Contents

CLO slew

Balance sheet deals ease warehouse risk

Seven US CLOs priced during September for a volume of US$6.9bn, five of them – all leveraged loan transactions – printing over the last week. This slew of paper marks a consolidated effort by the Street to manage its warehouse risk via static short-dated deals.

"The bulk of the assets backing the new deals are believed to be from existing arranger warehouses – it's unclear how much of the paper would have been sold to third-party investors," notes one buysider.

The transactions are characterised by their high concentration of first-lien loans and their tight pricing. For example, the US$1.7bn multi-managed Integral Funding – through Credit Suisse – priced blended triple-A to double-B paper at 65, 135, 225, 450 and 550bp over Libor. The US$965mn static Atlantis Funding 2007-1 from Citi priced triple-A to triple-Bs at 60, 150, 250 and 425bp over.

Both deals sourced collateral from various warehouses, with Octagon Credit Investors and Credit Suisse Alternative Capital among the managers assigned to select and dispose of assets for the Integral CDO.

Goldman Sachs, Lehman Brothers and UBS also priced balance sheet CLOs last week. Goldman's US$500m Muir Grove CLO will be advised by Tall Tree Investment Management; Lehman's US$500m Lightpoint CLO VIII by Lehman Brothers Asset Management; and UBS' US$400m Lime Street CLO by Feingold O'Keeffe Capital.

Other recent prints include Credit Suisse's US$600m Stone Tower CLO VII (managed by Stone Tower Debt Advisors) and Lehman Brother's US$700m Ares XII (Ares Management), the triple-A to double-B tranches of which priced tight to initial guidance at 63, 100, 200, 325 and 575bp.

Ashish Keyal, CDO strategist at Lehman Brothers in London, confirms that the CLO sector has achieved some stability in terms of pricing. "With generic spreads for triple-A CLO paper at 65-75bp, 120-150bp for double-A and even 250bp for single-A, current levels are prohibitive for many issuers," he says. "But these levels are no longer gapping out wider – and there has been an increase in secondary activity, with price determination now more apparent."

Indeed, whether the market sees any new warehouses opening depends on which level banks are willing to place paper at. "It's a tough call as to when the market will return: there's limited supply, which works in the market's favour. But any increase in volume will take a while to come through – it is proving difficult to turn the market around," adds Keyal.

While analysts at Citi don't expect triple-As and double-As to tighten dramatically in the near term, moderate tightening to a 50-60bp and 120-150bp trading range appears reasonable. Tightening in senior CLO spreads by around 10-20bp is expected to help clear the LBO pipeline.

Issuance is nevertheless improving in the underlying leveraged loan market, with over US$6bn of paper being placed with investors across several deals over the last week and KKR completing its acquisition of First Data.

But perhaps even more importantly for the US CDO sector as a whole, high grade structured finance CDOs are re-emerging after a dearth of issuance. Mizuho has launched the US$1.6bn Brighton 2007-1 deal for Delaware Asset Advisors, while UBS brought the US$1bn Highbridge 2007-2 for Merrill Lynch.

And in the last two weeks, six new CDOs have entered the US pipeline with a volume of US$3.3bn.

CS

Super-senior supply

ABS exposure on offer via bespoke correlation trades

A developing supply of ABS super-senior paper – a feature more commonly associated with the corporate tranche sector – is being selectively seen in the structured credit markets. Greater focus on risk management, together with balance sheet constraints, has resulted in many players seeking to offload this type of exposure.

The withdrawal of many CDO and ABS buyers has left dealers long exposure – both directly and indirectly – to US sub-prime MBS. This situation is allowing investors access to new axes, according to Citi CDO strategist Ratul Roy – in particular the most senior part of the capital structure of pools of ABS via public cashflow CDOs or bespoke ABS correlation trades.

"It comes at an opportune time for many bank investors, as the new Basel II ushers in substantially less onerous capital requirements for super-senior investors," he says. "For example, super triple-A tranches of CDO structures receive a regulatory weight of 20% using the standardised approach (or even a lower 7% using the IRB approach) versus 100% under Basel I."

Such trades enable investors to take positions on a tailored portfolio of ABS and/or ABS CDO names. The portfolio can be a synthetic replica of an existing CDO or a new portfolio created from a basket of credits that the investor selects.

Roy notes that other routes are also possible. "Because super-senior notionals can be large, participants may not want to risk the entire capital. They may want to take a thin super senior (for example, a 10-20% tranche within a high-grade deal), thus putting less capital at risk, or a junior tranche/first-to-default of a portfolio of super seniors," he says.

Long-short combinations are another alternative. The net payout of these trades will depend on where losses penetrate the capital structure and their timing. Timing is important because even if cumulative losses are high enough to impair the more senior of the long-short pair, the timing may be such that the interest differential between the two legs and the amount of time that the differential is earned could have a significant impact on overall economics.

Factors determining tranche returns are the coupons paid by the dealer to the investor and principal losses and payments through the life of the trade. The coupon that is paid by the dealer to buy protection on a slice of risk depends on: the risk of the portfolio; the likelihood of large losses, even if the expected amount of losses within the portfolio remains the same; tranche structure in terms of subordination and thickness; and any cashflow mechanisms which could lead to early repayment.

TABX is often portrayed as the index version of tranched risk, but it pools BBB/BBB- risk while many available super seniors relate to high-grade CDO portfolios where the average quality is closer to double-A minus. Additionally, TABX has little diversification in vintage and ratings; and the quality of collateral in high grade CDOs can be higher as measured by ratings or by the loan delinquency of the bonds within the CDO pool (after adjusting for differences in subordination below the securities).

Another qualitative difference with an ABX portfolio is the potential for large CDO buckets to be included in the collateral.

CS

Relative value trade

High implied correlation between CLO equity and LCDX tranches

As details of LCDX standardised tranches are clarified ahead of next week's launch, relative value opportunities are emerging. The new instruments are also expected to be of interest in the current environment to both corporate and CLO investors.

Dealers' indicative prices for the 0-5% LCDX tranche imply correlation significantly above that in the CDX.HY 0-10% tranche and above the 0-3% CDX.IG equity tranche. But investors need to be aware that the risk in each tranche is different.

