Contents

CDR Liquid Index data as at 15 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.

CDR 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.

New CDO-squared mixes assets

Credaris deal utilises cash and synthetics to maximise ABS opportunities

Credit-specialist asset manager Credaris has launched Tasman 2006-1, a hybrid CDO of CDOs. The US$300m transaction was lead managed by UBS and closed on 11 January. The deal has an 8-year expected maturity and issued six S&P and Moody's rated tranches, along with US$12m of unrated preference shares.

The transaction - which is Credaris' first broadly distributed CDO - leverages off the firm's competencies in active management of structured finance and synthetic credit assets, complementing its evolving suite of fund and securities offerings, according to Mark Brooks, portfolio manager at Credaris. "The primary public fund that we manage is in many ways a CDO-squared in that it has a large allocation to CDOs. Consequently, a lot of the technology we have put in place to run it could be easily utilised to drill down to multiple CDOs, so issuing a CDO-squared was a natural extension for us," he says.

At the same time, Credaris also recognised that there was an opportunity to use a CDO of ABS portfolio within the structure. Brooks explains: "As a result of the headline risk surrounding CDOs of ABS, the products' spreads have stayed wider than those for CLOs - the usual underlying for CDO-squareds. This enabled us to build a higher quality and more conservative portfolio than if we had been exclusively investing in loans, but have not needed to stretch to achieve the compelling returns the deal offers."

He continues: "In addition, the continued progress of the derivatives market and, in particular, the introduction of CDS on CDOs gave us the ability to ramp efficiently. Rather than just buy what comes along in the primary market, we were able to look into history to optimise the overall portfolio towards more seasoned ABS collateral and therefore enhance the diversification."

The collateral in the Tasman portfolio is exclusively mezzanine tranches of CDOs of ABS and CLOs, and targets a weighted average rating factor of 330 and a weighted average spread of 260bp. Building around an initial core portfolio of 2004 and 2005 vintage CDO collateral, Credaris completed the portfolio ramp with synthetic exposures.

The deal attracted strong interest across both small and large institutional investors, Brooks observes. Consequently, Tasman is likely to be the first in a programme of such deals from Credaris.

"Investors liked this deal because of its transparency and the fact that it is lightly managed, so they know what they are going to get. That is not to say that future deals will not involve more management and increased capability of trading - it will depend on the situation and investor requirements at the time. Meanwhile, we will keep looking for opportunities and value in markets" he adds.

MP

Unwind risk covered

Index and tranche markets should cope with any significant deleveraging

New research from Royal Bank of Scotland's structured credit strategy team in London has revealed what might happen if an unwind in structured credit leverage occurred in 2007. The report suggests that, bar an exceptional event, index and tranche markets should continue functioning uninterrupted in any period of market stress.

In the event of an unwind in leverage, RBS says that it would expect indices and tranches to become the focal points of liquidity as was the case during the May 2005 correlation crisis. However, it adds that those markets are much better equipped to deal with such scenarios, given how much more liquid they are today compared to 20 months ago.

RBS says that the nature of a credit unwind will be key to the effect on structured credit holdings and their different exposures to idiosyncratic and/or systematic risk. For example, the CDS index long-only CPDO can be vulnerable to idiosyncratic risk under certain circumstances.

Gregory Venizelos, structured credit strategy at Royal Bank of Scotland, explains: "A worst-case scenario would be downgrades and significant spread widening in only a handful of names in a CPDO's index basket; names which would be subsequently removed from the new basket. Thus the CPDO would not only suffer negative MTM due to the spread widening, but would also have to roll into a new index most likely trading tighter and therefore no extra spread [would be available] to counter the MTM loss. In addition, the more sudden such a widening the worse for the structure, as it may not have time to adjust its leverage."

In contrast, a (not too extreme) systematic widening could be benign to the CPDO, as the MTM losses are offset by the higher spread of the new index. In the event of a more extreme widening, any detrimental effect on CPDOs may also have an impact beyond the index market.

Venizelos says: "Given its exposure to idiosyncratic risk, the CPDO structure could be perceived as a long correlation instrument, despite having no direct sensitivity to implied correlation. So could a CPDO crisis spill over into the tranche market? Potentially yes, though that would depend on the size of the CPDO market if and when that crisis hits. As things stand currently, we don't think this market size is substantial enough yet."

The research also includes the following examples of the effects of different exposures to idiosyncratic/systematic risk:

• If driven by idiosyncratic risk, e.g. a concentration of jump-to-defaults or vigorous and synchronised LBO activity, structures which are long correlation/short idiosyncratic risk will suffer. First affected will be long equity risk exposures and subsequently short mezz risk exposures (either index or single-name delta hedged).

• If driven by systematic risk, delta hedged long equity risk positions should outperform, though they may initially overreact to any market spread widening. Any such outperformance will depend on the degree that correlation is repriced higher in response to the increase in systematic risk. Other changes in market characteristics, like curve term structure or spread dispersion within the iTraxx/CDX baskets, can also have an effect.

• Any sudden moves in tranche market pricing will result in significant shifts in the base correlation curves and disturb the valuation of bespoke tranches. Cracks in the hedging of bespokes with standard tranches are bound to appear, and any such dislocations will present opportunities for market participants, especially to the detriment of any investors who may be forced sellers.

MP

LCDX progresses smoothly

US loan CDX preparations proceed without controversy

Dealers working towards the introduction of the US loan CDS index, LCDX, report continuing progress. In contrast, discussions over the form of the European LCDS contract continue, but not yet with tangible results.

The LCDX working group met again last week and those involved appear confident that the latest deadline for the index launch will be met. "Last week was another week where a lot got done and it looks like we are close to putting the finishing touches on various aspects of the index. So it looks very promising for a first quarter launch," confirms one dealer.

