FIN(3)-PPP-011
A response to the Assembly Finance Committee’s Inquiry into Public-Private Partnerships
November 2007
PCS Wales, Phoenix House, 8 Cathedral Road, Cardiff CF11 9LJ.
Tel: 029 20 666363. Fax: 02920 666501. Website: www.pcs.org.uk/wales.
PFI - the experience in the civil and public services
About PCS
The Public and Commercial Services Union (PCS), the UK’s sixth largest union, principally represents civil and public servants in central government. It has more than 315,000 members - including 22,000 in Wales - in over 200 departments and agencies and in parts of government transferred to the private sector. In all, around 10% of its membership works within the private sector. PCS is and is affiliated to the TUC. Its general secretary is Mark Serwotka and its president is Janice Godrich.
The Private Finance Initiative (PFI)
Under PFI, private contractors pay for the construction cost of building projects in the public sector, and lease the finished project back to the public sector for periods of up to 30 years. Nine-tenths of new hospitals and schools built in the UK since 1997 have been built under PFI.
There has been widespread experience of PFI and PPP schemes in central government departments and other national public bodies, beginning soon after its introduction by the Thatcher government in 1992. IT support, facilities management and security have been among the functions most often contracted-out but PFI/PPP now affects areas as diverse as support services in the Ministry of Defence, the storage of our most private and personal data and National Savings account records. The present Government’s 'efficiency programme’ has seen the extension of privatisation in this and other forms: the 104,000 job cuts announced by Gordon Brown in 2004 have meant more outsourcing of services in order to plug gaps in service delivery and this announcement was accompanied by plans to sell off £30 billion of public assets.
Along with other public sector unions, PCS opposes PFI because of its effects on public services, jobs and working conditions, and the public sector ethos. We consider that it does not represent value for money; takes longer to complete projects; and leaves future generations with massive debts to service. PFI is attractive to the Government because it takes the cost of borrowing off the public sector balance sheet, thus improving the government’s finances. However, this is proving an ephemeral saving, whittled away over time. The cost of borrowing is much higher in the private sector, and so annual expenditures on PFI are higher than if projects were funded through conventional means. In addition, as contracts are of approximately 30 years’ duration, PFI locks future governments into huge, on-going spending commitments. PFI does not transfer risk to the private sector; the UK taxpayer must still bail out PFI failures (of which there are an increasing number) as essential public services cannot be allowed to fail.
The weight of evidence is that PFI does not offer long-term value for money for the government or the service user. It is ideological consensus, therefore, not 'evidence based policy making’, which drives the extension of privatisation and outsourcing into more and more public services, even where the returns are most dubious, as with London Underground and the NHS.
Risks in using PFI
Public sector - the ultimate guarantor
Nowhere in the world have universal public services been delivered solely by the private sector. There will always be a need for a public sector to provide those services that the market alone cannot adequately deliver. These services exist not for profit, but to support the social, economic and environmental well being of communities. The Government takes on responsibility for their funding and the regulation of their quality and delivery. It also has responsibility for the risk of service failure - a risk that as we have seen in the cases of Railtrack and National Air Traffic Services (NATS) can never be transferred, no matter who provides the service. The principal risks transferred to the private sector in most PFI projects are those met during the construction phase, risks that disappear at an early stage of the project. Despite this, the risks are treated as if they were spread over the whole length of the contract and it is therefore very profitable for contractors to refinance projects. In some contracts, the private contractor has profited from fluctuations in the interest rate. One company recently made huge profits from a prison contract when the interest rate dropped between the time it submitted its bid and when it was accepted.
Loss of control of service
Although contracts have penalty clauses attached to them, there is little evidence that they are effective. A penalty clause can only be invoked following service failure, so it is very much a question of shutting the stable door after the horse has bolted. Long term contracts also fail to provide the flexibility that is needed to alter the focus of resources or adapt the type of service required.
