PART THREE

It might have been possible for the Trust to have improved the original deal with greater competition and better defined requirements in the closing stages but the Trust is not convinced this would have brought added benefits as it sought to close a pathfinder deal which had already been assessed as value for money

3.1 The Trust's approved business case assessed this early PFI hospital deal as value for money when the contract was let in 1998 and when the additional works were commissioned in 2001 (Figure 20).

The approved assessment of the value for money of the deal with Octagon in 1998 and 2001

The Trust's full business case for the original deal with Octagon in January 1998, approved by the Department and the Treasury, was assessed as value for money after taking into account the following factors:

■ Octagon had been selected as being the preferred bidder following a competitive procurement in which the Trust ranked Octagon as being the best bidder on both price and quality;

■ The Trust took steps to benchmark the price variations which arose after Octagon became preferred bidder and was satisfied that these were reasonable;

■ The Trust was satisfied that the net benefits from a PFI procurement were at least as good as what might have been obtained from conventional procurement. This included an expected small cost saving from the deal with Octagon compared with a public sector comparator estimate of what the deal might have cost under conventional procurement;

■ The Trust's case for the additional works commissioned in 2001 was also approved as value for money.

Source: The Trust's approved Full Business Case and other analysis prior to contract letting

NOTES

1 The Trust's Full Business Case was approved by the Department and the Treasury.

2 In the scope of this examination, which focuses on the 2003 refinancing, the National Audit Office has not carried out a full examination of all aspects of the original deal.

3.2 Alternative financing solutions were not seriously explored to ensure the financing terms remained competitive during a two year deal closure, the Trust considering that it did not wish to further delay the project and that it was not convinced that the overall terms of the deal could be improved bearing in mind the relatively undeveloped state of the PFI financing market at that time.

1 The Trust chose not to request Octagon to seriously consider bond finance, although it was subsequently used as the financing method for a number of PFI hospitals, because the Trust did not want to delay closing the deal and it was not certain that suitable bond finance would be available for the deal (Figures 21 and 22).

21	Number of PFI hospital deals using bank and bond finance

Year in which contract let		Number of deals using bank finance	Number of deals using bond finance1
1997		2	1
1998		5	1
1999		3	4
2000		4	1
Source: The Department of Health
NOTE 1 Bond finance became the financing method used for a number of PFI hospitals in the late 1990s because it gave the prospect of lower financing costs as the terms of the finance were particularly appropriate for the index-linked contract payment method used in PFI hospital deals. Bond finance was the form of finance used on the refinancing of this project to maximise the refinancing benefits.

Reasons the Trust did not actively seek bond finance

■ The Trust and its advisers chose not to request Octagon to seriously explore bond finance in closing this deal in January 1998 as the Trust did not want to delay further completing the deal and it considered the bond market at that time was still immature so that it could not be certain that bond finance could be used for this deal to deliver improved financing terms.

■ A bond issue for this deal in January 1998 would have been significantly larger than any PFI bond issue that had been completed at that time and, although two smaller PFI hospital deals were funded by bond finance during 1997 and 1998, the first indexed PFI bond, now the main funding option for large hospital deals, was not issued until May 1999.

■ The Trust considered that exploring the possibility of bond finance would have delayed completing the deal, particularly as it considered it likely that time would have been needed to
arrange monoline insurance for the bond issue when monocline insurers were relatively inexperienced in this type of project compared to today.

■ The Trust was particularly concerned about the risk of Octagon seeking to increase the price of the deal if it was delayed to compensate for the effect of further construction cost inflation
which was then running at 6 per cent per annum. In addition, the Department considered that, as the PFI hospital market was in its formative stages in 1997 when this deal was being finalised, a late change to bond finance would have had an adverse affect on the interest of the banks which had supported this deal in financing other PFI deals.

Source: The Trust

2 A funding competition would have allowed alternative options to be tested more rigorously although at the time this was not an established approach (Figure 23).

Reasons why a funding competition could have helped to test whether the best financing solution was being obtained and the reasons why this approach was not used

Reasons why a funding competition could have helped to test value for money

■ There was a two year delay between preferred bidder and financial close (January 1996 to January 1998).

■ Where there is a delay in closing a deal a funding competition can be helpful before closing the deal to ensure the best possible financing terms are obtained (see NAO report on the Treasury Building funding competition, HC328 2001/02).

■ A funding competition could have helped test the decision on whether to use bank or bond financing on this deal.

■ It would have also put greater pressure on the cost of finance provided by the banks.

■ A funding competition was held by Octagon to get the best terms for the refinancing.