Another banker observes: "There is no controversy at all in the LCDX process: it's just been a question of getting the legal documentation right, which obviously takes time. What everyone wants to use is based on the net physical settlement protocol - and we want that to be hardwired into the LCDX documentation."

There had been some talk that LCDX would be cash settled and finalising that had caused some delay in the process. "You could call net physical settlement cash if you wanted, it's just semantics really. With net physical settlement, if you don't put in for loans then you will get a cash settlement," says the dealer.

Other than the legal framework, no other details have been firmly decided on LCDX yet. "Once we got closer from a legal documentation standpoint, we'll finalise the number of constituents and so on. Obviously, we've all had various meetings internally through time talking about such things. But we'll get more serious about that and resolve it externally as we get closer to feeling like we are about to launch," the dealer adds.

Meanwhile, a meeting of the smaller group of dealers working towards a consensus on the basis for the European LCDS contract took place on Tuesday, with no signs of significant progress. A number of future meetings are understood to be planned in order to continue going through the different issues that the two sets of dealers have. The next such meeting is scheduled for later this week.

MP

HSBC to roll-out unusual SIV

New three-tier vehicle to invest 90% in triple-A assets

HSBC is planning to launch its second structured investment vehicle (SIV), dubbed Asscher, in March. The vehicle is thought to be only the second ever to feature a three-tier liability structure, following the launch of Citigroup's groundbreaking Sedna Finance in June 2004.

As Dominic Swan, head of structured investment vehicles at HSBC, explains: "Asscher will have the ability to issue triple-A rated senior debt - both commercial paper and MTNs - as well as single-A rated mezzanine notes and triple-B rated income notes. Our first SIV, Cullinan Finance, has a two-tier capital structure consisting of triple-A senior and triple-B income notes. Furthermore, Asscher will invest 90% in triple-A rated assets, while Cullinan invests 80% at that level [the remaining portion is primarily made up of double-A/single-A rated bank debt]."

The attraction of offering a three-tier SIV is that such a structure should be of interest to a wider range of investors as the single-A capital notes carry a public rating from Standard & Poor's and hence are suitable for investors looking for higher rated or publicly rated instruments. In addition, while investors in SIV capital notes have traditionally been buy-and-hold fund managers and investment firms across Europe, the Middle East and Asia, both Sedna and Asscher are also looking to improve the secondary market liquidity of their single-A notes by issuing them in MTN format across a range of maturities, and by issuing notes with a pre-set margin rather than a share in the profits of the company.

Asscher should also be able to obtain a rating from Moody's as well as S&P on its Single-A notes as it maintains a higher percentage of triple-A assets than other vehicles in the SIV space. Most SIVs run with around 50-60% triple-A pools.

Swan expects that about 10% of Asscher's portfolio will be invested in triple-A cash CDOs, with around 40% invested in RMBS and the remainder in CMBS, student loan ABS, credit card ABS and bank paper. While Asscher is expected to reach around US$3bn in its first month, the final size of the programme depends on investor demand as SIV structures are typically open-ended, and continue to raise income notes over time. Cullinan, the previous SIV from HSBC has grown to over US$30bn in its first 18 months (making it the second largest programme).

MP

Muted response to high yield launch

Lukewarm buy-side reception for new benchmark indices

The International Index Company's launch of the iBoxx Euro High Yield index family (see last week's issue) has been met with a somewhat muted response by structured credit investors. Despite the fact that with multiple-contributor pricing the iBoxx consolidated price will be the most accurate reflection of the market, many of the high-yield fund managers contacted were still to pass judgment - with some yet to be even informed of the launch.

"It is simply another add-on to IIC's index family. It is still too early to say what kind of impact it will have on the market or the levels of uptake," says one French portfolio manager in Paris.

The new series of benchmarks comprise constrained and unconstrained indices, which are further broken down between crossover and non-crossover credits, totalling 193 bonds. Some investors believe the new index could be used as a tool to trade in its entirety but would be difficult on a single-name basis.

"Replicating any high-yield index is difficult because of the illiquid nature of the cash bonds. Unlike the iTraxx Crossover index, which references the CDS prices, the iBoxx underlyings are relatively illiquid," says Craig Abouchar, senior fund manager at Insight Investments in London and vice-chair of the European High Yield Association.

He believes the iBoxx indices will fill the benchmark needs for the traditional long only accounts and act as an alternative vehicle to gauge sentiment in the high-yield market. Others, though, favour the synthetic market.

"The iTraxx is the staple of the structured credit market. It's more liquid and transparent and so for us the Crossover is the perfect guide and better benchmark to follow," says a London-based high yield fund manager.

The iBoxx indices are entering into a market that contains more established and recognised benchmark indices, such as those provided by Merrill Lynch and Lehman Brothers. "The majority of the traditional long only benchmarked investors use the Merrill Lynch indices and I do not think that a lot of people will switch over from either them or Lehman as some mandates specify which benchmarks investors have to track. These indexes have a long history of data and research information, whereas you are starting from scratch with the iBoxx index," Abouchar says.

Despite early sentiment suggesting the new index family will not have much of an impact in the market, Abouchar feels the new series will find a place among some institutions. "There probably will be some firms that will standardise their investing of investment grade, government bonds, high yield and so on, across the iBoxx indices and for the contributing banks, it is effectively a common system to communicate and transact over," he adds.

