4. Most PFI contracts have arrangements to test the value of certain ongoing services through either benchmarking or market testing, these processes being known collectively as value testing. The implications of inadequate competition, which can affect the tendering process, are also relevant to value testing. Market testing is usually preferable to benchmarking because it seeks new bids for services from alternative suppliers, rather than simply comparing costs with similar services provided elsewhere. There may, however, be situations where benchmarking should be used, for example where there is insufficient competition to run an effective market testing exercise.7
5. Two of the three market tests examined in the C&AG's report had resulted in price increases with uncertain value for money. None of the market tests had been won by a new external supplier although in one case the service provision had been taken back in-house. The process was deemed to be competitive in each case with alternative suppliers putting in bids. The incumbent supplier is, however, often in a strong position to win the re-competition of a service contract as a result of its knowledge of the project and its relationship with the authority.8
6. If incumbent suppliers continue to win a high proportion of market tests, competition in future market tests could be discouraged, leaving the incumbents in a powerful position. They would then be likely to remain the supplier of services for the life of the PFI contracts, which may be as long as 35 years, with little incentive to improve the services they are providing. Recent data compiled by the Treasury showed that, where a number of services were being supplied within a PFI contract, some of the incumbent suppliers were exposed to competition. The Treasury's data identified that there had now been 13 market tests (on five projects, some of which had run separate competitions for a number of services) and the incumbent supplier on certain services had been replaced in six of the 13 tests. The Treasury is also working with Partnerships UK to facilitate a competitive market by publishing details of forthcoming competitions.9
7. If there are situations where project teams cannot secure strong competition through a market testing exercise then they will need to use benchmarking to test the value of services provided by an incumbent supplier. To replicate the competitive tension that arises in market testing, the public sector needs to be able to negotiate, based on good benchmarking data, on the terms offered by other suppliers for similar services. Although there are some departmental cost databases including non-PFI data, PFI projects have often had difficulties, however, in finding comparable data to test whether the prices proposed by the incumbent supplier are competitive. In two of the four building projects examined in the C&AG's report which had completed benchmarking, the outcome had been a price increase, in one case as much as 14%, and the value for money of these price changes had been uncertain. Better benchmarking data, combined with more effective negotiations, might have produced improved outcomes in these cases. To assist situations where benchmarking is used, the Treasury had asked Partnerships UK to collect data from 55 projects in order to improve the adequacy of benchmarking information currently available to departments. Partnerships UK is planning to roll out this data collection process across the PFI programme.10
________________________________________________________________________________________________________
7 C&AG's Report (2) para 1.1; Figure 4; p 9
8 C&AG's Report (2), paras 2.8-2.11