The recent distress in the UK and US power markets has not led to a flood of project finance debt onto the secondary markets - more of a trickle. And despite the claims of some investment banks - notably Deutsche and CSFB - that they are strategically pursuing the creation of a liquid secondary project debt market, the fact that most banks are handling project debt trades through distressed desks, rather than their loan trading operations, shows that there is still a substantial gap between the aims of the few and the overall market reality.

The tentative activity in project loan trading activity contrasts with the established secondary market in corporate loans. Corporate loans have been popular with loan funds, hedge funds, and money market accounts. Equity funds that want to gain control of distressed corporates also buy heavily into bank debt. The growth in corporate loan trading has convinced investment banks that it is time to widen the universe of project debt buyers.

And the secondary project debt market is beginning to claw its way into existence. Indeed, some project loans now come from bank arrangers with the proud promise that they will make a secondary market in them. While the banks still assure their clients that they would be on the hook for that client's risk, they also promise to support the ability of participants beneath them to sell out.

And the logics for a secondary project debt market are already in place. "I think the smarter investment banks may have sold out of their clients' credits without needing to inform them," says Blair Thomas at TCW's global project fund. "The derivatives market has developed to such an extent that banks can nominally hold loans, while transferring all of the credit risk to a third party."

The idea that a bank can hold the attention of a client by piling into its debt has to a large extent gone the way of the universal bank template - history. Those banks with the best results - Citigroup aside - have been those most stingy with their balance sheets. And even Citi is re-examining its project loan business.

"The idea that banks have to buy and hold sponsor debt is looking very antiquated," adds Thomas. "The more sophisticated banks realise that to prevent project finance becoming irrelevant you need to find a way of widening the universe of loan buyers," he adds. But few banks are yet putting great resources into making markets in project debt.

Deutsche opens shop

One that has started looking at the possibilities is Deutsche Bank, which says its Latin American credit traders have used sovereign and corporate default swaps to manage project debt risk for many years. According to Sean Bates, managing director for Latin Amerian integrated credit trading at the bank, this experience has borne fruit in the current low interest rate environment. "The picture is one in which investors are increasingly willing to spend time understanding the structure and additional risks involved in project debt, in their search for regional assets offering an attractive yield pick-up," he says.

Put briefly, Bates' traders look for project finance debt with risk factors that can be mitigated to a degree by available credit default swaps. The example he uses is that of a Mexican project bond, which, post construction, has parastatal obligations (CFE or Pemex payment risk) as its defining characteristic. "In many cases, Mexican project debt has minimal performance risk post-completion, and we hold these positions against United Mexican States default protection in order to hedge what we consider to be the primary medium-term credit risk. Our target return on the asset above funding and default protection costs is then a function of its liquidity and ancillary risks."

Bates also says that the growing investor interest in project finance assets has led to the application of more sophisticated financial techniques. "Using credit derivatives technology, we have been able to provide buyers with customised exposure to project assets. This activity ranges from simple financing trades, which enhance investors' return, to more complex re-tranchings of long-dated project assets, designed to reduce the reinvestment risk in a project whose debt rapidly amortizes."

The operation is for the time being limited to Latin America, and ideally to those issues that are liquid and have liquid markets in the relevant ambient sovereign default swaps. It also works best with older project finance bonds that carry comprehensive government guarantees. Early Turkish deals, in particular, were testing grounds for this structure. But it could be adapted to non-recourse debt dominated by a single corporate credit, say a project bond backed by a tolling agreement. A buyer of a Tenaska bond, say, might be both willing and able to purchase default protection for a Williams or a Shell offtaker.

This last is an important consideration for buyers of most project bonds, which do amortize. Some high-yield, and most mini-perm, debt is less susceptible to this risk. Likewise, Term B debt has been a key factor in the development of secondary trading in project debt. Term B debt is structured to trade, and priced to attract investors looking for yield.

Astoria changes Term B dynamics

Until very recently, most Term B debt had been issued for portfolios of assets comprising subsidiaries without recourse to parents. However, they have been large enough to run as corporates in their own right and been covenanted as such.

Credit Suisse First Boston's $690 million financing for Astoria Energy, a greenfield power project, has changed that dynamic (see Deal Analysis in this issue for details). This financing, bought heavily by money market accounts, mutual funds, and hedge funds, was the first single asset financing in this market, and the first on which there was construction risk.

