Last week the UK's National Audit Office (NAO) released a report on the rationale, costs, and benefits of the country's much acclaimed Private Finance Initiative (PFI) to build infrastructure, the pioneering PPP experiment. The report is a very damning indictment of PFI, concluding that the country has incurred billions of pounds in extra costs for no clear benefit.
The PFI was introduced in UK in 1992 as an innovative way to leverage private capital to supplement governments investments in infrastructure. In 2010, the UK government initiated PFI2, or PF2, with some learnings from PFI and greater role for the government, focusing on smaller deals, involving facilities and services rather than buildings. Over the past 20 years, PFI investments in social and economic infrastructure have averaged around £3 bn every year, small compared to the total £50 bn capital investments of the government. There are currently 716 PFI and PF2 projects operational or under construction with a total capital value of £59.4 bn.
Consider this on the presumed efficiency gains,
The Department for Education is currently collecting data and developing methodology and has, so far, found that the financing route has little or no effect on the construction costs of schools being built as part of the Priority School Building Programme (PSBP)... Our work on PFI hospitals found no evidence of operational efficiency: the costs of services in the samples we analysed were similar. Some of those data are more than 10 years old. More recent data from the NHS London Procurement Partnership shows that the cost of services, like cleaning, in London hospitals is higher under PFI contracts.
On the illusion of budget flexibility, or "fiscal illusion" as a July 2017 report of the Office of Budget Responsibility described,
Private finance increases departments’ budget exibility and spending power in the short term, as no upfront capital outlay is required. But departments face a long-term financial commitment – any additional investment will need to be paid back. For example, in the first 12 years of PFI use in the health sector, PFI resulted in extra capital investment for the Department of Health and Social Care (the Department) of around £0.9 billion each year on average: £0.5 billion a year more than the a average annual spending of the Department on operational PFI projects over the same period. However, in recent years PFI has been used much less by the Department and the operational PFI contracts, which cost over £2 billion a year, have reduced the Department’s budget flexibility.
On the higher cost of capital with private finance,
Private finance procurement results in additional costs compared to publicly nanced procurement, the most visible being the higher cost of nance. The 2010 National Infrastructure Plan estimated an indicative cost of capital for PFI as 2% to 3.75% above the cost of government gilts.Data collected by IPA on PFI and PF2 deals entered into since 2013 show that debt and equity investors are forecast to receive a return of between 2% and 4% above government borrowing. However, some 2013 deals, agreed when credit market conditions were poor, projected an annual return for debt and equity investors of over 8%; this was more than 5% higher than the cost of government borrowing at the time.20 Small changes to the cost of capital can have a signi cant impact on costs – as an illustration: paying off a debt of £100 million over 30 years with interest of 2% costs £34 million in interest; at 4% this more than doubles to £73 million.
There are other ways in which PFI adds to cost. Insurance, holding surplus cash to meet stricter lender requirements, use of advisers by both government and private partner given the complex nature of private finance procurement, arrangement fees for raising money, and so on are all additional layers. The net result in terms of total capital cost,
The higher cost of finance, combined with these other costs, means that overall cash spending on PFI and PF2 projects is higher than publicly nanced alternatives. The Department for Education has estimated the expected spend on PF2 schools compared with a public sector comparator (PSC). Our analysis of these data for one group of schools shows that PF2 costs are around forty per cent higher than the costs of a project nanced by government borrowing. The Treasury Committee undertook a similar analysis in 2011, which estimated the cost of a privately nanced hospital to be 70% higher than the PSC.
Consider these snippets of the costs inflicted by way of operational inflexibility,
The PFI structure means that changes in contracts can be expensive with lenders and investors charging administrative and management fees. For example, additional capital works of approximately £60,000 in a local authority PFI school increased to over £100,000 once fees were factored in – the local authority challenged this and the SPV agreed to reduce some of the management and approval fees although bank fees of £20,000 will still have to be paid. Department of Health and Social Care papers similarly highlight that some trusts with PFI facilities have to use alternative forms of procurement for capital variations. Government can also be locked into paying for services it no longer requires: for example, Liverpool City Council is paying around £4 million each year for Parklands High School which is now empty. Between 2017-18 and the contract end in 2027-28, it will pay an estimated £47 million, which includes interest, debt and facilities management payments, if no changes are made to the contract. The school cost an estimated £24 million to build.
On pricing in life-cycle risks,
The Department of Health, in a paper on PFI prepared for HM Treasury in 2012, noted that “there is an inbuilt incentive to price cautiously for lifecycle risk, requiring the build up of significant reserves. This may not necessarily result in optimum value for money for the public sector, although data illustrating out-turn costs for lifecycle is scarce”. It also reported that bidders were currently pricing the cost of insurance at a 20% premium to the market price in order to provide protection against future price rises. To mitigate this, HM Treasury introduced insurance gain-share arrangements in the standard PFI contract (paragraphs 2.12–2.13). There are also other risks, for example potential tax increases, that investors may factor into the prices they bid at the outset. These risks may not materialise and in some cases subsequent changes, such as reductions in corporation tax rates, have increased rather than reduced investor returns.
And on the importance of the discount rate used to assess value for money, or comparing with public sector comparators,
The VfM assessment compares private nance costs with a government discount rate of 3.5%, which is 6.09% with inflation, known as the Social Time Preference Rate (STPR), which is higher than government’s actual borrowing costs. The higher the rate applied, the lower the present value of future payments. For example a payment of £100 in 12 years will have a present value of just £49 when discounted by the STPR. Discounting using a lower discount rate, which compares private finance with the actual cost of government borrowing, results in fewer private nance deals being assessed as VfM... HM Treasury does not consider the cost of government borrowing to be relevant in making financing decisions on PFI and PF2 deals. However, other countries, such as Germany and the United States, do compare the cost of private finance with government borrowing costs when assessing financing options like PFI.
And its legacy on the fiscal balance sheet,
The public sector will still be making PFI unitary charge payments to private nance companies in the 2040s. Future payments for existing projects are forecast to total £199 billion from 2017-18 onwards – an average of £7.7 billion a year over the next 25 years. In 2016-17, total payments amounted to £10.3 billion, of which 59% related to four departments (Health and Social Care; Defence; Education and Transport). These payments cover financing costs (debt and interest payments and a return to shareholders) and operational costs. Public bodies also have to pay for maintenance and operational costs of publicly financed buildings.
This blog has been a very consistent critic of the indiscriminate use of PPPs. It has argued that PPPs should be deployed with great discrimination, the cost of capital is much cheaper for governments than the private sector for the same project, construction and commissioning risks are best borne by governments, O&M poses massive risks of skimping on investments and cutting corners on quality, and the near inevitability of renegotiations.