Reduced labour costs
Private contracts tend to make savings by reducing salaries and cutting jobs. We have often seen the emergence of a two-tier workforce, with terms and conditions much worse for new employees than for staff transferred under TUPE, causing disunity and lower morale. According to a PCS survey, pay differences between administrative staff working in privatised areas can vary by as much as £7,000, despite staff doing the same, or very similar, work and less than a quarter of private sector contractors offer 'defined benefits' pension schemes for all their staff. One in seven also fail to make employer contributions into staff stakeholder pensions schemes. The higher staff turnover associated with private employers also makes it difficult for employers to meet civil service vetting standards; this can be a particular problem when work is secondarily outsourced to a temporary labour agency.
Private sector inefficiency
There is no reliable evidence that the private sector is more efficient than the public sector. Private companies are not producing the anticipated improvements in delivery time or cost, nor are they meeting quality standards - as the court cases involving Railtrack and Balfour Beatty testify. PPPs do not give long-term value for money. Critical to the case for PPPs are a number of adjustments to ensure that the public sector alternative can never win the value-for-money competition. When making comparisons PFI schemes exaggerate the delivery time and overrun expenses of public projects, making PFI schemes appear to be better value for money. Commercial confidentiality has often been used to stifle public scrutiny of PFI proposals.
Imperfect Competition
One risk is that too much is thrown into the contract, with the inclusion of highly specialised areas that will present additional problems if privatised. The risk has two areas. Where there is limited provision of a specialist service, privatisation can actually increase costs. This can be through a monopoly supplier driving up the costs of the contract if any new requirements become apparent, especially if the contract is for longer than 3-5 years. The price is fixed against known requirements at the time of transfer. If these change, the contractor can insert additional charges. If there is not enough competition for the department to challenge this, then they will have to pay whatever it costs.
Consortia
In some instances employers may work together to win a contract, but revert back to their normal competitive state post-transfer. Where staff had previously worked together for a government department, they are now divided by the competing employers, and sharing routine information becomes difficult.
As the employers compete, they fight over the more lucrative aspects of the privatised contract, and this can lead to further TUPE transfers between the employers.
Accountability and democracy
When public services are controlled by government departments, the service remains accountable to the public. But once services are privatised they become accountable to share holders and big business.
Take the Defence Training Review in the Ministry of Defence for example. This is a 25 year contract worth £19 billion, making it the largest private finance initiative (PFI) - yet the contract cannot be reviewed for at least 15 years.
HMRC and Mapeley (UK) Ltd: a case study of problems with PFI
In 2000, Mapeley - a consortium of a dozen banks and finance companies based in a tax haven, the British Virgin Islands - won the 20-year contract to run the entire estate of Customs and Excise and Inland Revenue (which merged in 2005 to form HM Revenue & Customs) as part of STEPS - Strategic Transfer of the Estate to the Private Sector. It now owns, manages and maintains the entire estate of 600 buildings - worth an estimated £1 billion. In addition to the fact that UK tax offices are owned by a company that pays no corporation tax to the exchequer, there have been a number of problems with the PFI deal and serious failings in Mapeley's management of the estate:
Seven months into the contract, the Inland Revenue wrote a note to Mapeley effectively giving assurances that any future losses would be made good by government after financial problems in other parts of the company which threatened Mapeley with bankruptcy, in which case the buildings would have reverted to the company's bankers, the Royal Bank of Scotland.
Within three years, payments had rocketed to an average of £307 million a year rather than the anticipated facility price of £170 million, in addition to over 40 firms of consultants and advisors being used on the Mapeley deal at a cost to the taxpayer of £13 million to April 2004.
The Government expected to secure savings by giving up accommodation as staff numbers were cut, but the terms of the PFI contract impose a financial penalty, in the case of 71% of the HMRC estate, if the department should vacate those offices before the end of a 20-year period.
In one office the lights failed. Rather than fix the problem, Mapeley used a fairy light style arrangement of bulbs suspended from the ceiling.
