Contents

CDR Liquid Index data as at 8 January 2007

Source: Credit Derivatives Research

CDR Liquid Indices
The CDR Liquid indices represent the CDS levels of the most-liquid names in their respective markets and ratings classes. Liquidity is measured by the number of bid-offers a credit receives. Index values are simple averages of on-the-run five year CDS levels.

DR Global Market Depth™
The CDR Global Market Depth Index is a daily measure of how many names are actively traded. Liquidity is measured by the number of bid-offers a credit receives. Index values are counts of the number of names that exceed CDR's Liquidity Floor.

CDR Global Market Activity™
The CDR Global Market Activity Index is a daily measure of activity within the global CDS market. Liquidity is measured by the number of bid-offers a credit receives. Index values are simple averages of total bid-offers of all names that exceed CDR's Liquidity Floor multiplied by CDR's Global Base Liquidity Constant.

Invicta CDPC established

First structured credit investor-backed vehicle launched

The Massachusetts Mutual Life Insurance Company and its subsidiary Babson Capital Management have established Invicta Credit, a new credit derivative products company (CDPC).

The move marks the debut of a CDPC managed by an existing major participant in the structured credit market (Babson Capital), as well as the first such vehicle to be wholly-owned by a major financial institution (MassMutual). AXA Investment Managers and Deutsche Bank announced their plans for a CDPC last month (see SCI issue 18, but the vehicle – NewLands Financial CDPC – is yet to launch.

Invicta says it will specialise in taking on long-term risk with the intention of holding this risk through credit cycles. Initially, the CDPC will focus on corporate credit before expanding into the ABS market later in Q1 2007.

Ian Hawkins, president of Invicta, explains: "We assume credit risk by writing protection via credit default swap contracts, so will be dealing in synthetic investments rather than cash. As will be the case when we move into ABS, we'll be selling protection on CDS referencing individual ABS and tranches referencing pools of either cash or synthetic ABS. Our credit range is double-A and above, and we would expect our credit average to be around triple-A given a portfolio mix of double-A, triple-A and super senior deals."

While Invicta appears similar to other CDPCs in terms of its activity, its structure gives it an advantage, according to Hawkins. "Our credit analysis infrastructure is provided to Invicta through service agreements with Babson Capital. That allows us to take a bottom up view of structured credit in a way that is difficult to achieve if that entire infrastructure has to be created and paid for purely within the CDPC," he says.

Furthermore, Hawkins adds: "The activities that we carry out inside Invicta with dedicated personnel are simply those that are unique to the CDPC rather than the normal investment processes that Babson already manages for other accounts. So, overall we have a structure that can perform well even in a tight credit spread environment. That means we have less pressure than other entities might have to reach for yield targets by extending maturity or moving further down the credit curve."

The arrival of Invicta to join existing CDPCs Primus and Athilon was welcomed by credit market participants, who suggest that the move might bring further such vehicles into the market. There are reportedly 20 companies at various stages of readiness for launch: while not all of them are expected to see the light of day, those that are more advanced – particularly the bank-backed CDPCs – could well be spurred on by Invicta's appearance.

A further positive benefit of the new company's structure was noted by market participants. "The fact that it's not backed by a private equity company, as is the case with Athilon, might be a positive. Private equity firms understandably have one goal – to get a good return on their investment and look to sell at the right moment – whereas Invicta may be in the market for the long term," says one source.

Meanwhile, Primus Guaranty has closed its debut CDO, Primus CLO l. The US$400m deal is primarily composed of leveraged loans and is managed by the company's Primus Asset Management subsidiary and was underwritten by Lehman Brothers.

"The completion of our first collateralised loan obligation marks another step forward in our strategy to leverage our core credit competency, build our asset management business and diversify our revenue base," comments Thomas Jasper, ceo at Primus Guaranty.

MP

Euro CMBX launch moves nearer

Four new CDS indices being prepped

The long-awaited introduction of synthetic indices for the commercial mortgage-backed securities market in Europe moved a step closer this week. The new instruments - which will come under the banner of the European CMBX Index - are intended to further advance the development of secondary market trading and related structured trades based on the indices.

The indices are still only at the proposal stage, with the start of trading still some way off. "Ideally, the target launch date for the European CMBX Index is beginning of Q2 2007; however, this will depend on having all documentation and systems in place," says Rob Ford, md of European ABS Trading at Barclays Capital.

Nevertheless, there appears to be market-wide acceptance for the proposal. "There was an initial group of eight dealers who devised the basic structure. However, once that was finalised, all other dealers in the marketplace were contacted and therefore effectively all dealers have now become involved," explains Ford.

"For practical reasons some further development and documentation work is continuing among smaller groups. This is being coordinated by Markit, who will be the index administrator," he adds.

While such details have not yet been finalised, the index is expected to comprise a basket of approximately 20 equally-weighted CDS on ABS securities referencing recently issued European CMBS bonds. The eligible bonds are likely to have weighted average lives of from three to eight years and be rated by at least two rating agencies.

There will likely be both sterling- and euro-denominated indices, each with tranches at triple-A and triple-B. The products are expected to trade on standardised ISDA documentation, adopting a pay-as-you-go (PAUG) format, and will probably roll every six months.

Dealers expect there to be strong two-way flow in the market once the index is launched. CMBS conduit programmes are likely to naturally adopt the role of protection buyers on the indices, as they are long commercial mortgage assets ahead of CMBS bond issuance.

The US CMBX Index launched in March 2006, consisting of five sub-indices rated triple-A, double-A, single-A, triple-B and triple-B minus. Each sub-index comprises 25 equally-weighted single name CDS, also using the PAUG template. A double-B rated tranche was added to the index in October 2006.

MP

Structured bid still has it

Potential CPDO flows look strong, but fundamentals could bring fireworks

New research from Citigroup's European fixed income strategy and analysis group has shed further light on the possible impact of the structured bid going forward.

