The Basel II accords have been several years in the making and are designed to increase the transparency and stability of the world banking system. They also present the biggest threat to project finance in its current form. Within the overall proposals, rules on specialized lending take a very dim view of the risk profile for project finance loans.

Yet the industry has been very slow to present its case to a group of regulators with little in the way of experience of project lending. Submissions have been sporadic and piecemeal. In part, this lack of consensus can be put down to the normal healthy competition between banks. Many have different ways of analysing project risk internally. Alarmingly, the current proposals may allow those banks that can demonstrate they have sufficiently rigorous internal rating systems some capital relief. Good luck to these top tier institutions in syndicating deals.

Another obstacle is the marked difference in culture between US project lenders and those working elsewhere. Competition in lending is so sluggish in the US market that pricing has already been edging up. Enron and the new rules could eliminate non-recourse lending as a useful financial instrument for all but the most desperate sponsor. There are, after all, a number of other tools available.

In Europe the picture is different in that the syndications market is relatively buoyant. One professional has suggested that the new rules might even bring much needed discipline to a world of razor-thin coverages. The emerging markets look like being the chief casualties, especially if (as is possible), new capital treatment rules kill the multilaterals' B loan programmes instantaneously.

Project Finance magazine recently held two roundtables to discuss the proposals. The first, held in New York in December, is presented here. An account of proceedings at the London event will follow next month. The aim of the meetings was to share ideas but they also form the start of a consensus that could present a unified and accurate picture of project risk to Basel.

Implementation of the Accords is some way off and is, in theory, voluntary. Nevertheless, there are opportunities for an effective lobbying effort. The first of these, a recent Models Taskforce set to consider project lending, has passed. One option, apparently co-ordinated by Citigroup, is for banks to pool historical data for presentation in February. This has met with a mixed response from some institutions. Some at the top end, with Basel teams in place, seem unwilling to duplicate efforts, or possibly trust in their institutions to allow project finance groups to continue in business. Reaction at the participant end has been sluggish at best.

Evidence collected so far is that default rates, and particularly recovery rates, compare very favourably with corporate lending. Enron's project lenders are now taking steps to sell off assets at reasonable prices while its corporate lenders are trying desperately to make themselves heard in the Manhattan bankruptcy court. Losses on telecoms project financings have been manageable.

But perversely, at a time when the case for project lending has never appeared stronger and the returns never more enticing, several banks are looking for ways out of the project business. Few of them have been giving their clients much warning. The case for more lenient treatment will have to be made within institutions as well as to the regulators. Piecemeal fragments of portfolios may not be enough.

The discussion below looks at the mechanics of the proposals' possible effect on lending. It looks, however unwillingly, at the possible mitigants ? and few of them are particularly palatable. Banks and ratings agencies have a formidable task on their hands.

Project Finance: What effect do you imagine that the Basel proposals will have on your institutions' portfolio?

Chris Beale (Citigroup): The effect on the industry could be very adverse. The early indications were that the Basel Committee regarded the project finance asset class as riskier than corporate assets. We don't agree and our experience has been different. We have a better record on project assets than on corporate assets. Essentially, the recovery rate on project assets is much better than the recovery rate on corporate assets across the full credit spectrum. In fact, the lower you go down the credit spectrum, the better project finance assets perform relative to corporates. For investment grade assets, we are better than corporates, but for non-investment grade assets, we are much better than corporates. I think that this is because we do a lot of structuring to box the risks. We usually have security in the assets, we often have contractual supports, and we sometimes have political risk insurance or offshore accounts. Above all, we have transparency on the inputs and outputs and we are able to do intensive analysis of projections and sensitivities to key variables rather than just taking a static picture of the corporate balance sheet.

Sandra Bell (Deutsche Bank): Our institution's portfolio, like most, is primarily corporate exposures. As such, the focus of the process, internally as well as externally has been on developing a capital regime for corporate exposures. Of necessity, the people involved in the Basel process have come from a corporate lending background. Project finance lending is specialized and results in a different set of experiences than corporate lending. Witness, as an example, the specialized work and the teams that the ratings agencies have put together in recognition of the different analytical process necessary to evaluate project finance exposures. Simply by acknowledging the number of years experience ? a decade plus since Bill Chew set up a separate practice ? which has been invested by the agencies to develop a sophisticated basis for analyzing project finance, I think we have to all recognise that it is very difficult for this task force to garner the same level of understanding of the nature of our business in such a short period.