In order to more meaningfully compare correlation across these different products, analysts at JP Morgan have used a multiple of expected loss in each portfolio. "The results show that LCDX indicative prices suggest higher correlation across the equity and mezzanine tranches compared with CDX.IG and CDX.HY tranches, even when adjusting for differences in expected loss across the products," explains Eric Beinstein, head of credit derivatives research at the bank.

He adds: "The flipside of this is that the 15-100% tranche of LCDX has relatively more risk than the other two indices. One could conclude therefore that investors are more comfortable with LCDX equity compared to high yield equity."

As part of their effort to establish whether LCDX tranches would make reasonable alternatives for CLO investors in the current environment, the analysts found that the collateral, ratings and expected recoveries are similar for both instruments. Loss-given default is around 30% for both LCDX tranches and CLOs.

But while LCDX tranches typically reference 100 names, CLOs have more diversification. It is estimated that there are around 1500 leveraged loan issuers and around 600 to 900 reference entities in a given vintage.

The LCDX index exhibits covenant-lite loan concentration, but all reference entities are first lien and therefore such portfolios will be more homogeneous than CLOs. CLOs typically have buckets comprising second lien and high yield collateral.

Structurally, the senior-most LCDX tranche attaches at 15-100% and is therefore wider than a typical triple-A CLO tranche (which attach at 25-30%). LCDX carves out two tranches of equity – the 0-5% and 5-8% portions – making it more leveraged than a typical CLO, according to Kedran Garrison Panageas, a CDO research analyst at JP Morgan.

"LCDX has a linear write-down structure, whereas AA/A CLO noteholders benefit from overcollateralisation tests if a CLO performs badly – therefore offering more protection than an LCDX tranche does," she adds.

LCDX reference entities aren't publicly rated, so the analysts made their own assumptions about where rating agencies would likely rate LCDX tranches. Because of their greater leverage and lack of interest diversion, implied ratings are lower for LCDX tranches than CLOs. The 15-100% tranche would be equivalent to a single-A rating, the 0-5% and 5-8% to single-C, the 8-12% to single-B minus and the 12-15% to triple-B minus.

In terms of relative value, based on the fact that the 15-100% LCDX tranche is likely to trade at around 50bp and CLOs trade at around 90bp all-in, CLOs are more attractive. But, says Garrison Panageas, the funding costs of a CLO could be problematic in the current environment.

Looking ahead, there is pressure on senior CLO tranches to tighten and for LCDX tranches to widen. The analysts warn that the already-high implied correlation at the equity level could therefore move even higher.

CS

Cat bond bounce?

New US legislation could drive issuance volumes

Proposed legislation that would help US states to manage their catastrophe risk and facilitate the issuance of catastrophe bonds moved a step closer last week. While the move may generate much-needed diversification for structured credit investors, not all insurance market participants are convinced about it.

HR 3355, the Homeowners' Defense Act of 2007, cleared the US House Financial Services Committee on a 36-27 vote last Wednesday. The measure would create a lending facility for states that experience massive storms or earthquakes and would allow states or regions to aggregate risk and sell catastrophe bonds to help finance coverage for a portion of the losses that would result from a catastrophic event.

However, HR 3355 has received something of a mixed response. "Massive natural catastrophes, by definition, are so devastating that they require more than a lending facility and bonding mechanism, according to ProtectingAmerica.org co-chair Admiral James Loy (USCG retired), the former commandant of the US Coast Guard and former deputy secretary of the Department of Homeland Security.

"Adequate protection from massive natural catastrophes requires before-the-fact mitigation, homeowner education and preparation, real-time resources for first responders and the accumulation of substantial private funds to assure that affected families can repair, rebuild and recover after catastrophe strikes," Loy says.

"Congress has coalesced around the need to address the urgency for improvements to our catastrophe response programmes. But adopting legislation that is less than an integrated and comprehensive catastrophe programme presents the possibility that we will be lulled into a false sense of security," he adds.

Equally, Justin Roth, the National Association of Mutual Insurance Companies' (NAMIC) senior federal affairs director, comments: "We are concerned that this legislation would expand the federal government's role to the point that it could potentially crowd out the private insurance market. It could also encourage unwise residential and commercial development in high-risk coastal regions."

The legislation would create a non-governmental entity – the National Catastrophe Risk Consortium – through which participating states would be able to pool their resources to purchase reinsurance. "The bill leaves unanswered the question of how this arrangement would compete with the private reinsurance market and who would bear liability from the consortium's actions, since the bill indicates the federal government would assume no liability," Roth says.

Roth has urged the House to instead consider a recent bill passed by a Senate panel. "The legislation passed in the Senate Banking Committee would establish a bi-partisan commission to look into and make recommendations on natural disaster issues," he concludes.

Meanwhile, the Florida Hurricane Catastrophe Fund (FHCF) – which is currently the only fund of its kind – has made its first foray of 2007 into the capital markets with a US$3.5bn issue. FHCF Finance Corporation Floating Rate Notes, Series 2007A, were issued to provide the FHCF with pre-event liquidity and were met with strong investor demand to ensure pricing within the issuer's guidelines.

The US$3.5bn five-year, non-call one notes were priced with a coupon of 78bp over Libor through leads Citi, Goldman Sachs, Merrill Lynch and Morgan Stanley. The notes were rated Aa3/AA-/AA- by Moody's, S&P and Fitch.

MP

Structured credit hedge funds continue slide

Latest index figures show a further drop in returns, but not for all sub-strategies

Both gross and net monthly returns for August 2007 in the Palomar Structured Credit Hedge Fund (SC HF) Index dipped sharply. The latest figures for the index were released this week and show a gross return of -2.5% for August, while the net return was -2.6% for the month. Eleven of the 40 funds were able to report positive results, however.

Despite the continuing financial market turmoil in August, most structured credit hedge funds reported better figures than in July. Once again, the best performing sub-strategy was 'short biased,' with an average return of 9%. No sub-strategy returned less than -7%, on average.

The worst performing funds in July were among the worst performers in August, including two funds that were liquidated. Otherwise, the dispersion of returns was lower than in July.

The bottom quartile returned, on average, -26% (-13.5%, weighted by AUM), while the top quartile returned on average +2% (+4.5%, weighted by AUM). For more Index data click here.

The objective of the Palomar SC HF Index is to produce an index that represents the risk and return of investable hedge fund investments in the structured credit area. The index aims to provide a monthly measure of the performance of the universe of open, investable structured credit hedge funds. The Palomar SC HF Index is calculated in two formats - as gross asset value and as net asset value.

