Contents

Cash no longer king

Slew of synthetic single-tranche structures expected

Dealers report growing appreciation among investors for synthetic single-tranche CDOs. With the supply of cash CDOs unlikely to meet growing demand next year, synthetic structures are expected to fill the gap.

"The cash market has dropped off significantly. By end-2008 we expect 75% of the CDO sector to be traded via synthetic single tranches," says one dealer.

CLOs, for example, are still sound from a credit perspective and appetite for loan product is expected to return to the same levels seen in the first half of this year – but the supply won't be there, given limited warehouse capacity at investment banks. Consequently, says Tony Venutolo, global head of synthetic credit structuring at SG CIB, there is a significant opportunity for dealers to modify the CLO asset class to include single-tranche synthetic loan deals.

"A number of dealers launched ST CLOs at the beginning of the year with limited success because of tight CDS spreads and the prevalence of rating agency arbitrage," he adds. "But now conditions are better for this type of product and we expect activity to increase in synthetic loans going into 2008. Investors were so used to the spread in cash CLOs that they didn't see the need for synthetic instruments, but the new supply and demand dynamics will change this."

In addition, the concept of correlation and its pricing intricacies has historically been a challenge for potential investors in synthetic CDOs. But, given the opacity of CDO of ABS valuations that has transpired during the sub-prime crisis, there is growing recognition that correlation allows valuations to be explained and sensitivities to be measured in an objective manner.

The route of synthetic structures should also appeal to cash CDO desks because they all but remove warehousing risk. Sourcing the risk synthetically can be reduced to hours, days or weeks at worst, down from months of cash asset gathering.

Hedging spread and correlation risk remains an issue for dealers – mark-to-market changes can be larger compared to cash assets. Yet these risks can be managed much more efficiently because the standard tranche market is that much more transparent.

Moreover, the mark-to-market discipline that synthetic product entails is healthy for the market as a whole, according to Venutolo. "Problems associated with the cash sector – illiquidity in terms of pricing – can quickly be overcome with synthetic instruments. Dealers are incentivised to make markets in the underlying index tranches because of their own hedging activity," he notes.

Transparency and volatility can be seen as the two sides of the same coin, dealers say. While the transparency of correlation and standardised tranches meant that it was essentially the only structured credit market that remained open during the summer credit crunch, CDS spreads have been fluctuating faster and in a wider range than before.

"Spread volatility has provided good opportunities for nimble investors: the blow-outs over the past few months have generated some interest in highly rated bespoke CSO tranches offering very attractive spreads," confirms Gregorios Venizelos, structured credit strategist at RBS in London. "Although it has been difficult to print big-ticket deals as investors remain very cautious, marking these deals to market has been fairly straightforward – trading in the corporate correlation market happens hour-by-hour and its transparency means that you can practically price at any time because the marks are continuously in front of you."

However, market infrastructure needs to improve further for synthetic CDO structures to reach an even broader audience. Pricing systems have to become speedier and more widely available to investors.

A change in the way correlation desks operate could also be on the cards, according to Venizelos. "Correlation desks could look to print opportunistically by picking the peaks in spread and correlation value, in order to fulfill the requirements of their clients. This would entail managing the associated operational risks – the fact that value opportunities may come and go out of sync with client demand," he concludes.

CS

Gloomy forecast

Second SCI annual compensation survey shows negative expectations

Respondents to SCI's 2007 compensation survey paint a bleak picture. Nearly all of the more than 30 headhunters polled declined to put figures to expectations for the forthcoming bonus round, on the basis that averages are unrealistic given the range of anticipated pay-outs for structured credit professionals this year.

All survey respondents agreed, however, that good news would be rare. Of those headhunters offering numbers, even the most optimistic forecast average bonus declines against last year's survey (see SCI issue 18). Across both structured and flow disciplines, drops of 15% for the best traders and 90% for the hardest hit salespeople are suggested.

The actual amounts that individuals receive will depend on their institution. Each firm appears to be dealing with staff compensation in a different way, based on how they strategically view their CDO and broader structured credit businesses going forward.

The range of bonuses awarded this year could be huge, confirms a consultant at Watmough Mallett. "Banks which have involved themselves heavily in correlation tranche trading or which have derived a lot of their revenue from CDO/ABS sales will do far less well than those who have sold primarily synthetic bespoke credit."

She adds: "At the latter houses, some banks are still recording great numbers for Q4 2007 and bonuses are expected to be commensurately high. Moreover, all banks will try to look after their top performers as well as they can, while those salespeople whose P&L was below budget can expect a much rougher ride."

With both clients and candidates still at a loss as to when investor demand and liquidity will return, it is even fair to say that some banks will not be paying bonuses at all and that the primary concern of individuals at the moment is keeping their job. It is certainly extremely difficult to arrive at figures that are representative of average compensation packets.

As Alan O'Farrell, consultant at Kinsey Allen, notes: "There have been significant losers, less significant losers and – by some feat of magic – the odd opportunistic winner among banks involved in structured credit. In some cases, businesses have been closed due to losses, or due to never achieving critical mass. In other cases, teams have rationalised their broader execution capability and kept faith with the existing leadership. In other cases still, non-guaranteed and rather expensive desk heads have been rationalised, as they are felt to have suffered reputation damage – with bright vps being asked to step up as part of a wider more integrated structured credit group."