HD

Emerging issues

This week, Søren Rump, md, EM & structured credit at Sydbank, answers SCI's questions

Søren Rump

Q: When, how and why did your firm become involved in the structured credit markets?
A: Sydbank got involved in the structured credit markets as an investor in CDOs in 2001. We mainly did this to get exposure to sectors of the credit market where we didn't have a lot of expertise – predominantly US and European leveraged loans and ABS.

In 2004 we took the step of combining our expertise as a manager of emerging market debt with our knowledge of CDO technology and became a manager of emerging market CDOs. Our first CDO was the first ever managed synthetic CDO done in emerging markets.

Today we manage three CDOs. Our recently closed CDO called Evolution EM CDO was the first CDO to reference emerging market local currency debt. We have also brought a fund-linked CPPI product to the market.

Q: In your view, what has been the most significant development in the credit markets in recent years?
A: The development of credit indices and standardised tranches on these indices has been impressive and extremely important for liquidity and volatility in the credit markets. The tight credit spreads that have followed from this has fuelled more and more client demand for structured credit investments – meaning that the market has been in a virtuous cycle.

Q: How has this affected your business?
A: Unfortunately the development of credit indices and standardised tranches has not really spilled over in to emerging markets, which is the main area of our asset management business. The liquidity of tranches of the CDX.EM Diversified index is extremely limited and the number of bespoke EM transactions has also been quite limited. Therefore the structured credit market has not been a particularly important driver of EM spreads.

Q: What are your key areas of focus today?
A: Within the structured credit market our key focus is to continue developing CDOs of emerging market local currency debt. Given how tight emerging market external debt generally is, local currency debt is the area of emerging markets that has the largest upside.

Q: What is your strategy going forward?
A: Very much more of the same – to continue to focus on the opportunities provided in the EM local currency sphere. I think this is a very promising area for EM CDOs to develop within and there are many interesting trades/structures to be done. In the future I expect to see more structures like Evolution and it will also be interesting to look at credit/FX hybrids.

Q: What major developments do you need/expect from the market in the future?
A: I would like to see a continued increase in the liquidity and transparency of CDOs/structured credit. At the same time, structuring banks and asset managers need to spend more time educating investors.

About Sydbank
Sydbank was founded by a merger in 1970 and is today Denmark's 4th largest bank with total assets of €12bn. Sydbank has over 2,100 employees and is mainly based in Denmark, but also maintains offices in Germany and Switzerland.

Asset management plays a key role in Sydbank's business strategy. Within Sydbank Asset Management, emerging markets and structured credit is the main focus area. A team of 12 people co-headed by Søren Rump and Morten Bugge manages US$6bn invested in emerging market debt and equities. Sydbank is also a leading manager of emerging market CDOs.

Arora arrives at Stanley

The latest company and people moves

Arora arrives at Stanley
Priya Arora, who left her position of head of cash CDO structuring at Calyon in December last year has resurfaced at Morgan Stanley as senior CDO structurer in the structured products group. She reports jointly to Dorothee Fuhrmann and Hiram Hamilton, co-heads of the CDO business in Europe. She joined the bank on Thursday January 11.

Varotsis becomes Codefarm non-exec
Codefarm, the structured credit technology company, has appointed Paul Varotsis as a non-executive director. His long view of the structured credit market will be invaluable in the development of new products and services, Codefarm says.

Over the past twenty years, Varotsis has become one of the most highly experienced and respected figures in structured credit. He initiated Chase Manhattan's credit derivatives business in London and then internationally in the 1990s.

Varotsis then joined Lehman Brothers and was head of CDOs at Barclays Capital. During that period he was European chairman of the ISDA committee that drafted the 2003 Credit Derivatives Definitions, and advised the Bank of England and other regulators to help develop a trusted credit derivatives marketplace. He is currently a partner at Reoch Credit Partners.

Citi shuffle
Priya Nair, head of European credit structuring at Citigroup in London, is soon to change roles within the bank. An official announcement as to what her exact position will be is expected by the end of the month.

Additional moves at Citigroup include Mike Diyanni, a member of Nair's European credit structuring team, who has transferred to the bank's structuring team within the global structured credit products group in New York. In addition, Varya Nuttall (nee Orlova), who left the emerging markets credit structuring team in October last year has moved over to Société Générale, as structurer, director, structured credit product group. Citigroup corporate communications declined to comment on the moves.

Rousseau moves
Emmanuel Rousseau, head of European structured credit derivatives trading at Bear Stearns has changed roles, moving off the desk, although his final destination is still to be confirmed by the bank. Market sources believe that one of the bank's existing structured credit traders has moved into Rousseau's role and that Sergio Ravich has been hired in a structuring and marketing capacity. Ravich was formerly head of fixed income, Spain at CSFB.

MSIM expanding
Morgan Stanley Investment Management is expanding its structured products group in the United States and will transfer Michael Nolan, part of the London structured products team, to New York for several months to aid the process. This is another step in the firm's expansion within structured products, coming soon after last year's setting up of a London team, of which Nolan was a founder member. He will help develop the firm's product offering that ranges from collateralised debt obligations, currency and commodity funds to portable alpha and liability-driven products.

Artun to Deutsche?
Deutsche Bank is believed to be about to hire Mehmet Artun as a structured credit marketer. Artun has been Commerzbank's London-based head of structured credit trading for only 6 months, joining the firm last July from WestLB, where he was executive director in structured credit trading.

His potential hire still leaves the role of head of European structured credit trading at DB up for grabs after Nitin Prabhu's transfer to head up the bank's European commodity sales and structuring business in mid-December. Deutsche declined to comment.

Rafferty switches roles at BNP Paribas
Although no official announcement has been made, BNP Paribas have moved Peter Rafferty from UK bond sales into credit hedge fund sales. This follows the departure of Caroline Abensour to Dresdner Kleinwort's credit hedge fund sales team in December.

