Public Accounts Committee: Treasury must set out clear position on PFI

REPORT SUMMARY

PFI has been used as a finance vehicle to fund public infrastructure off balance sheet for reporting public debt.

The initial gain has been to the Treasury as the expenditure has been kept off the books. However, the ongoing costs to the institutions at the front line have been high and the contracts inflexible, and the subject of much public debate.

Yet the Treasury has no plans to assess the value for money of PFI and there is little evidence of a strategy for working with local public sector bodies to co-ordinate sharing of best practice about managing existing PFI deals.

The PAC raised concerns in 2011 and it is disappointing that so little progress has been made.

Liverpool City Council is paying £4 million a year for an empty school which underlines the worst extent of the inflexibility of PFI. This flawed deal will see almost £55.5million of taxpayer funds wasted since the school became empty in 2014.

Offshore funds have bought up about half of the equity in PFI and PF2 projects so that the projects’ owners are increasingly remote from the public service being delivered. In addition, offshore owners of these projects pay little tax, thereby reducing one of the benefits used to justify the contracts in the first place.

Add to this the additional costs (and profits) generated by variations to the contract and the deal is not working for the taxpayer.

The Government’s reform of PFI was ultimately fairly limited, with the new PF2 largely allowing greater transparency on a method that didn’t change fundamentally.

We note that there are only a handful of PF2 projects in the pipeline which suggests that the Government has lost faith in its own usage of PFI. If this is the case the Government should provide a clear explanation of its position.

The Infrastructure and Projects Authority has identified a need for the UK to spend £300 billion on infrastructure by 2020/21, yet PF2 is only being proposed for two projects which require total public and private investment of up to £7.8 billion.

The Treasury needs to be much clearer about its position on PFI – it maintains that the UK’s level of debt and accounting treatment of PFI and PF2 do not form any part of the decision about whether to use PFI, but it is changing the design of PF2 to prevent capital costs of future projects from counting towards UK national debt statistics.

COMMENT FROM PAC CHAIR MEG HILLIER MP

“Much has changed in the past quarter-century but Government’s inability to answer basic questions about PFI remains undimmed.

“It beggars belief that such apparently institutionalised fuzzy thinking over such large sums of public money should have prevailed for so long.

“The Treasury simply cannot support its assertion that PFI represents good value for money. Yet while Government is now seeking to collate the PFI data that does exist, it does not intend to publish the results of this work.

“This is unacceptable. Government must level with taxpayers about the value of PFI.

“The Treasury and Infrastructure and Projects Authority must also be far clearer about how it expects public bodies – many of whom are already locked into costly and inflexible PFI projects – to use PF2.

“We are not convinced the design of PF2 has fixed the underlying weaknesses of the previous method.

“Government’s own figures suggest the UK needs to spend some £300 billion on infrastructure over the next couple of years. It is critical that taxpayers are not further lumbered with excessive costs arising from poor contracting.”

CONCLUSIONS AND RECOMMENDATIONS

It is unacceptable that after more than 25 years the Treasury still has no data on benefits to show whether the PFI model provides value for money. PF2 is fundamentally the same model as PFI, with some improvements around the transparency of returns to equity investors. Treasury claims that PFI and PF2 provide a range of benefits, including greater certainty over construction costs, improved operational efficiency, and higher quality and well maintained assets, although these benefits can also be achieved with methods other than PFI. PFI and PF2 deals can only be value for money if these claimed benefits exceed the higher financing and other costs of the PFI model. When asked whether the benefits have justified the higher financing costs, the Treasury acknowledged that it is “an impossible question answer” because it does not have the facts needed. The Treasury told us that it considers collecting data on the benefits of PFI to be the responsibility of individual departments. It acknowledged that it has not attempted to quantify the benefits of using PFI, despite telling the previous Committee in 2011 that it would introduce benefits realisation assessment into its value for money guidance, for PFI projects that are underway. Without quantifying the benefits it is impossible to know whether PFI offers value for money, yet the Treasury continues to assert that it does. IPA told us that they have recruited a single member of staff, to look across the entire stock of PFI and PF2 projects and see what data exists, but the exercise will only collate rather than create data, and it is unclear how the Treasury and IPA will use the results. The Treasury wrote to the Committee after the hearing to say that this work will not culminate in a published report or a review. The Department for Education has been collecting data as part of the Priority School Building Project comparing PF2 and conventional public financing, and the Treasury expects this will be completed in summer 2018. 

Recommendation: The Treasury and IPA should, by April 2019, publish the results of their work in collecting data on the benefits of PFI, and set out what they will do to evaluate the value for money of PFI projects currently in operation in the absence of benefits data.

In addition to the Priority School Building Programme (PSBP), the Treasury and IPA should take a representative sample of PFI projects with a public sector comparator, for example roads and hospitals, and undertake in depth analysis of the suggested benefits of PFI. They should publish their findings by December 2018.

Some private investors have made large returns from PFI deals, suggesting that departments are overpaying for transferring the risks of projects to the private sector, one of the Treasury’s stated benefits of PFI. Private investors expect a financial return as reward for taking on risks transferred to the private sector under PFI deals. Investors in the M25 PFI deal made an estimated annual return of over 30% after selling their stake in the project after 8 years, more than double the 12%-15% annual returns expected over the life of most PFI deals. Equity returns this large may reflect errors in the pricing of risk transfer to the private sector at the time contracts were signed. We previously recommended that the Treasury should introduce arrangements to share the profits from PFI deals between private investors and the public sector, which the Treasury rejected. It instead sought to limit excessive returns to investors through equity funding competitions, which are relatively untested, and do nothing about the rate of returns from the existing stock of PFI projects. We heard from Professor Whitfield, Director of the European Services Strategy Unit, who told us that offshore infrastructure funds owned around half of the equity in PFI and PF2 projects, with the five largest of these offshore funds paying less than 1% in tax on their PFI profits. The Treasury told us that public procurement rules prevent discrimination against non-UK domicile investors. The amount of tax an investor pays is relevant, however, because the assessment of whether a prospective PFI deal is likely to provide value for money should include the corporation tax an investor is expected to pay as a benefit of PFI. The assessment may therefore overstate the benefits of the deal, and could lead to an incorrect value for money decision, if equity is subsequently purchased by offshore investment funds and corporation tax receipts are lower as a result.

