5. A public sector comparator is a costing of a conventionally financed project delivering the same outputs as those of the PFI deal under examination. It is just one of a number of ways of evaluating a proposed PFI deal. It is directly relevant only when the publicly financed option on which it is based is a genuine alternative to the PFI deal. This is most likely to arise at the outset of a project.3
6. The use of public sector comparators has been the subject of considerable debate about their reliability, accuracy and relevance in the contexts in which they have come to be used. We have seen many cases where the public sector comparator has been incorrectly used as a pass or fail test. In these cases the desire to show that the PFI deal is "cheaper" than the public sector comparator has led to manipulation of the underlying calculations and erroneous interpretation of the results. There are likely to be qualitative and non-financial differences between the options that cannot simply be subsumed in a difference in forecast cost.
7. The accuracy of public sector comparators is limited. They are prone to error because of the complexity of the financial modelling that is often used. They are also dependent on uncertain forecasts. This places a limit on the accuracy which can be achieved, however much work or analysis may be done. Further work takes time and money without necessarily adding to the value of the public sector comparator as a decision tool. There is also a risk that the users of the public sector comparator will believe that it is more accurate than it could ever be. Decisions can be made on the basis of small and spurious differences between the public sector comparator and the PFI option.
8. Examples of some significant weaknesses in the use of public sector comparators are set out in Figure 3.
Figure 3: Weaknesses in the use of public sector comparators
PFI deal | Committee's findings |
Dartford and Gravesham Hospital (12th Report, Session 1999-2000) | The NHS Trust did not detect significant errors in the public sector comparator. The Trust also did not quantify the full effects of changes in contract terms and of the sensitivity of the deal to changes in key assumptions, as the deal went forward. Had the Trust known that the savings were marginal when negotiating the deal, it might have made different decisions and achieved better value for money. |
A public sector comparator was not prepared until late in the procurement, and after a decision to use the PFI had already been made. It is therefore doubtful that the use of a comparator added to the decision-making process. | |
The public sector comparator gave a central estimate for the cost of a conventionally financed alternative to the PFI deal as £746.2 million, compared to an expected deal cost of £746.1 million. Such accuracy in long term project costings is spurious, and the small margin in favour of the PFI deal provided no assurance that the deal would deliver value for money. | |
The NHS Trust's advisers strove to make slight adjustments to the calculations, well within the range of error inherent in costing a 35 year project, so that the PFI cost appeared marginally cheaper than the public sector comparator. |
___________________________________________________________________
3 Ev 1