Select Committee on Public Accounts Sixty-Third Report


2  Time and cost implications of tendering and negotiations

TIME TAKEN TO TENDER NEW DEALS

8. The average tendering time from initial advertisement to the close of a deal for PFI projects in 2004-2006 was 34 months, compared to 33 months for projects that closed prior to 2004, although there were significant differences even within sectors (Figure 2). The shortest overall tendering period was 16 months, whilst the longest was 73 months. The average cost of external advice for each project was £3 million, reflecting the length of the process, and delays to the schemes (as well as putting bidders off) cost the taxpayer at least £67 million in total.[11]

9. Such extended procurement periods and high costs highlight the complexity of negotiating PFI deals and the fact that reasons for delays are sometimes unforeseeable. The C&AG's report concluded that in most cases, the major causes of delay were either avoidable or could have been mitigated by the public sector without compromising value for money.[12]

Figure 2: The length of tendering varied widely within sectors

Source: National Audit Office

10. Causes of delay that could be avoided by the public sector include insufficient specifications prior to going to market, changes to the scope and design of projects, and poor process management. The Treasury agreed that its performance in working to reduce tendering times could be improved. However, it pointed to sector-specific delivery models such as Building Schools for the Future and Partnerships for Health, which have average tendering times of 22 and 20 months respectively, to demonstrate that there have been some signs of progress. Sector-specific targets had also been introduced in schools, street lighting and waste PFI deals.[13]

TIME TAKEN TO VALUE TEST OPERATIONAL DEALS

11. Some projects had completed value testing in between nine and 16 months but two projects had taken longer. In the case of an office accommodation project (St John's House, Bootle), the benchmarking process took just over two years because there was some difficulty in agreeing comparative data for the benchmarking exercise and the public sector rejected a proposed 16% price increase. Hereford and Worcester Magistrates Court had been trying to complete a benchmarking exercise for over three years because of difficulties in finding comparable data and there was little incentive for the private sector to complete the process since, at the time of the first benchmarking exercise, the contract only allowed for a price reduction. Generally, however, the public sector experience of the time taken to renew services within PFI contracts is very similar to that of non PFI contracts, where competitive tendering typically takes between six months and two years.[14]

NEGOTIATIONS WITH A SINGLE BIDDER FOR A NEW DEAL

12. As PFI projects tend to be complex procurements, it is usual for the public sector to invite bids and then choose a single, preferred bidder with whom to negotiate in detail. Even if there is a strong competitive process during tendering or value testing, the way in which negotiations are handled by the public sector after the competition has ended can be crucial to the ultimate value for money of a PFI deal. One third of the projects examined in the C&AG's report on tendering made significant changes to the project scope during negotiations with a single, preferred bidder. On average, the value of these changes (both upwards and downwards) was £4 million per project per year, equivalent to 17% of the value of each project.[15]

13. In the past, public authorities have requested significant changes to the design solution, the addition or removal of major equipment components, changes to the agreed services to be provided and major changes to the agreed allocation of risk. The process of Competitive Dialogue now specifies that bidders can only be asked to 'fine tune, specify and clarify their bids' once the competitive phase has concluded. The Treasury told us that it was not desirable for changes to be made after competitive tendering had ended but considered that the introduction of the Competitive Dialogue process would dramatically reduce the period where there was only one bidder (Figure 3), although it was too early to say whether total tendering times would also reduce.[16]

Figure 3: Preferred bidder negotiations for projects were often lengthy

Source: National Audit Office

NEGOTIATIONS FOLLOWING THE VALUE TESTING OF OPERATIONAL DEALS

14. The Treasury expects public authorities to decide, on a contract by contract basis, whether to include facilities services in a PFI contract or to procure them through conventional outsourcing. In some cases, design issues may make it sensible for a construction company to both build PFI facilities and provide related services. In other cases, conventional outsourcing may allow greater competition between service providers. To date, however, conventional outsourcing has not been compared overall with the cost and quality experience of facilities services procured under the PFI.[17]

15. In all the building projects which had been subject to benchmarking or market testing, the public authorities have had to enter into negotiations to improve the price changes initially proposed by the private sector. In four of these seven building projects the final outcome after negotiation was still a price increase of between 1% and 14% in addition to the annual price increase for inflation allowed by the contracts. In these cases, it was uncertain whether value for money had been achieved (Figure 4) given that the other building projects had achieved price reductions or kept the price unchanged. It is possible, however, that some bidders may have initially set prices at below market rates knowing that price increases could subsequently be negotiated through the benchmarking and market testing process.[18]

16. During negotiations with suppliers as part of the value testing of services, some public sector authorities had agreed to reductions in the services being provided to keep the price of their PFI contracts affordable. These authorities considered the service levels were previously over-specified and did not expect the reductions in specifications to compromise the service delivered to the public, although it is too early to judge the outcome conclusively.[19]

Figure 4: Key features of the seven PFI building projects that have completed value testing of their ongoing services
Project Name
Value Testing Method
Change in Supplier (after market testing)
Annual Cost of Services (2005-6)
Final Agreed Price Change (note)
NAO assessment of value for money
Service change
Debden Park High School Benchmarkingn/a £0.2m+ 14% Uncertain
Queen Elizabeth Hospital, Greenwich Market TestingIncumbent £5.7m+ 6% UncertainSome planned enhancements to service not taken up to keep contract affordable
Sussex Partnership NHS Trust Market TestingIn-house £0.9m+ 5.7% Uncertain
University Hospital of North Durham Benchmarkingn/a £3.5m+ 1.2% Uncertain
St John's House, Bootle Benchmarkingn/a £0.8mNo change Yes
Norfolk and Norwich University Hospital Market TestingIncumbent £9.8m- 2.2% Yes
Darent Valley Hospital Benchmarkingn/a £5.1m- 2.4% YesA reduced office cleaning regime and cessation of two dedicated porters for each operating theatre but the Trust does not expect this to adversely affect users. Reduction in service costs also achieved through separate decision, as part of the clinical strategy, to close a number of beds

Note: The price changes exclude service enhancements and also the effect on the costs of the NHS Trusts of the NHS Agenda for Change which would have happened regardless of the PFI value testing process

Source: National Audit Office


11   C&AG's Report (1), paras 3.1, 3.2-3.24 Back

12   C&AG's Report (1), para 3.7; Q 14 Back

13   Qq 9-10, 63, 72 Back

14   Qq 48-49; Ev 13 Back

15   C&AG's Report (1), paras 3.27-3.29 Back

16   Qq 14, 62-63, 85-88, 130; C&AG's Report (1), paras 1.6, 3.29 Back

17   C&AG's Report (2), para 4(xi), 2.15; Qq 22-23,58, 61,77-81 Back

18   C&AG's Report (2), paras 2.3, 2.17; Qq 16-20 Back

19   Qq 18-21, 53-55; C&AG's Report (2), paras 2.19-2.26  Back


 
previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2007
Prepared 27 November 2007