Remuneration will consequently be both politically and economically motivated, and will likely be used as a tool to rationalise staff even further – particularly as cuts in Europe have not been as heavy as many feel the outlook for 2008 now requires. "In some cases the wider bank will carry the burden, while in others the teams will be held more directly accountable," continues O'Farrell. "Where desk heads have been retained, their cash packages will likely be much reduced in line with their business revenues – with potential upside through shorter deferred stock options being used by some banks, should they turn things around in 2008 and 2009."

Worse still, institutions may begin second-guessing each other as to what the market as a whole will do. Some institutions are thought to have been holding informal candidate conversations with this in mind.

Judging from the calibre of people let go so far, it appears that some institutions have decided that the opportunity cost of not being in the structured credit business is lower than originally thought. Remuneration levels for 2007 are therefore expected to reflect this realisation.

CS

Index overkill?

Value of new CDS indices questioned

S&P is prepping three US CDS indices, one of which – with constituent weightings based on float-adjusted equity market capitalisation – will be a first for the market. But, while some observers believe they may challenge the dominance of iTraxx/CDX, others question the value of creating another set of IG/HY CDS indices.

Slated for launch in Q108, the indices comprise the S&P US Investment Grade CDS Index (consisting of 100 equally-weighted investment grade corporate credits), the S&P US High Yield CDS Index (80 equally-weighted high yield corporate credits) and the S&P 100 CDS Index (80-90 members of the S&P 100 Equity Index that have sufficient liquidity). The weight of each constituent in the S&P 100 CDS Index will correspond to its weight in the S&P 100 Equity Index.

S&P says that, since it is designed to track the S&P 100 Equity Index, the index will be useful for monitoring the relationship between CDS spreads and equity prices. The equity returns of the S&P 100 with CDS components correlate 0.99 with the S&P 100 Equity Index for the last two years ending October 2007.

However, Bradley Golding, md at Christofferson, Robb & Co, suggests that the creation of industry-specific CDS indices (based on financials, industrials, autos and so on) may be more useful than a new IG/HY CDS index. "Industry-specific indices would enable investors to buy a basket of names rather than a number of single names – thereby allowing them to go long/short a sector, which isn't easily done unless you are a customer of a big house."

The agency plans to offer two types of indices: one which incorporates credit events into its pricing; and another which effectively removes the affected credit when a credit event occurs. The indices have a five-year maturity and will roll every six months to maintain sufficient liquidity and reflect changes in the credit market.

CMA DataVision will be used as the primary source of pricing for the indices. The firm draws on 30 buy-side contributions for its data.

"The S&P US CDS Indices are designed to be liquid and efficient enough to support investment products such as index funds, index portfolios and derivatives," says David Blitzer, md and chairman of S&P's Index Committee. "We expect a good amount of interest in these indices by both institutional investors and dealers alike."

At inception, the IG CDS Index and the HY CDS Index will each receive a fixed annual coupon which is represented by the weighted average spread of the associated basket of CDS. Each day, the coupon, the US dollar swap curve and the hazard curve implied by the latest aggregate spreads of the underlying CDS will be used to create a risk-adjusted present value for the index. In the case of a credit event, the reference entity in question is either removed from index calculations or the resulting contingent payment is incorporated into index calculations.

S&P couldn't provide any further details about the indices – such as whether they'll be exchange-traded or the extent of dealer involvement in their development – by press time. However, there is speculation that the agency will expand coverage to Europe in the future.

There is nevertheless an argument for S&P performing a public service with the launch of its CDS indices, given that smaller investors complain of exclusion from the iTraxx and CDX indices. One portfolio manager wonders whether the move may put pressure on Markit and the banks to licence and support a future or other more readily tradable product.

"Why isn't there any support for a futures contract?" he asks. "Due to contracting bid/offer spreads, dealers aren't making that much money on the iTraxx/CDX indices anymore. But if they were democratised and a futures contract was established, dealers could make money by being on the other side of bespoke exchange-traded transactions."

CS

Local trends

Opportunities emerge in European sovereign CDS

Sovereign CDS spreads have tended to diverge from bond spreads during the current credit crisis. A new study shows that both Spanish and German CDS levels look relatively high, compared to bond spreads, and may offer trading opportunities going into 2008.

The average spread between Germany and other major Euro-zone sovereign bonds was highly correlated to both the iTraxx Crossover index (at 93%) and the UBS Risk Index (91%) over the last year, according to a study prepared by analysts at UBS. Key to their forecast for Euro-sovereign spreads is that financial risk will remain at least at current levels, and possibly rise further as leverage is progressively curtailed.

The analysts cite a recent IMF paper which shows that, while the effect of credit crises (the collapse of LTCM, the IT crash and 9/11) on short-end spreads was temporary, the impact on longer-term credit spreads tended to be progressive and long-lived – extending at least a year after the initial event. "We believe the current crisis will cause a similar long-term effect affect on Euro-sovereign spreads," explains Meyrick Chapman, head of fixed income derivatives strategy at UBS. "We would add that an important difference of the current situation from previous episodes is the prolonged dislocation in short-term spreads. This effect in itself is likely to lead to wider 10-year spreads as credit differentiation at the short-end is transmitted through the curve. In addition, spreads entered the crisis at tighter levels than during previous episodes."

Sovereign spreads are assumed to remain correlated with risk appetite and credit indicators, such as iTraxx XO – and also that the spread widening seen so far is a secular shift in valuation which will remain a feature of the market for the foreseeable future. It is also assumed that sovereign spreads will eventually move out to reflect at least the average recent highs of the XO and UBS Risk Index. Based on coefficient levels from the last twelve months, these assumptions indicate a spread widening for average Euro-sovereign spreads of around 8bp relative to German benchmark 10-year bonds.