Michael Hampden-Turner, a director in the credit products strategy group at Citigroup, explains: "A lot of people were pointing to CPDOs as the reason why spreads were tightening in the fourth quarter last year. What we wanted to look at was what would happen if CPDO popularity grew as many expect."

He continues: "While only about US$2bn of CPDO deals have been publicly rated so far, that represents US$30bn of protection selling when you take into account an average of 15 times leverage. Such an amount was certainly enough to convince the market that the 'structured bid' still had the potential to squeeze the market even at tights, but we wanted to go beyond just the psychological."

The core of the structured bid to date has been collateralised synthetic obligation (CSO) issuance, which Citi estimates takes dealers short by around US$40bn per month. This implies that dealers would need to buy nearly 100% of every bond issued just to stay neutral.

Citi then assessed the impact of a proportion of triple-A investors switching to CPDOs. If that proportion were 10%, it would mean US$17bn in CPDO issuance – translating into more than US$200bn delta-adjusted. This alone would add 25% to dealer CSO shorts and create significant tightening pressure.

While the 10% figure is an arbitrary one, growth in CPDO volume seems inevitable. "What will keep that story going is that in addition to regular CPDOs there will be a series of variants on the structure in the same way that the leveraged super senior ran for quite some time with various different formats, all trying to squeeze the last drops of value out of the structure," says Hampden-Turner.

However, he adds: "With a CPDO it's harder to do that because it involves a much broader set of assets than LSS. Nevertheless, there is talk of CPDOs with a wide variety of underlyings, such as the ABX or CMBX indices, as well as offering them on a managed basis."

At the same time, Citi's report found that increased rated equity activity would also generate further tightening pressure. If 10% of non-triple-A mezz (3-6% and 6-9%) buyers were to switch to rated equity, they would create the need for US$3.6bn in issuance – which delta-adjusted is equal to US$90bn, or an 11% increase in dealer shorts.

That is not to say that the credit market will continue to see only one-way traffic indefinitely. As the Citi report notes, structured credit flows have until now been firmly supported by solid credit fundamentals – decreasing leverage and a declining default rate.

It continues: "Leverage, though, is already beginning to rise, and should do so further this year. And if the default rate follows suit – in line with a slowing economy – we think those structured credit flows will suddenly and miraculously reverse. Banks, for example, now have an enormous incentive to start buying [loan] protection under new fair value accounting rules – and loan books are massive."

So, the report asks, what does happen when you put together irresistible forces and immovable objects? It answers: "Best guess, not much – or in our case, a small drift wider. But we'd say there's a decent chance of a 98-style blowout in the middle somewhere. After all, if you play with fire, you've got to accept the possibility of fireworks."

MP

LCDS stalemate remains

European contract still in the mix

The past few weeks have failed to produce any resolution to the protracted negotiations over the future path for European loan CDS (LCDS) documentation. The hoped-for meeting and vote over the contract specifications (see SCI issue 19) has not yet been re-scheduled - and even the alternative discussions among a smaller group of dealers (see SCI Issue 20) over what is believed to be the outstanding issue surrounding the scope of deliverable obligations are thought to still be far from resolved.

As one dealer outside of the process says: "You still hear conflicting views - even from the same firm. Some dealers say there will be a definite move towards the US model, while others say there are still so many problems with the US way of doing things that it is increasingly more likely that Europe might win out."

Another source who is more closely tuned into the talks suggests that any reports of a firm decision either way are greatly exaggerated. "At present, sentiment appears to be moving towards a very central position - some kind of compromise between a reference obligor and a reference entity approach. Anyone suggesting otherwise is probably representing one extreme viewpoint or the other and just talking up their own book."

The area of dispute revolves around whether European contracts will reference an individual obligor or specific loan underlyings.

In recent weeks a handful of dealers, representing both sides of the debate, have been exploring the areas of consensus among them and working towards a possible compromise solution to the issue. The aim of these meetings is to develop a proposal that can be put to the wider group of firms involved in LCDS contract discussions to date.

A meeting of that wider group is expected to be scheduled as soon as the smaller group's proposal is finalised, but no firm timeframe is yet in place.

MP

"The best spread achievable"

Online CDO structuring service provider Codefarm is the subject of this week's Provider Profile

Codefarm, which started business in 2002, offers new efficient methods for CDO arrangers to structure their portfolios using technology called 'evolutionary computing'. Similar in some ways to artificial intelligence, evolutionary computing is nevertheless highly unusual as it is 'bio-inspired', meaning that it employs a learning process that mimics phenomena that occur in nature.

Borrowing heavily from Darwinian Theory, Codefarm applies evolutionary-type processes to CDO portfolio construction so the most efficient portfolios, offering the best possible spreads, are constructed. Or, to paraphrase Darwin: only the fittest CDOs survive.

Jeremy Mabbitt

Jeremy Mabbitt, Codefarm's founder and MD, explains further: "When the firm started up we knew there were optimisation problems in finance, and particularly in the structured credit market, where our technology could deliver real measurable value."

'Galapagos Structurer' – named to underline the program's Darwinian roots – today serves five of the top ten CDO arranging banks, together with a host of other banks of varying sizes.

Galapagos Structurer is focussed on structured credit, explains Mabbitt. "CDO models are not readily optimisable; there are particular aspects of CDO structuring that make optimisation difficult. We have a compelling offering for this problem: that of guided search to find the optimal structure. Not only can Galapagos automate structuring, but it also produces enhanced spreads" he says.

Galapagos Structurer constructs underlying candidate portfolios to find optimal CDO tranches and capital structures based upon the parameters defined by the arranger and investor. Each candidate CDO is then evaluated, and the best structure and underlying portfolio are selected.

