Select Committee on Public Accounts Twenty-Eighth Report

2. Striking a good deal

11. Good procurement practice needs to be followed for all purchases of goods and services, however they are financed. In this respect the PFI is no different from other forms of procurement. Some of the details may differ but the basic procurement philosophy as regards ensuring competition, properly evaluating bids and controlling the costs of the procurement remains the same.

Ensuring competition

12. Competition is a fundamental requirement for getting good value from PFI deals. A procuring department needs to survey the market to establish how many companies would be interested in the project and to assess whether its proposals are likely to be attractive to potential bidders. If too few bidders are interested there may be problems with the design of the project and the department should think again. Competitive tension amongst a number of bidders needs to be maintained for as long as possible. A single preferred bidder should not be chosen prematurely or before outstanding issues have been resolved. When it is no longer possible to maintain competitive tension and exclusive negotiations with a single bidder begin, departments should aim to manage these negotiations as effectively as possible. Changes to the project at this late stage are likely to increase its cost.

13. Figure 4 shows some examples of cases where the procurement process has not been fully competitive and value for money is unlikely to have been achieved. In some cases departments have ended up with a single bidder but have still pressed on despite evidence that there were problems with the design of their projects. In other cases there have been protracted negotiations at the preferred bidder stage and the cost of the deal has crept up.

Figure 4: Examples of inadequate competition
PFI deal Committee's findings
Immigration and Nationality Directorate (7th Report, Session 1999-2000) Key figures, on which future increases in productivity would be measured and payments to the contractor calculated, had not been finally agreed until more than a year after the contract was signed. Such important issues need to be finalised before a contractor is selected and the benefits of competition fall away.
Dartford and Gravesham Hospital (12th Report, Session 1999-2000) The NHS Trust selected two firms to submit final bids but one of the firms did not submit a bid. The Trust therefore ended up with only one final bidder on this major pathfinder project for the use of the PFI in the NHS. The bidder's final bid was 33% higher in real terms than its indicative bid. The Trust did not undertake a detailed analysis of the reasons for the increase in the final bid, especially given the absence of other bids. Such action might have helped the Trust to secure a greater price reduction in the subsequent negotiations.
Newcastle Estate (19th Report, Session 1999-2000) In this deal the Department of Social Security appointed a preferred bidder whilst important issues remained unresolved. Exclusive negotiations with the preferred bidder continued for 18 months.
Royal Armouries Museum (4th Report, Session 2001-02) There had been a lack of market interest in the deal when it was put out to the market and only one bid had actually been received. When withdrawing from the competition for this project, one company had expressed concern over the practicality of the proposals for joint working between the public and private sectors in certain areas. The Royal Armouries were not given access to the contractor's financial records and there were disagreements between the two parties over issues which were of fundamental importance to the museum's future.
West Middlesex Hospital (19th Report, Session 2002-03) The preferred bidder agreed to hold its price for seven months but it took the Trust eleven months to close the deal. The price increased after the commitment period had expired so the price commitment had only limited effect. The principle of securing a price commitment to deter "deal creep" is good, but a department using this approach needs to be sure that it can close the deal whilst the commitment still holds.

Evaluating bids

14. The full evaluation of bids should seek to identify the bid that offers the best combination of financial and non-financial factors. It should include an assessment of bidders' financial and technical competence, including their performance on other government projects. Information to help evaluate bids can be obtained by attempting to cost the bidder's proposed solution, which requires access to the bidder's financial model. The model would show the financial outcome of a particular set of estimated costs, revenues and charges for delivering the service over time.

15. Benchmarking the prices offered by contractors is highly desirable in a competitive situation but is absolutely essential in a single bidder scenario. In the Airwave deal (64th Report, Session 2001-02), when the procurement went down to a single bidder, the contractor recognised that there was a need to change its approach and proposed using a should-cost model. The model would describe the components that make up the system, the contractor would make its own estimate of what it would cost and the Department could check whether it felt the estimates were appropriate.

The cost of negotiating deals

16. The procurement of PFI deals is inherently more complex than the procurement of conventional deals and can involve departments and bidders in heavy administrative costs. For example, on the Newcastle Estate deal (19th Report, Session 1999-2000), the cost of the procurement to the Department of Social Security rose from an initial estimate of £0.4 million to £4.4 million, an eleven-fold increase, reflecting the complexity of this type of procurement and the Department's inability to undertake many of the tasks required to negotiate the deal. On the Prime deal to transfer the Department of Social Security estate to the private sector (41st Report, Session 1998-99), the Department's costs totalled £10.9 million, compared with an initial budget of £1.7 million, and the final three bidders spent around £27 million in preparing their bids.

The cost of advisers

17. Advisers' costs in PFI deals can exceed budgets by significant margins. For example, on the Newcastle Estate deal (19th Report, Session 1999-2000), the cost of legal advice increased from an initial estimate of £70,000 to an outturn of £2.3 million. On the Dartford and Gravesham Hospital deal (12th Report, Session 1999-2000) the Trust incurred advisers' costs of £2.4 million, which exceeded the initial estimates by almost 700%. After a series of hospital PFI deals, the Trust spent £2.3 million on advisers on the West Middlesex Hospital deal (19th Report, Session 2002-03), virtually the same amount as at Dartford and Gravesham four years earlier.

previous page contents next page

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

© Parliamentary copyright 2003
Prepared 19 June 2003