Watercraft Capital, a Luxembourg-registered special purpose vehicle, priced its Eu1.43 billion ($1.9 billion) bond issue on 25 July 2013. This refinancing of the Castor underground gas storage project in Spain marks the first use of the European Investment Bank’s (EIB) project bond credit enhancement (PBCE).

Watercraft Capital SA

STATUS

Priced 25 July 2013

SIZE

Eu1.7 billion

DESCRIPTION

1.9 billion cubic feet gas storage facility located 15km off the coast of Spain

SPONSORS

ACS (67%); Castor UGS (33%, with Dundee Energy representing 74% of that shareholding)

EQUITY

Eu252 million

DEBT

Eu1.43 billion

BOOKRUNNERS

Bankia, BNP Paribas, La Caixa, Crédit Agricole, Natixis, Santander, SG

MATURITY

2034

COUPON

5.76%

PROJECT BOND ENHANCEMENT

EIB

MONITORING ADVISER

Trifinium

BOOKRUNNER LEGAL COUNSEL

Allen & Overy (Luxembourg, English and Spanish law)

EIB LEGAL COUNSEL

Clifford Chance

TRIFINIUM LEGAL COUNSEL

Ashurst

INDEPENDENT ENGINEER

Gaffney Cline

INSURANCE ADVISER

Willis

MODEL, ACCOUNTING AND TAX ADVISER

Deloitte

The PBCE is part of the European Union’s 2020 project bond initiative, which is designed to improve European infrastructure borrowers’ use of capital markets. During the product’s 18-month development period the EIB has faced scepticism about the utility of the enhancement.

Both Castor’s existing lenders and the EIB were anxious for the bond refinancing to work. The EIB was anxious to create a template, while the Castor lenders were eager to rid themselves of exposure to a troubled asset. But enabling an infrastructure issuer in the eurozone’s troubled periphery access to the capital markets suggests an application of the PBCE that did not feature highly in its development.

The project bond credit enhancement involves the EIB providing a funded or unfunded subordinate debt commitment to a financing that allows sponsors to gain cost-effective leverage and gives bond investors an improved debt rating. The product draws on the loan guarantee instrument for trans-European transport network (LGTT), under which the EIB offers an unfunded subordinate debt commitment that can be drawn in the event that revenues from an infrastructure project featuring demand risk are insufficient to meet debt service.

The PBCE can be used in other scenarios, too, particularly those relating to cost overruns, and essentially can be employed to meet any kind of weakness with debt service. The product can be used either to make scheduled debt service or pay down a portion of the senior debt.

The development of the PBCE worked from the assumption that it could be used to enhance triple B-rated credits to single-A, the level at which investor appetite is strongest. The most obvious candidates were PPP projects with minimal demand risk in highly- rated jurisdictions, where the enhancement would counteract some of the notching downwards from a project transferring risk – particularly construction risk – from the sovereign to the private sector.

But Castor features neither demand risk nor much completion risk. It is a 1.9 billion cubic metre gas storage system located 15km off the eastern coast of Spain and the town of Vinaròs. Its construction has been complete since June 2012, and it waits primarily on acceptance by the Spanish gas grid.

Castor’s sponsors are ACS (67%) and Castor Limited Partnership (33%), of which Canadian-listed Dundee Energy owns 74%, and private investors the rest.

The construction financing for Castor closed on 22 July 2010 and consisted of a 10-year Eu1.276 billion mini-perm, a four- year Eu32 million letter of credit facility, a four-year Eu9.5 million VAT facility, and Eu334 million in equity.

Banesto, Caja Madrid, Crédit Agricole, Santander, and Societe Generale (SG) led the construction financing, which benefited from a completion guarantee, and a commitment from the Spanish government that the project could amortise its capital costs over a mini-perm-friendly 10-year period. Pricing on the bank deal, originally 350bp over Euribor, stepped up to 450bp.

But as the project neared completion in March 2012, and Spain’s condition worsened, the Spanish government changed the remuneration structure. The project would now recover its costs over a 20-year period, in order to reduce the burden of charges on Spanish gas and (indirectly) power consumers. Lenders stopped draws on the construction financing, though by this point the remaining proceeds were largely destined for purchases of cushion gas, which would be required to allow Castor to operate efficiently.

