Project lending looks set to be hit hard by the proposed new capital adequacy provisions being prepared by the Basle Committee, and project financiers are getting ready to fight their corner as they await the latest draft due to be published within the next couple of months.

To some extent it is inevitable that project finance will be negatively impacted under Basle II. The main benefit for corporate lenders is that triple-A corporate loans will now be assessed at 20% risk weighting, correcting the current anomaly where loans to even the best blue chip borrowers are weighted at 100%.

This means that some lower rated loans are going to be weighted at more than 100%. Under the expected draft proposals for asset based lending that may mean a 150% risk weighting kicking in at either the single B level, or even the double-B level, with triple-B loans remaining at 100%. The actual capital adequacy ratio will stay at its current 8%.

The weighting game

Given the fact that much project lending around the world takes place in the range between triple B and single B, the project finance sector was always bound to feel the squeeze. But the direction in which the Basle Committee has been heading with regard to several crucial issues makes matters even worse, and is set to force banks to modify their internal risk weighting models in such a way that project lending is classified as inherently risky.

These issues revolve around the probability of default, and the probable amount that a bank will recover from a loan in default. Bankers say that the Basle Committee is proceeding in such a way that the probability of default on project loans may be exaggerated, while assumptions on losses in the event of a default will also be on the high side. Add the two together, and the tendency will be to assign high risk weightings to projects- jacking up capital adequacy provisions in a way which will make bank lending to projects more expensive.

?There has been a great deal of confusion in the market over how project finance is going to be treated, and one important issue will be how they treat construction risk,? comments Andrew Blease, credit analyst at Moody's Investors Service in London. ?I don't think the project lending banks will on balance benefit, and they may find themselves in a worse position than at present. Even a good project loan may come in around the Baa level, and will be 100% risk weighted, and an average project loan may be in the speculative grade Ba category, which will be 150% weighted.?

?So Basle may hit you twice,? says Blease. ?First because project loans are down at the bottom end of the credit spectrum anyway, and then hitting you again because in the eyes of Basle project loans may be regarded as an even lower credit risk than the banks may think.?

Certainly project bankers feel that the proposed assumptions on default and loss recovery will need to be altered. ?The majority of restructurings in project finance involve the reworking of the payment schedule, or the leverage, to improve the ability of the deal to repay, and may often involve additional equity coming in from sponsors,? says Sandra Bell, Managing Director in the Project Finance Group at Deutsche Banc Alex.Brown in New York.

?The premise of the Basle Committee is that there is a high correlation between these defaults and loss given default for project finance loans,? says Bell. ?As a result, they view the higher incidence of restructuring of project finance loans as a negative, rather than a positive. In reality, restructurings have increased the recovery of project finance loans.?

?You often hear lawyers talk about a big D default and a little D default,? explains Craig Orchant, Head of Project Finance in the Americas at Deutsche Banc Alex.Brown in New York. ?A big D default is a payment default or something very material to the transaction which creates irreparable damage. These big D defaults track pretty well to the kinds of default you would see in corporate obligations.?

?In contrast, a project deal often has a very broad number of triggers (little D defaults) incorporated into the structure, that give banks early warning signals and the opportunity to restructure the loan to accommodate changes in collateral value or operating and market conditions,? says Orchant. ?Most project finance bankers consider this ongoing loan management process a great strength of project structuring, but right now the Basle Committee is undecided on how to characterise that early default system that is incorporated into many project deals.?

And having concluded that projects must be judged to have a high default risk, the Basle Committee looks set to compound this by assuming that the severity of loss in the event of default will be high. ?It is the exact opposite of where the rating agencies have come out,? comments Bell. ?When you look at what the Basle Committee is proposing for losses in default, project finance loans are to be allocated the highest loss given default of any asset class, which is the exact opposite of how the rating agencies look at project finance, where they assign the lowest loss given default of any asset class.?

These issues are likely to be fiercely contested over the next twelve months, as the new capital adequacy rules are discussed and commented upon by industry groups. But as Alexander Batchvarov, fixed income analyst at Merrill Lynch in London, notes, the attempts to put in place new frameworks for assessing the risks of different types of loans was always bound to throw up many complexities.

