The outcome of the initial benchmarkings varied with difficulties in finding comparable data

2.4  In the first eight early examples of benchmarking, two projects (the MOD's Defence Fixed Telecommunications Service and the Foreign and Commonwealth Office's Telecommunications Network) had achieved clear benefits by using benchmarking to secure price reductions in competitive markets where communications prices were falling. This was assisted by a good working relationship between the Authority and the private sector and input from specialist advisors.

10

Information on the projects we visited during our fieldwork in Summer 2006

 

Project name

department

Project

date of Financial close

capital Value £ million

Estimated lifetime value of services subject to value testing (cash prices in 2006 prices unless otherwise stated) £ million

Annual cost of services at time of benchmarking/ market testing (2005-06) £ million

Operational length of the contract

Services benchmarked/market tested

 

Sussex Partnership NHS Trust

Health

Mental health facilities

24 June 1999

22

27.81

0.9

30 years

Cleaning and catering

 

Darent valley Hospital

Health

First major hospital contract to be let under PFI

30 July 1997

94

153.01

5.1

30 years

Catering, switchboard, laundry & linen, waste management, porteringportering other (transport/post), domestic, pest control, accommodation (e.g. homes for nurses, doctors)

 

Norfolk and Norwich university Hospital

Health

main site of the Norfolk & Norwich university Hospital NHS Trust

9 January 1998

229

428.02

9.8

35 years (for soft Fm services, 60 years for hard Fm services with break options at 35, 40 & 50 years)

Domestic, catering, portering, laundry & linen, security, car parking, waste and grounds maintenance.

 

Queen Elizabeth Hospital, Greenwich

Health

The redevelopment of a former military hospital

8 July 1998

113

170.01

5.7

30 years (extendable to 45 years or 60 years)

Portering, catering, domestic window cleaning, pest control, security, car park, residential accommodation, help desk, switchboard

 

University Hospital of North Durham

Health

New district hospital for North Durham

31 march 1998

96

50.51

1.93

27 years

Portering/courier service, catering service, domestic service, security/car parking, linen service, helpdesk service, telecom service, estates services

 

Debden Park High School

Education

A new secondary school in Loughton

28 march 2000

15

5.11

0.2

25 years

Cleaning, catering, ict support, caretaking and groundskeeping

 

Defence Fixed Telecommunications Service (DFTS)

MOD

Telecoms service

1 July 1997

70

1,505.04

40.25

15 years6

Telecommunications

 

Foreign and Commonwealth Office Telecommunications Network (FTN)

Foreign & Commonwealth Office (FCO)

Provides voice data and messaging links for the Foreign Office worldwide

10 may 2000

74

180.0

32.0

10 years

All services subject to benchmarking and market testing

 

National Savings and Investments

National Savings

Provision of services for the administration of retail financial savings and investment products (including Premium Bonds) to NS&I customers.

circa April 1999

N/A

The agreement with Siemens is a Public Private Partnership (PPP) currently in the 7th year of its 15 year term and NS&I estimate that savings to be realised by the agreement against the Public Sector Comparator amount to some £540 million. To acquire a true understanding of the cost savings arising from the agreement, a comparison would need to be undertaken between the actual investment Siemens has made to the business of providing the NS&I services, with the investment that NS&I would have had to provide in-house in order to deliver the same services, as well as quantify the risk and chances of success had the whole operation been undertaken in-house. Given the changes that have arisen since the start of the agreement, this comparison would be difficult to quantify.

 

Hereford and Worcester magistrates' Court

Courts

New courthouses in Kidderminster, Hereford and Worcester. Refurbished courthouse in Redditch

29 march 2000

25

20.0

0.8

25 years

 

Security, cleaning, portering, helpdesk, vending and catering telephone/switchboard and waste

 

St John's House, Bootle

HMRC

Office accommodation

1 February 1997

12

20.0

0.8

25 years

 

Cleaning, waste management, grounds maintenance, vending etc

 

Source: National Audit Office and Partnerships UK database

NOTES

1  This is a proxy figure from the annual cost of the contract multiplied by the years of the contract. This method of calculating the cost of the services for the whole contract life was agreed with the projects when they were unable to identify a figure from the financial model.

2  This sum is taken from the financial model as at December 2005 and shows the projected market tested service values as at December 2005. (These values do not include contract variations which have to be adjusted). The projected total is derived from the addition of the service fee costs across the requested period – August 2001 to August 2036 – to achieve a comparison for a typical 35 year contract.

