Price certainty
5 Our census showed that most PFI projects were delivering price certainty to departments with 29 out of the 37 projects surveyed reporting no construction related price increase after contract award. Where there had been a price increase it had been due to changes led not by the contractor but by the department or other parties. The price changes mainly related to further work, which had not been part of the original specification, on additional or improved facilities or changes to the function of a building. But construction cost increases had been mainly borne by the consortium with no increase to the department's payments. This, together with the fact that there has been publicity of some construction companies incurring losses on certain PFI work, is evidence of risk transfer working. Construction companies say that risk transfer is also working in that it places pressure on them to manage risks more effectively.
6 The census results compare well with historical experience of construction contracts in the public sector. In our 2001 report 'Modernising Construction' (HC87, Session 2000-01) we reported that some 73 per cent of departments' and agencies' construction projects had run over budget for the public sector. In our PFI census only 22 per cent of those surveyed had overrun (see Figure 1). These price increases were generally relatively small and not due to the consortia charging more for the work originally specified.
1 |
| Improved project delivery under the PFI | ||
|
|
| Previous experience (1999 Government survey) | PFI experience (2002 NAO census) |
|
| Construction projects where cost to the public sector exceeds price agreed at contract | 73% | 22%1 |
|
| Construction projects delivered late to public sector | 70% | 24%2 |
|
| NOTES |
|
|
|
| 1 None of the increases in PFI price after contract award were due to changes led by the consortium alone. For example, in some cases the department changed some of the specifications from those for which the consortium had bid, so the price increased to reflect the changes. Some of these specification changes arose due to new factors affecting the department's needs after contract award. These changes would also have led to price increases under traditional procurement. | ||
|
| 2 In only eight per cent of PFI projects surveyed was the delay more than two months. No comparative data for this statistic are available for traditionally procured projects. Previous studies of traditional projects had referred to the percentage of time overruns rather than the number of months.2 | ||
|
| Source: National Audit Office | ||
7 Where prices had increased, we found that departments had carried out benchmarking in less than half the cases to satisfy themselves that the price increase was reasonable. This is important as such changes take place without competitive pressure. Further, as PFI building contracts are often for 25 or more years, departments may find they need to change or add to the built asset during the contract period. Where they do so it is important that they have procedures in place to demonstrate that the pricing of the change is value for money.
8 This report has focused on the delivery of PFI construction projects to the public sector. As noted above, the private sector experience is that risk transfer is working. But the available information on the level of rewards to construction companies from PFI work is limited and rather mixed. Whilst these issues have not been the focus of this report we hope to return to them in future reports.
___________________________________________________________________________________
2 The 2002 Mott MacDonald report found that traditionally procured standard building projects examined had taken between one and four per cent longer to complete than expected at business case stage, before contract award. Standard buildings are those not requiring special design considerations. Non-standard buildings examined had taken between two and 39 per cent longer. Non-standard buildings involve special design considerations and may include specialist hospitals, innovative prisons, high technology facilities, other unique buildings or refurbishment projects. The 1999 report, Benchmarking the Government Client found that construction programmes overran by an average of 13 per cent compared to the tender stage.