In the wake of the Carillion collapse it ispertinent to ask whether PFI is still an attractive and sensible option for theprovision of infrastructure and services. Much political debate is centred onthe topic and this report will look at the arguments surrounding not onlyCarillion, and the government’s handling of its demise, but also the policiesand doctrines surrounding PFI itself and what place it has in the UK andbeyond, going forward. 1 What is PFI According to the Infrastructure and ProjectsAuthority (IPA) over £300 billion is required to be invested in infrastructure by2021. There are various ways by which the government can pay for infrastructure.Historically, most of the finance for investment in infrastructure is drawnfrom tax receipts and/or government borrowing, and the government intends tospend up to 1.
2% of gross domestic product (GDP) each year on economic infrastructurebetween 2020 and 2050.1 However, a significant proportion of theplanned infrastructure will also be privately financed. One private financingroute is a Public Private Partnership (PPP) such as PFI and PF2. There are over700 PFI and PF2 projects in the UK. Over the last 20 years capital investmentusing PFI and PF2 has averaged around £3 billion a year – this isrelatively small in comparison to publicly financed government capitalinvestment which currently amounts to around £50 billion a year. The fundamental difference between conventional publicprocurement and PFI procurement for capital investment relates to which partyraises finance for the asset’s construction. In conventional procurementthe private sector is still involved (private contractors build the asset) butthe public sector provides the finance. When the public sector procures anasset using PFI, a private company – a Special Purpose Vehicle (SPV) – isformed and it raises finance from debt and equity investors to pay forconstruction.
Once the asset is constructed and available for use the taxpayermakes ‘unitary charge’ payments to the SPV over the contract term, usually 25to 30 years. This charge includes debt and interest repayments,shareholder dividends, asset maintenance, and in some cases other services likecleaning. These payments will be agreed at the start of the contract and someor all of them will be linked to inflation. All of these aspects remain in thePF2 model which replaced PFI in 2012; the costs and benefits of PFI discussedin this report also apply to PF2. HM Treasury made the introduction of PFI possible in1989 when it retired the ‘Ryrie-Rules’ (which had discouraged public sectorprojects from being privately financed) and announced that it would allowadditional privately financed investment in roads. In 1992, the use of PFI wasextended to other sectors and the name ‘Private Finance Initiative’ was usedfor the first time.2 Other changes were later introduced to allowfor PFI to be used within local bodies, for example the Department ofHealth and Social Care provides a Deed of Safeguard for PFI health deals whichguarantees PFI payments. (HM Treasury – National Audit Office, 2018) Many countries around the world have adopted PPP’sand PFI’s in similar forms to that seen in the UK.
Paul Drechsler, chair of theConfederation of British Industry (CBI), however, questioned recently why PFI”does not work here in the UK but does elsewhere. This may lead us to assumethat issues encountered with PFI in the UK are less an issue with the modelitself and more in its implementation and management by government bodies. 2 How did we get here andwhere are we going PFI was created in order to address the belief thatconstruction engineers are better trained to understand risk in constructionthan civil servants in government departments or local councils An early example of PFI being used was for thechannel tunnel, which encountered numerous financial and contractualdifficulties before major restructuring allowed it to eventually operate asintended. Since then PFI has often been used for hospitals and schools,projects which typically have large capital costs that struggling governmentdepartments would struggle to meet. During the years of the Blair government,it became incredibly attractive to Labour to use PFI as a funding model formany of these projects. PFI allowed large eye-catching construction projects togo ahead whilst keeping the treasury balance sheets free of huge short termspending liabilities. The Labour party, under Jeremy Corbyn and JohnMcDonnell, has made a pledge to nationalise contracts currently under PFI withthe UK government.
The National Audit Office (NAO) has just produced a reportgiving them plenty of justification, investigating the basis for PFI and findingit lacking. They reported that they could find no financial benefit for thetaxpayer in PFI. If implemented competently and for the correct reasons, however,PFI is not in itself inherently bad. Unfortunately, this is far from what hashappened.
Due in part as a response to criticism, the number and value of PFIprojects has fallen sharply in recent years. The annual charges for the UK’s 700 operational PFIdeals added up to £10.3bn in 2016-17 (0.5 per cent of gross domestic product).Even if no new deals were launched, future charges (which will continue untilthe 2040s) will amount to £199bn. (NAO 2018) Yet this experiment, which originated with a changein Treasury rules in 1989 and continued under subsequent governments, isinteresting. It is not unreasonable for government to contract for servicesfrom private suppliers. It is certainly reasonable for the UK and othergovernments to attempt to learn from this influential experiment.