PF: How close are we to establishing some kind of common methodology?

Bill Chew (Standard & Poors): We have been developing methods of analysis of default risk, as part of our process of rating CDOs, as well as recovery rate analysis. So there is information out there. There is a decent-sized portfolio that we have built up to test empirically. But we would argue that there is not a single black box that deals can fit in. You really need to have a much more open process for analysing projects. The attraction of the efficiency is there, but we can't come up with a single framework.

Eric McCartney (KBC): I think that the proposals have brought not just a little, but probably too much discipline to the market. There is a certain level of discipline in risk analysis but I'm not sure there's a black box. I'm not sure that a sufficient amount of credit is given to a secured loan as opposed to a corporate credit? It seems to me that sometimes even though you're secured you're somewhat penalised because the risks are more complicated.

Susan Abbott (Moody's Investors Service): One of the other issues is that it is very hard to construct a single framework within which you can stick a project. My experience is that every single deal is different and we spend a lot of time saying, ?Well this deal is like this and how does it compare to that one, and it's like this, and its very unlike the other one in that place?. And so we end up looking at how many things weigh down on one side and how many things weigh down on the other to come up with something even.

Sandra Bell: The current methodology proposed by the Basel task force is based on a basic assumption ? that ratings and recovery are highly correlated for project finance exposures ? which the experience of project finance lenders would show otherwise. Project finance professionals would argue that rating and recovery don't necessarily go together, with the latter very much a function of the type of asset in default and its relative industry position. Unless the project finance industry can demonstrate to the Basel task force that this fundamental assumption is flawed there will continue to be a very large gap between the proposed capital regime and the reality of experience.

Bruno Mejean (NordLB): Should we have been approached to provide historical data? It seems to me that one reason for this conservative approach is that there is a lack of data regarding defaults and recovery rates and so on.

Bill Chew: Well we've had inquiries and we've responded, that there is in fact some data available on defaults. There's still a low volume of public deals, but as part of the CDO process we have done a lot of work. In addition to those there are another 500 deals that aren't public.

Susan Abbott: Just like S&P we have two levels of discussion. And a lot of project finance research has been bundled up in the other ratings activities we've carried out. We have had a series of meetings with them, and intend to have a few more, where we've tried to educate them about what project finance is all about. I would like to stress that we have not published anything publicly to date.

PF: Do you think that increased use of corporate finance-type structures ? especially in US power ? may muddy analysis? We've started to call this ?corporatization?.

Chris Beale: I don't think this leads to muddy analysis or whatever. When we finance pools of merchant power assets on a non-recourse basis we do exactly the same sort of analysis we would do on a single asset project, and more. We analyze market dynamics, competitive cost position, and any barriers to entry. Our analytical work is far more intensive than on a typical corporate transaction. We can demand and get more data.

Jay Worenklein (SG): Whether the analysis of risk becomes more difficult in a corporate finance-type structure depends a lot on the structure. In general, the corporate finance structures employed by project financiers in the last several years have primarily involved the pooling of multiple projects (for example, ANP, TECO Panda, Calpine Construct), and these deals are no more complex than the projects themselves. I do not comment on the share trust deals that have gained notoriety other than to observe that the credit support for these deals was the corporate obligation of the sponsors to deposit their stock, rather than an analysis of the quality of the projects that some of these deals financed.

The question raises the broader issue of whether it is fair to say that project finance is a risky business in light of the California mess last year, the Enron debacle, the fallout effects of Enron on other project finance sponsors, the problems of Argentina, etc. None of these situations is likely to lead to any losses at the project finance level. The Enron experience in fact suggests that lenders are often better off lending to projects they can analyze and understand than to corporate credits that are better rated but are affected by a much wider variety of risks.

The portfolio performance of most of the major project finance banks demonstrates that in crisis situations, such as the Asian crisis in 1997 and I would predict the Argentina crisis of 2002, banks have much higher recoveries on their project finance loans than on their corporate loans. This is because of the combination of the well-structured and secured nature of most projects, the amount of equity underlying the project debt, the need for the output of the project, the sponsorship of projects by companies with a long-term commitment to their industries and to the countries involved, the careful analysis of downside scenarios relating to commodity pricing and other factors. Even the long maturity of some project finance deals, which many people consider to be the source of significant risk, needs to be understood in practical terms. A deal with a 16-year final maturity might have an 11-year average maturity but ? because of project refinancings ? the actual average maturity in some banks' project finance portfolios between 1994 and 2001 was less than three years.