Next, it breeds and mutates the most promising candidates together, constantly testing new combinations, all the while learning from the outcome of the previous combinations to produce an even more efficient portfolio. The vast computational horsepower needed to perform these trials comes from a powerful compute grid (available through Codefarm and provided by HP).

John Mooren

"The way the technology combines portfolios is key; it keeps improving the spread by learning from each previous generation of portfolios. We call this process 'intelligent search'," says John Mooren, the firm's new Director of Marketing.

This iterative learning process is the firm's unique proposition. "Head-to-head tests with banks have shown that the Galapagos Structurer portfolios are consistently superior to those generated by conventional methods. With Galapagos, the structurer's skill is deployed to translate what the investor wants, and, given the room for manoeuvre the structurer has within the mandate provided, use our technology to explore the search space to locate the CDO with the best possible spread," says Mabbitt.

Even at the cutting edge of the structured credit market, there was an opportunity to be seized. "Initially some of the banks didn't recognise they had an optimisation problem, but once we showed the potential spread improvement they became more enthusiastic," claims Mooren. "The product has quickly developed: for example we now offer a 'variable notional feature' which is increasingly popular: each entity in the portfolio can have a different exposure."

Codefarm claims that there is plenty of unlocked potential in the CDOs currently being arranged using conventional methods, even in today's tight spread environment. "We help choose the underlying credits better. The result is a higher spread; a 20-30bp improvement on even the most vanilla mezzanine single-tranche structure is possible, with no additional risk – it is simply a more efficient CDO," adds Mabbitt.

In fact, Mabbitt doesn't accept the claims that opaque pricing by banks has led to reduced spread for 'gullible' investors. "Savvy investors are not going in blind and realise they are free to place their own demands on arrangers," he argues.

Mabbitt continues: "Given the constraints that are increasingly demanded by investors, arrangers need tools like Galapagos Structurer to ensure they can keep up. Arrangers can, using a powerful trade specification language, detail the investor's requirements to the nth degree – setting precise limits on any number of sector, country, rating and spread concentrations, mutual exclusions, WAS, WARF, underlying notionals, buffers on attachment points, and much more – to create a portfolio that's perfectly tailored to the investor and offers them the best spread achievable."

As a result, Codefarm suggests, Galapagos Structurer is replacing conventional methods of structuring. "We offer a quick, low-entry cost online service, flexible enough to meet arrangers' individual needs," claims Mabbitt. "Banks can now focus on innovation and let Galapagos Structurer take out the grind."

JW

ABN enhances ABS CDOs

The latest company and people moves

ABN enhances ABS CDOs
ABN AMRO has created an ABS exotics desk within its credit exotics group in London. The new operation aims to develop the bank's ABS CDO business and its correlation pricing framework for ABS in particular.

Dean Atkins will head up the ABS exotics desk. He was previously head of structured finance trading at ABN, which included the ABS flow trading desk. Mark Economides is now in charge of flow ABS trading.

Morgan Stanley hires Kenna
Morgan Stanley has hired Alex Kenna for its New York credit proprietary trading group. Kenna, formerly a portfolio manager at Lyon Capital Management, joins the bank as an executive director reporting to Michael Pohly, md proprietary trading.

Ballard joins ABN Asset Management
ABN AMRO Asset Management has appointmented Simon Ballard as credit strategist in a move to strengthen its European fixed income team in London. Ballard, who joins from Arc Securities, takes up the day-to-day decision making on overall top-down credit policy. He will be part of the global credit team with particular responsibility for formulating policy on sectors, ratings, maturity and currencies based on investment grade and market sector research.

Ballard was head of research at Arc Securities and prior to that was senior credit strategist at BNP Paribas.

GFI and ICAP buy into Clearing Corporation
Interdealer brokers GFI Group and ICAP have agreed to acquire a minority ownership interest in The Clearing Corporation. The firms intend to jointly develop new clearing services for OTC derivative products, including credit and interest rate derivatives.

The Chicago-based Clearing Corporation, one of the world's oldest independent clearinghouses for derivative instruments, is known to have been exploring the possibilities for CDS clearing for some time.

BNP Paribas adds credit structurer
BNP Paribas has hired Ronald Nittenberg to its structured credit group in London. He joins the bank as a senior credit derivatives structurer and reports directly to Herve Besnard, head of credit derivatives structuring.

Nittenberg is the most recent of a large number of appointments made by the bank's structured credit group. The group continues to have ambitious headcount and revenue growth targets for 2007 and beyond, BNP Paribas says.

Nittenberg joins from Royal Bank of Scotland in London, where he was a director in the structured capital solutions team since 2003, responsible for new product development as well as developing and implementing a correlation product platform. Prior to RBS, Nittenberg spent three years at JP Morgan, where he worked within the structured solutions and capital markets businesses.

Radian promotes credit twosome
Radian Group has promoted Scott Theobald to svp, corporate risk and capital management, while David Rockwell has been named svp, risk management, Radian Asset Assurance.

Theobald will continue to lead Radian's efforts to enhance its credit risk management systems and will play a central role in enterprise-wide capital analytics and transaction management. Rockwell will continue to be responsible for managing Radian's CDO portfolio and provide counsel to the business units to ensure that all transactions are consistent with the company's risk profile.

Timmis joins Calyon
Mark Timmis has joined Calyon's credit trading desk and will be responsible for all TMT trading on both the CDS and cash sides of the business. He previously worked at Bear Stearns, where he was in charge of building up the firm's trading capacity in CDS. Prior to that, Timmis ran the CDS trading desk at Credit Suisse in Europe.

Based in London, he reports to David Cohen, global head of credit trading.

Markit acquires ABSReports
Markit Group has acquired ABSReports, an electronic deal reporting service that focuses exclusively on the performance of European deals in the ABS, MBS and CDO markets.

Markit will integrate performance data from ABSReports into its European ABS pricing service with a view to bringing greater efficiency and transparency to the secondary European ABS market. Its ABS pricing service allows clients to track ABS values using independent, consensus pricing drawn from 25 market makers and provides data for over 4,000 securities.