The Palomar SC HF Index is compiled and run by Palomar Capital Advisors and published exclusively by Structured Credit Investor. Palomar Capital Advisors is a financial advisory firm specialising in structuring, managing and placing alternative investment products, specifically credit-related securities. It is an independent firm based in Zurich, Switzerland, owned and controlled by its investment professionals.

CS

All is not loss

The latest company and people moves

All is not loss
With banks ranging from Citi to UBS reporting significant structured credit losses it is perhaps unsurprising that headhunters are reporting a slow down in interbank movement. Nevertheless, many say they remain extremely busy and that it is not all about pre- and post-cutback counselling.

"A lot of the banks are doing internal management reviews and budgets for next year, and everyone is wondering: 'If we move early, can we get someone really good at 20% less than market rate?'. Structured credit is still a terribly thin market in terms of people – I can think of half a dozen banks that don't have anyone really decent at a senior structuring level," explains one recruiter.

While most of the interest surrounds the synthetic market – and its offshoots, such as insurance-linked securities and fund structuring – there are understood to be a number of banks considering bold moves in the cash market. The plan now appears to be to staff up and establish a cash business by warehousing loans at a discount and ABS with wider than average spreads, and wait for the market to come back in 12 months.

Overall, long-term structured credit job prospects a far from terrible, according to one head-hunter. "Sure, a lot of the banks have taken huge losses and made some big mistakes. As is becoming increasingly typical, it will regularly emerge that there have been lapses in controls. Something has to be seen to be done – there will be reductions across the board, particularly with the US banks because that is their way. But next year if the market start moving back, there certainly will be plenty of places for these people to go."

D&T promotes
Nathan Abegg was recently promoted to director in Deloitte & Touche's Securitisation Services group. He is based in New York, and serves Deloitte's CDO clients globally, assisting them in their issuance and surveillance activities.

FSF instigates turmoil task force
The Financial Stability Forum (FSF) warns that the turmoil in global financial markets in recent months has raised important concerns that require careful consideration by financial policymakers. Some weaknesses will be addressed through adjustments in the private sector, FSF says, but in other areas authorities will need to prompt or take action.

To help formulate an appropriate and coordinated international response, the FSF will form a working group of the relevant national authorities and chairs of international bodies. The working group, which will be chaired by FSF Chairman Mario Draghi, will analyse the underlying causes of the recent market turbulence and make proposals to enhance market and institutional resilience. It will draw on the expertise and ongoing work of the FSF and its member bodies.

An outline of the work plan will be provided to G7 Finance Ministers and central bank Governors for their October meeting. A final report and recommendations by the FSF will be submitted to G7 Ministers and Governors for their April 2008 meeting.

Crunch cases
Details of two law suits being brought as a result of the credit crunch have emerged.

Prudential Retirement Insurance and Annuity Co has filed a suit against State Street Global Advisors and State Street Bank and Trust Company, seeking recompense for losses Prudential claims were as a result of "highly leveraged" credit investments made by State Street on its behalf. Those investments, Prudential says, were undertaken without its clear approval to do so.

Meanwhile, German Bank IKB is being sued by 40 shareholders who claim that they were deceived about risks related to investments in US sub-prime linked securities. Both the shareholders' cases and Prudential's are expected to be contested.

Moody's appoints senior managers
Moody's Investors Service has announced that Michel Madelain has been named evp, responsible for global fundamental ratings, and Noel Kirnon has been appointed evp responsible for global structured and US public finance ratings. Andy Kimball has been named chief credit officer and chairman of credit policy.

Madelain will assume responsibility for all global fundamental ratings, including corporate finance, banking, insurance, financial guarantors and non-banking finance. Previously, he was group md with responsibility for global banking.

Kirnon will now be responsible for the global structured finance ratings business, including asset finance and derivatives, as well as US public finance ratings. Most recently, he was senior md, Moody's structured finance group, responsible for global derivatives, managed funds and US commercial real estate.

Ethical credit fund launched
F&C Investments has launched the latest part of its strategy to further develop its sustainable investment franchise – the F&C Ethical Bond Fund. The fund, which will be structured as a UCITS 3 open-ended investment company and launch with in excess of £40m of existing assets, will be available to both institutional and retail investors. The fund will leverage the expertise of F&C's 16-strong governance and sustainable investment team to provide research on environmental, social and governance issues.

The fund's universe of eligible securities will be screened on a comprehensive range of both "positive" and "negative" criteria, selected by an independent policy committee. The F&C Ethical Bond Fund will actively target credits from issuers which contribute to sustainable development. These includes companies providing products that are basic necessities of life, such as housing and water, or that help to meet sustainable development challenges like renewable energy, as well as those with good business practices in the environment and human rights.

The fund will also avoid exposure to issuers engaged in areas such as tobacco or alcohol production, gambling, pornography and the weapons industry - as well as those with poor practices towards the environment, human rights or labour relations.

The lead manager of the fund will be Rebecca Seabrook, with Ian Robinson as the alternate manager. Seabrook and Robinson co-head F&C's UK credit team, which manages in excess of £10bn of assets, including a number of screened segregated mandates for institutional investors.

MP

Taiwanese CBO hit by Westways downgrade

A round up of this week's structured credit news

Taiwanese CBO hit by Westways downgrade
Fitch Ratings has downgraded the ratings of the Class C bond issued by E. Sun Bank 2007-2 CBO from triple-B to single-B minus and kept it on rating watch negative (RWN). The move reflects the rating action taken by Fitch on 25 September when the income notes of Westways Funding XI were downgraded to triple-C (SCI issue 57).

The rating of E. Sun's Class C bond is linked to the rating of Westways XI income notes, as the principal repayment is reliant on the principal repayment of these income notes.

One Taiwanese dollar asset, representing 2.7% of the original asset pool, was repaid and the proceeds were used to reduce the size of the class A1 bond. Otherwise, the portfolio quality remains largely unchanged.

Financial names dominate September trading
Credit default swaps for the financial services sector dominated the most active trading in the US during the month of September, according to GFI, led by Countrywide Home Loans, Lehman Brothers, Bear Stearns, SLM Corp and CIT Group. The banking and auto manufacturing sectors rounded out the most active list.

