Calyon didn't dominate the global power sector in 2007 by playing with kid gloves. The bank's hardball strategy was illustrated by that year's Rijnmond 1 refinancing in Holland, Calyon securing the deal with ultra-tight pricing and a bridge loan structure that would soon be refinanced again.

And sure enough, Intergen's Rijnmond 1 CCGT has now been refinanced again, but by the time financial close was reached last month the deal had taken a path rather different to that envisaged in 2007. Chaminda Jayanetti looks at a refi quite different to any other.

Backstory

Primary financing of Rijnmond 1 - known as the Rijnmond Energie Centre in the days before Rijnmond 2 - closed in 2002. Built in a Rotterdam industrial estate as a 790MW CCGT plant, sponsor Intergen hailed it as the first IPP in north west Europe since the EU Electricity Directive of 1996. BNP Paribas and Societe Generale underwrote a €405 million senior debt facility for the €622 million project.

At the time, Intergen had concerns that it was losing out on equity returns by opting for a fixed-price long-term gas supply agreement (with Gasunie) and offtake agreement (with Nuon). Intergen needed some form of hedging to make the deal bankable, but the company secured a deal under which these fixed prices would phase out over time to improve the upside.

Under this phased deal, both the gas supply and the offtake operate at a fixed price until June 2009. After that, 50 per cent of the offtake agreement will continue at a fixed price, but the other 50 per cent will be tied to market index prices. The original financing was flexed twice due to this merchant risk.

Nuon was later required to divest the Rijnmond PPA under the conditions of a separate acquisition deal, with Eneco taking on the offtake agreement. Meanwhile, Gasunie was restructured such that Gasterra became the gas supplier.

It was against this backdrop that Intergen launched a refi and small expansion project that reached financial close in July 2007, with Calyon 'warehousing' a bridge loan priced at just Euribor +20bp. Calyon's strategy was to refinance the loan with a syndicate of banks a few months later.

Imperfect storm

The 50 per cent merchant risk that would arise in July 2009 was a natural frightener for the banking market. Add to this the subprime crisis pushing pricing back up, and it became clear the refinancing would have to be priced far higher than the warehouse loan.

Calyon entered talks with Intergen on how to mitigate the merchant risk, and over potential compensation payments to the sponsor for the loss on equity created by increased debt pricing.

The result was a deal that would offer two separate structures. Intergen would try and agree a new extended PPA with Eneco, with a fixed offtake price for the life of the deal together with the fixed gas price - a de facto toll.

Alternatively, should Intergen and Eneco fail to agree a new PPA, the old PPA would continue with 50 per cent of the offtake linking to market prices from July 2009 as the full fixed-price offtake structure phases out - the merchant risk option.

Accordingly, Calyon headed to the market to form a bank group for the new €514 million refi with the following target debt pricing:

• merchant risk - Euribor +105bp to June 2009, rising to Euribor +150bp flat to June 2019
• full toll - Euribor +105bp to June 2009, rising to Euribor +145bp in step-ups to Dec 2024

Both tenors broadly match the length of the respective offtake agreements.

Banks were invited to take one of two tickets, with higher MLA fees on the underwriting tickets:

• €85m underwriting
• €65m take-and-hold

Marginal issues

Banks had a number of risk factors to consider on the deal, varying between the two very different deal structures.

First of all, the merchant risk in the absence of a new tolling agreement was clear - the Dutch power sector is not undersupplied, with a number of power plant initiatives that mostly cater to their own clients.

Indeed, Intergen is virtually the only developer in the Dutch market that does not have its own client base to sell to - it has no vertical hedge, no distribution company of its own. Should Rijnmond 1 take on merchant risk, prices could go down - Intergen could find itself recovering marginal costs but not the average costs needed to service the debt.

Another issue was the forthcoming restructuring of the offtaker, Eneco, which will unbundle itself into a commercial power generation company and a separate distribution network company within the next three years.

Rijnmond 1 would be selling its offtake to the supply and generation wing of Eneco rather than the distribution company. While power distributors generally have stable revenue streams, cashflow for a generation company is rather more volatile.