Local risk factors are also expected to emerge as the drivers for spread widening. Principal component analysis shows that Spain and Portugal have been driven almost exclusively by credit concerns in the last twelve months, with these bonds following the evolution of the XO index more closely than other sovereigns.

"We interpret the current credit sensitivity of Spanish and Portuguese bonds to specific concerns about housing exposure in those countries. On the other hand, Italian bonds have shown much less direct sensitivity to iTraxx XO, although these bonds too have underperformed Germany," notes Chapman.

Six factors are likely to dominate spreads between sovereign issuers in 2008: current fiscal position, outstanding debt/GDP, housing debt, house prices, current account position and non-resident holdings of debt. Investors will remain sensitive to debt and housing risk exposure, as well as the risk any slowdown poses for traditional fiscal positioning of sovereign issuers.

Germany, Finland and Austria emerge in the study as the strongest credits, while potential spread vulnerability is revealed for a number of issuers previously regarded as 'core' issuers. In particular, Ireland, Belgium and France are flagged as at risk of spread widening, with scores similar to those of Portugal.

"There is currently a large divergence between the asset-swap levels and CDS prices of Euro-sovereigns. We believe this is an early indication of increased differentiation of risk between issuers," says Chapman.

Such a divergence appears to reflect a different type of investor group involved in CDS compared to asset-swap markets – notably the high proportion of hedge funds involved in sovereign CDS indicates a rearrangement of the old comparison 'core' versus 'non-core' Euro-sovereign issuers. In addition, the UBS analysts believe that the relative cost of shorting cash bonds has inhibited longer-term adjustment that is likely over the medium term. In particular, the cost of shorting some issues (for instance, Spanish bonds) in repo tends to slow the adjustment of spreads.

Chapman says that French, Belgian, Irish and Dutch bonds are currently expensive in asset swap terms, while Belgian CDS are cheap and Spanish and German CDS expensive. "The biggest anomaly is Italian BTPs, which look cheap in asset swap terms, particularly relative to CDS," he concludes. "We would caution that, although BTPs look attractive on a relative basis, our overall concern for a continued widening of spreads means we prefer to adopt a strategy of selling overvalued bonds."

CS

Citadel buys E-Trade's ABS...

The latest company and people moves

Citadel buys E-Trade's ABS...
In a US$2.5bn transaction led by affiliates of Citadel Investment Group, E-Trade financial corporation is selling off its ABS portfolio. The investment fortifies the company's balance sheet, allows the company to focus on its core retail business and provides additional capital to manage credit risk, E-Trade says.

The terms of the Citadel transaction include:

• E-Trade will receive US$1.6bn of capital in exchange for 12.5% senior unsecured notes and common stock. This includes a contribution of capital by investment funds managed by BlackRock.
• Citadel has acquired E-Trade's entire ABS portfolio, including CDOs, for US$800m in cash.
• Upon final closing, it is expected that Citadel will invest an additional US$150m in exchange for 12.5% senior unsecured notes and common stock.

As a result of the sale of the ABS portfolio, E-Trade will take a charge of US$2.2bn. E-Trade also expects to take a provision in the fourth quarter related to its portfolio of home equity loans in excess of the quarter's expected losses that will result in an ending allowance of over US$400m.

E-Trade also announced that, effective immediately, Jarrett Lilien has been named acting ceo, succeeding Mitchell Caplan, who has stepped down from the position. Caplan will serve as an advisor to the company on transition matters through the end of the year. Lilien has been E-Trade Financial's president and coo, leading the retail business, since 2003.

...and raises MTM questions
The figures involved in the E-Trade sale have quite a few significant implications, according to the European credit portfolio strategy team at BNP Paribas.

"Most investors would be stunned by the 73% haircut on the portfolio, signifying a severely distressed asset. We believe Citadel has bought this portfolio estimating a 30% return per annum for the next three years, which is in line with the historical returns produced by the hedge fund and that would signify not a 73% loss severity on sub-prime but a 45% loss severity (Citadel would approximately double its investment on 27% in three years to 55% at a 30% return per annum) on sub-prime," the strategists say.

If this is a true mark-to-market event, it raises two critical questions regarding sub-prime, BNP Paribas suggests. "Firstly, should investors now be revising up their loss estimates from US$250bn to US$500-600bn and secondly, will this sale be considered a legitimate marker for investors to mark-to-market their sub-prime assets (Level 2) rather than mark-to-model or mark-to-management (Level 3) at the end of 2007 for their audited yearly accounts?"

The strategists continue: "A simplistic analysis done by another investment bank suggests that, using this mark-to-market price, Merrill Lynch would have to a take a further hit of US$9bn, while Citigroup would have to take a hit of US$26bn."

Bowles joins RZB
Raiffeisen Zentralbank (RZB) has appointed Mark Bowles as general manager of its London branch. He joins from Bayerische Hypo und Vereinsbank London (part of the UniCredit Group), where he was head of the fixed income investment portfolios group.

Patrick Butler, RZB's managing board member responsible for treasury and investment banking, says: "Mark is an all-rounder with more than 20 years' experience in the City of London. He has been instrumental in negotiating, structuring and managing securitisations of credit portfolios, and planning and executing a number of growth strategies. His strong background in asset-backed investment will be highly beneficial for the development of RZB's London branch."

Markit agrees to acquire SwapsWire
Markit has agreed in principle to buy SwapsWire, the electronic trade confirmation network. The acquisition is expected to complete in early 2008, subject to agreement of definitive documents, due diligence, shareholder and regulatory approval.