In Europe, the telecom sector dominated again, led by Telefonica, Telecom Italia, Vodafone and Deutsche Telecom, while the banking and food retail and wholesale sectors retained their positions on the most active list.

In Asia, the banking and financial services sectors dominated, led by Kazakhstani banks JSC Kazkommertsbank and Bank TuranAlem, as well as Japan's Aiful, Takefuji and Mizuho Corporate Bank. The fixed line telecom sector entered the most active category.

Meanwhile, Turkey, the Philippines, Brazil and Russia were among the most active sovereigns, with Ukraine replacing Argentina, according to GFI.

The inter-dealer broker defines 'most active' as the sector, reference entity or sovereign with the most trades executed by GFI.

Cov-lites to face high refinancing risk
Covenant-lite structures will face equally high refinancing risk when compared to loans subject to a complete set of financial maintenance covenants, according to Fitch in a new report, especially in the event of a prolonged market downturn. However, covenant-lite structures can impair recovery prospects given default for creditors overall, as they can potentially influence the timing of addressing a borrower's deteriorating financial performance.

"As LBO exits become more problematic due to tighter liquidity conditions, covenant-lite structures will be required to show evidence of excellent operational track record in order to be able to refinance in later years. In the meantime, creditors incur the risk of lower control on such investments," says Pablo Mazzini, director in Fitch's leveraged finance team.

Because Fitch's default definition does not take into account the covenant structure of a loan in its analysis of the Issuer Default Rating (IDR), the agency treats covenant-lite loans and fully covenanted loans equally, and considers the risk of payment default the same. The agency notes, however, that a lack of covenants can indirectly affect the IDRs of borrowers experiencing difficulties if they affect management's freedom to swiftly implement reformative measures and creditors' ability to force a change in strategy.

This is captured in Fitch's Recovery Ratings analysis by way of potentially lower distressed enterprise value available for creditors upon default. However, junior creditors and equity, rather than the covenant-lite senior secured creditors, may in such instances absorb the potentially higher losses.

The lack of measures to keep the borrower on check therefore implies that lenders have no recourse to force management to refocus financial and operational strategies such that the interests of both the lenders and the borrower are aligned.

Fitch notes that the majority of its rated European CLOs is based on asset-specific ratings and will thus not be affected by the presence of covenant-lite loans within the CLOs. However, in the event that Recovery Ratings have not been assigned to such loans (as it tends to be for loans included in US CLOs), a 10% haircut to the standard Fitch VECTOR recovery rate will be applied. This stance may therefore have a more pronounced impact on the overall recovery expectations of US CLO portfolios, depending on the proportion of covenant-lite loan exposure.

Fitch also notes that, while covenant-lite loans are in retreat given the retrenchment in liquidity in the leveraged credit markets, the return of demand for syndicated loans and the type of investors will determine whether covenant-lite loans will experience a comeback or be seen as the pinnacle of the last credit cycle.

Additional maturities for XO
With effect from 6 November 2007, 3-year and 7-year maturities will be introduced for iTraxx Crossover Series 8, in addition to the existing 5-year and 10-year maturity contracts, following a vote by iTraxx Crossover market makers. The new 3y and 7y tenors will have a 20 Dec 10 and 20 Dec 14 maturity respectively.

Though the new maturities may cannibalise some of the 5y activity, the potential for very liquid Crossover curve trades – as well as the fact that the 5y point is the best 'duration' match for hedging cash corporate bonds – should contribute to an overall increase in Crossover traded volumes across all maturities, note analysts at RBS.

Fitch revisits CAM ratings
In light of current credit developments in the RMBS sector and their consequences for structured finance CDOs, Derivative Fitch will perform a focused CDO performance review of all structured finance CDO Asset Managers (CAMs) to which it has assigned a 'CAM' rating. The objective of these reviews is to ensure a consistent framework for evaluating managers' performance at this point in the credit cycle.

Fitch's normal CDO surveillance practices, which include regular dialogue with each manager concerning portfolio performance, will continue to apply for both rated and non-rated CDO asset managers.

The review will commence for the following Fitch-rated Structured Finance CDO Asset Managers:
- BlackRock;
- Cairn Financial Products;
- Collineo Asset Management;
- Declaration Management and Research;
- Duke Funding Management;
- E*TRADE Global Asset Management;
- Faxtor Securities;
- GE Asset Management;
- GSC Partners;
- Gulf International Bank (UK);
- Omicron Investment Management;
- Pacific Investment Management Company;
- Prudential M&G;
- Rabobank;
- Redwood Trust;
- Solent Capital;
- Trainor Wortham;
- Trust Company of the West;
- UNIQA Alternative Investments;
- Vertical Capital, LLC;
- Western Asset.

CDO performance assessments for each manager are expected to be completed over the next 45 days. For managers for whom Fitch has completed a timely on-site review in 2007, the agency will announce updated rating decisions as each manager's performance analysis is completed during this time period. For managers who have not received a timely on-site review in 2007, a rating decision may be postponed pending the completion of this step.

Fitch rates CDO asset managers by asset type, on a scale of 1 to 5, with 1 being the highest rating. The rating scale includes plus and minus designations, along with flat ratings. The ratings are based on a standardised scorecard methodology, which are subsequently validated in committee.

Fitch affirms StaGe Mezzanine
Fitch has affirmed StaGe Mezzanine Societe en Commandite Simple's (StaGe Mezzanine) Class A and Class B notes, and removed the Class B from rating watch negative (RWN). The transaction is a cash securitisation of subordinated debt instruments to German SMEs.

Following the initiation of insolvency proceedings by one of the portfolio companies on 22 January 2007 and a payment 'holiday' granted by StaGe Mezzanine to another portfolio company, the Class B notes were placed on RWN. Subsequently, the insolvency was accepted by the German courts, triggering a principal deficiency ledger (PDL) event.

This triggered a diversion of excess spread to repay the most senior outstanding note. As a result Class A has amortised to 97% of its initial balance as of the June 2007 payment date. Given no further defaults, the PDL balance is expected to be fully paid off by the payment date in June 2008.

Regarding the payment holiday, it is expected to be resolved by the end of the year with the deferred payments scheduled to be made in Q3 2008. According to the financial advisor, Deloitte & Touche Corporate Finance and the recovery manager, Deloitte & Touche Wirtschaftspruefungsgesellschaft, the financial condition of the company placed under payment holiday continues to improve and is not expected to trigger a PDL event.