In another workplace, the fire alarm system became dysfunctional. Mapeley declined to repair or replace the necessary equipment because this would be too expensive and PCS was forced to raise the matter as an emergency health and safety issue in order to resolve it.
Mapeley banned third-party organisations from premises including organisations long associated with the civil service including Benenden Health Care and the Civil Service Insurance Society.
When ceiling tiles began to fall down at Portcullis House, Southend, Mapeley took out all the ceilings, leaving just bare pipes and concretes.
Some other examples of problems with PFI deals in the civil service
The Home Office's Immigration and Nationality Directorate engaged Siemens Business Services (SBS) in April 1996 to develop an IT-based Casework programme which would improve the effectiveness of UK immigration control and make the Directorate more efficient, flexible and responsive. However, the implementation of the system was delayed by nearly a year after the target date of October 1998, contributing to an increase in the backlog of cases awaiting processing and decision, from 151,000 in May 1998 to 219,000 a year later, and requiring the employment of large numbers of additional staff.
In May 1996, the Benefits Agency and Post Office Counters Ltd jointly awarded a contract to Pathway, a subsidiary of ICL, to develop a magnetic stripe payment card to replace paper-based methods of paying social security benefits and to automate the payment of these benefits at post offices across the UK. The project took far longer than expected, however, and was scrapped after three years with an estimated expenditure of £1 billion, by which time the card technology employed was already outdated.
In 1997, SBS won another PFI contract from the UK Passport Agency to introduce a new computer system, which was intended to improve the efficiency of the passport-issuing process and increase security. The Agency had planned to roll-out the new processing system to all its offices within a tight timetable in late 1998, before the busy season, but the implementation of the new system caused serious delays in dealing with passport applications. Processing times reached 50 days in July 1999 (against the Agency’s target of 10 working days) and there was widespread public anxiety as the Agency's telephone service became overloaded and queues grew outside offices. The Agency had to employ additional staff to cope with a bottleneck of 565,000 passport applications and the cost was passed on to the public, who saw the price of passports rise by 33 per cent.
Andersen Consulting secured a PFI contract to provide a National Insurance computer system (NIRS 2) for the Inland Revenue, at a cost of £170 million. Delays in implementing NIRS2 led to the calculation of thousands of short term and long term benefits on an interim or emergency basis. By January 1999, it was estimated that potentially some 172,000 pensioners could be being underpaid in respect of their SERPS by between £0.01 and £100 a week. Delays in payments to personal pension holders meant that the Government had to pay an estimated £38 million in compensation for loss of investment income and recipients of Retirement Pension, Widows Benefit or Incapacity Benefit also received a flat-rate compensation payment of £10 for 'unreasonable delays’.
The Air Traffic Control System installed at Swanwick in Hampshire by Lockheed Martin was originally due for delivery in 1996 but was plagued by problems and delays and finally became operational in 2002, by which times costs had more than doubled. Even then, the software needed lengthy debugging and the text on the screens was too small to be readable. National Air Traffic Services (NATS) was itself the subject of a private-public partnership in July 2001, with a 46% stake in the organisation being sold to the Airline Group, a consortium of seven UK-based airlines. When NATS was hit by the severe downturn in air traffic that followed the terrorist attacks of September 11th 2001, the Airline Group was unwilling or unable to invest more money in the service and the Government had to secure new investment of £130 million, half of which it provided itself, while the other £65 million came from BAA plc, which was brought in as a new investor.
