The main items on the agenda include Borders Railway, Intercity Express Programme (IEP), Thameslink rolling stock, Crosslink and High Speed 2 (HS2). These projects have attracted the attention of the project finance market because PFI-derived financing structures have been an increasingly prevalent way to pay for that £70 billion ($110 billion) shopping list. But a series of failed concessions has created a legacy of cynicism in the market.

“The main issue with any UK rail PFI is the need to ring-fence the project and define the interface risk clearly, such as providing the land and right of way,” says Manish Gupta, head of Ernst & Young’s transport infrastructure practice. “This has proven to be hard to isolate and investors have yet to become comfortable with the risk. The only real example is the Channel Tunnel rail link, which failed miserably in the 1990s, was nationalised, and was later sold as HS1 post-completion.”

Borders blows up

The £295 million Borders PPP struggled to manage this interface risk. The 30-year design-build-finance-maintain contract was for the construction of 50km of line between Newcraighall in Midlothian to Tweedbank in the Scottish Borders region. It launched in December 2009 using an availability-based model, with three bidders shortlisted in 2010.

During the full tender, however, several complications emerged. “The main problem was that the risk allocation and risk transfer aspects, which changed from the initial bid, and this made the project less attractive,” one consultant suggests.

One sticking point was the operational interface with Network Rail. The contract said that Scottish Rail would enter an agreement with the project company, which in turn would have to enter into an agreement with Network Rail. But Network Rail, the non-profit but independent owner of most of the UK’s rail infrastructure, would not be able to assume risks such as engineering works or cost overruns on a project built by a third party, so the project company had to take on the burden. This was, say sources familiar with the process, not outlined in the initial procurement, and the increased risk, and hence increased costs, meant bidders no longer found the project to be financially viable.

First Fluor pulled out of the consortium it had formed with Miller and Lloyds’ Uberior Infrastructure, and then Carillion left its IMCD group, which consequently also withdrew from the tender. This left BAM as the sole participant and, following the withdrawal of the other bidders, the tender was scrapped in March 2011. Borders has since been transferred to Network Rail, which will deliver it using internal resources.

Separate ways

Borders illustrated the challenges in allocating construction risks on rail projects but HS1 does show that private concessions can work post-completion. Appetite for the contract, a 30-year concession for a 110km rail line running between St Pancras station and the Channel Tunnel, was strong, with Borealis and Ontario Teachers’ Pension Plan winning the deal with an offer of £2.1 billion. The winning bid was said to be £600 million over the seller’s asking price.

Revenues on HSI are linked to an inflation-indexed investment recovery charge levied when train operating companies (TOCs) use the line, as well as an operation, maintenance and renewal charge. HS1 was originally awarded as a £5.18 billion PFI contract to London & Continental Railways (LCR) before it became insolvent. As such, the successful sale of HS1 was more through accident than design, but it does suggest that a PFI model could be deployed for infrastructure in some of the larger projects once up and running.

“The whole issue centres on the financial viability of a project,” Gupta comments. “In the case of HS1, it was sold for £2 billion but the development costs were around £6 billion. This means that the financially viable part of the project that could provide stand- alone revenue was worth £2 billion.”

For Crossrail, like HS1, there will be guaranteed revenues from train operating companies to cover the cost of the infrastructure. Unlike the original HS1 plan, though, the government seems keen on retaining most of the risks of the pre-construction infrastructure, rather than attempting a full-blown PFI from the project’s inception. The Crossrail financing already has a £1 billion loan pledge from the EIB, the Greater London Authority has issued a £600 million bond to cover part of its share of the costs, while the Crossrail development levy – applicable to all real estate developments over 100 square metres – is estimated to bring in £300 million.

While the suitability of PFI to rail infrastructure is up for debate, rolling stock may present a more suitable candidate. The issue of rolling stock procurement has been controversial, ever since the Office of the Rail Regulator and the Competition Commission launched investigations into the sector in 2006. The conclusion of the reports was that the three main companies – Angel Trains, HSBC Rail and Porterbrook Leasing– were not always the best possible solution. “Since the commission’s investigation into rolling stock, there have not been successful closings for alternative financing models,” Gupta comments.

As such, the latest idea is to deploy PFI to increase competition. TherearethreerollingstockPFIsinthewings:IEP,Thameslinkand Crossrail. IEP is a 30-year, £4.5 billion contract for the construction and maintenance of 600 new rail carriages for the Great Western and East Coast main lines. A group comprising Hitachi, Barclays Private Equity and John Laing is preferred bidder, and while little is publicly known about the proposed funding, financial close is scheduled for mid-2012.

Thameslink is currently out to market, while a PFI has been proposed for Crossrail’s rolling stock component. “Though Thameslink and IEP are structured as PPPs, they have taken a huge amount of time and have not closed yet,” Gupta continues. “There is on-going debate on Crossrail as to whether this should be funded as a PPP or funded on TfL’s balance sheet. There is no consensus view yet in the market on what is the best structure.”