Two additional payment holidays were granted; the first payment holiday started in Q1 2007 lasting until Q3 2007. The second started in Q2 2007 and will last until Q4 2007.

The combined deferred interest for the June 2007 payment date amounted to €248,323. These events were incorporated into the cashflow model analysis.

Since January 2007, there have not been additional PDL events and the portfolio is performing within expectations. Given the robust levels of excess spread available to support the structure, the current levels of credit enhancement are sufficient to justify the current ratings.

Fundamental credit focus for leveraged finance
There are signs that the European leveraged finance market is focusing more on fundamental credit and will differentiate deals much more on the basis of default risk and recovery prospects than has been the case in recent years, according to a Credit FAQ published by S&P.

"The leveraged finance market is in a holding position right now as market participants wait to see how the first transactions brought to market after the liquidity crunch will fare," says Taron Wade of S&P's leveraged finance and recovery group.

The new leveraged transactions – as well as those already on banks' balance sheets – that have been getting to market are typically midsize deals able to tap local bank liquidity. Arrangers of debt have been exploring all sources of possible liquidity, including mezzanine investors, the asset-backed loan market and the repackaging of loans to high-yield investors. Structural enhancements to transactions are also being considered, such as converting debt to non-cash-pay facilities.

"The market is trying to find a new equilibrium as the balance of power shifts to investors from issuers," Wade adds. "This may take some time as arrangers anticipate that the first few deals out to syndication may suffer unduly from the dislocation, resulting in the need for severe concessions to investors on pricing and structure."

There is now much greater sensitivity to the ratings being assigned and the supporting rationale, which suggests to S&P that fundamental credit analysis will carry far more weight in structuring and selling deals in the future.

The Credit FAQ, entitled 'Considering Investors' Renewed Focus On Default Risk And Recovery Prospects In The European Leveraged Finance Market,' addresses some of the most frequent questions the agency has been receiving in relation to rating activity, credit quality and recovery prospects.

Latest permacap moves
Cambridge Place Investment Management's Caliber vehicle last week reported on its net asset value, while Cheyne Capital's Queen's Walk began its share buyback programme (see also SCI issue 51).

As at the close of business on 31 August 2007, Caliber's unaudited adjusted net asset value per share was US$1.64, against an unaudited adjusted NAV of US$2.94 at 31 July 2007. The fall in the unaudited adjusted NAV reflects further market disruption during August.

The unaudited adjusted NAV excludes those non-recourse special purpose vehicles which contributed negative net assets to the unaudited consolidated NAV as at 31 August 2007. As previously reported, the unaudited consolidated NAV was US$1.61 at 31 July 2007. Under this methodology, the NAV as at 31 August 2007 was US$0.08.

Meanwhile, Queen's Walk Investment Limited has sent a circular to eligible shareholders detailing its proposed tender offer to purchase up to 24.99% of its ordinary shares and thereby to return a maximum of €25m in cash to shareholders. The tender offer closed at 3pm on 27 September 2007 and the strike price for ordinary shares tendered was set at €5.55.

560,742 ordinary shares that were tendered at the strike price, being approximately 66.6% of the ordinary shares that were tendered at such price, were conditionally repurchased pursuant to the tender offer. The total consideration conditionally payable under the tender offer is €24,999,975.

The tender offer is conditional on the approval of shareholders at the extraordinary general meeting of the company to be held at 12.30pm on 8 October 2007. A further announcement will be made after the EGM.

The company has entered into an irrevocable, non-discretionary arrangement with its brokers, Citi and Goldman Sachs, to repurchase on its behalf ordinary shares for cancellation during the close period commencing on 1 October 2007 and ending on or around 27 November 2007. The maximum price to be paid shall be not more than 105% of the average of the middle market prices paid for the company's shares for the five business days before the day on which purchase is made.

Goldman Sachs and Citi will not repurchase on each trading day more than 50% of the average daily trading volume of the company's shares traded over the 20 trading days preceding that date. Yesterday, 2 October, the company confirmed that it had purchased 17,000 ordinary shares at an average price of €5.26 per share.

The highest price and lowest price paid for these shares were €5.27 and €5.25 respectively. The purchased shares will all be cancelled.

Markit and Reuters join forces on CDS data
Reuters and Markit have announced a wide-reaching content distribution deal for their customers. Users of Reuters' premium desktop, 3000 Xtra, will now gain access to Markit's comprehensive range of CDS data, including Markit's new Intraday service. Markit Intraday is a streaming pricing tool for the leading credit indices, iTraxx and CDX, as well as their constituents, which draws from the major CDS market makers globally.

Reuters will also provide mutual customers with Markit's End of Day and Sameday pricing for the full universe of approximately 3,300 credits.

Lance Uggla, ceo of Markit, comments: "Markit's arrangement to distribute credit content via Reuters is unique and a first of its kind. This is the first time that we have allowed our data to be redistributed by a third party, and we chose Reuters as a strategic partner because of their extensive global footprint. We look forward to bringing greater transparency to the credit markets by delivering Markit's benchmark CDS pricing to Reuters' broad client universe."

Credit derivatives top US$45.46trn
ISDA has announced the results of its Mid-Year 2007 Market Survey of privately negotiated derivatives. According to the survey, notional amount outstanding of credit derivatives grew by 32% in the first six months of the year to US$45.46trn from US$34.42trn. The annual growth rate for credit derivatives is 75% from US$26trn at mid-year 2006.

For the purposes of the survey, credit derivatives comprise credit default swaps referencing single names, indexes, baskets and portfolios.

CS

Trading ideas - big MAC

Dave Klein, research analyst at Credit Derivatives Research, looks at a curve steepener trade on SLM Corp.

When we hold a deteriorating fundamental outlook on a name, the positive carry curve steepener is our trade of choice as it pays us to be short the credit. With the current turmoil in the credit markets, we have found this relationship to be a bit tenuous.

One only needs to look at how the brokers performed over the past two months for an example of curves inverting much sooner than expected. However, LBO names have not followed this trend, staying steep even as spreads rise. Anecdotally, the thinking is that LBO names do not invert because default risk is pushed out to the longer end of the curve rather than concentrating on the short end, as happens with operationally-challenged credits.

Today's trade, a curve steepener on SLM Corp. (SLM), is a bet on the completion of the SLM LBO.