A large part of the pitch on a Term B financing is that the arranger will structure a loan to sell quickly and effectively, making it much more liquid than a standard project finance bank loan. The second part of the B loan's appeal explains why borrowers cannot complain - the market is open to even the worst credits. Some observers question how long the B lenders will remain in the market.

According to Steven Greenwald, CSFB's project finance head, "the institutional loan market is an important new source of funding for power projects, and is enjoying the most favourable conditions in years. The market is open for even lower-rated credits for up to 12 years." Asked whether the current boom in the market could continue beyond the current environment of low interest rates, Greenwald says: "rates will probably not shoot up right away, and we believe this market is here to stay." The sheer numbers of providers active in the market bear this statement out.

But, the fact that Astoria sold even during construction also eliminates one of the key comforts that the banks can hold close - that they are the best placed to handle construction risk.

There have been retreats from long-dated lending by US banks before, but few where they have abandoned construction lending altogether. While the institutional market was the place for long-dated funding before the entry of European lenders began to dominate the US market - few would have predicted that a hedge fund, armed with little more than a ratings report, would enter the construction game.

Distressed tests

Karl Miller is a senior partner at MMC Energy, a distressed asset buyer that keeps up regular contact with hedge funds with sizeable distressed debt or equity positions. Miller describes MMC as "an energy merchant bank with asset management capabilities," and says the firm has demonstrated a sizeable book of potential equity providers. He adds that this equity has lined up behind 12 bids totalling $8.5 billion in market value in 2003. Miller says that MMC receives a market in project debt daily from the main trading intermediaries.

The main trading operations with a capability in project debt are at the investment banks, with Goldman Sachs, CSFB, and Deutsche Bank the names emerging most frequently. CSFB's presence comes from its bookrunning and arranging franchise, while Deutsche's background has already been highlighted. And Goldman Sachs has a private equity and energy trading arm that complements its existing business.

One source at Goldman Sachs says that, "we make markets in project finance debt that we arrange, such as Term B issues, but not in other project finance debt." Goldman has, however, been one of the most active traders in the debt behind the AES Drax acquisition. Indeed, distressed desks are the players whose activities are hardest to quantify, since they often involve trading on banks' own accounts. They are the main reason for the liquid nature of UK power debt, and, as Miller notes "the US is around 18-24 months behind the UK, in terms of distressed debt".

The UK experience is instructive - since it shows that the bid-ask spread on project debt has narrowed so little that banks decided to become distressed buyers themselves (for more details, search for 'CGE' on www.projectfinancemagazine.com). And large-scale sellers of project loans have done so more for strategic reasons than through any assessment of a given project's virtues. The sale of Dresdner's stakes in project loans was achieved as part of a larger auction of its loan portfolio, although distressed credits had a heavy representation within that subsection.

"If distressed desks remain the homes for secondary trading in project debt, there is every reason to believe that the markets currently open to distressed borrowers will not be open forever," says one banker active in project lending. This has to be the hope of traditional project lenders wanting to reclaim lost business. Miller is sceptical, however: "I think the lesson we learned from the real estate crisis of the late 80's is that troubled loans should be moved to a dedicated asset recovery group as quickly as possible, since there may be too much attachment to projects in the project finance groups. Once in the workout group, banks are free to make the necessary provisions, and, say, build up a book of equity warrants that might be available to banks that sell below par."

But this is still a vision of the market for project debt rather than the reality. As Thomas at TCW puts it, "I think that the secondary market is still characterised by spotty growth, and its difficult to say whether there has been any improvement in liquidity beyond the presence of a few large sellers and a number of non-bank players chasing yields." Some lenders, such as TCW, Beal Bank, and GE Capital's newly aggressive debt business, are looking to hold debt. The sceptical players will probably conclude that such a universe will suffice for the piecemeal disposals banks are willing to make.

But in power, at least, there is still an industry that needs to restructure, beyond the development of any debt product. As Miller notes, "the pure financial arbitrage opportunities in the sector are over - it is about getting physical control of the assets. You need to identify the debt that you are looking for and work out which sets of creditors you need to deal with." Miller's words are another challenge to the bank lenders. He believes that future restructurings will demonstrate that non-bank creditors will align with physical asset players such as MMC to realise the remaining value in distressed energy credits.

But, even if non-recourse finance does make a comeback in the power sector, bankers might be wondering whether they are in line to provide it. One comforting thought to banks sitting on sour loan portfolios was that the current situation of distress might give lenders more leverage when negotiating with their clients. The current non-bank demand, however, means that they may be frozen out.