Advocates of the Third Way may be surprised to find that the private finance initiative (PFI) is close to taking a step back into the public domain. The Treasury is currently piloting the credit guarantee finance (CGF) scheme, whereby the government assumes the role of capital provider whilst the private sector is confined, in the shape of a monoline, to wrapping project risk. The sponsor is effectively sandwiched between one part of the state, the concession awarder, and another, central government as the lender, with the private sector's involvement confined to insurance.

Rationale

Set against the Treasury's other goals for PFI and the aggressive market for senior debt mandates, the rationale for CGF is surprisingly conservative and the scheme's inception seemingly ill-timed. In the Treasury's 'PFI: Meeting the investment challenge' July 2003 report the government set down three goals for private finance initiatives (PFI) against the oft-repeated best value mantra. These were: 1) to maintain a variety of funding for PFI projects; 2) explore the provision of framework funding, particularly in bundled small PFI schemes, and; 3) explore through pilot projects and consultation the potential of a form of credit guarantee finance (CGF).

Rather than pushing out the envelope of viable PFI projects that would otherwise be funded by conventional public procurement or not at all, the government 'is committed to securing the best value for its investment programme by ensuring there is no inherent bias in favour of one procurement option over the other.' Also stating in its CGF technical note that CGF 'may be best applied to PFI programmes that are well established with well-established commercial and financial arrangements' and that it 'may not be appropriate where the risk profile of the project is untested in the financial markets.'

The banking community's reception to the scheme has warmed slightly from cold to lukewarm. On first hearing of CGF, many were highly sceptical, fearing that the wholesale adoption of CGF would totally eradicate the PFI market. However in light of the documentation released by the Treasury, some view CGF as a clever device that could save the government money and less a paradigm shift in PFI and more as another funding tool. This may cool again on further reflection.

A major bugbear in PFI is its cost inefficiency on smaller deals, yet CGF is unlikely to pare down costs to any great extent in these deals. Treasury research into PFI projects with capital values below £20 million has shown that the procurement process for small projects was of comparable length to that of major capital schemes. It found that although construction and operational performance were good and in line with larger projects, the cost and procurement times were disproportionate.

However, the CGF scheme will provide better value on larger projects that are more likely to issue bonds, rather than small bank deals. This is because monolines, which provide the cheapest guarantees, are reluctant to wrap small amounts of debt - this is normally left unwrapped or to more expensive bank guarantees. A banker has estimated that the rough figure that CGF is likely to save on small PFI projects over commercially arranged debt would be a meagre 10 basis points.

CGF has come at an unfortunate time for banks with strong demand for PFI mandates. Competition has been particularly fierce for lead arranging mandates - it is said that on recent PFI deals, banks have bid as low as 6 basis points for lead arranging. Peter Morawitz, from the public private finance team at BNP Paribas, said: "Bidding for lead arranging mandates has been incredibly aggressive, especially for the plain vanilla deals. The cost of manpower that is required to issue and sell a bond issue outweighs the money recouped by a 6 basis point fee. It is becoming less viable to offer services at these fees, with profits slim to none at all. To improve their margins banks could be looking to take on genuine underwriter risk, say, by taking on deals that struggle to sell down and take a reward. There is a move towards providing a more specialised service."

How it works

Despite lead arranging profit margins being squeezed to vanishing point, CGF should provide a return by the inherently lower cost of the government in borrowing funds through issuing gilts. The CGF scheme is centred on removing the need for private lending, and thus the lending premium, by installing the government as lender and contracting out risk to the private sector. This will be achieved by requiring the private sector sponsors to arrange for a financial guarantee to be provided by creditworthy financial institutions to support loans made to the project by the government.

Effectively the efficiency is found between the differential spread of AAA monoline paper versus how much it costs the Treasury in gilts plus monoline spread, for the same financing. This difference is a product of the difference in cost of capital between the government and private debt providers. Given that monoline credit spreads for PFI are currently 50 to 60 basis points for fixed rate PFI wrapped bonds, a generous estimate would value this gain at 40 to 60 basis points and conservatively at 20 to 30 basis points.

A principal aim of the scheme is to reduce the overall cost of PFI project. This means the benefits of the CGF programme are intended to accrue to the public sector as a whole, i.e. central government, and not to the procuring authority. The primary cost saving will be made by excising the lending premium, but the government also hopes that under CGF the PFI scheme may also benefit from lower transaction costs, for example the up-front issuing fees, and increased flexibility in accessing the financial market.