MP

WestLB brings LSS ABS trade

A round up of this week's structured credit news

WestLB brings LSS ABS trade
WestLB has closed Khamsin Credit Products Series 6, a small but unusual public leveraged super senior (LSS) swap on an outstanding ABS tranche.

Moody's has assigned triple-A ratings to the US$55.6m trade, which references the US$890m Class A notes from Brightwater Capital Management's Blue Heron Funding II transaction due in 2041. The deal transfers to investors not only credit risk but also market risk.

If the average credit quality of the portfolio underlying transaction exceeds a certain time-dependant threshold (the WARF trigger), investors will have the option to either purchase additional notes and de-leverage the transaction (in which case additional principal will be issued and additional collateral posted), or unwind the transaction. In the event that investors choose to unwind the transaction, they will bear the market risk of the referenced tranche.

The management of the portfolio is performed according to the existing criteria set in the Blue Heron II transaction. The deal's reinvestment period will, however, end in March 2007, with a weighted average life constraint on reinvested assets of approximately three years at that time.

The transaction involves WestLB entering into an asset swap with Khamsin, with the SPV's obligations being collateralised by an amount equivalent to the notional of the Blue Heron Class A notes. Series 6 will also amortise proportionally to Blue Heron notes. Under the asset swap, WestLB pays Khamsin the coupon due under the notes less the interest due under the collateral held by the SPV.

IIC offers high yield indices
International Index Company (IIC) has launched its iBoxx Euro High Yield Index family. The firm says the indices are structured to provide high yield investors with accurate benchmarks of the sub-investment grade fixed income market for euro-denominated corporate bonds and to provide a platform on which research and structured products can be based.

In an effort to address different investors' requirements, the indices are calculated both with and without crossover bonds. In addition, constrained and unconstrained indices will be published. Under the constrained indices, issuers are limited to a maximum weighting of 3%, with the weighting being adjusted monthly.

A full range of sector, maturity and rating indices will also be published. As at 1 January 2007, 193 bonds qualified for index membership.

Daphne Asia 1 CLO launches
Calyon has become the latest bank to offload the credit risk associated with a portfolio of Asian corporates via its SPV Daphne Finance.

The synthetic balance sheet CLO comprises a US$2.1bn triple-A rated super senior CDS; US$96.3m triple-A rated Class S notes; US$132.5m triple-A rated Class A notes; US$25m triple-A rated Class B notes; US$20m single-A plus Class C notes; US$15m single-A minus Class D notes; US$32.5m triple-B minus Class E notes; and US$12.5m double-B Class F notes.

The synthetic balance sheet CLO references a diverse pool of senior and predominantly unsecured loans. Under the CDS agreement, Calyon has bought US$2.5bn protection on an initial pool of 211 reference obligations from 172 predominantly Asian corporate groups.

The three largest country concentrations are Hong Kong (at 20.78% of the portfolio), Japan (11.76%) and Singapore (8.34%). Names from Asia ex-Japan represent 55.2% of the initial portfolio.

Calyon has the option to redeem the notes two years after closing. Should this occur, the bank will be required to repay noteholders any accrued interest and outstanding principal in accordance with the transaction's documentation.

It also has the option to extend the scheduled maturity of the notes by one year to 31 March 2013 by giving notice to the issuer at least 20 business days prior to 31 March 2008 and a further option to extend the deal by one more year to 31 March 2014, provided the first extension option has been exercised.

Algebra score for SG
SG CIB has launched Algebra, a synthetic CDO managed by Société Générale Asset Management Alternative Investments (SGAM AI). The transaction aims to provide an attractive return for investors, despite the current low spread environment.

Rated by Moody's, the 10-year deal comprises €50m Aa3 notes and US$70m A2 notes. It incorporates a trading account through which the manager can monetise trading gains and use the proceeds as a buffer in case of trading losses or defaults.

The balance of the trading account is paid at maturity, with 75% of the total expected to be paid to investors and 25% to SGAM AI. The dollar-denominated notes offer a step-up subordination in year five, providing investors with increased protection in the final years of the deal when defaults are most likely to occur. These notes also have a call option at year five at par, thereby boosting returns.

A variety of notes across different currencies and maturities are available under the transaction.

According to Béatrice Benhamou, responsible for structured credit in France at SG: "This project results from strong collaboration between ourselves and SGAM AI. We believe there is still strong demand from investors for such structures, and our sophisticated structuring and trading capabilities allow us to propose an attractive return despite the current low spread environment."

MP

Trading ideas - hot yet?

Tim Backshall, chief credit derivatives strategist at Credit Derivatives Research, looks at a curve steepener trade for the Dow Jones CDX NA HY Series 7 index

We start the New Year with a somewhat ambivalent view on the credit markets. While we remain net bearish, we do not believe spreads will 'blow out' anytime soon with a high likelihood. A gradual spread widening on the back of 'turn-in-the-cycle' flows, LBO risks, and drop in technical bid effect is most likely.

We remain committed to idiosyncratic 'relative-value' trade ideas as a main source of alpha but we feel a core market position is necessary – and ours is 'bearish'. We have a number of index-based trades outstanding currently – mostly skewed to steepeners and technical-bid mean-reversion ideas. We now expand that to the HY credit market with a positive carry, positive roll-down bear steepener.

Big picture
Our December Market Update calls gave our view of the credit markets for 2007 and two main themes came out from discussions with clients. There are two supporting columns for the market's apparent indifference to default risk – demand for spread and fundamentals.

Corporate profitability is at all-time highs (over 2.5 standard deviations above historical averages) and structured credit issuance is at all-time highs. We do not see dramatic unravelling of this in the short-term but certainly the increased shareholder-friendliness and M&A activity is starting to drag even the best fundamentals down.