Centerbrook appoints new ceo
Centerbrook Financial, the credit risk products subsidiary of CharterMac, have appointed Nicholas Mumford as ceo. Mumford is succeeding Robert Maum, who left Centerbrook at the end of December to pursue other opportunities.

Prior to joining Centerbrook, Mumford was an md at IXIS Capital Markets, where he was responsible for managing the firm's credit business. Prior to joining IXIS' predecessor in 1999, Mumford was with both Lehman Brothers and Citibank, where he started and managed the credit derivatives desk.

Black joins McDermott Will & Emery
The London office of law firm McDermott Will & Emery has appointed Steven Black as a partner in its securitisation and structured finance group. Black will be working closely alongside existing structured finance partners, Paul-Michael Rebus and Nick Terras, both of whom joined the London practice during 2006. Black has extensive experience with CDO and CLO structures representing arrangers and managers.

T-Zero hires Harrington
T-Zero has hired George Harrington as coo to expand its senior management team. Harrington joins T-Zero in its New York office after spending more than two years at Thomson Trade Web, where he was most recently product manager of its credit default swap index trading business.

Harrington has expertise working with dealers and buy-side market participants and has experience working to develop and implement system enhancements. In addition, he has held senior roles as an analyst and accountant with various financial organisations since the launch of his career in 1994.

Moody's reorganises US derivatives group
To manage what it describes as the tremendous growth experienced in its US derivatives business, Moody's has increased resources and reallocated the responsibilities of its US derivatives management team. Notably, Jonathan Polansky returns as a team md.

Polansky joined Moody's in 1997 as an analyst in the US derivatives group. In 1999, he left Moody's to co-head the structured products group at Triton Partners. He rejoined Moody's derivatives group in 2003 to manage the enhanced monitoring service and to develop and manage the CDO surveillance group. In April 2005, he was promoted to team md of Moody's US asset backed commercial paper team.

The business line responsibilities within Moody's US derivatives group have been reallocated as follows:
• William May – cash flow, synthetic & hybrid CLOs/CBOs, SME CLOs, emerging market CDOs and project finance CDOs
• Yvonne Fu – structured finance operating companies, market value CDOs, trust preferred & REIT CDOs and catastrophe bonds
• Eric Kolchinsky – cash flow, synthetic & hybrid ABS CDOs and muni CDOs
• Jonathan Polansky – CDO surveillance, synthetic correlation trades & esoteric synthetics, CDO repacks and structured notes.

HD & MP

ILS market reawakened by Vita Capital

A round up of this week's structured credit news

ILS market reawakened by Vita Capital
Swiss Re has become the latest issuer to take advantage of the revision in September of Standard & Poor's rating methodology which enables unwrapped insurance-linked securities (ILS) to reach investment grade ratings. Previously, catastrophe bonds were typically rated in the double-B/single-B area.

The US$700m Vita Capital III transaction comprises seven series of notes: €100m five-year double-A minus paper, together with single-A rated notes split between US$50m five-year, US$90m four-year and €30m four-year maturities. There are also three series of notes wrapped up to triple-A: US$100m MBIA-guaranteed four-year Class As; US$100m FSA-guaranteed five-year Class As and US$50m Class Bs; and €55m CIFG-wrapped four-year Class As and €55m Class Bs.

Maren Josefs, credit analyst at Standard & Poor's, says: "This issue shows that there is still strong investor appetite for what is effectively a new asset class. The investment grade ratings make the notes attractive to a broader investor base. Market conditions suggest that this demand will continue throughout this year, as new investors continue to enter the insurance-linked securities market."

Investors in the transaction are at risk from an increase in mortality that exceeds a specified percentage of a predefined index of age and gender-weighted mortality rates on an accumulated basis over four years from January 1 2007 to December 31 2010 (or over five years from January 1 2007 for the five-year maturities). The index is defined on a rolling two-calendar-year basis, and the probability of a loss attaching and the magnitude of the loss in principal depends on the extent to which the index for any two-consecutive-year period exceeds the respective attachment point for each class of notes.

The index value for 2005 has been chosen as the reference year against which index values corresponding to future measurement periods are measured. The index is constructed using published population mortality rates from official reporting sources in the US, the UK, Germany, Canada and Japan, weighted by country, age and gender.

The transaction involves the total return swap counterparty converting the investment return earned on the permitted investments to the reference rate consistent with that of the notes. Together with the proceeds from the total return swap, the protection premiums are used to make the scheduled coupon on payments to the noteholders. If the index rises above the attachment point for the respective class of notes, the assets held in the trust are sold to fund the payment to Swiss Re, and the principal amount of the notes is reduced.

Derivative Fitch updates RAP CD
Derivative Fitch has launched an updated version of its Risk Analytics Platform for Credit Derivatives (RAP CD). Enhancements in RAP CD version 1.2 include daily mark-to-market pricing and risk analysis and the platform's full support for CDO-squared structures.

"We recognise that in today's market investors sometimes have a basic model for vanilla synthetic CDOs. However, investors rarely have models for more exotic deals, even though it is often these deals that are more frequently traded. Their challenge now involves being able to analyse all market risk exposures for all deal types, including the more exotic ones, on a portfolio basis", says James Wood, md at Derivative Fitch and co-head of RAP CD.

"Derivative Fitch is therefore addressing the market's need for transparency through a dynamic series of releases with future versions to include support for variable subordination, long/short CDOs, zero coupon CDOs, CPPI and CPDO. We intend to have coverage of many of these products in our next release scheduled for March 2007," he adds.