Reasons why this approach was not adopted

The Trust and its advisers acknowledge the potential benefits that a funding competition can contribute in closing a deal. For the following reasons they consider that a funding competition would not have been feasible in closing the Trust's deal in 1998:

■ Decisions at the time were influenced by a wish to close this pathfinder deal, which had already been significantly delayed, without further delay so as to promote the development of the market for PFI hospitals. The market was then in its formative stages and needed completed deals to create confidence in this new form of procurement.

■ As the market for PFI hospital deals was not fully developed at the time this deal was being closed, and this was the largest health PFI project then in the market, there would have been constraints on the extent to which alternative sources of funding could have been sourced. In respect of bank finance, the Trust and its advisers consider that as a significant number of the banks then involved in the PFI market were already funding this deal there were only a few other banks who could have been invited to participate in a funding competition.

■ The concept of a funding competition was not, at the time the Trust was closing this deal, an established approach in PFI procurement. It gained prominence when the Treasury completed a funding competition in early 2000 for the Treasury Building PFI procurement which had previously been deferred.

■ In particular, to avoid extensive delay in running a funding competition, the funding competition needs to be based on market acceptance of the project documentation. It was only in the closing stages of agreeing this deal in late 1997 that the contract terms for this pathfinder deal were finally accepted by the funders.

■ The Trust was concerned that any benefits from running a funding competition might have been offset by Octagon increasing the contract price to take account of construction cost inflation, which was then around 6 per cent per annum, during any further delay in closing the deal as a result of running a funding competition. The Trust estimates that construction costs would have increased by around £700,000 for each month of delay and it is likely that Octagon would have sought to recover these additional costs from the Trust.

Source: The National Audit Office and the Trust

3.3 The annual charge increased by a fifth in a non-competitive situation due to specification changes, including an increase in the number of beds of over 40 per cent, although the Trust took steps to test through benchmarking that the pricing of this additional work was reasonable.

1 The annual price increased by 20 per cent after Octagon became the preferred bidder up to the time of the refinancing (Figure 24).

24	Increase in annual contract price since Octagon became the preferred bidder

		Annual price £m	Increase in beds £m	Other variation £m	Refinancing £m	Removal of IT contract2 £m
When Octagon became preferred bidder		35.6
Additional 108 beds3		2.8	2.8
Other changes		0.6		0.6
At contract letting		39.0 (+10%)
Additional 144 beds3		3.4	3.4
Other changes		0.3		0.3
Prior to the refinancing		42.7 (+20%)
Refinancing		(3.6)			(3.6)
Price following the refinancing		39.1 (+10%)
Price movements since refinancing		(1.3)		0.9		(2.2)
Current price		37.8 (+6%)
Total price movements since Octagon		2.2	6.2	1.8	(3.6)	(2.2)
became preferred bidder		(+6%)
Source: The Trust
NOTES
1 Figures expressed in March 2005 prices.
2 The provision of IT services are now under separate contractual arrangements.
3 Bed numbers have increased by 41 per cent from 701 at preferred bidder to 989 as a result of additional Trust requirements. The increase arose from the two main variations shown above which added 252 new beds and a further 36 beds arising from conversion of office space to ward usage.

2 It is possible that a better pricing could have been achieved if the Trust's current requirements had been the basis of the original bidding competition although the Trust took steps to test through benchmarking that the pricing of additional building work has been reasonable (Figure 25).

Factors which may have affected the extent to which the price variations have been value for money

■ The price variations (which have included an increase in bed numbers from 701 to 989) occurred in a non-competitive situation after Octagon became preferred bidder.

■ There can be risks to value for money in any non-competitive procurement because, in pricing the work, the contractor is not under pressure from other bidders. In these situations,
partnering and joint working arrangements, where new building projects are taken forwards by a pre-selected contractor can produce value for money where the terms of the new building work are subject to a pricing basis developed under a previous competition or good benchmarking arrangements.

■ Including the Trust's increased bed requirements in the original specification which was competitively tendered may have produced keener pricing for these additional beds through the strength of competition between bidders to win the PFI contract.

The Trust's approach

■ The Trust drew up its specification within the then NHS guidelines on bed numbers for new hospitals. Commissioning additional bed numbers then may also not have been possible within the Trust's limit of what would be considered affordable.

■ Octagon considers that, in respect of those additional requirements which the Trust requested during the construction of the new hospital it would not, at that time, have been practicable to put the work out to competition as the construction contractor was already on site. Octagon also considers that deferring the additional work until a competitive tender could take place may have increased the pricing of the work because of construction cost inflation.

■ The Trust took steps to test, through benchmarking, that Octagon's price changes were reasonable. Davis Langdon and Everest, the Trust's professional advisers, benchmarked and technically reviewed Octagon's proposed capital costs.

Source: The National Audit Office and the Trust