Paul David of MBIA is credited with being the originator of the scheme, yet the treasury has made clear that if CGF were adopted, it would look to consortia of banks as well as monolines as possible guarantors. Geoffrey Spence of the Treasury, and formerly of Deutsche, takes some credit for putting together the scheme. Spence conducted a seminar and question and answer session when the government's mandated legal advisers, Linklaters, presented the CGF scheme.

Implications

In practice, because of their higher cost of capital, banks are unlikely to win many CGF mandates. In a feat akin to propping up a three-legged table, the Treasury has stated that it intends to place a quota on the amount of business underwritten by each sector. "Without any apportionment", said a PFI banker, "monolines could be expected to win every guarantee. Banks are not well equipped to provide guarantees: they have neither experience nor the inclination, since it is not part of their core business, the opportunity costs are high and banks prefer not to hold assets over the long term. To compound the situation, a bank providing a single guarantee will need to be counter guaranteed by other banks, and will be required to meet minimum guarantor credit criteria, still to be agreed by the government and consulting parties."

Nevertheless, since the government is willing to ring-fence some projects for bank guarantees, effectively paying to spread the risk from monolines, there is likely to be appetite from the banks. James Miller, a director in corporate securitization at Royal Bank of Scotland, said: "If senior lending banks, acting as guarantors, remain exposed to project risk then I can't see why margins, which already reflect a mature and competitive market, should be any different. Senior lenders will still need to consider their overall exposure to a single project, which CGF will also need to take account of." As a consequence of the scheme, the government exposes itself to the risk of the guarantor's ability to meet interest payments when due, and repay the loan. The Treasury will in effect become a portfolio manager of credit risk.

On the upside, as well as saving the lending premium, CGF will provide greater flexibility by tapping another pool of financing, which, subject to quotas, has limitless liquidity. This would have the biggest affect on the relatively immature commercial index-linked bond market. The Derby hospital financing, which featured a switch from index-linked to fixed bond at the eleventh hour, perhaps highlights the need for greater flexibility.

The Derby hospital PFI bond was originally slated as an index-linked transaction, but after finding extremely limited available funding, bookrunner BNP Paribas and co-manager Barclays went back to the client and persuaded it to reprice the contract to accommodate a fixed-rate bond. The £446.5 million ($811 million) bond was priced on 3 September 2003 paying a tightly priced semi-annual coupon of 5.564%. The maturity on the notes was 30 June 2041 and the pricing at issue was 75 basis points over the reference Gilt (the 4.25% June 2032 Gilt). The deal carried an underlying rating of BBB by Fitch and a Baa3 rating by Moody's, with MBIA wrapping.

It is a moot point whether the index-linked market was as shallow as appeared, but an impartial observer would most likely conclude that Derby was a temporary aberration and simply a matter of poor timing. Excepting 2003, in the previous 20 years, no more than £300 million to £400 million has been raised on a quarterly basis through index-linked bond issues. However, exceptional volume in 2003 - for example the London Underground £1.05 billion Metronet project, the Southern Water bond issue, the Moyle Interconnector, the Dartford and Blackburn hospital PFI projects and the BBC's £813.32 million fixed-rate bond with RPI swap - had tapped out the market in the short term.

Derby was a milestone for hospital PFIs, being the first to use a fixed-rate bond market. Health trusts are loath to depart from index-linked bonds because of how the NHS is funded. Trusts are awarded a grant on an annual basis, which, from the public purse, is inherently index-linked, so coupon payments in the long term are more-easily accounted for. If possible, the type of financing chosen is one that best reflects the nature of revenue streams, thus the ability of the borrower to meet repayments.

It is unlikely that the threadbare index-linked market met by Derby was decisive in CGF's inception - the fixed-rate financing was tightly priced and proves the flexibility of commercial funding - but the scheme's adoption would sidestep the need to raise funds in a saturated market.

What rate, what scope?

The outlook for PFI commercially-provided senior debt is dependent on the amount of money the government is willing to throw behind the scheme, and which projects it chooses for CGF. A figure of £1.5 billon per annum was installed as a nominal figure, and the Treasury has said that to preserve the mixed economy of funding sources, CGF would comprise a minority of PFI financing, and at maximum between 20% to 50%. The former would see CGF accommodate three fair-sized projects, whereas half of all PFI would have a substantial detrimental effect on banks.

As the benefit of CGF will accrue to central government and not to either the sponsor or the procuring authority, the CGF will be provided in line with PFI market rates. In short, the CGF loan will not be a cheaper source of finance for a project company compared to commercially arranged senior debt. This will maximise the returns to government and, so the government hopes, sidestep the charge of state aid.