RAP CD v1.2 also includes the addition of portfolio reporting and an improved graphical interface. Portfolio reporting offers the ability to evaluate the risk of a single name to an entire portfolio of synthetic CDOs.

The v1.2 CDO-squared analytics include two models from Reoch Credit, whose credit derivatives analytics business was acquired by Derivative Fitch in July 2006. The first approach uses the supertranche method in which a CDO-squared is treated as a closest equivalent CDO.

The second model explicitly takes into consideration the base correlation skew of each inner CDO tranche and also that of the outer CDO tranche. This is based on David Shelton's proxy integration approach and David Li's concept of overwriting a copula-modelled loss distribution with one implied from the tranche market.

S&P releases December SROC numbers
Standard & Poor's has published its latest monthly global SROC (synthetic rated overcollateralisation) report. A further 74 tranches have been added since the November report and there have been 129 rating actions in that time, the agency says.

The report provides SROC and other performance metrics on 2,341 individual CDO tranches. This represents publicly rated transactions in Europe, North America, Japan, Hong Kong, Singapore, Australasia and Canada, for which currently outstanding ratings were assigned on or before November 30 2006.

The analysis is based on portfolio compositions as reported to Standard & Poor's to the end of December 2006. Class rating information is also correct to that date.

Transactions with a SROC figure below 100% are being reviewed and appropriate rating actions will be published later this month.

CMA data now available through Sophis
Sophis has developed an interface to enable users of its VALUE buy-side platform to directly access CMA DataVision, which collates the real-time credit price discovery services and data providers' same-day closing CDS prices.

CMA DataVision provides data for more than 2,000 single name CDS, indices and tranches. Daily updated CDS spreads based on real-time indicative quotes are combined with bond prices, terms and conditions and ratings data. CMA's data consortium of 30 buy-side firms continuously provides average spreads based on observed quotes.

MP

Trading ideas - high yield prescription

Dave Klein, research analyst at Credit Derivatives Research, looks at a long butterfly trade on Rite Aid Corp

We most often trade a single-name credit curve with a view towards an improving or deteriorating outlook. This view translates into an expected flattening or steepening of part of the curve. However, if a company is trading wide enough, the curve may flatten rather than steepen making it difficult to predict curve movement even if we have a deteriorating view of the credit.

In this case, we look for dislocations from fair value of parts of the curve and construct trades to take advantage. Butterfly trades are good candidates because they are hedged against both parallel shifts and twists of the credit curve.

Although Rite Aid Corp. (RAD) is not the most liquid of CDS names, its volatility in pricing levels across the term structure has triggered a number of our fair-value models. We find that a 3s-5s flattener has significantly positive carry and a 5s-7s steepener has excellent roll-down. Both positions are trading wide to fair value. Furthermore, when employing our butterfly fair value model (which judges fair pricing for five-year risk given threes and sevens), we find that RAD's five-year sits wide of fair value compared to its wings.

Given the reasonable liquidity across the term structure of RAD and the positive economics of the package, we feel a duration-hedged 3s-5s-7s butterfly offers good relative value while being generally hedged against parallel or twist moves in the curve.

Hot wings
To estimate the fair value of the 3-5-7 butterfly, we model the 5 year bid level as a linear combination of the 3 and 7 year offer levels. This is undertaken across the universe of credits whose 3's and 7's are close in value to RAD, and the output of the regression is shown in Exhibit 1.

Exhibit 1

For each issuer, we produce an expectation of the five-year CDS bid level. This is compared to the current market five-year CDS bid level in Exhibit 1. The regression has an R-squared of over 99% and so we can be quite confident of its findings.

The Exhibit clearly shows that RAD is trading (y-axis) wider than our fair value model would imply. If the modelled bid level is below the 5 year market level, we expect the 3-5 leg to flatten or the 5-7 leg to steepen.

This is an indication that the five-year body is trading cheap to the three-seven wings and a butterfly is potentially applicable. In order to consider the potential driver of any relative-value pick-up we next look at the individual legs of the trade – a 5s-7s steepener and 3s-5s flattener.

Nice legs
Our fair value butterfly model has shown that there is some relative-value across the term structure of RAD but we analyze each leg below to discover where the expected change might come from.

The fives-seven curve is slightly flat to fair value, as seen in Exhibit 2. Our fair value curve model for US credit shows that the current levels for a steepener (which is the equivalent leg) are just above the solid black line (as seen by the green square). Although this leg is very close to fair value, its positive roll-down strengthens our confidence in this leg.

Exhibit 2

The threes-fives curve, on the other hand, is significantly steep to fair value. Exhibit 3 shows that compared to its peer groups, RAD sits over 50 basis points too steep for a flattener (our butterfly's equivalent leg). Whatever the reason for the steepness, our flattener leg is extremely steep to fair value and offers significant relative value as well as (minimally) positive roll-down.

Exhibit 3

This tends to make us believe that from a technical perspective we would expect the threes-fives leg to be the bigger driver of the butterfly's performance.

Before moving on to the economics of the trade, we consider the behavior of RAD within our 3-5-7's Fair Value model over time. Exhibit 4 charts the difference between market 5 Year bid and the model expected value. First, we note that, as shown earlier, our model indicates that the 5's are trading above fair value. Second, we see that, over the past few months, RAD's 5's have traded both above and below the model's fair value. Thus, we do not see any credit-specific trend that keeps this name trading away from the model.

Exhibit 4

Trade economics
Three of our models indicate that the trade is warranted, but we must consider the actual economics of the trade. The carry of the DV01-neutral butterfly is often more important than the expected behavior of individual legs of the trade. In the case of our DV01-hedged butterfly, we find around 75 basis points of positive carry, which along with around 23 basis points of MTM relative value we would expect provides some significant upside, given the nature of the butterfly trade.