How flat is too flat?
With news (or at least speculation) that SLM is in negotiations with its buyers to change the terms of the deal (after the news that the buyers were no longer willing to pay US$60/share and that SLM would put up a court fight), both the company's share price and CDS levels have jumped. However, we still find SLM fairly flat to fair value – although how flat is too flat is difficult to judge these days, with the lack of 10-year liquidity in most names.

A DV01-neutral curve steepener, selling 5-year protection and buying 10-year protection, currently has positive carry and positive rolldown. Exhibit 1 charts the 5s-10s absolute differential over the past year.

After the LBO announcement in April, the curve steepened dramatically. Since then, it has trended flatter.

With SLM trading flatter than it has when previously at these levels, we believe there is a good probability of curve steepening in the short term. If the LBO deal goes through, we further believe there is the potential for further steepening as levels widen.

Risk analysis
This trade is duration-weighted to ensure positive carry, as well as to reduce our exposure to absolute levels. We are therefore hedged against short-term movements in absolute spread levels, profiting only from a curve steepening between the fives and tens.

The carry cushion protects the investor from any short-term mark-to-market losses. This trade has positive roll-down, thanks to the curve shape and tightness of bid-offer.

Exhibit 2 plots the potential six-month P&L of the trade under three different scenarios based on changes to the absolute 5s-10s differential. The three scenarios are: no parallel shift in 5s-10s; a 100bp parallel shift up; and a 100bp parallel shift down. Transaction costs, as well as carry and roll-down are taken into account.

We note that current transaction costs are quite high, much higher than past history would indicate, and expect them to come down as the market settles down. Even so, we'll need about 20bp of widening in six months time to break even on the trade. If the LBO falls apart, we would expect to lose money on the trade.

The doomsday scenario for this trade is that the LBO does not happen and SLM runs into credit problems, causing the curve to invert. Given the company's current guidance, we believe it is more likely that the curve would flatten slightly in the absence of the LBO, limiting our downside.

Entering and exiting any trade in these maturities carries execution risk, but this is not a major risk with this credit in these maturities as they are increasingly liquid.

Liquidity
Liquidity is a major driver of any trade – i.e. the ability to transact effectively across the bid-offer spread in the CDS markets. Our data on liquidity, created from the volume of bids, offers and trades we see each day, provide us with significant comfort in both the ability to enter a trade in SLM and the bid-offer spread costs.

SLM has excellent CDS liquidity, but we're currently seeing 10bp of bid-offer in 5-year and over 20bp of bid-offer in 10-year. We expect the size of the bid-offer to drop, especially in 10-year, as it is far wider than previously seen.

Fundamentals
This trade is significantly impacted by the fundamentals. The technical flatness of the credit and negative economics are helped by a potentially negative systemic outlook which, in our view, overwhelms the stable idiosyncratic performance of the credit.

Kathleen Shanley, Gimme Credit's Home Financial Services expert, maintains a deteriorating outlook for SLM. She notes that SLM's proposed LBO would add financial risk to the company.

Kathleen further notes that, although SLM's buyers do not want to pay the deal price, it is unclear whether there is a lower price that would be acceptable to them. Further, given that SLM claims that the buyers have no contractual basis to cancel the deal, Kathleen concludes it is too soon to assume the deal is off.

Clearly, we have no extra information about whether the LBO will go through or not. What we see is a flat curve, a positive-carry steepener with nice potential upside in the case of an LBO. If the LBO does not happen, we see limited upside and would look to exit the trade as soon as possible.

Summary and trade recommendation
After weeks of rumors, it appears the SLM LBO is under threat. With talk of MAC-clause triggers and deal re-pricing, it is too soon tell whether the deal is truly off. Indeed, today's trade bets that the SLM LBO will go through.

We do not possess a crystal ball, but note that the market seems to be betting in favour of the LBO today more than earlier in the week. If the LBO is completed, a DV01-neutral curve steepener should return a nice profit as default risk is pushed to the longer end of the curve.

Positive carry and positive rolldown aid the economics of the trade. If the LBO falls apart, we expect a quick exit.

Sell US$10m notional SLM Corp. 5- year CDS protection at 232bp.

Buy US$6m notional SLM Corp. 10-year CDS protection at 265bp to gain 73bp of positive carry.

For more information and regular updates on this trade idea go to: http://www.creditresearch.com/.

Copyright © 2007 Credit Derivatives Research LLC. All Rights Reserved.

Note: This article is intended for general information and use, and does not constitute trading advice from Structured Credit Investor (see also terms and conditions, below, section 12).

To square, or not to square?

The stability of CLO-squared tranches is discussed by JP Morgan's structured credit strategy team

It may be somewhat surprising to consider CLO-squareds (CLO2s) in the current market environment when, generally speaking, liquidity is at a premium. However, we write this piece because the combination of historically wide CLO liability spreads and near-zero default rates makes it an optimal time for buy-and-hold investors to efficiently monetise the current illiquidity created in the spread rout of 2007.

With CLO spreads dramatically re-priced, risk-reward has changed. We moved from Overweight to Neutral on March 15 at market tights, followed by a move to Underweight in early July, and more recently, back to Overweight at these multiyear spread wides.

In our view, loan credit fundamentals remain solid, especially given successful investor pushback on issuer-friendly LBO structures. Wide spread levels price in overly severe loss scenarios, especially given a more accommodative Fed.

If current significant illiquidity premia make CLO tranches attractive, the obvious next question is whether CLO2 tranches are even more attractive. Investors may be able to pick-up 50bp to 400bp of spread, from triple-A to double-B, versus already wide spreads in underlying CLOs.

Just as with regular CDOs, the basic CDO2 premise is to lock-in fundamentally cheap assets and monetise illiquidity premia. Indeed, since 1999 opportunistic managers have brought CDO2s to market in various asset classes. Zais brought the first, a repackaging of HY CBOs and CLOs, while RBC, Mariner and Babson have been the next most active managers.

Still, fully distributed CDO2s (as opposed to synthetic single-tranche CSO2s done as correlation trades) have remained a niche sector, due partly to their greater leverage and complexity. While a CLO2 is no doubt more complex than a standard CLO, we point out that conceptually it is more similar to an ABS CDO than an ABS CDO2. In each case the base underlying asset class is loans, repackaged twice: residential mortgage loans in the case of ABS CDOs, and HY corporate borrowers in the case of CLO2s.

