Renewed deal flow in Europe’s PPP sector makes the outlook for new business positive. The sector generally performed well during the credit crunch, but weaknesses in its funding model became painfully apparent. The repercussions of these weaknesses were averted, but have not disappeared.

The danger is that the success that some European governments – most notably the UK and France – have had in alleviating problems with short-term funding is impeding a revival of capital markets activity in the sector. Even before the credit crunch, capital markets never featured very prominently in the European PPP sector, outside the UK. But the number of deals that have closed featuring soft mini-perms since the credit crunch, means that bond activity in the sector will need to pick up rapidly, and soon, if stable and long-term funding solutions for these projects are to be secured.

Before the credit crunch, tenors were long and margins were low largely because banks were so flush with liquidity. But the other major contributory factor was that the monoline insurers were aggressively pursing PPP business, and this competition kept the banks on their toes. Since the monolines disappeared from the PPP scene, no substitute has emerged for that role. Government involvement, in guises ranging from guarantees to direct loans has plugged the monoline-shaped hole in the market. But with “austerity” replacing “stimulus” as the buzzword of 2010, this too will probably be scaled back before long.

While the depth and liquidity of the market for bank debt for PPP projects has been remarkably resilient, the last couple of years have shown us its volatility, and what the situation will be like when the current round of recently closed deals need refinancing is impossible to predict. Most take the view that solid availability-based deals should have no problem refinancing on better terms once the construction period is out of the way, but refinancing risk is an issue.

Unlikely as it may seem, a nightmare scenario might be a PPP market similar to Australia’s, which is currently dominated by five-year hard mini-perms – and where the A$1.1 billion Lane Cove Tunnel project this year went into receivership after its lenders refused to extend its debt. Spain also provides a cautionary tale, as many of the real toll projects that closed near the beginning of the last decade have performed badly and the refinancings for their mini-perms are now causing headaches.

Can we rest easy again?

But troubled Spanish and Australian deals are based on traffic risk, and with notable exceptions in France and Germany, most current PPP concessions are availability-based. Plenty of bankers stress the unsuitability of imposing mini-perm structures on availability-based revenue streams. For now the tide is receding, even for soft mini-perms in PPP transactions, as projects can increasingly find long-term amortizing debt again.

Notable large deals to close recently with long-term project debt include the 28.5-year SEK5 billion (Eu532 million, or $684 million) of debt for the SEK14.5 billion New Karolinska Solna University Hospital PPP and the Flemish Schools financing, which featured a Eu700 million six-year revolving credit facility and a Eu1.5 billion 30-year term loan underwritten by BNP Paribas, which carries a Flemish government guarantee. The D1 motorway in Slovakia was also set to close without a mini-perm as soon as a remaining environmental permit had been obtained, though changed political circumstances mean few now expect that deal to close this year.

But these deals also highlight some of the limits to the improved funding market. After generating local controversy, the amount of debt on the Karolinska deal had to be restructured to reduce interest payments. The debt achieved its long tenor after it was cut it in half from SEK10 billion, causing Deutsche and WestLB to drop out of the deal. The deal also features margin step-ups, with pricing ranging between 250bp and 300bp, and fees of 250bp to 275bp.

The Flemish schools deal points to one possible new PPP model, mooted last year by the European PPP Expertise Centre (EPEC), the EIB’s think-tank. But the presence of a government guarantee to support the long-term debt facility if it is not refinanced after seven years diminishes the risk BNP took in underwriting it. The highly unorthodox model highlights the important crutch that government support still is in today’s market.

The D1 only secured so much debt on a long-term amortizing basis because half of it was coming from the EIB. That project might now be restructured after the election of a new coalition government led by a right-wing party that campaigned on a platform of cuts and expressed hostility to PPP while in opposition.

Cover required in the UK

Economists looking for a demonstration of the efficacy of the multiplier effect might well consider the Treasury Infrastructure Finance Unit (TIFU), which in its year of existence before it was merged into Infrastructure UK, in March 2010, provided just £120 million of public funding, to just one project. “TIFU did a sterling job during the credit crunch in maintaining the market, and they have also been a useful sounding board, listening to sponsors’ concerns,” says the head of project finance at one PPP sponsor.

By lending to the Greater Manchester Waste project, some have credited TIFU with preventing a collapse of confidence in the UK PFI market. It did not lend on the much larger M25 project, but the mere threat of doing so concentrated minds among the commercial bankers that did, allowing the market to believe that despite the credit crunch the UK could still muster £956 million in debt on a PFI deal.