Exhibit 1, below, shows the last credit cycle, in terms of equity and credit spreads, highlighting the fact that we seem to be turning the corner in the cycle. After a 'recovery' phase since 2003 (restructuring efforts to boost cash flow, margin expansion, and increasing positive FCF), we have moved to an 'expansion' phase in 2006 and into 2007. The recovery phase saw equity and credit rallying but as we move into and through the expansion phase, equity is likely to outperform credit as margins decline, FCF turns negative and leverage starts to rise on the back of management decisions to 'pay-back' shareholders for their patience.

Our view is that this expansion phase may be short-lived and swiftly shift into a downturn phase on the back of any market-sized event – 3-6 more months of flat rates from the Fed. The sharper global slowdown (as opposed to the soft landing we expect), will drive volatility up sharply and defaults will rise gradually. This downturn phase will have equity and credit selling off.

The sell-off in credit is likely to be lead by the HY space in our view - not only due to their 'proximity to peril' but the relationship with equity and volatility markets. Exhibit 2, below, shows the relationship between credit spreads and VIX - a strong relationship when spreads are wider and no relationship when spreads are tight. The last time HY spreads were tight and volatility was this low (VIX is around one standard deviation below its historical average) was in 1996/97.

The disconnect in HY spreads and VIX in 96/97 – where spreads remained tight as volatility increased is unlikely to happen this time as 'things are different this time'. Far more risk is held by levered investors and hedge funds (as opposed to banks in 96/97) keeping the debt-equity-vol relationship far tighter than before and the short-side of credit is now available for trading (CDS). Volatility is the biggest risk to the second support of the credit markets – demand. Any increase in vol (risk) will change the characteristics of the risk-reward and potentially upset the inexhaustible demand for credit (structured or otherwise).

We feel that volatility will rise. As M&A activity increases, LBO fears become ubiquitous (Home Depot???), and corporate profit concerns increase (December retail sales), we expect to see VIX rise from the ashes and drag spreads wider with it as seemingly cheap shorts abound in the CDS world. It is this background that pushes us towards a bearish HY spreads trade but, as usual, we prefer to structure the trade with positive carry, and opt for the steepener. With its positive theta (roll-down), we can be wrong for a while and still get paid as breakevens look good for us and given the huge liquidity in the indices, we are not hindered by the typical chasm of bid-offer spreads HY traders face.

Spreads steepening
The DJ CDX NA HY Series 7 index has excellent liquidity in the CDS market across the curve and now trades actively in five-, ten-, three- and seven-year maturities. Bid-offer differentials and liquidity are excellent in the 5's and 10's. As spreads have moved tighter over the past three months so the curve differential (between five- and ten-year CDS) has moved steeper, as seen in Exhibit 3, below.

This behaviour is not entirely unexpected given the technical bid demand in five-year credit (as investors use the HY to hedge IG movements and structurers focus on 5Y HY). The steepening has definitely been driven more by 5Y CDS rallying than by 10Y widening. The last few weeks (albeit quiet) have seen the 5s-10s differential pull back modestly – providing us with a great opportunity to enter this bearish trade.

Our simple rationale for the trade is that as vol rises, so HY spreads will sell-off from their all-time lows and as spreads widen, so the curve will steepen. In the meantime, we will earn carry and roll-down. Interestingly, the two-faced behaviour of the curve as credits drift into HY space is a help to us in the case of the index trade.

As we have discussed with many of our clients, our research has found a split between poor-performing HY credits and LBO-like (highly-levered) HY credits. As IG spreads widen, so curves steepen to reflect the increasingly marginal risk of default and uncertainty. In the case of HY, as we move through around 250-300bp in 5Y, we notice that the curve continues to steepen as spreads widen for highly-levered LBO-like names, but for weak HY credits, we see the curve begin to flatten back again from steeps at around 80-90bp in 5s-10s.

The outcome of this is that if one is bearish on HY credits, a flattener is a good trade for weak performing fundamentally deteriorating credits as opposed to a steepener for LBO-like credits. However, the index, being composed of a diverse mix of good, bad, and ugly credits, trades down the middle of this bifurcated curve.

Exhibit 4, below, indicates the 'best-fit' relationship between curves and spreads for a broad universe of IG, XO, and HY credits in the US. Notice the right hand side shows that curves generally begin to flatten as spreads widen, but a number of highly-levered names remain excessively steeper (TSG for instance at over 140bp differential in 5s-10s). The green arrow shows the differential between the current CDX7 HY 5s-10s steepener (at around 75bp) and the blue curve (fair value at around 95bp). This indicates that in aggregate the index is trading very flat to its components curves.

Fundamentally, we note that more of the HY CDX names have stable or deteriorating outlooks as opposed to improving. This idiosyncratic fundamental view combines with our cyclical credit view to provide a solid background for the bearish trade.

The 5s-10s steepener offers good risk-reward characteristics and benefits from being very flat to its fair value. One more analysis provides a final nail-in-the-coffin for us – the 5Y index level is trading wide of the component-based fair value. When we adjust for maturity mismatches (post the December roll), DV01 adjustments, and liquidity, we find that the 5Y CDX7 HY index is around 5bp wide of its fair value.

So, selling 5Y protection and buying 10Y protection seems to offer a number of good relative value opportunities as well as benefiting from time and our fundamental views on credit weakness next year. The duration-neutral package offers around 55bp of carry and excellent roll-down.

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.

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 CDX7 HY and the bid-offer spread costs.

Bid-offer spreads have narrowed to around four basis points in five- and ten-year CDS respectively.

Fundamentals
This trade is significantly impacted by the macro- and micro-fundamentals. While the technical flatness of the curve and relative cheapness of the 5Y help the position, the fact that more credits have stable or deteriorating fundamental outlooks than improving helps reassure us from an idiosyncratic risk perspective.