A further point for banks to consider was the ownership of Intergen itself. AIG was in the process of selling its 50 per cent stake, and there was a question mark over who would acquire co-ownership of the firm.

In particular, were a private equity firm such as Blackstone to acquire the stake, some banks were concerned an aggressive strategy towards assets would ensue. However, in June the AIG stake was sold to Indian power plant developer GMR.

All these factors had to be considered by banks as they decided what debt margins to press for. However, as significant as any of these was the simple fact that there was no regional post-subprime power sector pricing to act as a guide - meaning the R1 refi would essentially be navigating a new path for Dutch power pricing.

Up, up, to get away

The only Dutch power deal that had closed since the subprime crisis started to bite was Sloe Centrale - but the pricing on this deal had been settled before the debt market tightened, well below Euribor +100bp. The R1 refi would be setting the benchmark for Dutch power debt pricing.

Some of the banks that were invited to come in on the refi simply took the line that they were not going to help Calyon out of the hole they felt it had dug for itself, and walked away from the deal.

Others looked to force the pricing up - and up, and up. The target pricing was just that - a target, trying to draw banks downwards in their margin demands. All parties knew that final pricing would be higher.

However, aware that neither Calyon nor Intergen was in a position of strength, and given the lack of guideline pricing in the regional power sector, some banks consistently demanded ever higher pricing. Calyon found itself regularly asking Intergen to approve higher margins in order to bring more banks in, get the debt away and reduce its exposure.

For its part, Intergen knew that the final refi pricing would set the standard for the country's power sector - not least for the Rijnmond 2 debt, which is to be repriced and relaunched to syndication based on the R1 refi margins.

Intergen agreed to a number of price increases, but then shut the door - with the debt ultimately priced as follows:

Merchant risk:

• Euribor +135bp - to June 2009
• Euribor +180bp - thereafter to June 2019

Full toll:

• Euribor +120bp - years 1-4
• Euribor +130bp - years 5-8
• Euribor +145bp - years 9-12
• Euribor +160bp - thereafter to Dec 2024

The merchant risk structure incorporates a partial cash sweep towards the end of the loan.

As expected, the pricing was higher than the target - but remains respectable nevertheless. Margins certainly did not sail out of control, and it would be wrong to say that Intergen was held to ransom. In particular, the pricing under the full toll option reflects the fundamental security of the deal under that structure.

Intergen did have to give ground on debt service cover ratio, however. R1 has a debt cover of around 1.20 until summer 2009. Under a full toll, the ratio would remain at a similar level after that.

But the banking market compelled Intergen to alter its repayment profiles under the merchant risk structure to boost the debt cover ratio to around 1.50, to compensate for the merchant risk and make the deal bankable.

The seven MLAs that formed the final bank group all opted for take-and-hold tickets, though some went for the larger €85 million option on the basis that it would likely be scaled back should enough banks enter the deal.

Come financial close, the bank group comprised:

• Calyon - €124m
• Export Development Canada (EDC) - €85m
• ING Bank - €85m
• BayernLB - €60m
• Fortis - €60m
• Société Générale - €60m
• National Australia Bank (NAB) - €40m

MLA fees were 120bp.

Intergen's refusal to accede to bank demands for higher pricing kept margins in check, but also limited the number of banks that were willing to come in on the deal.

As a result, EDC and ING's take-and-hold tickets were not scaled back. Calyon was also left with more debt than intended. All three banks switched from take-and-hold to underwriting positions to sell in secondary.

ING and EDC are under no pressure to downsell - ING for one is not planning an imminent syndication. Calyon is thought to have launched to the secondary market already, and has flex on its debt for this purpose.

Clifford Chance and Consilium advised Intergen on the deal, with Linklaters acting as lenders' legal adviser.

Financial close was reached on 18 July.

Conclusion

The history of Rijnmond 1 is perhaps not a model that developers or banks may choose to follow, and the refi itself was not without its risk factors - but should Eneco and Intergen agree a new de facto tolling arrangement, both sponsor and lenders will be left with a solid and secure deal priced at a respectable level that will set the benchmark for the Rijnmond 2 relaunch and other upcoming Dutch power deals, such as the Enecogen IPP in Rotterdam.