Stock and roll

The Department for Transport (DfT) is procuring Thameslink on behalf of the line’s train operating company, First Capital Connect. The company will enter a lease-and-pay agreement for 1,200 trains that will last for 26 years. The tender introduced some much-need risk transfer into the process, in a bid to move away from the notorious hell-or-high-water leasing payments, which saw companies paid irrespective of the performance of their trains. On Thameslink if the trains do not hit targets, some payment is withheld.

“The hell-or-high-water provision is certainly something that the private sector and funders prefer, but the contracting authority needs to understand whether this gives it adequate security on availability of rolling stock,” Gupta says.

Thameslink’s main difficulty has not been risk allocation but the terms on offer on the financing. The financing package includes a Eu300 million ($387 million) guarantee commitment from the European Investment Bank alongside a proposed £200 million loan and an equity input of £140 million. The commercial tranche has caused some difference of opinion over its 20-year tenor and aggressive pricing.

The starting point was said to be 175bp, rising to 250bp over Libor, which some banks said was too low in light of current market conditions and the lack of precedents in rolling stock PFI. But sponsor Siemens has a strong credit rating and is providing extensive support to the financing during both construction and operations. Lenders are still exposed to the residual value of the trains at the end of the concession, but this is partially offset with a number of cash sweeps towards the end of the financing’s tenor.

“It is not the first time a sponsor has come to the market with an underpriced financing and it wouldn’t be the first time that banks complained it was underpriced,” one observer notes. The initial pricing put Thameslink in line with the seven-year £450 million tranche on the HS1 purchase (a term loan priced even lower, at 125bp), but bankers point to the Nottingham Express Transit financing as more comparable deal. Nottingham involves the construction of two new lines and expansion of an existing tram system by 17.5km. It is a smaller deal and involves building infrastructure, but margins are said to start at around 300bp.

Some bankers felt that factoring in both deals should see pricing on Thameslink (which is linked to a ratings grid) start at between 200bp and 250bp. The sponsor, which is already providing considerable support to the project economics, is still in negotiations with lenders, and close is not expected until later in 2012. A bond has also been floated as an alternative solution, should a loan fail to materialise.

The £1 billion needed for the 60 Crossrail train carriages is set to follow the Thameslink model of being procured using a PFI structure, after the Department for Transport knocked back the attempt by London’s mayor, Boris Johnson, to get central government to finance the fleet directly. The tender for this contract is earmarked for 2014, with Siemens, Bombardier, CAF and Hitachi competing.

Government’s belief is that once IEP and Thameslink close, a precedent will be set and rolling stock contracts would become the easiest parts of ambitious projects to procure as a PFI. “Banks seem to more comfortable with the idea of rolling stock PFIs because the underlying risk profile is lower than they would be if paying for a project in its entirety,” one banker adds. “In addition, the sponsors usually offer very strong support, as is the case for Siemens with Thameslink, for example, which helps.”

HS2 hopes

The current fiscal environment means the over £32 billion HS2 project, which will run between London and Birmingham, and then on to Leeds and Manchester, will probably have some PFI component. While the deal is years away, experts say that carving up the project into digestible parts will increase investor appetite, though Gupta believes the core infrastructure should at least remain the responsibility of the government.

“Given the scale and risks involved, we believe HS2 should be developed and implemented by a Crossrail-like entity with significant government funding. However, discrete elements such as rolling stock and stations could be developed on a private basis, though these represent a small portion of overall costs. PFI, or any similar structure, is not an appropriate starting point for projects of this scale and complexity.”

But some form of private component is in keeping with the government’s current rail strategy, which involves enticing a wider set of investors into the process to spread risk and costs. This strategy involves financing construction, tracks and related infrastructure on-balance sheet and finding ways to allocate risk to private entities where possible. The Borders project, even though it involved sinking costs into a failed tender, at least made sure that Network Rail was the best option to build and operate the track, and established some limits for the private sector’s involvement.

Lenders, both bank and institutional, still like rolling stock credits. Indeed, 2011 saw a £850 million bank refinancing for rolling stock company Angel Trains, split between a £450 million five-year tranche and £400 million seven-year tranche. The bank debt complemented a £400 million 15-year bond deal that priced at 275bp over gilts. That deal refinanced debt connected to the acquisition of the lessor from Royal Bank of Scotland by a consortium of Access Capital, AMP, Arcus, Amber-managed INPP and PSP. Bankers claim that newbuild rolling stock deals will achieve widespread acceptance once the terms are given a bit more thought.

“There is not a huge number of large-scale infrastructure projects coming in the UK, so lots of investors will be looking at the few deals that do come along,” one banker concludes. “But market conditions at present are delicate, so sponsors will need to revise their pricing and tenors, while banks will be looking for smaller tickets too.”

Even so, the 2011 report into the rail sector by former Civil Aviation Authority chairman Sir Roy McNulty will also be fresh in many minds. His recommendations included an expansion in the ownership of rail infrastructure, as well as its integration with train operations, and decentralisation and devolution within Network Rail. Were the recommendations to be implemented there would be opportunities for greater private sector involvement in infrastructure, but there is little indication that the private sector is willing to take on more operational risks at a price acceptable to government. Whether the preference is for public procurement or PFI, rolling stock and concession companies have yet to discover the best route available route for delivering these deals.