Sandra Bell: The Basel task force has taken the concept of restructuring in the project finance area as a negative. On the contrary, those of us who practice this every day view restructuring as a basis for ensuring ultimate repayment and recovery. In particular, project finance lenders put a great deal of emphasis on one aspect as particularly important, and that is the covenant package. Covenant packages are deliberately designed to allow a lot of things to happen early, providing the opportunity for projects to restructure. This provides project finance lenders with a powerful risk management tool not available to lenders to equivalently-rated corporates. In the corporate world, restructuring connotes the concept of loss, as corporate restructurings often result in lenders taking more subordinated securities or other less valuable compensation as part of a restructuring package. In contrast, in the project finance world, restructuring is associated with full recovery and often includes enhanced coverage and security. And that to me is a big distinction not recognized by the Basel process.

Bill Chew: The notion of covenants within corporate finance is much different. Many project finance covenants have as their aim the trapping of cashflow for the protection of lenders.

Tom Murray: What you're also dealing with a lot now is portfolio financings where ? even if you don't have an explicit guarantee from a corporate parent company ? you are financing several assets on a cross collateralized basis which are non-recourse but strategically important to the sponsor. In many cases it is a lot less likely, given the tens of million dollars in equity that they've put in and the cross collateralization, that they'll walk away. There is more of a corporatization because you're financing businesses which require procurement and marketing expertise. We view non-recourse portfolio financings as the best, most secure structures ? an example being the Calpine Construction Revolvers deal we led recently. It is better than corporate lending because the lenders have significant asset coverage and a tighter covenant package than they would get on a corporate basis.

Bruno Mejean: From our point of view there are projects which can no longer be financed on a standalone basis. The only way to do some is to aggregate them with projects in other regions to get diversity geographically. So I think the genco concept will continue, but single assets will face greater scrutiny.

PF: What sorts of instruments exist for mitigating adverse capital treatment? I guess we're mainly talking about project CLOs here.

Sandra Bell: The value of CLOs as a capital management tool ultimately comes down to how the BIS [Bank of International Settlements] treats first loss positions in CLOs relative to the capital treatment of the project finance loans which are being sold. The long and the short of it is that the size of the first loss position assigned by the rating agencies in project finance CLOs is often twice that of a corporate CLO, primarily due to the fewer number of loans. The cost of the lower-rated tranches is also so high that it is often not economically justified to sell all the way down to a BB tranche. The net effect of these factors is that, if retained interests in CLOs are charged one for one reduction in capital, you achieve very little capital relief without selling a portion of the first loss position. Given the limited universe for buyers for first loss positions, this can also be an expensive proposition. So, in order to truly get risk transfer based on the rules ? and ultimately capital relief (which at the end of the day we're all trying to do) ? it is an extremely expensive proposition even if you can find a buyer. People are looking to the use of these instruments to manage the use of capital, but expected treatment of retained interests in CLOs combined with the potentially higher capital treatment for BBB and BB rated project finance loans is unlikely to take the upward pressure off price for project finance assets in the future.

Eric McCartney: Here at KBC we've been evaluating the structure and we were one of the first institutions to do a European corporate CLO. The banks came to a rude awakening that, while when they first did them it was a way of managing risk, people now realise that it is more about managing capital. Sometimes when a loan in a CLO defaults in order to preserve the CLO and prevent it from collapsing you have to buy the asset out. Really it is not a way to diversify your risk. What it is doing is simply reducing your capital.

Chris Beale: We did two CBOs this year, one for our assets and one for a client's assets. The one for our own assets was done to recycle capital and increase our return on equity. The one for the client, Trust Company of the West, was a blind pool of assets to be accumulated by them, with the client intending to make a profit through good project selection. In both CBOs we were able to distribute the second loss mezzanine tranche (which is a hard piece to price and sell) to institutional investors and insurers. We want to do more CBOs for ourselves and for others to build an alternative distribution channel for project assets.