Now that TIFU no longer exists, the challenge for Infrastructure UK is to help facilitate the development of alternative sources of funding for PFI by spurring the sort of competitive pressure on commercial bank debt that the monolines once provided. But a change in government brings a change of priorities, and Britain’s new government has already made its mark on the sector by cancelling the Building Schools for the Future programme, the mainstay of PFI over the past year. If the new government is as uninterested in “multipliers” as it appears, the onus will be on the market to come up with  long-term funding alternatives to bank debt, which since Greater Manchester and the M25 has once again become remarkably competitive.

Greater Manchester and the M25 both closed with soft mini-perms, on the assumption that these structures were the only way to get deals closed in the depths of the crunch. But even then, mini-perms weren’t universally popular among lenders. A January 2009 survey by PricewaterhouseCoopers highlighted the contrast between lenders motivated by internal profitability constraints and those interested in maintaining a presence in the PFI market as part of a balanced portfolio. Though the survey showed strong support among UK lenders for shorter concessions and loans, a few dissenters were concerned that shorter concessions would make it harder to refinance the initial facilities in the capital markets.

“Cash sweeps on availability-based projects don’t make a lot of sense,” says one banker. “The minis are more rare than they were made out to be, even soft mini-perms. Banks with regulatory requirements made a big stink about having cash sweeps included, but it didn’t happen much in practice.”

Perhaps this is not all just down to the return of bank liquidity. Since the M25 there simply haven’t been that many large PFI projects closing, with most of the volume coming from smaller deals such as street-lighting, BSF and LIFT projects. But the one deal was big enough to gauge the state of the market for larger projects, the £625 million Bristol Southmead Hospital, closed in February with £375 million in long-term commercial debt – along with a £250 million EIB tranche – rather than with a mini-perm or in the capital markets, though both those options were considered.

The credit crisis hit both the banks and bond investors, but for PFI the banks were the more resilient and able to adapt. Between the first PFI bond issue in 1997 and the end of 2007, bond issues were used to finance 25 out of 47 projects with a capital value over £200 million, and bond investors bought roughly £15 billion of bonds in PPP issues in that time. Only two of these issues were without a wrap, and both of these were early ones. Since 2008 there have been no new issues.

Bristol Southmead priced at 237.5bp during construction, then 225-260bp during operations. The sponsors, Carillion and Lloyds, sounded out the bond market beforehand, and estimated that bond issue rated A or BBB would have cost them 25-35bp more than the long-term bank swaps.

Further indications of bond investor appetite for infrastructure are the train rolling stock companies – Porterbrook, HSBC Rosco and Angel Trains – which this summer have been refinancing their acquisition debt from 2008. The latest of these, the BBB-rated Angel Trains issue, priced in July at 295bp over the 4.75% 2020 gilt for a 10-year £300 million bullet bond, and 310bp over the 5.25% 2025 gilt for a 25-year amortising bond. Pricing is on the low end of what was expected, but still indicates that the pricing for a project bond would be higher than what banks are offering.

“There is a large and liquid bank market, so unless the bond markets can come up with a better solution, its unlikely that people will choose bonds,” says one syndications banker.

But the biggest obstacle to getting the capital markets funding PPP deals once more is not the pricing but rather their complexity. Once a PPP deal is up and running, it is easier to negotiate with a small number of banks any changes to the terms of the deal. With bonds the issuer won’t always even know who holds the paper, and negotiating changes would logistically pose a far greater challenge. The difficulty of resolving this problem in time to close before the May general election is as much a reason for why Bristol Southmead eschewed the bond route as better pricing.

But while PPPs continue to offer attractive risk-adjusted returns and long-term maturities, UK institutional investors will continue to be attracted to the market. Previously these investors helped push transform the market from one based on 18- to 25-year concessions with 15- to 18-year loans to one to one based on 30-year concessions with matching debt maturities. Government might again have to shorten concessions, if sponsors cannot tap the bond markets for sufficiently attractive terms, even if grantors do not reap as many life-cycle benefits.

Some sponsors hope, forlornly, that a monoline market will again sprout up, but the best-looking solution comes from funds that would provide first-loss junior debt on projects. The reasoning for their existence is that other lenders would assume those funds chose only the strongest deals, given that they are risking the most if the projects turn sour. Hadrian’s Wall Capital made waves in February with the announcement it had struck a strategic partnership with Aviva to provide just such a product, but it is still several months from closing a fund. The manager is raising both sterling and euros for investment in the UK and continental Europe, mostly for investment in availability-based projects but also in some with traffic risk. AMP Capital is looking to raise between Eu700 million and Eu1 billion for a similar purpose.

Miniperm heaven and hell in France and Spain

If the general mood among those involved in the UK PPP sector is upbeat, the same cannot be said for Europe’s next biggest PPP market of the last decade, Spain.