Despite its serious problems with the Benefits Payment Card project, ICL was chosen in July 1998 as the preferred bidder to provide a new computer system for Magistrates' Courts (the Libra project). Within three months, ICL had increased its bid from £146 million to £184 million and a year later, the company - now anticipating a £39 million deficit over the life of the deal - sought to renegotiate, securing a revised contract for £319 million over 14.5 years in May 2000. Little more than year later, however, ICL told the Lord Chancellor’s Department that its forecast losses were now so high that it could not continue with the contract unless it was substantially renegotiated. Having breached its contract by failing to meet the delivery date for core software at the first site, the company told the department that it would repudiate the contract unless the Department agreed to cover an expected loss of up to £200 million. Rather than terminate the contract and sue for damages, the department signed a legally binding Memorandum of Understanding with the company. It decided that it was no longer affordable for ICL to continue with the whole contract and so in July 2002 it signed a revised contract with ICL (now known as Fujitsu Services) for £232 million over 8.5 years to supply only the infrastructure element of Libra, going on to sign separate contracts with other companies for the core software application and for the roll-out of the programme. Ultimately, the project cost £134 million more than originally bid and was described by the Public Accounts Committee as "one of the worst PFI deals that we have seen".
The Criminal Records Bureau disclosure service run by Capita nearly collapsed at its launch because the Home Office accepted the cheapest quote for the service, ignoring warnings about its feasibility from the two unsuccessful bidders and assuming, without evidence, that 85% of those applying for access to criminal records would do so by telephone or online. When more than 80% of applications arrived by post, the service, which had not been designed to cope with such a load, collapsed, causing huge backlogs of checks and leaving tens of thousands of people unable to start work because they had not gone through the vetting procedure. The taxpayer had to bail out the service at a cost of £68m, the cost of standard checks more than doubled, from £12 to £28m, and performance targets had to be cut, after the CRB managed to issue only 19.4% of standard disclosures within one week, against a target of 95%.
EDS provided HM Revenue & Customs with an IT system for processing the new Child Tax Credit and Working Tax Credit. When the new schemes went live in April 2003, problems with the IT systems delayed payments, leaving hundreds of thousands of families without financial support for several months, and forcing HMRC to arrange for interim payments to be made by girocheque ('manual payments’) at local offices. The system was finally stabilised after around four months but problems continued: 455,000 households received duplicate payments when the system failed to recognise that the manual payment had already been made and there were £45 million of overpayments to 60,000 couples whose change in circumstance led to the second partners having their income zero-rated. HMRC sought compensation from EDS - whom it held responsible for £209 million of losses - and ultimately settled in November 2005 for £71.25 million, £26.5 million of which was dependent on EDS winning future work from the Government.
The head of the Child Support Agency had to resign in November 2004 over his role in overseeing a new computer system that had made payments to only one in eight single parents - 61,000 out of 478,000 - in the previous 20 months. The system, provided by EDS under a PFI contract, was introduced in March 2003, and cost £456m but the average waiting time for payments subsequently increased from the six-week target to 22 weeks. Twenty months later, 88% of claimants applying under the EDS system had still not received a payment and a quarter of the 742,000 applications under the old computer system had still not been processed. The new system had a backlog of more than 270,000 claims, with about 160,000 of these apparently "stuck", partly because of the IT system, which then Work and Pensions Secretary Alan Johnson described as "problematic and unstable". The Commons Works and Pensions Committee was very critical of the system and even EDS’ own review described it as "badly designed, badly delivered, badly tested and badly implemented". The CSA had to write off £1bn of debt and the problems have still not been fully resolved.
The Defence Training Academy
A recent example of a central government PFI deal affecting Wales is the award to the Metrix Consortium of a £19 billion contract to provide training for the armed forces at St Athan in the Vale of Glamorgan. This consists of a programme to:
Rationalise defence training across the Ministry of Defence (MoD);
Reduce the number of sites where training is conducted;
Use a PFI model to build the new training infrastructure to replace the current training and accommodation facilities; and
Privatise training support and delivery.
It has generally been welcomed in Wales because of the prospect of thousands of jobs coming to St. Athan and the surrounding area. There are, however, some significant question-marks over the project.