Jay Worenklein: Project finance CLOs are at the babyhood of their development. As we each do our first deals and experiment with various structures, we will get smarter. So we at SG will start with a synthetic CLO in the nature of a credit default swap, with various tiers of priority, and we will learn the lessons of this deal in the next deal we hope to do some months later. It is important for all of us in the business to increase liquidity, to find new investors, and to develop better structures for freeing up and rolling over our capital employed in the business. In this respect these are very exciting times for the project finance business as we develop techniques for increasing the liquidity of the project finance market.

PF: Might a repricing of assets have to happen?

Tom Murray: If you're referring to CLOs buying up project paper at Libor plus 150 the issue is the yield. The interesting thing is if the new Basel rules require higher capital to be set aside and/or higher ratings criteria then there will likely have to be a repricing of project finance loans. On the positive side, a repricing would likely stimulate a secondary market and more liquidity and would be positive from a syndications standpoint because you'd have an alternative market to go to. If such a repricing were to occur, I would also expect to see a shift to more bond deals because right now the pricing in the bank market is cheap relative to the bond market, even with the underlying treasuries the way they are.

Eric McCartney: If Basel comes to pass and these loans need to have a higher capital allocation, and the level of pricing is inelastic, a developer will see lenders going up from 150bp to 350bp because we need to allocate more capital to them. The question becomes whether that is the right alternative for that developer to do a project financing. They might prefer to do something on their balance sheet. It's all about competition between the players here in this room when we come to pricing in the bank market. Here in the US we're pricing a project finance loan at 150bp, whereas in Europe it is often nearer 100bp for a very similar risk. How do you make that difference? And it's all about competition. There's not enough business to feed all the mouths in Europe, so people become very aggressive and they price the asset lower. Does that really price it as a different risk? In my opinion, no.

Jay Worenklein: The fact that secondary market pricing usually results in higher yields than the pricing of the original loan in the primary market reflects the fact that loans have often been underpriced to our clients. This creates a vicious circle because we need to attract more investors into the primary market, but it is difficult for some investors to rationalize entry into the market on a primary basis when they see the losses they would have to take if they need or want to sell their loans on a secondary basis. The consequence is that pricing has risen in the primary markets and that innovative techniques are being developed, such as various forms of CLOs, to create greater efficiency and liquidity in the secondary markets

PF: Would an increase in bank loan ratings help?

Bill Chew: We've begun to see that already, although the new regime has yet to bite. Even in advance we are seeing more bank deals getting rated, either to ease syndication or to instill confidence, to carry out internal risk-weighting, or as part of the process of securitizing such loans.

Jay Worenklein: We at SG are getting nearly every bank deal rated as we prepare ourselves for our CLO and other secondary portfolio transactions. We ought to get to the stage where virtually every deal is rated at its inception both because it is helpful in the primary syndication and because otherwise many banks will proceed on their own to get ratings on the same project, which is highly inefficient.

Tom Murray: To bring new entrants like institutional loan buyers in quickly and avoid them having to become project finance experts straight away we would need to bring the ratings agencies in and have more bank deals rated. That said, when you push the envelope out you're not always going to get the required investment grade rating. A requirement of our Project Finance Bank Loan Program is that a project has to get a rating from either S&P or Moody's for inclusion. But a lot of clients sometimes feel that it's an extra step that they don't want to take. n

Attendees at Project Finance's New York roundtable were:

Susan Abbott, managing director, project and infrastructure finance, Moody's Investors Service

Chris Beale, global head of project and structured trade finance, Citigroup

Sandra Bell, managing director, Deutsche Bank Alex Brown

Michel Chatain, head of project and sectorial finance Americas, SG

Bill Chew, managing director, infrastructure finance ratings, Standard & Poor's

Eric McCartney, head of project finance Americas, KBC Bank

Bruno Mejean, senior vice-president structured finance, Nord LB

Thomas Murray, managing director, Credit Suisse First Boston

Craig Orchant, managing director, Deutsche Bank Alex Brown

Monique Palumbo, project finance syndications, KBC Bank

Richard Slocum, managing director global structured finance, WestLB

Jay Worenklein, global head of project and sectorial finance, SG

Project Finance would like to thank all attendees for their contributions and, in particular, Standard & Poor's, which hosted the event.