Additionally, over the next year we would expect to gain about 95 basis points of MTM roll-down, primarily due to the flatness of the 5-7's leg, even if the RAD curve did not move. Taken together, these gains more than offset the expensive 50-60 basis points of bid-offer spread we must cross on our trade.

The butterfly trade is hedged (for small moves) against both parallel shifts as well as twists to the credit curve. Although we have a deteriorating fundamental view on the credit, we are exploiting the relative 3-5-7 levels to pick up carry, roll-down and relative-value.

Risk analysis
The butterfly is constructed to be duration- (and slope-) neutral with 3-5-7 weights of 1.3-2.0-0.9 based on the respective DV01s. Given how wide the credit is trading, we prefer to be hedged against both shifts and pivots. The position has a minimally short default risk position (we are long protection in US$22mm and short protection in only US$20mm). The following table provides a more complete picture of the expected P&L impact of shifts and twists in the term structure of RAD.

As is clear, the parallel shifts and curve twists do generate some P&L (due to convexity differences) but these are minimal for moderate movements and the carry earned (and MTM relative-value realization and roll-down) outweighs this.

The trade is significantly positive carry – given the current levels. This carry cushion protects the investor from any short-term mark-to-market losses.

In general, the biggest challenge to a butterfly is the ability to unwind the trade profitably. With three legs, we must cross three bid-offer spreads. As stated earlier, we face about 50-60 basis points of bid-offer spread that is offset by healthy carry, MTM roll-down and MTM relative value gain. This is the most significant challenge to the trade.

Entering and exiting any trade in these maturities carries execution risk, but this is not a major risk as RAD has moderate liquidity in the credit derivatives markets. We see quite a bit of volatility in indicative levels for RAD. We only recommend the trade if the 3s-5s-7s levels are close to those stated here. If the 3s-5s flattens, the trade will lose much of its profit potential.

Liquidity
Liquidity is a major driver of any longer-dated trade – i.e. the ability to transact effectively across the bid-offer spread in the bond and CDS markets.

As stated earlier, RAD has moderate liquidity in the CDS market. We'd prefer to see lower bid-offer spreads, but are well-compensated with the healthy carry the trade earns.

Fundamentals
This trade is technical in nature and is not based on any fundamental outlook. Even without a fundamental view on the credit, we would be comfortable putting on the trade. That said, it is always good to keep an eye on the fundamentals of the underlying credit of even the most technical trade.

Summary and trade recommendation
We head deep into high-yield land to consider RAD's CDS curve. RAD has rallied significantly from its wides of about a year ago.

However, its aggressive expansion plans have pressured this credit. While we maintain a deteriorating fundamental outlook, the steepness of the 3s-5s CDS curve suggests an interesting relative-value play. Using our 3s-5s-7s fair value model, we find the 5 year CDS to be wide of fair value given its 3 and 7 year levels.

Given the steepness of the 3s-5s leg, we view the butterfly as a strong trade candidate. Excellent carry, roll-down and the potential profit of a return to fair value strengthen the economics of the trade. Given our deteriorating fundamental outlook and the current wide levels for the company, we feel this is the safest relative-value trade available for RAD.

Buy US$13mm notional Rite Aid Corp. 3 Year CDS protection at 225bp and

Sell US$20mm notional Rite Aid Corp. 5 Year CDS protection at 395bp

Buy US$9mm notional Rite Aid Corp. 7 Year CDS protection at 470bp to gain 75 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 2

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

The Greeks

Spread sensitivity – Delta

Delta describes the first-order sensitivity of the position to changes in spreads. The delta is defined as a ratio of mark-to-market changes in the equity tranche and the underlying index in the event of a 1bp change in the underlying. We explain in detail the behaviour of the equity tranche with respect to spread movements in our publication Delta-hedging: A primer, 12 April 2006.

What is at risk in equity tranches
The key to understanding why delta should differ for equity tranches structured with different coupon schedules is appreciating just what is at risk in each case. A change in the index or average portfolio spread level essentially indicates a change in the market implied default intensity (and hence expected loss), and so those equity tranche structures that put the most at risk in the event of default will be most sensitive to changes in the index/aggregate portfolio spread.
? All-running - Following our discussion of default time sensitivities, it should be clear that the structures that put most at risk when defaults occur are those paying all coupons on a running basis. This is because coupons will be lost/diminished in the event of default.
? Funded zero-coupon and unfunded all-upfront - Correspondingly, when all returns are traded upfront (funded zero-coupon equity or unfunded all-upfront tranche), all coupons are essentially 'locked in' at issue and so there are no coupons 'at risk' per se.
? Standard – The standard equity tranche lies in between these two extremes.
The amount of principal (as opposed to coupons) at risk does not depend on the coupon structure. As a result, of the equity trades that put principal at risk, those that trade all returns on an upfront basis have the lowest deltas, and those that trade all returns by way of running coupons have the highest deltas.
? IO strips - The only structures that do not put principal at risk are the IO strips. In the event of default, whilst tranches that do trade principal protection may see some future coupon payments lost or diminished, their main source of MTM loss will come from the fact that the outstanding tranche notional is diminished (especially at shorter maturities). As a result, at shorter maturities IO strip deltas are significantly lower than those of the other tranches, although this effect does diminish as time to maturity increases and the relative size of coupons with respect to principal increases.

Deltas as a function of time to maturity
As explained in Delta-hedging: A primer, 12 April 2006, we expect equity tranche deltas to fall as time to maturity increases. This is because the portfolio expected loss point moves further up the capital structure (and eventually sits above the detachment point of the equity tranche). As a result, the equity tranche is not so sensitive to spread changes.

