Committee of Public Accounts

Press Notice No. 23 of Session 2003-04, dated 15 June 2004


Mr Edward Leigh MP, Chairman of the Committee of Public Accounts, said today that there have been unwelcome cost increases on the programme to provide new accommodation for GCHQ and that it was astonishing that GCHQ did not realise much sooner what would be involved in moving its technology.

Mr Leigh was speaking as the Committee published its 23rd Report of this Session, which examined the increase in the costs of the PFI deal for GCHQ's new accommodation, why GCHQ did not recognise the full scope and cost of technical transition earlier, and whether expenditure on the programme represented good value for money.

GCHQ has occupied over 50 buildings at two sites in Cheltenham since the 1950s and these were becoming difficult to maintain and unsuitable for modern information technology systems. In 1997 GCHQ decided to relocate the whole of the accommodation on a single site within a ten mile radius of Cheltenham under a Private Finance Initiative (PFI) deal. In June 2000, GCHQ signed a contract with IAS to provide new, fully serviced accommodation at Benhall at a net present cost of £489 million of 30 years. The building was completed early and was ready for occupation by GCHQ in September 2003.

Independently of the PFI deal, GCHQ retained responsibility for moving its technical capability into the new building, largely for security reasons. Between 1997 and 1999 GCHQ's cost estimates for this technical transition increased from £41 million to £450 million. The Treasury would not fund such a large increase but contributed £216 million to a revised budget of £308 million. GCHQ is now undertaking the technical transition in stages.

The Committee found that it is unwise to enter negotiations with a preferred bidder when key requirements have not been settled and priced. In this case there was a 21 month period of exclusive negotiations before the contract was signed in June 2000 during which time the price increased by 21%, not wholly in a competitive environment. When a preferred bidder is selected before all key requirements have been negotiated, the contractor is inevitably at a price advantage in closing the deal.

Major change programmes need to be managed as such from the outset. Any move of headquarters is likely to require numerous projects in addition to the building itself and to changes in working practices. In this case, which obviously was bound to involve a major technological move, GCHQ continued to perceive it as a building project for far too long. It was nearly three years after a PFI solution for its accommodation was first explored that GCHQ recognised the move as a major programme and designated it as such.

GCHQ experts failed to spot that development of IT networking during the 1990s would hugely complicate technical transition which effectively evolved into a major systems upgrade. The GCHQ Board was principally concerned with the feasibility of testing the PFI market for a new building and lost track of the scope and cost of the technical transition.

A further consequence of GCHQ's failure to see the programme as a whole is the staging of the technical transition for budgetary reasons. GCHQ will be keeping one of its existing sites open for 7 years longer than planned and incurring extra costs of £43 million.

GCHQ made a highly uncertain assumption that a conventionally procured building would have over-run its budget by 24%. That alone accounted for the comparative cost saving that GCHQ estimated the PFI deal would offer, but was simply an average over past projects and hides the wide range of outcomes.

Departments should not uncritically accept that PFI is the only way to improve on past construction performance. The Treasury and the Office of Government Commerce have told us that they are seeking to modernise construction and that there are non-PFI approaches to construction procurement that offer potential advantages. The public sector could learn from the experience of the private sector and reduce the risks in construction projects to provide a better measure of competition with PFI bids and be a spur to securing improved value for money.

Mr Leigh said today:

"There have been unwelcome cost increases on this programme. The process of moving the technology so critical to GCHQ's business will take longer and cost well over £300 million more than initially estimated, including extra running costs during the transition period. It is astonishing that GCHQ did not realise the extent of what would be involved much sooner, given how critical these systems are to its core business.

Between selecting IAS as the preferred bidder and signing the contract, costs rose by £85 million on the PFI deal for the building. Departments should settle key requirements and negotiate prices before selecting a preferred bidder otherwise the contractor will always have the upper hand, and the taxpayer will lose out."

Click here to view Report