Markit plans to combine the SwapsWire confirmation capabilities with its trade processing workflow platform to provide the OTC derivative markets with a cross-asset trade processing solution with critical mass and a global network. The enhanced platform will have over 200 buy-side institutions, 50 dealers and 45 inter-dealer brokers as clients. The combined business will be co-headed by Jeff Gooch, evp and head of trade processing and valuations at Markit, and Chip Carver, ceo of SwapsWire.

David Lown, evp at PIMCO, says: "As a client of Markit's comprehensive workflow service, we believe this deal will be of immense value to the buy-side community. The integration of SwapsWire's real-time confirmation tool into the Markit service will give us a cross-asset platform with electronic straight-through-processing, allowing us to support our growing OTC derivative businesses with greater operational efficiency."

MP

Moody's completes SIV capital note review

A round up of this week's structured credit news

Moody's completes SIV capital note review
Moody's has confirmed, downgraded or placed on review for possible downgrade the ratings of 79 debt programmes (with a total nominal amount of approximately US$130bn) from across 20 SIVs. The move follows the completion of its review of capital notes which began on 7 November.

The significant additional deterioration in market value of assets across the SIV sector observed since 7 November has resulted in the expansion of Moody's original review to include the senior debt ratings of some vehicles. The agency will continue to closely monitor SIV ratings, taking actions on individual vehicles as warranted. In its monitoring of the ratings, Moody's pays particular attention to the evolving liquidity situation of each vehicle, changes in portfolio market value and the vehicle's prospects for restructuring.

In recent weeks, Moody's has observed material declines in market value across most asset classes in SIV portfolios. These asset classes include financials (which represent on average 38% of SIV portfolios), ABS (16%), CDOs (12%, including CDOs of ABS at 1.4%). Financial institutions debt suffered an average price decline of 1.6% from 19 October to 23 November, ABS 0.7%, CDOs (excluding CDOs of ABS) 0.5% and CDOs of ABS 22%. Furthermore, the continued inability to issue ABCP or MTNs causes mark-to-market losses to be realised when assets are liquidated to meet maturing ABCP and MTNs.

The restructured vehicles affected by Moody's rating actions include: Carrera Capital Finance (whose senior notes are now fully protected against market-value risk by committed liquidity provided by HSH Nordbank); and Harrier Finance and Kestrel Funding (which have been provided with a funding facility that permits the redemption of senior debt as it falls due).

Vehicles with ratings downgraded or placed on watch negative include: Asscher Finance; Beta Finance Corporation; Centauri Corporation; Cheyne Finance; Cullinan Finance; Dorada Corporation; Five Finance Corporation; Hudson-Thames Capital; Links Finance Corporation; Nightingale Finance; Orion Finance Corporation; Premier Asset Collateralised Entity (PACE); Sedna Finance Corporation; Tango Finance; Victoria Finance; Whistlejacket Capital; White Pine Corporation; and Zela Finance Corporation.

In addition, Fitch Ratings downgraded the second priority senior notes issued from the MTN programmes of Sedna Finance to triple-C on watch negative from single-A.

Fitch reports on CDO EODs
Unprecedented negative performance in both the US sub-prime RMBS and SF CDO sectors will lead to an increasing likelihood of CDO event of defaults (EODs), according to Derivative Fitch in a new report (see also this week's Research Notes for more on EODs).

Derivative Fitch has received two EOD notices to date, with the number likely to increase. SF CDOs and CDO-squared transactions originated in 2006 and this year are the most vulnerable to EODs, according to Fitch senior director Beth Nugent, although EOD risk does not stop with these vintages.

"The rights and remedies of an event of default are unique to each transaction; therefore, investors should properly examine and interpret the documents of their investments in order to fully understand the effects of an event of default," she says. "Better understanding of EOD documentation is especially important at this juncture. Should US sub-prime RMBS performance continue to deteriorate, Derivative Fitch expects to see further collateral downgrades that could put more SF CDOs at risk of triggering event of defaults."

In an EOD, the controlling class on a transaction can vote to either accelerate payments to noteholders or to liquidate assets. Both scenarios are accounted for in Derivative Fitch's CDO analysis. For transactions where an EOD has been triggered or is imminent, the agency assumes that the controlling class will vote to accelerate note payment.

Fitch will model the priority of payments that is affected upon an EOD acceleration in its cashflow analysis. This approach may yield a more favourable credit opinion of the senior notes if the priority of payments directs all interest and principal proceeds to redeem these senior notes. However, if the controlling class does not vote to accelerate or votes to liquidate the transaction, the agency will revise its assumptions, which may result in a rating adjustment.

Freeze for sub-prime interest rates?
The US Treasury secretary has, together with a few banks, embarked on a new project called the 'Hope Now Alliance', which seeks to temporarily (for as long as five to seven years) freeze sub-prime interest rates at current levels. Financial institutions will likely set the criteria that divide sub-prime borrowers into three groups: those who can continue to make their payments even if rates rise; those who cannot afford their mortgages even if rates stay steady; and those who could keep their homes if the maturity date of their mortgages were extended or the interest rates remained at the teaser rates. Only the third group would be eligible for help under the plan.

The US Treasury secretary rightly acknowledges that the loan servicer does not have the legal right to modify loans, some of which are distributed globally, and that the right to loan modification lies squarely with the investor. In the case of sub-prime and Alt-A mortgages, US$700bn lie with the banks as whole loans, while the other US$1.5bn lie packaged as CDOs.