In January 2013, the Spanish authorities approved a ministerial order allowing the project to use gas from a third party. Enagás, the operator of Spain’s gas grid, and Castor’s largest customer, had already agreed to buy half of ACS’ stake at completion. It now took over from the project the role of buying cushion gas.

The project benefits from a regulated asset value regime, under which the project can recover its capital plus an allowed rate of return, and escalates at 2.5% to take account of inflation. Under the restructured framework, Castor receives a rate of return on unpaid balances of 8.76% (or 350bp over the 10-year Spanish government bond at the time), as of 5 July 2013, the end date for capital investments.

Castor’s revenues come from charges that it levies on grid users, which in turn are passed onto consumers. Grid users’ payment obligations are joint and several, making the counterparty credit that of all Spanish gas users. The obligations lack a government guarantee, unlike Spain’s tariff securitisations, in which it promised to step in if the electricity system was unable to increase collections to meet the gap between user charges and generation costs.

But Castor, thanks to the EIB’s enhancement, achieved ratings of BBB (S&P) and BBB+ (Fitch), one notch above each agency ’s sovereign rating for Spain. S&P cited the project’s ability to recover costs directly from users, and the way that the PBCE minimised the effect of stresses on revenues, as reasons for the one-notch difference.

The EIB’s enhancement is a Eu200 million unfunded commitment that starts out as equivalent to 14% of senior debt, but will eventually decrease as the bonds amortise, and will then stay equivalent to 20% of outstanding principal. In the event that the project draws on the PBCE, the draws become subordinate obligations of the project. The PBCE amounts are repaid after senior debt and the replenishment of the project’s debt service reserve, and any shortfalls in payment of principal and interest to the EIB are capitalised.

The lead bookrunners for the financing, Bankia, BNP Paribas, La Caixa, Crédit Agricole, Natixis, Santander, and SG, brought the Eu1.43 billion in bonds to market with a price target of 50bp-100bp over the Spanish 10-year government bond, and the assumption that the EIB would buy between Eu200 million and Eu300 million of the issue.

The final pricing, as well as the EIB’s take, ended up at the top of their respective bands. The bonds have a 21-year maturity, and a 12-year average life, and priced at 100bp over the government benchmark for a coupon of 5.756%. Insurers and pension funds accounted for 61% of the issue’s buyers, and this group included institutions that do not normally buy project bonds, but the EIB, with a Eu300 million order, bought around 25% of the issue.

“There were few direct comparables for the transaction, though the major Spanish utilities were one possible option,” notes Matthew Vickerstaff, global head of structured finance at SG. “We were pleasantly surprised at the level of engagement from non-traditional buyers.” That broadening of the investor base will be validation enough for the PBCE.

The EIB is anxious to highlight the usefulness of the product, but places increased stress on the value of having itself present in a transaction. Left unsaid is the promise that the EIB’s presence may deter host governments from making retroactive adjustments to concession structures.

There is still some uncertainty about whether the PBCE could sit alongside both banks and bonds. The EIB seems wary of offering the enhancement to bank lenders, whether because of the inter-creditor issues or because such an extension undermines the product’s stated purpose of increasing infrastructure borrowers’ use of the capital markets.

One banker familiar with the Castor refinancing said that attempts to introduce a bank tranche into the structure were firmly rebuffed. As sponsors and grantors grapple with how to minimise the negative carry issues that plague greenfield bond financings, this stance may soften.

The contours of the non-bank infrastructure finance market have shifted in the period since the PBCE entered development. The Dutch government tried holding sponsors to a stapled institutional financing – the N33 template – before deciding the inflation protection that it had to offer to lenders was not good value for money. Ballast Nedam is working with ING and NIBC on a bank-to-bond deal using ING’s PEBBLE structure.

But the UK’s project bond market is now in full flight. Assured Guaranty has written two primary polices on project bonds in July – its first since 2008, when its competitors imploded. ULiving Hertfordshire, comprising Meridiam, Bouygues, Derwent Living and Legal & General, closed a single A-rated unwrapped issue for a student accommodation project in May.

Northern European projects still appear among the list of PBCE candidates, though one of the most obvious ones, the Netherland’s A1/A6, eventually closed with a bank group earlier this year. Belgium’s A11 and the UK ’s offshore transmission operator (OFTO) projects are still potential users. But the fact that the product has been proven to work for an operational asset with a regulated revenue profile means that Italian assets, both gas storage facilities and toll roads, are the obvious potential beneficiaries.

The Castor project structure