The extent of the jobs gain and savings
The Metrix project originally consisted of two separate packages, which together accounted for an estimated total of 4,000 training staff who would have been in scope for transfer to the private sector, and relocation to St. Athan from numerous locations across the country. It was announced recently, however, that 'Package 2’ - covering logistics, security, intelligence, personnel and policing - is no longer considered economical and has been removed from the scope of the project. This accounts for around 2,000 staff and more sites than Package 1 and therefore calls into question the £2 billion savings that the deal was supposed to deliver
Loss of experience
PCS is extremely concerned about the assumption that the staff providing the services within and around facilities should automatically transfer. In discussions with the MoD, PCS has consistently raised concerns over the risk to defence training of instructional and support staff not transferring to the preferred bidder. Recent transfers indicate that between 80% and 90% of staff will not relocate. Whilst the MoD recognises this as a risk, it has not addressed the scenario, nor indeed offered any solutions. Instead the MoD states that such a risk will transfer to the bidder and is therefore not the concern of the department.
The belief that a single training specialist included in the bidding consortium can deliver the same high standard provided by the current training delivery staff is questionable. If large numbers of training staff were to decide against moving to the private sector, as our indicators suggest, we believe it is unlikely that the bidders could replace them with equally experienced instructors. The project therefore raises the spectre of an incalculable skills drain in specialist training.
Lack of accountability
PCS is concerned that the department will not to be able to control the costs of the project. The contract’s length and the fact that the first breakpoint - unprecedentedly - will not be until 15 years into the contract, will almost certainly lead to spiralling costs. Numerous factors could influence training requirements, including future deployments, new equipment and the quality of new recruits. The changing nature of Britain’s defence response will also impact upon the training requirement, and for these changes the private sector will exact a high financial price. A recent NAO report into MoD contracts noted that over 50% of contracts had to be altered due to changes in specification. In a twenty five year contract in an area as fast moving as defence and its associated training requirement, it is obvious that a large number of contract amendments will be made.
VOSA
Another potential PFI project, about which PCS currently has concerns, involves the Vehicle & Operator Services Agency (VOSA), which is responsible for monitoring the issue of MoT certificates, as well as carrying out annual inspections and roadside checks on buses, heavy goods vehicles and coaches. The Agency’s management is currently considering proposals to outsource the bulk of its activities, including training services; operator licensing administration; lorry & bus testing; prosecution and legal services; part of its administrative and support services; and estates rationalisation.
This Agency has vital functions with regard to road safety, such as safety checking buses, heavy goods vehicles and coaches; in regulating and policing the MOT, it catches thousands of unsafe vehicles each year. PCS has a particular concern about the impact of outsourcing on the agency’s responsibility for bus safety inspections, given the implications for the safety of school children who use buses for daily travel to and from school and for school trips. We would have far less confidence that road safety remained the overriding priority if these inspections were to be carried out by a commercial enterprise, concerned to make a profit. These concerns are borne out by the experience of mainland Europe, where the public sector has a far lesser role in vehicle safety testing. According to a written parliamentary answer provided by the then-Minister of State for Transport, Stephen Ladyman, on 17 October 2006, checks on HGVs entering ports in the south-east of England led to almost half of the vehicles in question having prohibition notices served on them. We would therefore prefer to see safety inspections continue to be carried out by independent and publicly-accountable Government inspectors employed by VOSA.
Conclusion
The view of PCS is that PFI is fraught with difficulties as way of delivering public services. While it is superficially attractive, as it allows funding to be raised quickly without immediate resort to taxation or government borrowing, it has less attractive long-term implications in terms of service quality, employee well-being and value for money. While some schemes may have been implemented relatively smoothly, there have been a number of disastrous experiences in relation to projects within the civil service, particularly those involving IT systems, and we have given numerous examples of these above. We would therefore strongly caution the Assembly against the adoption of PFI as a method of providing services in Wales.
___________________________________________________________________________________________
Further Information
Further information about this submission can be obtained from:
Darren Williams
Campaigns Officer (Wales)
Public and Commercial Services Union
Phoenix House, 8 Cathedral Road, Cardiff CF11 9LJ.
Tel: 029 20 666363 / 07921 490261
Email: darrenw@pcs.org.uk.