Summary and trade recommendation
Our end-of-year discussions with clients have left us somewhat ambivalent about credit in 2007. Cyclically, we believe that spreads move gradually wider as M&A activity picks up, LBO fears become ubiquitous (HD!), and corporate profit concerns increase (Dec. Retail Sales?).Tactically, we see idiosyncratic relative-value trades as the way to generate alpha.

With spreads at their tights, vol at its lows, and corporate profitability at its highs, credit seems well-supported, but we feel the more negative/stable versus positive fundamental outlooks of the index components, significant flatness relative to its peers, and cheap to fair-value nature of the 5Y index, provides us with a great opportunity to enter a bearish position. The positive carry and roll-down of the duration-neutral steepener provides us time to be right and offers good upside as the diverse components of the index widen and steepen as vol mean-reverts.

Sell US$15mm notional Dow Jones CDX NA HY Series 7 5 Year CDS protection at 264bp (US$102.4375) and

Buy US$10mm notional Dow Jones CDX NA HY Series 7 10 Year CDS protection at 339bp (US$99.125) to gain 55 basis points of positive 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).

Trading equity tranches - part 1

In this series of three articles, Richard Huddart and Domenico Picone of the structured credit research team at Dresdner Kleinwort look at some of the different ways to trade equity tranches and analyse them

Value in 0-3% equity tranches
The demand for equity tranches of synthetics and bespokes from real and leveraged accounts has been growing steadily since 2004. By looking at the table below, we see that a total of US$56.3bn of credit risk was distributed into global markets through synthetic and bespoke CDOs in Q1 2006 and US$87.6bn in Q2 2006. On a delta-adjusted basis, this translates into US$197.2bn in Q1 2006 (US$313.6bn in Q2 2006) of credit risk, of which 45% (42%) was distributed in the form of an equity tranche, up from 38-39% in 2005 and 32% in 2004.

We delta-adjusted the numbers so that we could compare tranches regardless of where they sit in the capital structure: equity tranches, for example, are much more sensitive to changes in the underlying index spread than senior and super-senior tranches. This is reflected in the fact that equity tranches have a much higher delta. Therefore, multiplying tranche volumes by their deltas gives us a measure of issuance weighted by spread risk sensitivity.

The ratio of this delta-adjusted issuance to the unadjusted volume gives a weighted average tranche delta of 3.5 for Q1 2006 (197.2/56.3) and 3.6 for Q2 2006 (313.6/87.6). The same measure for the last seven months of 2005 (after the correlation dislocation) was 2.5. As delta tends to be higher for tranches with less subordination (loosely assuming a fixed tranche width), this suggests to us that investors have recently been selling protection further down the capital structure than they were in 2005. This is once again unsurprising, given current low credit spread levels.

The two charts above show the relative loss allocation profiles, which explain how the underlying level of risk in the index (given by the index expected loss) is allocated, or priced in, across the tranches of the capital structure, for 5y iTraxx and CDX (on-the-run indices). In both cases it is clear that the proportion of losses in the index being allocated to the 0-3% regions has increased markedly since early 2005, meaning that the 0-3% equity tranches are offering much greater relative return opportunities (compared with the rest of the capital structure) than was the case around 18 months ago.

The indices have both tightened over the last year or so, and whilst this should naturally lead to a higher proportion of losses being allocated to equity, the tighter indices cannot explain all of the increased relative value in the 0-3% region. This is illustrated in the fact that the compound correlation skews are now significantly lower than they were in March last year – especially as far as the 0-3% tranches are concerned. Long equity tranche positions are long in correlation, and so spreads will widen out offering greater value in comparison to the rest of the capital structure when correlations fall.

The charts below show the normalised dispersion of these 5y iTraxx and CDX indices respectively, and how these measures have developed over the past year. Normalised dispersion is defined as the standard deviation of the CDS spreads underlying the index divided by their mean.

The fact that the current on the run baskets have normalised dispersions around the same level as (or, in the case of CDX, lower than) the on-the-run baskets in early 2005, suggests that the portfolios now display a similar or even increased level of homogeneity. This means that the extent of outlying names, with very wide spreads (that can be seen as large idiosyncratic risks), has not increased.

This is partially due to the fact that troublesome names tend to be rolled out whenever a new series comes into force as being on-the-run. Idiosyncratic risk is particularly of interest to equity protection sellers, as it is they who are essentially exposed to one-off, independent defaults.

Similar patterns are observable at 7 and 10 year maturities.

Funded and unfunded tranche markets
Over the past few years, there has been massive growth in synthetic CDO tranche trading, driven largely by the advent of CDS indices such as iTraxx Europe and CDX.NA.IG. Such trades can be done on a funded or an unfunded basis.

Notes and swaps
In the funded case, the investor funds the purchase of a credit note (usually issued by an SPV) linked to the performance of a reference portfolio. The proceeds from the note are used to buy collateral, which has virtually zero credit risk.

When a default occurs, part of the collateral is sold to cover the loss, and a portion of the note is written off. In return for the credit risk taken, the investor receives regular interest payments as LIBOR plus a spread. We often refer to such a position as a Credit Linked Note (CLN), and the position is essentially similar to a floating rate bond, in that the investor pays par at issue, receives a regular floating rate coupon based on the outstanding tranche notional and receives back par minus any losses at maturity.

In the unfunded case, the trade is structured as a swap, rather than a note. That is to say, the protection seller does not put up any upfront cash, however just receives a regular premium from the tranche protection buyer. In the event of default, the protection seller must then fund any tranche losses, which are paid for at the time of default.

Funded and unfunded – two different types of participant
In no strict way, the funded and unfunded tranche markets have grown to appeal to different types of investor groups. Buy-to-hold investors have been the primary users of the funded tranche market, as the structures here are more like traditional cash bonds.

The unfunded market has been used as more of a tool for hedge funds, looking to play relative value strategies.