?The Basle Proposals emphasise an internal over external risk assessment approach, and allow for more individual approaches within a broader framework proposed by the Basle Committee, and to be enforced by the national regulators,? Batchvarov explains ?With regard to the internal approach, for the first time I see the Committee highlight project finance as a separate type of product, for which the banks can use different types of risk weightings internally. It basically means they will be differentiating the risk weightings for project finance, but we are still waiting for details when the next draft is published.?

Iain Barbour, global head of the structured finance research team at Commerzbank Securities in London, is also awaiting clarification in the next draft, but feels that structured risk is being treated too harshly versus corporate risk.

Capital redistribution ? but where?

?The objective of Basle is to keep roughly the same amount of capital in the system, but to distribute it differently, to better reflect the risks inherent in the banking system,? he says, but structured finance throws up its own complexities. ?For example, the probable loss severity of a project loan changes over its life,? Barbour notes. ?And if an issuer did a project finance bond rated BB, there are a lot of questions in the market as to why this structured risk should be weighted at 150% while a BB corporate risk would be rated at 100%.?

Loans where external ratings are available fall under what is known as the Standardised Approach. Where rating agencies are not involved, risk weightings will depend upon the Internal Measurement Approach, using internal bank systems. The plan is for these internal systems to include separate assessment methodologies for project finance loans.

In order to address the wide spectrum of capabilities and risk management sophistication of various banks, the Committee anticipates that there will be two variations of the internal ratings based approach. The first is known as the Foundation Methodology, where the bank uses its own estimates for probability of default and uses other regulator-provided inputs to determine the potential future loss amounts that it faces on the asset side.

The second approach, known as the Advanced Methodology, is for larger and more sophisticated banks. Under this system the bank, with the regulator's blessing, will use its own estimates as inputs to determine future potential losses. National regulators will take on a role of monitoring the systems of banks under their jurisdiction.

But a major problem is the lack of historical data on the probability of default, and loss given default, specific to project finance transactions. In a report published in May, Moody's noted that ?without such data, it is difficult for any lending institution to demonstrate the accuracy of its risk weights applied to a particular transaction. In our view, this problem is not easily rectified by a bank demonstrating the accuracy of its estimates for corporates, because of the large differences between project finance and corporate lending.?

Describing historical default loss or data specific to project finance as being ?sorely lacking? Moody's says ?we believe that it is currently difficult to verify whether many lending institutions can accurately predict default or losses for a project finance loan. Our experience suggests that few institutions currently record separately extensive default or loss data for project finance lending, and that such data- if available- will most likely form part of corporate loss and default data.?

?In addition, the long gestation period for projects from lending to operation means that there is likely to be little useful time-series data observing the behaviour of project finance loans. Many current project loans are either still not operational or, even if operational, have not yet been through a full economic cycle.?

Bankers also note that if one national regulator is less onerous in the way it oversees the commercial banks, then banks in one jurisdiction will be hit harder than others with respect to capital adequacy provisions. And they suggest that there will be an important role for the rating agencies, as internal bank assessments may be matched to external ratings, providing some sort of common yardstick. Indeed in July of this year Fitch Ratings published a paper where it looked at how internal risk grading systems might be assessed for quality and efficacy, and came up with a summary of ?best practices? for such systems.

Bankers note that discussion on new rules for asset based lending are running well behind those for corporate lending, partly because corporate lending is a bigger slice of the business, and partly because of the technicalities and structures that make project lending more complex to assess.

The Basle changes are not scheduled to kick in until 2004, but the big project finance arrangers and lenders are already looking ahead to a time when capital charges will be more expensive, and there is already a feeling that the Basle rules will reduce the role of bank debt in project finance. Which means that bringing in new sources of capital is becoming even more important.

?The pricing of bank credit versus other capital providers of project finance has historically been lower, and therefore the banks have been the largest creditors of project finance,? says Orchant. ?That may well change going forward. If the cheapest source of capital is shrinking, it will tend to increase the cost of project finance.?