To promote transparency in the CLO2 sector and help analyse the leverage/spread trade-off, we model a CLO2 with individual loans and realistic CLO tranche waterfalls, enabling us to capture leverage effects.

Basic terms and parameters of the model
In our CLO2 model, loan-level default simulation drives child CLO and parent CLO2 cashflows. By controlling the bottom-layer collateral performance, we model realistic CLO2 performance, reflecting tail risk from such factors as asset correlation and overlap. Although capital structures vary, our basic assumption is that the child CLOs each hold 160 equally-weighted names and the CLO2 holds 50 mezzanine (triple-B) tranches of the child CLOs.

The universe of assets
The starting point for our analysis is a large universe of N loan assets which are placed into CLOs as underlying collateral. N is typically around 400 to 1000, depending on the degree of overlap desired across the child CLO tranches.

We posit three broad loan types: 30% double-B first lien; 60% single-B first lien; and 10% "other", which generically represents second lien, unsecured HY bonds or covenant-lite assets.

For each loan type we make assumptions regarding:
1. The average default rate
2. The recovery rate
3. The spread over Libor.

The number of buckets in the model could be increased but we find three buckets sufficient for our analysis. We also make a single broad assumption about the prepay rate (25%) and asset correlation.

Assigning assets to CLOs
For each child CLO, we randomly draw 160 loans from the underlying N loans. Since the child CLOs draw from the same set of loans, issuer overlap results when two CLOs draw some of the same names. We control the average overlap via the total number of loans: the larger N is, the lower the overlap.

Note that when we use the term 'overlap' here, we refer to the average pairwise overlap between CLOs. Thus, not every pair of CLOs has the same percent of overlap.

Moreover, the overlapped names vary: the same loans are not shared by every CLO. We feel our modeling of asset overlap is reflective of market conditions.

Default simulation
We assume each loan has some probability of defaulting in every period. Loosely speaking, this is the "hazard" rate.

This probability is the average default rate for that kind of loan. However, the actual default time is random.

To capture the fact that defaults cluster in time, we model correlated firm asset values. This produces correlated default times. The (base case) correlation among the assets of the firms is 12%, roughly consistent with the findings of Moody's.

Because of CDO cashflow complexity, we use a Monte Carlo simulation to go from the theoretical probability distribution of loan default times to probabilities of various events for the CLOs and the CLO2. In a Monte Carlo simulation, several thousand scenarios are run and then probabilities of various events are estimated based on how often each event happens in the scenarios.

Specifically, we draw 10,000 times from the probability distribution for defaults for all the loans. Exhibit 1 graphically illustrates the procedure. A default time beyond the maturity of the loan (quarter 41) corresponds to an event of no default for the purposes of our analysis.

After specifying these 10,000 scenarios, we map the cashflows from the underlying collateral to the child CLOs and finally to the CLO2, according to the respective capital structures and priority of payments (described below).

Capital structures and fees
In both the child CLOs and the CLO2, 2% of the initial collateral value goes to pay structuring, placement, rating agency and other upfront fees. We also model an ongoing senior management fee of 20bp, which is paid before the senior tranche's interest.

We model a combined AAA/AA tranche for simplicity, since neither can PIK (that is, accumulate unpaid interest as principal) without it constituting a default. Theoretical principal impairment and interest disruption rates would be lower for a pure triple-A tranche and slightly higher for a pure double-A tranche than the averages shown in our results.

Our child CLO capital structure, including the OC tests, is indicative of recent CLOs (Exhibit 2). Since few CLO2s have priced, there is no 'indicative' capital structure.

While we feel our CLO2 structure is reasonable, it is quite possible that actual transactions will vary. We do not model OC triggers in the CLO2; this makes our results conservative insofar as CLO2s with OC triggers may have better debt performance than shown in our results.

Impact of spread assumptions
Our overall spread levels are around where CLO transactions were pricing in July; we also assume a US$95 overall purchase price for triple-Bs. This can be interpreted as the CLO2 manager buying some combination of recent vintage CLOs in secondary at a discount and new-issue tranches at par, although format (cash/synthetic) may alter executable levels. For the CLO2 we assume a certain discount to the CLO, although given limited issuance these levels are unclear.

Overall our results are fairly insensitive to spread assumptions; they do not impact the fundamental default /return-of-principal simulation. There is a limited effect on the chance of interest disruptions; principal performance is more sensitive to pricing discount (US$5 on triple-Bs), so we elected to remain conservative in this regard.

Cashflow mechanics
The child CLOs have a five-year reinvestment period and each loan has a maturity of 10 years (40 quarters). This maturity is longer than that of typical leveraged loans.

We make this assumption to avoid introducing new loans during the reinvestment period (and having to draw new default times). Instead, during the reinvestment period, principal proceeds from prepays and default recoveries (that are not required to cure OC trigger failures or interest shortfalls) are invested pro rata in surviving loans. Proceeds recovered from defaulted securities become available two quarters after default.

To simplify computation, in the event of an OC failure, all cash beyond required interest goes to the top tranche, not just enough to cure the OC failure. For example, if the deal fails the single-A OC test, the order of cash flows is AAA/AA tranche coupon, single-A coupon, then AAA/AA principal with no reinvestment regardless of the magnitude of the OC failure.

This is somewhat beneficial to debtholders, more so the higher the prepayment assumption. However we think benefit accrues more to seniors (AAA/AA), as juniors might benefit more from reinvestment and generation of excess spread.
Single-A to double-B tranches are the most likely to be included in upcoming CLO2s.

There is no reinvestment period for the CLO2 (although the child CLO tranches do not frequently prepay or default anyway). Since we assume the CLO2 has no OC triggers, the principal and interest payments are always sequential and any excess interest always goes to the equity tranche.

Throughout we assume that Libor is 5.36%, but our results are not very sensitive to this assumption because all of the assets and liabilities are floating rate.

Do CLO2s have nine lives? One benchmark and eight stress cases
In this section we define a benchmark case and eight additional cases that are meant to correspond to key risk factors. The idea is to illustrate the severity of the scenarios which are required to cause principal losses or interest disruptions, and also to highlight the most consequential risk factors.

Exhibit 3 summarises the underlying collateral assumptions for our benchmark case. Based on the findings in our recent studies of overlap in CLOs, we set the average pairwise overlap in the child CLOs to 25%.