According to analysts at BNP Paribas, the first hurdle of the plan is therefore one of legality, while the secondary issue is how to modify loans packaged within CDOs. By modifying the loans and keeping rates constant at a lower level, the holders of lower rated tranches could be starved of any cashflow, driving the deals rapidly into default. Loan modification would also alter the terms of default and it is likely that senior tranches would loose their credit enhancement.

Thirdly, say the analysts, it is likely that a borrower falling under group one may prefer to qualify under group three and play the system to their advantage. Perversely, this may encourage more borrowers to delay their interest and principal payments, in the hope that the banks will offer better refinancing terms. How this lower interest rate gets accounted for on bank balance sheets from an accounting perspective is also debatable, as the loss in income is almost non-recoverable.

From a long-term perspective, the scheme could forestall foreclosures, such that impaired assets do not keep increasing on bank balance sheets. By not allowing foreclosures, the banks may inadvertently put a temporary floor under house prices.

Moody's reports on ABX
In the first in a series of reports that will examine the credit drivers of the home equity ABX index, Moody's says that the performance of the index closely tracks the credit performance of the larger universe of sub-prime mortgage deals. Summarising ABX credit performance through October 2007, the agency says that there has been a continued rise in serious delinquencies across the three 2006 and 2007 series of the index examined in the report – and more are likely on the way.

"Interest rate resets in an environment with limited refinancing opportunities might lead to a steeper jump in delinquencies in the coming months," says Moody's analyst Karandeep Bains, author of the report.

At the same time, the tighter lending conditions and a scarcity of credit that have limited the ability of borrowers to refinance their loans have led to a slowdown in prepayments. For now, cumulative loss levels on the loan pools underlying the ABX deals continue to be low despite a large pipeline of delinquencies, says Moody's.

However, a growing number of deals are falling below their overcollateralisation targets as losses start to materialise. Summarising rating actions on the deals that underlie these indices, Moody's notes it has not taken any negative rating actions on any of the Aaa-rated tranches in the index.

But the number of rating downgrades increases not only for the lower rating levels within each index, but also at the same rating level for each successive version of the ABX index. For example, only 10% of the Aa-rated tranches backing the 2006-2 ABX index are on review for downgrade, as opposed to 65% of the tranches backing the 2007-1 index.

As for loan pool credit characteristics that may have an impact on the performance of the deals that underlie the ABX, Moody's finds little correlation between either FICO scores or LTV ratios and the level of future delinquencies in the pools. By contrast, there is stronger correlation between deal delinquencies and the percentage of borrowers in the deal with full documentation loans, as well as between delinquencies and the loan originator.

Rating action for REIT CDOs
Moody's has placed or left on review for downgrade notes issued by 11 REIT-related CDOs. The rating actions were prompted by severe and continued credit deterioration in the residential mortgage REIT and homebuilder sectors.

The affected CDOs (Taberna Preferred Funding II through to VII, Attentus CDO I through to III, Kodiak CDO I and Trapeza CDO X) have significant exposure to these sectors, ranging from approximately 25% to 50% of their aggregate portfolio balances. Moody's rating actions also reflect liquidity concerns for some of the issuers, primarily residential mortgage REITs, whose securities are contained in the portfolios underlying these CDOs, as well as uncertainties over final workout values for defaulted assets in the underlying collateral pool.

The transactions also have exposure to real estate through equity REITs, real estate operating companies (REOCs) and CMBS. This exposure can range from 25% to 50% of the aggregate portfolio balance of the CDOs.

To date, Moody's has not seen any material deterioration in these sectors. Furthermore, the agency currently has not seen any substantial deterioration in the bank and insurance Trups sectors.

Moody's has rated 103 US Trups CDOs. The size of the US Trups CDO market is almost US$55bn, the portfolios of which are comprised of 70% bank assets, 12% insurance assets, and 18% REIT, REOC, and CMBS assets.

Rating actions for SIV-lites
S&P has lowered and retained on credit watch negative its ratings on the CP, Mezz Tier 1 and Mezz Tier 2 notes issued by Sachsen Funding I from A1+ to A3, from triple-B minus to triple-C minus and from single-B minus to triple-C minus respectively. It has also lowered and retained on watch negative its ratings on the Class A2, B1 and B2 notes issued by Duke Funding High Grade II from triple-B to triple-C minus, from triple-B minus to triple-C minus an from single-B minus to triple-C minus.

The rating actions on Sachsen Funding and Duke Funding are based on updated current portfolio values and the increased likelihood of enforcement. If the vehicles were to liquidate all their assets at the current market valuations reported by the manager, it is highly unlikely that all tranches would be repaid in full.

In addition, Moody's has downgraded the Mezz notes issued by Duke Funding High Grade II, as well as those issued by Triaxx Funding High Grade I. The rating action reflects continued deterioration in market value of the portfolio, which consists of prime RMBS with a preponderance of Alt-A mortgages, together with uncertainty about the renewal of existing repurchase facilities that provide liquidity to the vehicle and the potential restructuring of the transaction.

The US$149.4m Class C notes issued by Triaxx Funding account for 55.4% of the underlying pool of a Taiwanese transaction, which has also been subject to Moody's rating action as a result of the downgrade. The Class A1, B1 and B2 notes (worth NTD8.96bn) from E.Sun Bank 2007-1 CBO have in turn been downgraded.

More German mezz SME CLOs hit
Moody's has placed under review for possible downgrade multiple classes of notes from across five German mezzanine SME CLOs after defaults occurred in their underlying portfolios. The move follows the negative watch placements of PREPS 06-1 and 05-2 (see last week's round up).