A few notable deals have been closed there this year, such as the Eu410 million Eje Diagonal shadow toll, a 33-year concession that closed with a Eu249 million eight-year mini-perm. But banks do not just have to contend with weakened sovereign finances in southern Europe, but also the hangover from poorly-performing real toll roads that were financed with hard mini-perms that are now beginning to expire.

Earlier this year the R3+5 became the second of the three Radiales toll roads into Madrid to agree a short-term extension of its loan, after the R4 refinanced last year with a three-year extension. The Eu650 million R3+5 refinancing includes a Eu350 million 2.5-year bullet from around 40 commercial banks and three-year Eu300 million EIB facility guaranteed by the banks.

The lengths of the R3+5 and the R4 concessions are 50 and 65 years, respectively, and the refinancings are clearly stopgaps until some resolution is reached on what is to be done with these projects. The private sector participants are hoping for some form of government support for the projects, which the government is reluctant to give. Unless the government finds some way to keep lenders whole, say bankers, it will struggle with its plans to use private sector funding for its upcoming infrastructure developments.

In July the Spanish Ministry of Public Works had to cancel 32 projects that had formed part of its Extraordinary Infrastructure Development Plan, aimed at mobilising Eu15 billion in 2010 and 2011, including much of the development of its high speed rail network. The cancellations were necessitated by the country’s austerity measures, since lenders had demanded levels of support to projects that Spain’s government could not credibly provide.

But Spain’s bad experience doesn’t necessarily mean that mini-perms are a bad solution for PPP deals, and, on the other side of the Pyrenees, France used a similar mini-perm model on the 55-year A41 and A65 concessions, which closed in 2006 and 2007 respectively, without any subsequent indications of trouble. But the A41 mini-perm does not mature until 2014 and can be extended to 2018. Bankers with a fondness for mini-perms think that, where a robust refinancing market exists, and traffic forecasts are pessimistic enough, miniperms can be a harmless way to allow sponsors to cash in on a project’s improved risk profile.

Spain and France have both also announced ambitious plans to use infrastructure to spending to stimulate their economies out of the tough times, but so far France’s plans have been far more successful, largely because it still has the resources to prop up long-term PPP deals with guarantees.

Whereas the UK propped up the PFI sector by forming TIFU, France went down the road of providing guarantees, unveiling its Plan de Relance, a stimulus package that included a state guarantee programme for PPPs and concessions of up to Eu10 billion. Like TIFU, having the option there was useful to bidders on projects, some of which incorporated them into their bids. But given the cost of the facilities versus long-term debt, the debt proved cheaper, and the plan is unlikely to be extended once it expires at the end of the year.

France has now developed Europe’s largest PPP pipeline, and in July finalised negotiations with the Vinci-led LISEA consortium over the Eu7.8 billion concession to build the South Europe Atlantic High Speed Line (SEAHSL) between Tours and Bordeaux.

New alternatives

In a paper published in August 2009, the European PPP Expertise Centre analysed a number of ways forward for PPP market, which was only just starting to recover. One of the more radical solutions was for the procurers to separate completely the bidding competition from the funding competition – holding separate DBM/O and F competitions – and possibly later merging them under the bidder’s control.

The Flemish schools PPP and SEAHSL are both versions of this model. The Flemish Schools deal was structured by BNP Paribas Fortis, KBC and Dexia, with the 211 schools themselves to be bundled into lots of four to six and tendered to construction companies. On SEAHSL, Reseau Ferre de France, the concession awarder, plans to sign the concession with Vinci later this year before going on to secure the financing early next year.

Holding separate DBM/O and F competitions is not a solution that would be popular with all sponsors, which have built up expertise in the F department and want to use it. A bigger threat to the model’s viability, however, is the extent to which it would require the government to guarantee the F. In today’s economic environment very few people expect such a model could possibly take off.

Ruling out other forms of government backing for the debt – such as a public bond insurer – in the capital markets, means that there are still few alternatives to flexible but flaky commercial bank debt. “I would say there has been a missed opportunity in the last 10 years to cultivate an institutional investor base for public projects the way the US over 100 years has developed the municipal bond market,” says one sponsor. “A lot of the bonds that were launched went to straight to banks like Dexia and Depfa.”

One more model is a new “German model”, and given how much influence Germany has had on European fiscal policy it could provide the surprise template for many future European PPP deals. In November 2008, Germany’s A5 road concession closed financing with steep margin ratchets but a 30-year concession length. In March 2010 Germany launched the Eu128 million A9 Hermsdorf-Schleiz expansion PPP, which has a prequalification deadline of September. The concession length for the A9 is 20 years. Shorter, suggests the German model, could be sweeter.