The possibility of IG rated equity
Recent evidence has shown us that, so long as a relatively high quality reference portfolio is used, it is possible to sell protection on a tranche which both attaches and detaches within the traditional 0-3% 'equity' region, whilst achieving an investment grade rating. This opens up the possibility of a wider range of investors being able to invest in a product that has equity-like behaviour (i.e. high leverage and coupons).

In order to achieve an IG rating on such an equity tranche it will most likely be necessary to choose a highly correlated portfolio. This is because the equity tranche is long in (i.e. benefits from) correlation.

The logic behind this is as follows: In a highly correlated portfolio, the possibility of 'extreme' events is increased – that is to say, both the possibility of widespread defaults and the possibility of no defaults at all occurring are increased. An equity tranche investor likes the increased probability of no defaults, as he is selling first loss protection, however, is unaffected by widespread defaults above his detachment point. This means that, in comparison with the case of low correlation, the equity investor has a lower expected loss.

Expected loss is also an increasing function of time, and so IG ratings should be more easily achievable on shorter dated tranches.

Funded equity tranches: four ways to trade
The following four trades are what we consider to be the benchmark ways to trade equity in a funded form.

The 'standard' equity tranche
The standard arrangement for the 0-3% tranche of indices such as iTraxx or CDX is to trade the tranche with a 500bp running spread and the rest of the return on an upfront basis. In the funded case, this means the 'standard' CLN is issued at par minus an upfront payment, and running coupons are paid at 3 month LIBOR (or EURIBOR) plus 500bp.

The Zero-Coupon equity tranche/Principal Only (PO) strip
In the zero-coupon case, no running coupons are paid throughout the life of the deal, and so the note is deeply discounted and issued below par at a price that reflects what the investor expects to receive back at maturity. The only other cash flow comes at maturity, when the investor receives back par net of any losses sustained by the tranche. Zero-coupon equity tranches are proving very popular. As we will see later in this report, this is because they are insensitive to the timing of defaults in addition to their simplicity and low mark-to-market volatility relative to tranches with other coupon structures.

The all-running equity tranche
In the all-running case, the CLN is issued at par and all returns are paid in the form of running coupons. The running coupons in this instance are therefore larger than those paid in the 'standard' case. The all-running funded tranche has the highest upside potential of all the funded-tranche coupon structures that put principal at risk, and tends to outperform the alternatives in cases of low or back-loaded defaults. The downside risk and MTM volatility are also higher than for the funded alternatives however.

The Interest-Only (IO) strip
The IO strip is the only one of the four trades in which the investor does not put principal at risk – rather he pays at issue the PV of the 3mL+500bp running coupon stream received as in the 'standard ' case and then actually receives this stream throughout the life of the deal. As the coupon payments are calculated based on the outstanding tranche balance during the relevant coupon period, the IO strip achieves its best returns when defaults are low or back-loaded.

In the event of default, the investor only risks losing future coupon payments. The sum of the zero-coupon (PO) tranche and the IO strip is the same as the standard tranche. As such it outperforms the standard tranche whenever the zero-coupon tranche underperforms the standard tranche, and can be a vital hedging tool for dealers.

Each of these different equity tranche trades have different risk-return profiles, and it is the purpose of this report to explain how the value of each trade type reacts to changes in certain key variables. In particular, we consider sensitivities to:
? The timing of defaults
? First order spread risk (delta) and second order spread risk (gamma)
? Interest rate risk, correlation risk, idiosyncratic risk and time decay

Unfunded equity tranches: four ways to trade
Each of the funded trades that we saw in the previous section also has an unfunded 'analogue'. Whereas funded tranches are structured as Credit Linked Notes (CLNs), their unfunded equivalents are essentially credit default swaps written on the tranches. Each trade tends to over- or underperform the others in the same situations that their funded analogues over- or underperform.

The 'standard' equity tranche
The unfunded standard equity tranche is the simplest to reconcile with the standard market quotes. In the swap agreement, the investor (protection seller) receives the upfront coupon at issue and then a running coupon of 500bp based on the outstanding tranche notional throughout the life of the swap. On the loss leg of the swap, he must make payments to cover losses resulting from defaults as and when they occur.

The 'all-upfront' equity tranche (PO strip)
The all-upfront unfunded equity tranche is akin to the zero-coupon funded variety. In this swap agreement, the protection seller receives all return in the form of a large upfront coupon, and on the loss leg must simply pay for losses as they hit the tranche. Unlike the zero-coupon funded note, however, this trade is not insensitive to the timing of defaults, although its default-time sensitivity is significantly smaller than for other unfunded trades.

The 'all-running' equity tranche
In the unfunded all-running case, no exchanges are made upfront – rather the protection seller receives all returns in the form of a running premium (larger than the standard 500bp spread) and makes payments to cover tranche losses as and when defaults occur. Once again, this is the strategy with the biggest upside and downside potential, performing best when defaults are low and back loaded.

The IO strip
Once again, the IO strip completes the picture of benchmark structures, and can be a useful hedging tool for dealers. The difference between the funded and unfunded IO strips is that in the unfunded case, the investor receives (running) and pays (upfront) for a stream of 500bp coupons on the outstanding tranche notional, as opposed to a stream of 3mL+500bp coupons.

Whilst the presence of this trade 'completes the picture', some may not really consider it to be an unfunded trade given that the coupon stream needs to be paid for upfront. We introduce an alternative unfunded IO strip with no upfront payment in the section. Some more exotic equity trades.

Of course, there are many other combinations of coupon structures that could be adopted by such trades (both funded and unfunded) – for example any mix of running and upfront coupons could be used, not necessarily in line with the 'standard' 500bp running convention. The comparative risk-return profiles of such trades should be clear from the analysis of the four funded and four unfunded cases that we consider in this report.