On a project-by-project basis, the projects to be financed by CGF will be identified prior to their procurement. This is unlikely to have any substantial effect on sponsors, save for the potential time saving from dealing with the government as lender that could mean more transparent terms, standardised forms and fast transaction times. The form and processes will not change significantly from commercial lending.

Adele Archer, an analyst at Standard & Poor's, says: "For credit implications we do not see a huge amount of change with the adoption of CGF: the risks will continue to be transferred to the private sector and transactions will be structured as they have been. The only change is likely to be the public sector term sheet, with the Treasury including covenants regarding reputational issues, but otherwise they will rely on the guarantor."

Archer added a slight caveat: "There is a slim risk that transparency will be diluted under CGF, as the driver for disclosing underlying project risk is diminished in the presence of a guarantor. Risks could be buried in the project because of the Government's guarantor protection."

The financing will mirror those available commercially: a loan with either a fixed rate of interest, calculated as a spread over the relevant gilt rate, or an inflation-linked rate of interest - the choice dependent on the revenue streams of the project.

In terms of the wider PFI market, the rationale behind CGF does not bode well for PFI banks. With the principal aim of CGF to save costs for central government, and, crucially not to support projects unviable in the commercially provided debt market by offering cheaper loans, the government will target projects where the margins for savings will be highest. These will be medium to large deals, conventionally bond-backed and monoline wrapped. A saving grace for the commercial institutions is the stated reluctance of the government to enter into inter-creditor agreements, which could deny CGF exposure to large deals that require multiple sources of funding.

Still many will view CGF as creaming off the best projects - those with the highest margins - that would decapitate the market and may trigger a wave of consolidation among banks.

Morawitz at BNP Paribas adds: "Even without CGF, we would expect to see consolidation in the PFI bond lead manager market, which has always been a specialist niche." He adds, "There are higher margins in the less mature PPP markets in Europe. In particular we are looking at Italy, and the healthcare markets of Portugal and Spain - both Portugal and Spain are looking to use the UK PFI/PPP model."

Road to adoption

Before CGF is put before ministers, the scheme will undergo three or four pilots, to see if the debt can be put in place with the projected margin of savings and to ensure that the government can meet or better the commercial banks' timetable. Portsmouth hospital is the first project to be picked as a pilot with the Treasury shortly to announce two or three others, of different sizes, across different sectors. Originally, the Treasury had planned to have the pilots completed and put before the National Audit Office before autumn, but with only one pilot formally announced this now seems improbable. It is rumoured that, despite the Treasury's assurances that those schemes involved with the CGF pilot can proceed unhindered, a number of PFI projects have rejected the Treasury's advances for fear of delaying financial close.

Parties to these projects may have been overly cautious, indeed, it is understood that the Treasury's contact with Portsmouth has not been sufficient to create certainty at the sponsor and advisers as to the CGF process. From what has been gleaned from Treasury literature, the CGF scheme will twin-track alongside the conventional commercially arranged debt, and the preferred option chosen prior to financial close.

The Hospital Company (THC) won the £210 million Portsmouth Hospital scheme last year, with a 30-year plus construction concession. THC is a joint venture comprising Carillion and RBS. Given that the concession awarder is a health trust and the length of the concession, it is likely that the commercial option will be an index-linked bond.

Aside from the successful completion of the pilots, the Treasury has many lengthy rounds of consultation to endure before any hopes of adoption. Still unclear at present is whether the government will account for the money it lends under CGF off balance sheet or on. And the government must be careful not to fall foul of state aid and Commission intervention. Also, it is unlikely the Treasury has the resources to run CGF and is certain to look at procuring an outside agency or outsource the business processes to the private sector.

If CGF is adopted it will mark a sea change for PFI. Banks will feel aggrieved that they have set down the guidelines for practice and efficiencies in the market, only for the government to cream off those projects offering the highest margins. In the role of lender, the government changes from being a provider of services that is concerned about the inherent value of a project to the community to a self-interested lender, concerned purely with monetary returns to central government. Adoption, although likely to be backed by limited funds at first, will undermine banks' confidence in viable returns for PFI lending when lead arranging mandate fees are at an all time low. CGF could see banks leaving PFI lending, pushing up fees, and having the opposite effect than that intended: pushing up the overall cost of PFI.

The government's commitment to a mixed economy of funding sources is in tension with its desire to lower the cost of PFI, since CGF uses the government's ability to borrow at zero cost, a level with which commercial banks cannot compete. Fortunately for senior debt providers and bond managers the government is constrained by funding limitations, risk exposure, and the need to preserve the commercial sector for future investment. Banks will carefully scrutinise further developments but may be powerless to prevent this unwelcome precedent.