The rating action affects the Class B notes issued by H.E.A.T Mezzanine I-2005 (which has suffered its second default since inception, totalling 3.6% of the portfolio); Classes A to E of CB MezzCAP Limited Partnership (third default, at 15.5% of the portfolio); the Class Bs of PREPS 2004-2 (second default, at 4.1%); the Class Bs of StaGe Mezzanine (second default, at 9.1%); and Classes A to D of FORCE 2005-1 Limited Partnership (second default and first repayment, at 6.7% of the portfolio). Moody's will actively monitor these deals in the coming weeks.

TriOptima reaches four million
Provider of post-trade processing services TriOptima has announced that more than 15 institutions were reconciling over four million OTC derivative transactions through triResolve, its portfolio reconciliation service. Launched earlier this year, triResolve has seen smaller dealers and hedge funds joining the largest OTC derivative dealers subscribing to TriOptima's ground-breaking service.

"TriResolve offers us the advantage of easy, automated reconciliations against some of our largest counterparties. We are able to identify and resolve discrepancies in both positions and valuations," comments Chris Harris, md of operations at KBC AIM, a global multi-strategy hedge fund manager with over US$1bn in assets under management.

While all triResolve subscribers utilise the automated, web-based service to reconcile with each other on a weekly or bi-weekly basis, some also regularly submit portfolios of 100 or more non-subscribing counterparties as an alternative to manual spread sheet analysis. Portfolios regularly submitted to triResolve typically achieve very high auto-match levels, allowing institutions to focus exclusively on exceptions in their collateral management process.

Collateralisation of OTC derivative portfolios is a major risk mitigation technique for substantially reducing credit risk and managing counterparty credit exposure. Accurate portfolio reconciliation is the foundation for collateral management.

"In periods of market turbulence like we have experienced recently, triResolve provides precise information about the status of counterparty portfolios for easy resolution of collateral management issues," comments Henrik Nilsson, head of business development at TriOptima. "Information from the service allows counterparties to zero in on specific areas of dispute, whether it be trade population or valuation related, quickly and efficiently."

TriResolve transforms portfolio reconciliation into a proactive process. Its unique network community approach facilitates close collaboration between institutions and internally within an institution. Regular, ongoing triResolve reconciliations reduce collateral disputes, operational costs and errors in credit exposure calculation.

Another tender offer for QWIL
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 €15m in cash to shareholders. The tender offer will be open from 4 December to 17 December 2007 and is being made at a price per existing ordinary share in issue of up to €6.30.

The maximum price represents a premium of approximately 35.48% over the middle market closing price of €4.65 per ordinary share on 30 November 2007. The tender offer is conditional on the approval of shareholders at the extraordinary general meeting of the company to be held on 8 January 2008.

CS

Trading ideas: skewed madness

Byron Douglass, senior research analyst at Credit Derivatives Research, looks at a positive carry short on the CDX IG Series 9 five-year 3-7% tranche

Unwinds of leveraged super-senior trades have had two recent effects on the correlation market: super-senior spreads have traded wider; and an outperformance of the lower part of the capital structure. This has caused base correlation to reach new highs and correlation skews to steepen significantly.

We believe the market has taken off a good piece of the outstanding super-senior risk and now is a good time to initiate a short on the CDX five-year 3-7% tranche delta hedged with the index. Both of our tranche fair value models point to this being an overvalued tranche and if we see any sort of reversion to the mean of correlation skews, we will profitably exit the trade.

As this trade has positive convexity, it fits in well with our view that credit market volatility will continue. The trade has 56bp of positive carry to buffer any losses.

Tranche index multiple (TIM) model
A simple, yet effective way to gauge an index tranche's richness/cheapness is by looking at its relationship with the index itself. For our TIM model, we regress the tranche spread on the index level and the index level-squared, allowing for a non-linear fit.

Exhibit 1 shows the relationship of the index and the tranche spread. This model relies on a historical relationship to continue into the future and by no means is the answer to all of our problems.

Exhibit 1

In order to have a long enough time series, we combined both the CDX Series 8 and Series 9 in the regression using the current on-the-run series, enabling us to estimate our model back to March 2007. We feel that there has been a significant regime change in the credit market over the past few months, thus we did not incorporate any other CDX Series into our dataset.

Exhibit 2 shows the time series of the estimated model. We used an expanding window regression, thus giving us out-of-sample predictions. As you can see, the difference between the predicted and actual tranche spreads reverts between -100 and 50bp. Currently we see a mis-pricing of -20bp.

Exhibit 2

Multi-factor tranche (MFT) model
The MFT model is another multiple regression model. However, instead of using the index level as the independent variable, we use proxies for what we believe to be drivers of tranche value.

For the 3-7% tranche, our two factors are the average of the 14 worst credits (measure of tail risk) of the index and the VIX Index level (forward-looking correlation risk). Again, we use expanding window regressions, thus giving us out-of-sample results shown in Exhibit 3.

Exhibit 3

Since our betas are time-varying, we are very careful with settling on specific betas for the predicted tranche spread. We use multiple sets of betas, giving us a full range of predicted values. For this tranche, our models show this tranche to be overvalued by between 10bp to 30bp.

All things must come down
The base correlation skew for the five-year 3-7% tranche has seen an unprecedented steepening (Exhibit 4) over the past two months. Though it's hard to gauge when to step in and go against the flow, we think this steepening has been overdone and now is a good time to short the 3-7%.

Exhibit 4

We may experience some mark-to-market volatility if the skew continues to widen or correlation levels drop without a flattening of the skew, but our expectation is for there to be a flattening of the skew over the next three to six months.

