000. All civil central government projects were required to go through that process and a similar process for local authority projects was under consideration. Part of the Gateway Review process is to give the OGC a much sharper insight on where good things are happening so that it can encourage other authorities to replicate them. The Gateway Review process focuses attention on the early life of projects, where the OGC considers there is greater scope for management to take corrective action. The OGC told us that accounting officers were paying a lot of attention to the recommendations that were emerging from the Gateway Reviews.
- Asked whether authorities were allowed to go ahead with a project unless they had been given clearance and understood the risks, the OGC told us that the Gateway Review process would enable projects to be reviewed before contracts were signed. If it felt a department was doing something which was fundamentally not conducive to value for money, then the OGC had escalation routes to draw the matter to the attention of the Chief Secretary of the Treasury or the Chancellor of the Exchequer. It preferred, however, to deal with these matters on a face to face basis with departments, winning over their hearts and minds rather than instructing them what to do.
- 47 per cent of authorities surveyed by the National Audit Office transferred less than a quarter of the staff employed on the contract procurement team to the contract management team. A number of authorities and contractors said they had experienced problems because of the lack of staff continuity within each other's contract management teams. The OGC said that its Gateway Review process explicitly tested what the authority's contract management plans were for an acceptable level of continuity.
- The OGC said it strongly encouraged continuity between those involved up to the point of contract award and those involved post contract award. If there was loss of memory about what had led to particular agreements in the negotiations, then it must increase the risk that the partnering arrangement would not get off to a good start. The Major Contractors' Group said that although it would expect some links between before and after contract award, the business skills needed to put the deal together were very different from those needed during the progress of the contract. The people building a hospital would not be those who had negotiated the contract.
- Contractors responding to the NAO survey highlighted the need for authorities to understand the operations of commercial organisations. The OGC agreed that departments had to respond to the business needs that they faced and the market conditions at the time they are doing business. It was looking at a range of factors that were necessary to ensure the successful management of the whole life cycle of a project. These included the need for specialists in procurement, project management, contract management and other key disciplines. It wanted greater professionalism and stronger capability on the part of practitioners who were involved in these disciplines, and to build up a network of people involved in contract management. Recruitment and retention of staff were affected by the fact that in a number of key skills areas there were wide pay differentials between what the private and public sectors paid for certain skills. There were also some pay differentials between departments.
1 These include all central government PFI contracts and those let by local authorities where there has been financial support from central government. Back
2 C&AG's Report, Managing the relationship to secure a successful partnership in PFI projects (HC 375, Session 2001-02) Back
3 C&AG's Reports: The Renegotiation of the PFI-type Deal for the Royal Armouries Museum in Leeds (HC 103, Session 2000-01); The Channel Tunnel Rail Link (HC 302, Session 2000-01) Back
4 C&AG's Report, para 7 and Figure 1, p2 Back
5 ibid, para 3.39 Back
6 Qs 14-15, 35, 56-62 Back
7 C&AG's Report, Figure 1, p2 Back
8 Q1 Back
9 Ev 22 Back
10 C&AG's Report, para 1 Back
11 Appraisal and Evaluation in Central Government ("The Green Book") HM Treasury, para 1.13 Back
12 Qs 2, 14 Back
13 C&AG's Report, para 3.39 Back
14 Qs 28-34 Back
15 C&AG's Report, para 1.18 Back
16 Qs 75-81, 86-87 Back
17 Qs 178-185 Back
18 Qs 21-23; 13th Report from the Committee of Public Accounts, The Refinancing of the Fazakerley PFI Prison Contract (HC 372, Session 2000-01), para 29 Back
19 Qs 22, 79 Back
20 Qs 115, 139, 146-177 Back
21 C&AG's Report, para 1.24 Back
22 Qs 16-19, 69-71, 202-205 Back
23 Q2 Back
24 C&AG's Report, Figure 8, p13 Back
25 ibid, para 1.30 Back
26 Qs 23-24, 200-201 Back
27 C&AG's Report, paras 1.37-1.38 Back
28 Q11 Back
29 C&AG's Report, para 3.39 Back
30 2nd Report from the Committee of Public Accounts, Improving Construction Performance (HC 337, Session 2001-02), para 3 Back
31 Qs 228-230 Back
32 Qs 48, 88-106 Back
33 C&AG's Report, para 1.22 Back
34 Qs 47-48, 109-111, 116-122 Back
35 22nd Report from the Committee of Public Accounts, Delays to the new National Insurance Recording System (HC 182, Session 1998-99); 24th Report from the Committee of Public Accounts, The Passport Delays of Summer 1999 (HC 208, Session 1999-2000) Back
36 Qs 191-192 Back
37 Qs 10, 25, 27, 112-114, 219 Back
38 4th Report from the Committee of Public Accounts, The Renegotiation of the PFI-type Deal for the Royal Armouries Museum in Leeds (HC 359, Session 2001-02) Back
39 Qs 231-239, 246 Back
40 Qs 227-229 Back
41 Q50 Back
42 Qs 9, 51-55 Back
43 Qs 115, 146-177 Back
44 Qs 240-245 Back
45 C&AG's Report, para 1 Back
46 Qs 8, 12-13, 65 Back
47 Qs 3, 206, 211, 215-218 Back
48 Ev 22 Back
49 Q214 Back
50 Qs140-141 Back
51 Qs 142, 207-213 Back
52 C&AG's Report, paras 2.11-2.12, and Figure 15, p19 Back
53 Qs 5-7, 66-68 Back
54 Qs 66-67, 144-145 Back
55 C&AG's Report, para 2.17 Back
56 Qs 4, 193-196 Back