3  The figure of £1.87 million is just for the soft services the hard services were due the following year.

4  The total cost of the services over the life of the contract is £1,505 million. About 80% of the services are reviewed quarterly and if a service is judged not to be value for money then it can be subjected to the formal contractual benchmarking process. So far, one service has been contractually benchmarked: Restricted Local Area Network Interconnect (RLI).

5  The annual cost of the RLI service before benchmarking.

6  The original length of the contract was 10 years (1997-2007) but has been extended via a renegotiation from April 2005 to July 2012.



11

Key Features of the PFI projects that have value tested their services

 

Project name

Type of services value tested

Value Testing Method

Supplier (after the market testing)

Annual cost of Services (2005-06)

Final Agreed Price change (excluding enhancements)

NAO assessment of whether vfm was likely to have been achieved through the value testing process (see Figure 16)

 

Sussex Partnership NHS Trust

Soft

market Testing

In-house

£0.9m

5.7%

Uncertain

 

Norfolk and Norwich university Hospital

Soft

market Testing

Incumbent

£9.8m

– 2.2%

yes

 

Queen Elizabeth Hospital, Greenwich

Soft

market Testing

Incumbent

£5.7m

6%

Uncertain

 

Darent valley Hospital

Soft

Benchmarking

Not applicable1

£5.1m

– 2.4%

yes

 

Debden Park High School

Soft

Benchmarking

Not applicable1

£0.2m

14%

Uncertain

 

Hereford and Worcester magistrates' Court

Soft

Benchmarking

Not applicable1

£0.8m

Not completed

Uncertain

 

St John's House, Bootle

Soft

Benchmarking

Not applicable1

£0.8m

0% (no change)

yes

 

University Hospital of North Durham

Soft and Hard

Benchmarking

Not applicable1

£3.5m

1.2%

Uncertain

 

Defence Fixed Telecommunications Service

Telecommunications

Benchmarking

Not applicable1

£40.2m

– 37.3%

yes

 

Foreign and Commonwealth Office Telecommunications Network

Telecommunications

Benchmarking

Not applicable1

£32.0m

– 19%

yes

 

National Savings and Investments

Financial

Benchmarking

Not applicable1  Difficult to quantify

Not yet quantified (Figure 15)

Uncertain

 

From the above 11 projects:

 

 

 

 

 

 

 

Service provider following market tests

 

Number of projects won by the incumbent

Number of projects won by in-house supplier

Number of projects won by new  external suppliers

Total

 

2

1

Nil

3

 

Price changes for those projects that have market tested or benchmarked

 

Projects

Number of projects which had a price increase

Number of projects which had  a price decrease

Number of projects where the price stayed the same

unresolved

Not yet  quantified

Total

 

Buildings

4

2

1

1

0

8

 

Financial/ Communications

0

2

0

0

1

3

 

Total

4

4

1

1

1

11

 

Source: National Audit Office

NOTE

1  There is no change of supplier in a benchmarking exercise.

2.5  In the other six projects, the effectiveness of the benchmarking was more varied. A key factor was that a number of project teams told us that comparable data for benchmarking soft FM services was sometimes difficult to find, and in some cases, data collected was of limited use. For example, the nature of the services provided will vary from project to project - cleaning will, for example, vary depending on the configuration of a hospital and the type of clinical services. There may also be regional variations in costs. For benchmarking to be successful therefore, comparable market data often needs to be adjusted to take account of the characteristics of the particular project. The Treasury, in liaison with departments, has started collecting information so that there will be a central source of data which can be used by projects.

2.6  A variety of information databases exist which contain benchmark costs although these will often include non-PFI data which is not directly comparable with PFI projects3. The Department of Health (DoH) is advanced compared with other sectors in providing a costs database, ERIC (Estates Returns Information Collection). However users of ERIC told us there are weaknesses within ERIC which can undermine the value of the data (Figure 12).

2.7  Based on our visits to the eight projects that had used benchmarking, although the projects had experienced these practical difficulties in applying benchmarking, the following factors in certain projects had contributed positively to the benchmarking process:

  A well structured PFI contract providing an effective means of price adjustment.

  The use of external FM consultants to collate appropriate data.

  The authority having a strong negotiating position enhanced by the option to instigate a market test if agreement cannot be reached.




__________________________________________________________________________________________________

3  Databases include the British Institute of Facilities Management, the International Property Database and Trade Organisations such as the British Institute of Cleaning Science. However, PFI projects have different contractual arrangements and standards to non-PFI projects. For example, service specifications may be set higher and payment risk regimes will reflect the fact that a service provider may face penalty deductions if the service does not meet the minimum standard.