Risk analysis
A short five-year 3-7% tranche hedged with the index is long correlation, long convexity and short theta (Exhibit 5).

Exhibit 5

Liquidity
The liquidity of the index and five-year tranches is very good, even during 'illiquid' times, with both having bid-offer spreads of 0.5bp and 3-4bp respectively.

Fundamentals
This trade is not necessarily based on fundamentals, but if we see a continued sell-off or downturn in the economy then defaults may rise and the 3-7% tranche will likely underperform the index hedge.

Summary and trade recommendation
As the credit market has sold off over the past couple of months, there has been an unwinding of super-senior risk causing dealers to hedge their exposures. This has created temporary relative value opportunities as CDX mezz tranches have seen their base correlation skews steepen to extreme levels. We recommend initiating a short on the CDX five-year 3-7% tranche hedged with the index, as both of our fair value models indicate this tranche to be overvalued.

Buy US$10m notional CDX IG Series 9 5Y 3-7% tranche protection at 260bp.

Sell US$40m notional CDX IG Series 9 Index at 79bp to gain 56bp in carry.

For more information and regular updates on this trade idea go to: 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).

Anatomy of a liquidation

Concerns about structured finance CDO liquidations are discussed by JP Morgan's structured credit strategy team

The prospect of wide-scale forced asset sales from a variety of sources – including SIVs, CDOs, dealers and ratings-sensitive investors – weighs down the market and keeps spreads wide. We do not think issuance will resume in earnest until this secondary supply is disposed of or absorbed. Distressed credit investors have raised funds and are waiting on the sidelines for firesale prices; holders of the paper are too busy assessing their valuation, restructuring and liquidation options to contemplate new investments.

Here we discuss the two main concerns relevant in typical SF CDOs – forced sales by ratings-sensitive investors and collateral liquidations, given an event-of-default (EOD).

Forced sales
Many Asian bank and insurance companies are ratings-sensitive, and potentially forced sellers if ratings fall below single-A or triple-B. Likewise, many of the hedge funds active at the mezzanine or super-senior levels bought with leverage and are sensitive to market value declines.

Given unprecedented ratings activity, virtually all junior paper has either already been downgraded or is likely to be in the near future. Exhibit 1 illustrates the extent of ratings stress.

Exhibit 1

We estimate that 10% of 06/07 vintage super-senior paper (US$28bn) and 45% of mezzanine (junior triple-A to triple-B) paper of all vintages (US$66bn) will be forced sale candidates by Q208. The former has probably already occurred (Merrill Lynch, for instance, was reported to have seized super-senior paper as repo collateral from BSAM after its implosion in June). The latter we think is still to come.

That said, given the extent of the sell-off, it is not clear that forced sales make sense in this environment. Distressed funds are focusing on super-senior as the only locus of potential value, and there is little bid for mezzanine debt.

Indeed, in deals facing imminent PIKing or the possibility of liquidation, prices have declined to US$0; the rest are trading as credit IOs. At some point, it makes sense to retain the optionality in mezzanine paper, however far out-of-the-money it is.

Events of default and collateral liquidations
Dealers who hold super-senior debt are perhaps 'motivated' sellers, but not necessarily forced. Some have already hedged a portion of their holdings. In EOD situations they also own control rights and have the option of accelerating payments or even collapsing the deal.

There are many triggers for EODs, but the most pertinent ones are:

- Failure to pay interest to non-PIKable tranches (AAA/AA)
- Triple-A par coverage ratio falling below some threshold (typically 100%).

S&P reports that 31 CDOs have experienced EODs so far, about US$32.5bn, all from the late 2006/early 2007 vintages. They are mostly mezz deals, but include five high grade transactions and four CDO-squareds (see Exhibit 2). All of the deals failed triple-A coverage ratios.

Exhibit 2

Most SF CDOs include ratings- and market value-based collateral par haircuts in their coverage calculations for purposes of cashflow distribution, but some also include these haircuts in EOD language. Strong patterns are apparent across dealers, who often re-used the same legal documents and if they intended to retain the super-senior risk, were motivated to bargain for stricter control rights.

Event of default remedies
When an EOD occurs, usually debt is accelerated (all principal and interest payments are shut off to junior investors). Whether junior triple-A and double-A investors continue to receive interest varies by transaction. There may be some protection for junior investors in the form of market value requirements or voting rights (in this environment, no junior investors would accede to a liquidation).

Liquidation provisions may also vary according to the cause of the EOD. However, if there are no impediments, then the controlling class (usually the senior triple-A investors or in the case of wrapped tranches, the guarantor) can direct a full or partial collateral liquidation, with proceeds distributed in order of priority.

To our knowledge, just two transactions have commenced a liquidation – one of which was rumoured to have been because the acceleration language was weak and not liquidating would have allowed value to leak out to junior investors.

For mezz SF CDOs in Exhibit 2, we estimate super-senior liquidation value at 30%, based on current asset values; for high grade SF super-seniors, 63%; and for CDO-squared, 17% (junior tranches would recover at US$0). To us, whether liquidation actually occurs depends on several factors:

1) who holds the super-senior tranche
2) whether acceleration language is strict
3) whether asset sales suppress prices to the extent that recovery would be even lower than what is currently indicated and
4) whether the controlling class feels there is fundamental value at given collateral levels.