For PFI to work, the asset being built needs togenerate its own revenue stream that can accrue to the contracting team that foundthe money to get the project built. Early examples were: the Channel Tunnel, besetby difficulties because of a flawed initial project structure and requiring significantfinancial and contractual restructuring in order to get it on track; and theDartford Crossing, which was successful on almost every front, mostly due tothe tolls levied on users who have little alternative to using it. Then the finance teams in the government realisedthat with a little creative accounting PFI could be used to shift publicspending off the balance sheet and massage the numbers to meet fiscal rules.PFI relies on a revenue stream.
Spurious revenue streams were produced for prisons,hospitals, schools and government offices by giving the SPV’s contracts for20-30 years to stick around and look busy, doing the washing up and emptyingthe bins and. Buy now, pay later meant some hefty up-front savings wereready to be made – but the whole life cost is huge. But the private ownersweren’t even the ones making the money. It was the creditors. The UK treasury is of the opinion that the transferof risk away from the public sector to the private can lead to benefits whichmay outweigh the increased costs of financing. The private sector has a strong incentiveto construct assets to a tight budget due to the risk it bears for costoverruns during construction. It is standard practice for the contracts toinclude operation of the asset and as such the SPV is incentivised to developplans to reduce long term running costs from the outset. PFI also requiresassets to be well maintained throughout the period of the contract which couldhave many knock on benefits including longer asset life.
In response, however, the NAO finds, that whilst theSPV has an incentive to deliver the project within budget, increasing costcertainty, this does not preclude that the lowest possible cost. In reality,and particularly for complex projects, unforeseen costs are likely to becovered by higher prices charged by bidders. There is also “no evidence” found by the auditoffice that assets are operated more efficiently. Some services may even bemore expensive, possibly due to increased standards. Lastly, standards of maintenance are found to behigher in projects under PFI. This appears to be because, under budgetarysqueezes, maintenance spending is quickly reduced by government departments,this is more difficult under PFI. One can argue that it is desirable forgovernment to have such flexibility.
Yet one can also argue that, as is thecase with the palace of Westminster, the government occasionally needs to berestricted from postponing necessary maintenance. These arguments for PFI, however, are weak. As theNAO also argues, the cost of capital in a PFI project is far higher than forthe government. An effective counter-argument is that government borrowing issubject to an implicit subsidy from taxpayers. This subsidy, as with theimplicit subsidy to banks which are seen as too-big-to-fail, represents anunpriced insurance contract.
That government funding looks cheaper due to thisallows us to expose an illusion. In turn, PFI begins to look less unreasonable. There is a politically important and inherentlywrong reason for PFI. This is that PFI is not included in conventionalstatements for public debt, in essence, off the balance sheet of the governmentdepartment. A “fiscal illusion” is what the Office for Budget Responsibilityand the International Monetary Fund have called this. Just as governmentborrowing establishes a long-term contractual liability, so too does a PFIcontract. Another implicit liability is that any publicly owned asset isobliged to be operated.
Accounting procedures which deal with these two methodsof financing services differently are deceitful. It is to the Treasury’sdiscredit that it continues to encourage misleading accounting that is likelyto lead to mistaken decisions. The Treasury should make serious attempts totreat its balance sheet and finances with due rigour and honesty. 3 Carillion Carillion was putinto liquidation in January after the money to continue running its widespectrum of government contracts, including road building, serving schooldinners and maintenance of prisons, finally dried up. The second-largestconstruction company in the UK was also a large beneficiary of the raft of PFIcontracts struck, over the past 25 years, between private companies and thegovernment. The collapse has cast doubt over the futures of its 30,000subcontractors, 43,000 employees and the future completion of three decadesworth of government contracts.
Carillion’s downfall has led the many detractorsof such deals — notably Jeremy Corbyn et al. — to trumpet the proposal to takecontrol of PFI projects back in-house. Carillion did not go bust because of privatisationof public services. If anything, from one viewpoint, it shows that thosecontracting out public contracts were getting a good deal. There will always bepublic projects contracted out to the private sector.
Much of the increase inpublic investment planned by Labour if it wins the next election will beundertaken by private firms. Getting the contracting relationship right isdifficult and fraught with dangers. It is a tale ofmiscalculation and commercial overreach. But also the tale of a political ideology.Carillion were the embodiment of small state Conservatism, pioneered byMargaret Thatcher and which swept the world.