The timing of defaults
The equity tranche is obviously very exposed to defaults, as it sits right at the bottom end of the capital structure and hence absorbs the first losses in the portfolio. In addition to the absolute level of defaults (and hence losses) that a reference portfolio experiences over the lifetime of a CDO, however, the return realised on an equity tranche will also be significantly affected by the timing of such defaults.

'Front-loaded' vs 'Back-loaded' defaults
When referring to defaults in a portfolio, we often say that they are 'front-loaded' or 'back-loaded'. For a given number of defaults, we say that they are front-loaded whenever they have tended to occur towards the beginning of the life of a transaction and back-loaded when the CDO has run the majority of its course before the defaults kick in.

The possibility of lost coupons
The reason that equity tranche returns will depend on the timing of defaults comes primarily from the possibility of lost coupons, although there is more to consider in the unfunded case. When paid on a running basis, coupon payments are calculated based on the outstanding principal balance of the tranche.

Therefore, if the tranche suffers losses as a result of defaults in the reference portfolio, the tranche investor will receive diminished coupon payments. If defaults are front-loaded, as opposed to back-loaded, then more will be lost in the way of diminished coupons throughout the life of the deal. Equity tranche coupons are typically quite large in order to compensate their relatively-substantial risk, and as such the timing of defaults can have a significant effect.

Funded case
In the funded case, the timing of defaults has no impact on principal cash-flows received back by the investor. This is because he receives all cash-flows of this type at maturity – in fact the investor receives the par value of the tranche minus any the size of any principal losses, regardless of when the losses occurred. Default timing risk therefore only influences funded equity tranche returns through the effect on running coupons. Of course, the absolute size of losses due to defaults is still of prime importance.

Unfunded case
The difference in the unfunded case is that the tranche investor (protection seller) must make payments to reimburse the protection buyer for losses as and when defaults occur. The earlier a given default occurs therefore, the earlier the protection seller must fund such a payment and as a result he will be subject to some default timing risk on the principal side due to the time value of money. For a given coupon structure, his coupons will be affected in the same way as the funded case.

Timing of default scenarios
Over the next few pages we present some cash-flow scenarios for the four funded and four unfunded trades that we have introduced in this report. For each 5y 0-3% equity trade we show the difference in cashflows for two different default time scenarios.

In Scenario A, we assume that the reference portfolio experiences two defaults, both of which occur in year 1. In scenario B, we also assume that the reference portfolio experiences two defaults, but this time in year 4. We assume a portfolio of 125 CDS names of equal notional exposures, a fixed recovery rate of 35% and a fixed LIBOR rate of 4%. We also assume that all defaults occur at the beginning of a period and for simplicity we assume that all coupons are paid on an annual basis. Whilst these are quite broad assumptions, the analysis serves to highlight the key ideas.

Timing of default scenarios: funded trades
In the charts below, the negative cash flows at year 0 are the cash outflows required to fund the purchase of the equity notes, whereas all positive cash flows represent the coupons received or the principal repaid at maturity.
The first point to notice in looking at these charts is that the present value of each trade falls when the defaults occur early in the deal (in year 1 – scenario A) as opposed to late in the deal (year 4 – scenario B), with the exception of the Zero-coupon equity.

? Zero-coupon As explained on the previous page, all default timing risk for funded tranches comes from the possibility of diminished or lost running coupons, and so this is eliminated in the zero-coupon case by trading all returns upfront.
? All-running coupon structure is hit most by early defaults as more is lost in the way of coupon income.
? The IO strip has a high sensitivity to default timing, given that everything received by the investor is in the way of running coupons. Also, given that the standard tranche is the same as the sum of the IO strip and the PO strip (zero-coupon), all default time risk taken on by the standard tranche must be heaped onto the IO strip.

Of course, we must note the higher returns achieved by all-running coupon structures in the case of late, or no defaults. This is because all tranches are priced with the same expected default profile, as implied by the underlying CDS curves. If the tranche 'lasts longer' in a sense than was priced in, then the all-running tranche will pay out more in the way of coupons than was initially priced into the all-upfront tranche, and the all-running tranche will outperform.

Timing of default scenarios: unfunded trades
The story in the unfunded case is unsurprisingly very similar to the funded scenarios, with the all-running coupon trade being more sensitive to the timing of a given set of defaults than the standard tranche, which is more sensitive than the all-upfront.

In the charts of the standard and all-upfront equity, the positive cash-flow at year 0 relates to the upfront premium received to sell protection, whereas in the chart of the IO strip, the cash outflow at year 0 is the upfront payment to sell protection with an IO strip.

? The obvious difference between the funded and unfunded cases comes when all returns are traded on an upfront basis. Whereas the funded zero-coupon equity tranche is insensitive to default timing risk, the unfunded all-upfront structure does show some sensitivity. As discussed earlier, this is down to the time value of money and the fact that principal losses need to be reimbursed as and when they occur in the unfunded case.

Two key conclusions
There are essentially two key conclusions that can be drawn from this discussion:

? Tranches which trade all return upfront are the least sensitive to the timing of defaults. Therefore, equity tranche investors who want to remove (or limit) the timing of defaults as a return effecting variable may be well advised to trade a zero-coupon or all-upfront tranche.

? All-running coupons mean higher risk but higher potential returns. When at least a part of the return is traded upfront, the investor's downside is limited – because even if a catastrophe scenario was to occur such that the entire equity tranche is wiped out immediately, he would still receive the upfront coupon. A zero-coupon tranche (funded) or all-upfront tranche (unfunded) therefore has the lowest downside potential. On the other hand, if no defaults occur then the all-running tranche investor will receive more in the way of coupon-returns than was initially priced into the zero-coupon tranche, as the tranche essentially 'lasts longer' than expected. All-running coupon structures therefore have the biggest upside potential.

© 2007 Dresdner Kleinwort. All Rights Reserved. This Research Note was first published by Dresdner Kleinwort on 6 October 2006.