Contents

CDR Liquid Index data as at 25 June 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.

ABX finds its level

Index market moves on as dealers discuss next roll

The ABX has plumbed new lows since last week, but traders now expect it to drift until further strong positive or negative news emerges with July's index roll coming in to focus. The index market appears to have moved on from the issues surrounding Bear Stearns' hedge funds (see SCI issue 44), but the mixed messages in this respect are still impacting the structured finance (SF) CDO market.

As one ABX trader observes: "We seem to be past the Bear Stearns hedge fund saga now, but on Monday remittances were weak as the market had expected – as was the Case-Shiller house price index, which continues to be negative. Having said that, once the ABX fell through its previous floors, it seemed to find support again. So there was a slight rally Tuesday on very thin trading."

He continues: "Basically, the market is waiting for more news one way or the other before it goes either up or down from here. In the meantime, the index seems to have found a new base level – of around 57 for the triple-B minus 07-1 – so we are in the same position as before, where everybody is as short as they want to be and nobody wants to go long."

In the meantime, dealers have held discussions over the construction of the ABX and TABX after their next rolls. "Disappointingly ABX will continue to only comprise 20 bonds and the servicer concentration level has had to be relaxed – in recognition of the fact that in the end there will probably only be five servicers in the US due to consolidation," says one.

Similarly, TABX will continue to be based on only 40 securities from the two most recent indices. There was talk about retranching the instrument to make it more closely aligned to the HY index, but the majority of dealers concerned voted against such a change.

At the same time, conflicting stories continued to appear surrounding the Bear Stearns Asset Management (BSAM) High-Grade Structured Credit Fund and BSAM High-Grade Structured Credit Enhanced Leveraged Fund. Last Friday, Bear Stearns announced that it had offered to provide up to US$3.2bn in secured financing to the High-Grade Fund in the form of a collateralised repurchase agreement.

The facility will enable the fund to replace current secured financing, improve liquidity and facilitate an orderly de-leveraging of the fund in the marketplace, the bank says. Meanwhile, BSAM will continue to work with creditors and counterparties of the Enhanced Fund to reduce leverage in an orderly manner and improve liquidity.

However, Monday saw rumours circulating that Bear has been able to sell down some of the underlying assets in the fund, resulting in it having to pay up around US$1.6bn instead. Nevertheless, analysts remained cautious, given the still substantial amount involved and the unclear picture concerning the Enhanced Fund.

Consequently, a report from JP Morgan's CDO research team in New York notes that pipeline activity in the SF CDO space has slowed considerably as the situation in the HEL ABS space evolves.

The US funded high-grade SF CDO pipeline stood at just over US$20bn about a month ago. Since that time, US$18bn has priced and the current pipeline figure is US$3bn. "We expect events surrounding warehousing liquidations last week to further slow, if not halt entirely, the new issue market," the report concludes.

MP

Cat bond flood

Three major deals launched, including first cat bond-squared

Significant catastrophe bond issues covering the world's major time zones have come to the market in the past seven days: the debut from a UK insurer; the inaugural issue from the largest US programme; and perhaps most notably a brand new structure.

The US$140m Fusion 2007 is partly sponsored by Kyoei Fire & Marine Insurance Co (KFMI) and arranged by Swiss Re Capital Markets. It is the first catastrophe bond to reference an existing cat bond as a trigger.

Through this transaction, Swiss Re has bought fully collateralised retrocession protection against high severity losses incurred from typhoons in Japan in connection with a reinsurance contract it entered into with KFMI, and against high severity losses incurred from earthquakes in Mexico. For losses incurred in Mexico, the notes are only triggered if there has been an event under the corresponding class of the existing CAT-Mex catastrophe bond.

S&P has rated the bond's US$30m Class A (which pay a coupon of 800bp over Libor) and US$80m Class B notes (600bp) single-B, and the US$30m Class C notes (300bp) double-B plus. The Class A notes are exposed to the occurrence of a typhoon in Japan and to a loss under the CAT-Mex Class A and B notes, the Class Bs are exposed to the Japanese typhoon trigger and a loss under the CAT-Mex Class B notes, and the Class Cs can only be triggered by a loss event under the CAT-Mex Class B notes.

Meanwhile, Brit Insurance has become the first UK listed insurer to hedge its natural catastrophe risk by issuing a CDO of natural catastrophe risk. The US$200m three-year deal was brought by ABN AMRO and is the first issue from a programme known as Fremantle Ltd.

Rated by Fitch and Moody's, the transaction comprises US$60m AAA/Aa1 Class A notes, US$60m BBB+/A3 Class Bs and US$80m BB-/Ba2 Class Cs.

Under the terms of the catastrophe swap, Fremantle will pay BIL if – during the next three years – the number of qualifying events from a basket of eligible events exceeds a designated threshold. The events covered are UK windstorm, European windstorm (excluding the UK), Japanese typhoon, Japanese earthquake, California earthquake, New Madrid earthquake and US hurricanes.

The programme was developed in conjunction with Guy Carpenter & Co and Risk Management Solutions, which also acts as the event calculation agent.

