The £1.4 billion ($2.08 billion) Balfour Beatty-led M25 London ringroad financing is officially meant to close in 2008, but the deal will not be out to market until 2009. Four coordinating banks – HSBC, Lloyds, Dexia and Barclays – have been appointed and are guiding a further 15 lenders into the project, as well as the European Investment Bank (EIB).

Banks will struggle with the proposed tenor of 27 years as they grapple with their own funding constraints and many are already calling for a miniperm tranche. But the M25 looks more likely to be a test of the solution to PPP liquidity being proffered by many project arrangers: that EIB and government participation combined with a large number of small-ticket commercial bank participations is enough to close a high-profile deal.

The problem with PPP financings is not project risk. Availability-based PPP deals are among the safest form of project financings (Metronet excluded), with visible income streams and relatively benign penalty regimes. The problem is bank liquidity. This is why banks rejected a proposed contingent guarantee, similar to that provided by Transport for London on the Tube Lines structure, offered by the Highways Agency (HA) for the M25. Lenders would still have to find liquidity for the project, but with reduced margins.

There is not a lot of fat in PPP structures, since they are usually geared at 90%, and the most obvious compromise would be to ask for more equity in the deal. However, a material increase of equity is unlikely on major deals because the cost of equity, particularly for listed sponsors, is high, as corporates look to pay down balance sheet debt. And infrastructure funds are unlikely to be tempted by the internal rates of return on deals with lower leverage. The higher risk associated with more equity is not reflected in the current UK guidelines for sharing refinancing gains.

PPP adversity is spurring innovation outside the UK. For the recently-closed Eu841 million ($1.13 billion) Liefkenshoek rail tunnel PPP in Belgium the state-owned rail infrastructure procuring agency Infrabel is taking on market disruption risk (funding cost risk) during the four-year construction period.

In Australia the A$250 million South East Queensland schools project is pioneering the supported debt model (SDM) where the Queensland government is refinancing the 70% of the senior debt. The government will take the last loss piece, which it views as notionally risk free, while commercial banks will provide the construction bridge and the remaining 30% senior debt in operation.

The rationale underpinning SDM is that the Queensland government can raise funds at or below the interbank benchmark.

Both Liefkenshoek rail and the SEQ projects do not offer a sustainable solution to a shortage of bank funds. For Liefkenshoek full bank funding is still required and the Australian SDM, according to one banker, is likely to see banks raise their margins on the construction bridge to compensate. And the mezzanine senior debt will be expensive, given that the loss given default rates will be much higher than a normal senior debt tranche.

What is to be done? A wholesale shift to placing refinancing risk on sponsors is difficult in the current environment. PPP projects in the US and Spain favour the miniperm approach, but this could leave sponsors with bridges that expire and leave them to stump up equity until a long-term funding market returns. Perhaps the biggest lesson for banks and advisers from the crunch is not to rely on the market improving.

Outside of PPP, project financiers have shifted refinancing risk by reducing the likely tenor of debt below the nominal term by structuring in cash sweeps around year 7 and increasing debt margins at the debt's halfway point. However, PPPs are less like standalone businesses and the models for most availability-based schemes would break with large increases in debt pricing without passing this cost increase on to the government.

In the short term the EIB can take up some of the slack through direct lending. The EIB's presence prevented changes to the terms of the underwritten bank funding for the Eu600 million Transmontana PPP highway that is due to close in Portugal. The EIB is providing Eu200 million, reducing the commercial bank tickets and keeping starting debt margins at around 125bp.

The EIB is also likely to take one-third of the senior debt, a £200 million ticket, for the upcoming Greater Manchester waste scheme and the Greater Manchester Waste Disposal Authority is putting the maximum grant allowed – £70 million or 10% of total cost – into the scheme. But this may not be enough in future.

One idea being mooted is for the commercial banks to provide project risk guarantees with another entity providing the funding, similar to deals under the old EIB lending regime. The now moribund credit guarantee finance (CGF) was also based on commercial guarantees. The CGF scheme was based on the same rationale as the SDM, where the UK government could take advantage of its cheaper cost of funds, but with banks or monolines providing guarantees. The reason CGF did not take off, however, was because the Treasury had to book the debt twice; once for the issuance of gilts and once for the project debt.

Rather than having banks acting as monolines, a tidier solution suggested by one banker would be for the Bank of England to provide an infrastructure-specific ABS funding solution, similar to the ECB ABS repo agreement, with the Bank of England taking a haircut for placed securities. This would circumnavigate the current difficultly banks are facing recycling their capital, because the buy-side for project debt CLOs has evaporated.

A Bank of England infrastructure ABS repo solution would catch the political zeitgeist and tie fiscal stimulus to investment in infrastructure. However, aside from increasing the UK's national debt, such a facility would postpone the return to a properly functioning ABS market. It may also act as an unnecessary crutch for banks and postpone a return for banks to act as banks – that is, lending money and accepting both funding and project risks.