But he notes that funding sources for project finance are already being broadened. ?Deutsche Bank is a proponent of looking for new sources of risk capital for this business, be it through institutional investors or other participants,? Orchant says. ?The expansion of the sources of capital used is already happening, from institutional investors in the bond market to other third party risk providers such as financial guarantors and insurance companies. The likely outcome of higher capital adequacy requirements for banks will be to accelerate these trends.?

CDO solutions

One solution is Collateralised Debt Obligations (CDOs), where loans are moved off the balance sheet via securitisation. This may be especially attractive since the capital weightings for banks buying and holding these securities on balance sheet will set at 20% under Basle II.

At present banks keep the first loss piece, and so the amount of capital set aside remains quite high. Thus CDOs are being mainly used as a way to ensure the liquidity of loans, and ability to manage your loan book is the major reason for project finance CDOs, because without selling equity you get limited regulatory capital relief.

However if a market develops where the first loss pieces can be sold at a price attractive to the issuing bank, then capital adequacy set asides will be reduced to zero, making securitisation more attractive. And if project lending has some of the harshest capital adequacy rules, then project loans will become prime contenders to be packaged up and sold off via securitisations.

?Without the sale of the equity piece, CDOs are less a regulatory relief vehicle than a balance sheet management tool,? says Bell at Deutsche. ?But an interesting market which is developing is the market for the equity in CDOs of project finance loans.?

?The investors may be the same people who are already typically looking at the mezzanine components of project financings, the lower rated, higher return investors who have a deep understanding of project finance. Right now there are a handful of players there, but more and more institutional players and insurance companies are looking at what is an alternative class of non rated or lower rated investments with a high relative return. At the end of the day, the availability of investors for the equity pieces will dictate whether CLOs are done for regulatory capital relief reasons.?

To date three deals have been done, with banks keeping the first loss piece in each case. The first two were for Credit Suisse First Boston, and in July Citibank launched its first deal. And Deutsche Bank and Chase have both been working on their own deals.

The Citibank offering was launched under the name Project Securitisation Co I Ltd, and comprised $350 million worth of floating rate notes, backed by a pool of 25 project loans originated by Citibank. The deal was underwritten by Schroder Salomon Smith Barney.

But the scope for widespread use of CDOs may be less in project finance than in other areas of bank lending. The main problem with such deals, versus other types of securitisations, is that project lenders have a relatively small number of loans on their books. They cannot achieve the diversity seen in other asset pools such as corporate loans, car loans or mortgage loans, where thousands of loans are commonly pooled together. In addition, project finance has high concentrations in certain sectors such as energy, which also makes for concentration risk.

In the case of the Citibank deal, all 25 loans in the pool were examined and rated by Standard & Poors. This process of having to rate each individual loan adds considerably to the expense of the transaction.

The individual results were not disclosed in the prospectus, nor in the offering prospectus was there any list of the actual projects involved. There was however some statistical information about geographical and industry concentration.

?These deals are very time consuming to put together, since the analysis is more complicated than a straightforward corporate loan CLO, and banks also have to go through a process of getting consent from each borrower to sell the participations in the loans to a Special Purpose Vehicle,? says Mike Wilkins, director at Standard & Poor's in London.

?We describe it as a grey box rather than a black box,? Wilkins explains. ?With a black box you don't know what is going in one side, you just know what is coming out the other. In the case of a grey box you know a little bit about what is going in, since we were able to give some information such as geographical spread.? There was however no list of projects or the countries they are located in on the Citibank deal rated by S&P.

Wilkins does envisage an increase in the number of project loan CDOs, though he notes that only the very biggest players in project finance may find it worthwhile. ?German banks have been very active in project lending, and European banks in general have lent very aggressively, though some have tended to see loans as tied to a relationship with a client who may be a sponsor or supplier to a project,? he says. ?Low loan margins do not help in putting together a CLO, because they rely upon cashflows versus risk,? he says, and so a larger amount of credit enhancement will be needed if some of the more aggressive European project lenders are going to get deals done.

CDO technology and a market for the first loss pieces could be well developed by 2004, which may give project banks a solution if Basle does come down hard on the sector. In the meantime the market is eagerly awaiting the next draft from the Committee, and between now and the end of the year will be stepping up their work to ensure that project finance does not emerge as one of the hardest hit areas under the new capital adequacy regulations.