Where once public services were providedby governments, now they are commissioned from private companies. The crux isto expose stale state monopolies to competition and the innovation of the freemarket. However, a host of failures across the UK, of which Carillion is onlythe latest, means the country which brought “contracting out” to the world isat risk of becoming a cautionary tale. Now with Labour courting ever moreinterest from voters whilst waving the banner of renationalisation theCarillion affair could be the end, not just of a company but of one of the mostinfluential ideas of the last 50 years. It can be argued thatCarillion is a good example of how it is better to have a failure ofcontracting-out than a public sector failure. Its losses not be borne bytaxpayers, but largely by its creditors and shareholders. However, dig a littledeeper and its demise shows how PFI and contracting-out more generally mustchange before it can truly live up to the expectations of its most ardentproponents. Regardless of whetherJoe Bloggs is picking up the bill for Carillion’s losses, there is still theprovision of vital services, project delays and contract transfers that will meancosts for the taxpayer of hundreds of millions of pounds.
Not to mention theplight of its subcontractors. The company’s fragility should have disqualifiedit from HS2, a monstrously complex rail project. Rather, ministers appear tohave thrown as a lifeline the £1.4bn ($1.9bn) contract. Serious misgivings mustbe addressed by the government when we hear that even after various profitwarnings they continued to award contracts. This smacks of ministerial failure. Governmentalshortcomings brings us to a key lesson of Carillion’s collapse: this continuedmisplaced obsession with lowest bid over whole-life cost.
Prequalificationsystems, two-envelope bidding and weighted scoring have gone some way toreducing unscrupulous bidding practices over the years, but still, the lowestbid can reliably be found at the top of the pile. Not without reason, governmentcommissioners are in a tricky position, if they are ever challenged, the lowestbid is the simplest for government purchasers to justify accepting. They cannotbe seen to be being frivolous with money that is inherently not theirs. Where was the ‘crownrepresentative’ (who was meant to be overseeing scrutiny of, among others,Carillion) during all this? The position had been left vacant? The governmentshould also ask whether companies should be allowed to pay large dividends whentheir own pension fund is underfunded. Carillion’s board recently changed itsrules to ensure that, in the event of collapse, managers bonuses could not beclawed back.
This serves in the mind of any onlooker to reinforce the commonimage of private greed. Questions must be asked of Carillion’s auditors, KPMG.How could a company, with a balance sheet showing quite clear signs of stressbe given the all clear? It should be notedthat there is an unhealthy concentration in the marketplace for largeall-encompassing service providers. For example, private prisons in Britain areoperated by just 3 companies. Only two bidders were involved in a billion poundcontract for the redevelopment of a London hospital. Routinely, the governmentcommissions huge, long-term projects based on fewer quotes than voters wouldget for refitting their kitchen. One of the mainbenefits of contracting out, as mentioned, is that it can save money. Any talkof ‘the market’ here is misleading, as, with just one “customer” and only a fewvendors for a contract which is agreed only once every 3-4 years, this canscarcely be passed off as a normal market.
It is incentives faced by workersand their managers, in realising efficiency and encouraging innovation. Incentivesystems to maximise profits in a private sector firm are clearer for managers,and that incentive is necessitated by the imperative to make a profit whilst managingto bid as low as possible. Carillion’s example shows that margins are verysmall on the majority of outsourced public sector contracts. In this, Carillionproves that the mechanism, at least in this sense, is working. Unless it is incompetition with private sector firms a single public sector body would not beable to match this advantage, competition improves incentives. One important qualificationto this argument involves information. The temptation of a bidding system basedon the lowest price is to cut quality.
So the public sector needs a clear meansof not just specifying quality in the contract, but of ensuring fulfilment ofsaid contract after it has been awarded. Occasionally politics stands in theway of that taking place. When quality is truly difficult to observe,contracting out of this kind should never be floated as an idea. 4 Conclusion Where does this leave PFI? Contrary to Theresa May’sPMQ insistence that the government is simply “a customer, not a manager”, thegovernment must display effective client representation and perform watertightdue diligence. No customer should walk into a contract blindly and when acontract extends into the billions of pounds and is looking to provideessential services such as a hospital or high speed rail link (if HS2 can beconsidered essential) then more robust “customer” oversight is crucial. If thegovernment can specify and monitor the contract, can be confident that theprivate sector will deliver, cannot find a superior organisational form andwants to bind itself to delivering the service over the term of the contract,then it can make good sense.
The argument put forward by some, that directborrowing should finance all government spending, because it is cheaper, is farfrom faultless: the government’s borrowing costs need to include the insuranceimplicitly provided by the taxpayer. However, PFI must not be used simply toshift a liability off the balance sheet.