Second-order spread risk – Gamma

As we explained in our report Gamma: The story of convexity, 23 May 2006, gamma explains the rate of change of delta for a given hedged trade position, or correspondingly it can be thought of as the convexity of the MTM function of the hedged position as a function of aggregate shifts in the underlying spreads in the basket.

Whilst a MTM position of a delta-hedged trade is theoretically insensitive to small shifts in the index spread, the same is not true for larger aggregate spread movements. Delta-hedged long equity trades are long in gamma, or have a positively convex MTM profile as a function of aggregate spread movements of CDS names in the underlying basket.

This means that delta-hedged long equity tranche positions make MTM gains in the case of widespread shifts in the spreads of the underlying CDS names – regardless of whether these shifts signal credit improvement or deterioration. For a full explanation of this, we refer readers to the above report.

Delta-hedged long equity tranche
An investor can therefore take advantage of this positive convexity by entering into a delta-hedged long equity tranche position – and the same is true for each of the alternative equity structures that we have considered in this report. In this section, we assume that the hedge instrument used is the index, although the ideas are transferable to cases where more senior tranches are used to hedge.

The reason that such convexity effects occur essentially stems from the fact that delta is not constant – it changes with respect to the underlying CDS spreads, leaving the position either under- or over-hedged, depending on the direction of spread movements. Understanding what makes delta change therefore allows us to understand why the different equity tranches exhibit differing degrees of convexity when delta-hedged.

High gamma positions
? All-running – Delta essentially has the highest sensitivity to correlated spread movements when all coupons are traded on a running basis. This is because when the underlying CDS spreads all increase, the possibility of certain coupon and principal payments being lost becomes more and more certain. As this certainty increases, the incremental effect of a widening in underlying spreads on this certainty becomes smaller – and hence the tranche value becomes less sensitive to an aggregate increase in the spreads of the CDS names underlying the trade. As a result the delta falls.
? Funded zero-coupon and unfunded all-upfront – A similar effect occurs in the zero-coupon (funded) or all-upfront (unfunded) case, however to a lesser extent because in these cases it is only the certainty of principal losses that can change (the coupon is paid upfront and therefore guaranteed).
? In sum, the delta-hedged equity tranche has a higher gamma (and hence exhibits more convexity) when more of the return is paid on a running coupon basis.

IO strips
As might be expected, the IO format embodies the lowest variation in delta for spread movement since there is only interest income at risk and the investor is not subject to having to make additional payments/lose principal in the event of defaults.

Examples
The convexity profiles as a function of correlated shifts in the underlying spreads are traced out for all four funded and four unfunded equity tranche cases that we consider in the charts below. Here we assume a seven year maturity – with a shorter tenor (e.g. five years) we find that there is less heterogeneity between the two 'extreme' cases – i.e. zero-coupon/all-upfront and all-running.

Similarly, as time to maturity increases the convexity profiles become more and more different from one another. As the IO strip can be expressed as standard minus zero-coupon (funded) or standard minus all-upfront (unfunded), this increased divergence sees IO convexity increase with time to maturity. The funded IO strip is also more convex than the unfunded equivalent as the LIBOR portion of the coupon stream also exhibits positive convexity with respect to the index.

Clearly, this difference in the gamma performance of the equity structures means that investors can engage in a gamma trade in ways more suited to their needs. If they want to limit their need to re-balance their delta-hedges when investing in equity tranches, then the zero-coupon structure is advisable.

On the other hand, when an investor aims to gain maximum exposure to the potential positive gamma convexity effect, then the all-running format is the most appealing. Certainly such a choice is also a function of carry considerations among other factors.

Managing other risks

The advantage to equity investors provided by the positive gamma effect is compensation for the idiosyncratic risk that equity investors face. Also, equity tranches are long in correlation but the effect once again varies according to the coupon structure.

Idiosyncratic risk
For the various equity structures detailed in this paper, idiosyncratic risk exposure varies. Idiosyncratic risk is risk due to a certain credit or a small number of credits that manifests itself when those spreads move independently of the basket. In this way, if a spread blows out due to an event or if it defaults, then the equity tranche investor is most heavily exposed to such a move, as it is the equity investor who absorbs the first principal losses.

There are two key points to consider when comparing idiosyncratic risk effects across equity tranche structures:
? Coupons at risk – Structures that pay coupons on an all-running basis have the largest exposure to idiosyncratic events. This is because both the running spread and principal (due in the event of a default) are at risk when an idiosyncratic event materialises. In the zero-coupon (funded) or all-upfront (unfunded) case there is only principal at risk, as the entire coupon is paid upfront and so is guaranteed.
? Across maturities, the idiosyncratic risk decreases for all formats of the equity tranche as time to maturity increases, except for the all-running structure. In the zero-coupon case, the idiosyncratic risk falls because the upfront payment by the investor falls and it is only this that is at risk. The coupon paid on the all-running tranche rises with time to maturity and so the idiosyncratic risk does not fall in that case, as we have the offsetting effect that there is more to be lost in the way of coupons.

iGamma

iGamma is essentially the effect of idiosyncratic risk on a hedged trade position. In the case of delta-hedged long equity positions, the risk of a single spread moving wider independently of the basket is described by iGamma. As explained in Gamma: The story of convexity, 23 May 2006, delta-hedged long equity tranches are short iGamma, reflecting their negative idiosyncratic risk exposure. A jump-to-default can be thought of as an extreme case of the iGamma effect, and therefore our analysis here suggests that the negative iGamma effect is larger when more of the return is traded in the form of running, rather than upfront, coupons.