Recommendation: The Treasury should calculate the returns to originating PFI equity investors when they sell-on their stake and use the information to inform pricing for future projects. This should include the domicile of project equity holders, and what impact a reduced tax take would have had on the original project value for money tests.

The Treasury and IPA are not doing enough to identify or address the impact of individual PFI projects on local budgets. Treasury has focused its attention on making changes to the PFI model for the rollout of PF2, rather than addressing continuing problems in the existing stock of over 700 PFI deals. PFI contracts are inherently inflexible, which can have considerable impact on budgets at a local level and in some cases wastes taxpayers’ money. Despite being empty since 2010, the Parklands School in Liverpool continues to cost the Council £4 million a year in PFI costs, and a total of £47 million by the end of the contract, unless a solution can be found. IPA has accepted that this is a “very unfortunate situation”, but asserted that PFI costs in general only represent a small fraction of departmental budgets. However, inflexible PFI costs can create immense pain at a local level. Departments and other public bodies have intervened in some cases where they decided that terminating the PFI contract represented value for money and best protected taxpayers’ money. The Treasury and IPA do not intervene in this way, and do not actively monitor where their intervention might be helpful, for example looking across all PFI contracts for other potential buy-out candidates. There are few signs of the Treasury and IPA adopting a centralised approach to addressing the problems of legacy PFI deals to help local bodies and benefit the taxpayer. For example the Treasury told us it was not specifically looking at the Barts Health NHS Trust’s PFI deal, where the Trust committed to payments of £7 billion over the contract, and paid £143.6 million in 2015 while running a deficit of over £90 million in the same year. The Treasury has taken no action to help local bodies recoup savings in insurance costs that PFI providers owe them, or to identify the scale of this problem.

Recommendation: The Treasury should publish how it monitors the impact of PFI at local level, how it shares good practice in contract management, and the circumstances in which it would proactively intervene to help those public bodies struggling with their PFI legacy.

The Treasury’s obvious desire to keep PF2 projects excluded from government debt statistics has created risks to value for money for the taxpayer. For a public body to use PFI and PF2, Treasury rules require it to demonstrate that this financing route provides better value for money than conventional government procurement. However, accounting rules create incentives for public bodies to use PFI for reasons other than value for money, and these incentives remain under PF2. Under national accounting rules, most PFI debt is recorded ‘off-balance sheet’ and excluded from public debt calculations, which is advantageous for the Treasury. The Treasury told us that the UK’s level of public debt played no part in departments’ decision to use PF2. Despite this, Treasury is nonetheless making changes to PF2 to ensure it remains off-balance sheet in the National Accounts. PF2 includes gain-share arrangements which entitle the public sector to a percentage share of any savings made when private PFI companies reduce their costs of borrowing through refinancing, reflecting recommendations of previous Committees. Under the National Accounts rules, the higher the percentage the public sector is entitled to, the more likely PF2 will be recorded on-balance sheet, and count towards UK debt. To avoid this, the Treasury is significantly reducing the amount the public sector will receive if savings are made, and it accepts that this will negatively impact the value for money of PF2.   

Recommendation: If it is no longer a Treasury requirement to keep PFI contracts “off balance sheet” for reporting public debt, it should revisit the original plans for PF2, and the subsequent adjustments made to keep it off balance sheet, and ensure the focus is on value for money and not accounting treatment. The Treasury should write to the Committee before any new PF2 deals are signed, demonstrating how the changes introduced under PF2 are not influenced by balance sheet treatment.

PF2 is hardly being used, and IPA hasn’t made clear when its use is appropriate. PFI has been used less over time, falling from a peak of around 60 PFI projects in 2007-08 to no new contracts at all in the last two years. Some departments say this is because of concerns about the cost efficiency and value for money of the model. Since its introduction in 2012, PF2 has only been used for six projects, with a combined capital cost of £920 million. IPA has identified a need for the UK to spend £300 billion on infrastructure by 2020-21, with over 50% of this to be privately financed. The Treasury and IPA were unclear about the scale of the pipeline and money involved. They subsequently wrote to us saying government has plans to invest between £6 billion and £7.8 billion in two new projects using PF2. It is therefore unclear how PF2, and other private finance sources, will meet the needs that the IPA identified. IPA told us that the decline in the use of PF2 is in part because government is getting better at designing and constructing projects on time and to budget, and therefore the hurdle for PF2 projects to demonstrate better value for money than public sector procurement is higher than it was for PFI. It is unclear which projects are suitable for PF2 investment, if the benefits of transferring the risks of delivering projects have fallen and only a small number of low risk projects are being considered. If the public sector can borrow more cheaply as well as manage the risk of project overruns and delays more effectively than before, then the future scope of PF2 investment appears limited.

Recommendation: The Treasury and IPA should set out more clearly the nature and level of risk they consider appropriate to transfer to the private sector in PF2 projects, and outline a clearer view of how it expects public bodies to use PF2 and other types of private finance in future.