Stress cases
The eight additional cases correspond to:
1. Back-loaded defaults zero in years 0-4 and 3.3% thereafter for an average annual default rate of 2%
2. Higher aggregate default rates (5%)
3. Lower aggregate recovery rates (55%)
4. Other bucket default probability increased to 15% annually from 2.5%
5. Other bucket recovery rate decreased from 40% to 10% to mimic cov-lite loans
6. Increased asset correlation (25% versus 12%)
7. Increased CLO overlap combined with Case 2; that is, higher aggregate default rates
8. Kitchen sink stress test: higher default rates, lower recovery rates and higher correlation in combination (Case 2 + Case 3 + Case 6).

Analysis
Child CLO performance
Overall, the frequency of principal impairment and the level of expected principal loss is negligible for triple-A to single-A CLOs for all the cases we consider, outside of the severe 'kitchen sink' scenario (8). Principal loss is a rare event even for triple-B tranches, implying very low probability of principal impairment for all CLO2 tranches.

The cases where we concentrate risk in certain small buckets (4, 6) or change default timing (1) prove almost negligible to the triple-B CLOs given the amount of subordination cushion. This suggests to us that focus on small CCC/second lien/HY bond/cov-lite buckets – while helpful in terms of holding the line of credit – is really more relevant for junior (double-B) and equity investors who bear more of the residual risk.

It takes broadly worse default (2), recovery (3) or correlation (6) assumptions to materially impact triple-B performance. Since these cases are generally much worse than historic averages, the good performance of triple-B CLOs even in these stress cases is comforting.

The probability of interest disruptions is higher than the probability of principal losses, and extends as high as the single-A tranche. The single-A and triple-B behave similarly (modest PIKing), while the double-B frequently PIKs.

Higher default rates are usually worse than lower recovery rates in this respect because they more quickly deleverage the transaction and reduce interest arbitrage. But in a principal context, they are roughly equally harmful as lower recovery rates.

Thin double-Bs are volatile
Our modeled double-B CLOs do not perform as well. They take large principal losses in stressed scenarios, and can be materially impacted in milder cases where triple-Bs are fairly immune.

For instance, in the benchmark case, the triple-B rate of principal impairment is 0.03% versus 8.68% for double-B. We attribute this to loss volatility, tranche thinness and, to some extent, structural over-simplification in our model. While most double-B CLOs would be able to withstand 2% annual defaults on a deterministic basis, given our default volatility and asset correlation assumptions, tail scenarios that cause double-B losses eventually arise over enough scenarios.

Also, we found that seemingly minor changes in subordination, tranche width, basecase loss expectations or OC levels could materially alter double-B performance. The greater volatility of double-B CLO tranches (versus more stable single-A or triple-B tranches) makes them less suitable for re-leveraging; we draw an analogy to thin triple-B and triple-B minus HEL tranches (often 1-2% wide) repackaged into SF CDOs.

Lastly, while we do not model double-B specific OC tests in order to speed computation, many actual transactions have OC tests, collateral reinvestment triggers or reverse turbos to help protect double-Bs. Any of these options would improve double-B CLO stability versus our modeled performance.

Parent CLO2 performance
In general we find that subordinate tranches of our triple-B CLO2 perform better than subordinate tranches of CLOs, while senior tranches (triple-A through single-A) perform worse. This is perhaps predictable, given the extra layer of leverage/tranching in the CLO2.

In benchmark or lightly stressed cases, while some of the collateral triple-B CLOs may take losses, the CLO2 subordination provides sufficient cushion. Indeed, the double-B CLO2 looks more compelling than a double-B CLO position, behaving more like the triple-B collateral it references.

However, principal losses can extend as high as AAA/AA when aggregate default rates are severely stressed (Cases 2, 7, 8). While probability of loss and expected loss on the CLO2 remains small, the expected loss given default is substantially greater than for the child CLO tranches.

PIK risk to senior CLO2s
Likewise, we find that interest shortfalls, while not a frequent event for CLO2s, happens more often for senior tranches than one would expect. For example, in our Case 7 (5% default rate and 37.5% overlap), the AAA/AA CLO2 tranche experiences an interest shortfall in 1.55% of the scenarios. We note that most of these interest disruptions were ultimately transitory.

The frequency of principal impairment for the senior tranche was 0.08% in the same case, indicating that interest (plus interest on interest) was largely repaid by transaction end. This meshes with historical performance observed for early-vintage triple-B and double-B CLOs.

Actual CLO2s may further alleviate this risk through CLO2 OC tests and asset PIK swaps (which we do not model). For the former to be effective, PIKing collateral tranches would need to be counted as defaulted, and material excess spread exist in the transaction.

With an asset PIK swap, typically a counterparty agrees to reimburse the CDO trust for unpaid interest in the event an underlying child CLO tranche PIKs, for a specified amount of time, usually two years. This costs the CLO2 a small running premium, but helps smooth the interest flow. Such a swap is often required by the rating agencies.

Ultimately we recommend that investors carefully consider the excess spread available in marketed transactions; levels may vary widely in the current environment and depending on the manager's particular strategy.

Overlap: not that important
While overlap is a key concern for market participants, we do not find it that relevant. We increased child CLO overlap from our benchmark 25% to 37.5% and it had little impact when we kept the average default rate at 2%. Even when combined with higher default rates in Case 7, higher overlap increases CLO2 principal impairment only minimally; stressed default rates cause most of the damage.

In our opinion, the subordination in CLOs protects well against idiosyncratic risk and it is asset correlation (affecting the 60% to 80% of non-overlapped assets) which is a more important variable. This suggests that overall name and industry diversification should be more of a focus item.

We make a comparison to early vintage IG CDOs which suffered from high defaults among overlapped names (large-market cap names Enron, WorldCom, Global Crossing, Tyco, Adelphia etc). These transactions suffered because the overlapped names turned out to be the riskiest ones, and on top of that subordination was thin.

In our opinion CLOs compare well since they are structurally well-enhanced and prior research indicates overlap amongst the riskiest buckets is modest. Still, keeping average overlap below 25% to 30% is probably a useful insurance policy and certainly achievable.

© 2007 JP Morgan. This Research Note is an excerpt from 'To Square, or not to Square?', which was first published by JP Morgan on 25 September 2007.