Dealers would be under more pressure to clean up their books and liquidate (especially those with new ceos). Monolines, on the other hand, probably want to delay any losses as long as possible (less capital base to absorb) and are also more inclined to rely on credit outperformance versus ABX marks. By Q208 we estimate that 25% of outstanding mezz SF CDOs will be eligible for EOD-driven liquidation (basically, all of those with ratings-based haircuts in their EOD language), meaning that approximately US$40-50bn of low-rated sub-prime collateral could come to the market.

In high grade SF CDOs and CDO-squareds coverage ratios may cause EODs, but we also think that interest shortfalls will become a problem in transactions with thin excess spread and large SF CDO buckets. The issue is that the vast majority of recent vintage single-A and below SF CDOs will soon begin deferring interest, and even junior triple-As or double-As could defer in an acceleration situation.

In a sampling of 15-20 high grade SF CDOs, we estimate that between 5-25% of collateral could stop paying interest before interest shortfalls hit the senior tranches. Risk would be accentuated in transactions with CP paper that has been put back to the dealer at higher than original spreads or those originated in 2006 when spreads were tight.

The main possible protection – in the near term – would be collateral PIK swaps. Sometimes transactions include a PIK swap counterparty, who for a nominal fee agrees to reimburse any deferral of interest on designated CDOs for a set amount of time, for example, 12-24 months. But if even if a small share of 06/07 vintage high grade CDOs become eligible – say 15% – that would imply another US$30-45bn of higher-rated RMBS, HEL and CDO collateral could come out (again pending the actual decision to liquidate).

Legal quagmire
If there is one safe prediction for 2008, it is that legal teams will be busy. While EODs and credit events in CDO CDS contracts are generally clear, the remedies and floating payments (respectively) are often involved or unclear.

Market participants would be well served (and most are already dedicating their time to this – see SCI issue 62) to review contract language in their CDS contracts and individual CDO exposures. CDS implications are simple in a liquidation, since a failure to pay principal event is triggered – otherwise the long position would have to begin reimbursing interest shortfalls at some capped level. The prospect of implied write-downs also means that CDO CDS notional will begin to decline, triggering pay-as-you-go principal payments.

Timing of losses

Ratings-based PIKing
Losses to junior investors will begin immediately in transactions with triggers. When overcollateralisation (OC) performance triggers fail, excess interest is generally re-directed towards senior investors and withheld from single-A to double-B investors (who instead accrue deferred interest as principal, or pay-in-kind).

Exhibit 3

Already, half to three-quarters of recent-vintage transactions appear to be failing one of their OC tests, while a third to a half are failing their triple-A OC test (see Exhibit 3). The severity of the failures (Exhibit 4) is such that for most deals, OC failure signals the end of cashflows, assuming that eventual HEL performance bears out the ratings downgrades (which we expect).

Exhibit 4

When actual triple-B HEL write-downs begin in 18-24 months or so, OC ratios will begin to reflect realised rather than forecast performance. The rating agencies are gradually progressing up the quality curve in terms of collateral downgrades (meaning earlier vintages and higher ratings are being affected), so OC test failure may be only a matter of time.

Transactions not currently failing performance tests will trade as credit IOs; ironically junior investors in triggerless deals will relatively outperform, until the point at which collateral write-downs and interest shortfalls cause a true interest shortfall in the CDO. Note that in transactions where ratings downgrades have caused EODs, the liquidation decision is only relevant to junior investors to the extent it preserves some minimal option value; acceleration will generally block cashflows regardless.

Super-senior losses
Super-senior losses have the potential to be extremely extended, if one measures to ultimate principal write-down or legal maturity. To illustrate potential loss timing, we ran indicative prepay and loss curves on two sample SF CDOs.

Exhibit 5

Under severe cases for mezz and high grade CDOs (95% and 80-83% super-senior writedown respectively), interest shortfalls begin accumulating in June 2009, and July and October 2012 (see Exhibits 5 and 6). Principal trickled in from mortgage bonds for up to 40 years (although one could conservatively assume clean-up calls or some sort of restructuring that pushes principal maturity up to 10-20 years). Because of the discounting, interest shortfalls dominated the NPV of the total loss term; timing of ultimate principal loss was less significant.

Exhibit 6

CDO buckets are a risk factor to accelerate principal losses. In synthetic transactions, implied write-downs could be assessed on a large portion of the exposure (if this contract option was used).

Also, in either cash or synthetic transactions, there could be EOD-driven liquidations. Lastly, interest deferrals could cause interest shortfalls to hit as high as super-senior in the next 1-2 years.

A potential mitigant for the latter is PIK swaps in which a counterparty guarantees interest on a portion of the collateral bucket for a certain amount of time. Super-senior investors also stand to benefit from early diversion of cashflows, although we estimate this is only able to mitigate losses by 2-5%, given the thickness of the senior tranche and the thin excess spread available for much of 2006 and 2007.

Concluding thoughts
All in all, it appears that we are in for a long slog of work poring through a distressed SF CDO space; we do not anticipate a recovery and stabilisation of the broader CDO market until mid to late-2008 at the earliest. Going forward, the emphasis will be on simple, analysable, 'back-to-basic' structures and assets, where the value-added through securitisation is clear.

Corporate assets (bonds, loans, trust preferreds) will be the CDO focus; certainly tranched real estate-related risk is an untenable investment proposition until home prices adjust sufficiently and some stabilisation is seen. Excessive leverage is out of bounds: even CLO-squareds, which we think are more stable and simple than SF CDOs, face a dubious future given the shortage of triple-A buyers for CLOs themselves.

© 2007 JP Morgan. This Research Note is an excerpt from 'CDO Monitor – 2008 Outlook', which was first published by JP Morgan on 30 November 2007.