Interest rate risk
The MTM value of an equity tranche will change with the forward interest rate curve – and once again this effect depends on the traded equity structure. Funded tranches also react differently to unfunded tranches. In this analysis we assume no correlation between interest rate and credit curve shifts.

Unfunded tranches and MTM risk
All unfunded tranches that put principal at risk exhibit a positive MTM interest rate sensitivity. The key here is to note that, in the unfunded case, the investor does not wait to receive back a redemption value at maturity, rather he pays out for defaults along the way. In other words, he has some cash in-flows and also potential cash out-flows. Therefore, whilst an up-shift of the interest rate curve will reduce the PV of his expected coupon stream, it will also lower the PV of the 'loss leg' on which he must pay out.

To see this, we have to look at the duration of the two legs of an unfunded trade. We begin with the all-running coupon case. The loss- and coupon legs have equal PVs at the outset of the trade; however, the duration of the fixed leg is lower than that of the loss leg. This is because coupon payments are at their highest before any defaults occur and their lowest after defaults occur.

As a result the weighted average time to coupon cash-flows is lower than the weighed average time to payments made out on the loss leg. Therefore, the protection seller will gain from rising interest rates as the PV of the loss leg drops by more than the PV of the coupon leg.

Moreover, when the running coupons paid are larger, the coupon leg is hit more heavily by increased discounting, and hence the standard tranche benefits more from an increase in interest rates than the all-running tranche, which pays larger coupons.

The all-upfront coupon case is the most sensitive to changes in interest rates, as there is only the loss leg whose value is sensitive to interest rates. It is in this case therefore that the investor gains most from an up-shift in the forward interest rate curve. In general, the message for unfunded tranches is that smaller coupons imply a larger positive reaction to interest rate increases.

In unfunded IO strips, the future cash flows involve a fixed stream of payments being made to the investor. They therefore have a negative MTM sensitivity to upward shifts in the interest rate curve as they are more heavily discounted.

Funded tranches
A floating rate coupon paying funded tranche that puts principal at risk can be thought of from the investor's point of view as the equivalent unfunded tranche plus a risk-free par floater that pays LIBOR. To see this we note that selling funded protection on an equity tranche has essentially the same economics from the investor's point of view as selling unfunded protection and investing an amount equal to the tranche notional in risk-free collateral himself. Moreover, the LIBOR par floater has very little interest rate sensitivity, and so the funded standard and all-running tranches have much the same positive interest rate sensitivities as the unfunded standard and all-running cases.

The funded zero-coupon tranche is not a floating rate note however, and is therefore the 'odd one out' among the trades we consider. The concept in this case is quite simple, however. The only future cash flow to be made is that the investor is returned par minus net losses at maturity. An increase in interest rates then sees this payment more heavily discounted, and as such the zero-coupon tranche has a negative interest rate sensitivity.

Treating the funded IO strip as the difference between the standard and zero-coupon funded tranches, we see that it benefits from an up-shift in the interest rate curve.

There are two key conclusions we can draw from this:
? All equity tranches that put principal at risk benefit in MTM terms from a rise in the interest rate curve, with the exception of the zero-coupon (funded) tranche.
? Funded IO strips benefit from a rising interest rate curve, whereas unfunded lose.

Buy-and-hold investors
As we explained earlier, the funded tranche market has been largely used by buy-and-hold investors. Whilst the MTM effect of interest rate curve shifts we have just discussed will no doubt be primarily important to leveraged accounts, buy and hold investors will also be interested in how changing interest rates may affect their yield to maturity. Positions may have to be liquidated before maturity, or MTM valuation may be required, and therefore MTM issues have to be addressed when investing in equity tranches.

Correlation risk

Equity tranches are long in correlation (i.e. they benefit in MTM terms from a rise in the equity tranche correlation assumption) as more highly correlated portfolios have higher probabilities of extreme events. That is to say, the probability of no losses occurring increases with correlation which benefits equity tranche investors. The possibility of very large portfolio losses also increases on the other hand, however this does not affect equity investors so much as they are only exposed up to their detachment point.

Correlation and coupons
The extent of this equity tranche correlation sensitivity also depends on the coupon structure, for reasons we have already discussed. As we saw when discussing delta, all-running coupon structures are most sensitive to a change in the level of expected losses because they have most at risk (i.e. principal and coupons). Increasing correlation essentially implies a decrease in equity expected loss, and so all-running structures benefit more when correlations increase than all-upfront/zero-coupon tranches. In the same spirit, they also have a larger downside when base correlation decreases.

Leveraged accounts and trading gamma
As in the case of gamma trading, if investors want equity exposure without much correlation risk, then zero-coupon equity can be appropriate. On the other hand, if investors want to take advantage of a long correlation view then they can use the all-running structure to enhance the benefit of an upward move in correlation.

Time decay

All-running tranches benefit the most
Time decay and curve roll-down effects are important considerations for tranche investors, in particular, equity investors. If market conditions remain stable with interest rates unchanged and no defaults or variations in underlying spreads, all-running formats gain the most from this time decay effect. This is because tranches that pay coupons can be thought of as the sum of the zero and the running coupon stream.

If time passes with no defaults then the risky coupon stream benefits in MTM terms from having survived this risky period untouched. The all-running investor therefore benefits more from the passage of time than the zero investor.

This result also satisfies the notion that riskier products should offer more return if no defaults occur. As we have seen, all-running tranches stand to lose the most in case of defaults.

Funded vs. unfunded tranches
Funded tranches also have higher rates of time decay than unfunded tranches. This is because, whilst they have the same curve roll down from a credit risk point of view, the funded tranche investor expects to receive back the outstanding tranche notional at maturity, such that there is an extra yield-curve roll-down effect.

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