James Gorman, structured credit marketing at ABN AMRO, comments: "Last year we launched Bay Haven, the market's first CDO of natural catastrophe risk, to very strong demand. Fremantle has built on this success and, once again, we have found significant demand resulting in heavy oversubscription across all three tranches. The notes were placed to a broad range of European and US investors, highlighting an increasing desire to diversify into rated insurance-based product."

Last week also saw the initial issuance from the largest-ever cat bond programme via AON Capital Markets, Citi and Merrill Lynch. The three-year deal from Merna Re totalled in excess of US$4bn.

Rated by Fitch, the capital structure consists of US$500m triple-A rated Class A notes; US$1.2bn double-A plus Class Bs; US$850m single-A minus Class Cs; US$690m double-B Class Ds; and US$780m single-B plus Class Es.

The transaction transfers to the capital markets a portion of the sponsor's – State Farm Mutual Automobile Insurance Company – exposure to natural catastrophe losses in the US and Canada, including hurricane, earthquake, tornado, hail, winter storm and brush fire. The structure features an indemnity-based trigger and provides cumulative aggregate excess of loss protection.

MP

Cairn and JPM close first fully managed CPDO

Deal meets with investor demand, but overall bottleneck in the sector remains

On Monday, 25 June, JP Morgan closed the first fully managed rated CPDO, managed by Cairn Financial Products. Cairn CPDO I was placed globally and attracted a diverse group of investors, according to the bank, but had been subject to 144A rules.

The transaction is structured around the iTraxx and CDX indices to take advantage of their liquidity, but can then be adjusted to become a bespoke portfolio based on Cairn's credit view through the use of single name CDS. Cairn can take long and short positions in single names, replicate the index through single names, skip a roll and change leverage multiples.

Rated by Moody's, the notes – all due in 2017 – were issued in three series: €68m Aaa rated Series A1-E1 (with a coupon of 130bp over Euribor); US$6m Aaa Series A1-U1 (130bp over Libor); and US$3.3m Aa2 Series B1-U1 Turbo (210bp over).

The notes have an early redemption feature designed to increase the chances of cashing in early for investors. The cash-in mechanism is such that indexation to the reference portfolio ends whenever the transaction net asset value is higher than the present value of all future liabilities of the issuer.

Coupons for the three series of notes step down to a floating benchmark minus 7bp after predetermined dates. After five years in the life of the transaction have elapsed (if the cash-in event occurs before year five) or at the next following coupon date (if the cash-in event occurs after year five), the level of coupons steps down. The issuer then has the right to call the notes – a right that is delegated to the manager.

Base case leverage for Cairn CPDO I starts at around seven times for the Aaa notes and 10 times for the Aa2 notes. The leverage mechanism is designed to take advantage of the structural positive gamma of CPDOs, hence leverage steps up with spreads. However, Cairn has the flexibility and discretion to choose, within certain limits, the exact leverage it considers appropriate based on its credit market view subject to a maximum leverage of 16 times.

Cairn and JP Morgan say they have worked extensively over the last six months to build upon the technology of the first generation of CPDOs to incorporate full management features. As Moody's analysts observe: "Cairn CPDO features many innovations to allow the manager to have discretion over the algorithm and to actively trade the underlying composition."

But while such an application has brought its rewards, there remains a sense in the market that the progress of the more complex CPDO deals is being slowed by their innovation.

The deal that had been touted as the first fully managed CPDO (and may yet be heralded as the first fully, fully managed deal, referencing as it does an entirely bespoke portfolio) – ABN AMRO's Degas CPDO managed by Fortis (see SCI issue 36) – could still be as much as a month away from pricing.

In addition, still to close are Cepheus CPDO from Rabobank, managed by Rothschild & Cie Gestion, and Barclays Capital's Alhambra – managed by Deutsche Asset Management (SCI 44). Meanwhile, Deutsche Bank's MACROS transaction – to be managed by Natixis – is probably even further away.

MP

Permacap problems

Listed structured credit vehicles continue to suffer

Evidence of the severe impact that events in the US sub-prime market are still having on permanent capital vehicles emerged last week. The planned Everquest IPO was pulled and Queen's Walk Investment Ltd confirmed a net loss in its year-end results.

On Monday, 25 June, Bear Stearns Asset Management filed a withdrawal of registration statement with the SEC for its proposed IPO of the Everquest permanent capital vehicle (see SCI issue 39). The company declined to comment, but the move was regarded by market participants as far from unexpected given the close structural links between Everquest and Bear's hedge funds (see this week's lead news story for more on these).

Also on Monday, Cheyne Capital's beleaguered Queen's Walk vehicle (see SCI issue 37) announced its preliminary results for the year ended 31 March 2007. The company confirmed a net loss for the year of €67.7m, principally as a result of fair value adjustments made to investments in 2007. This equates to a loss per share of €1.67.

In a statement, the company went on to say that fair value write-downs in the fourth quarter totalled €108.4m. These write-downs resulted principally from significant developments affecting the UK and US mortgage markets.

Since 31 December 2006, the company has sold three US investments and four UK investments for aggregate sale proceeds of approximately €116m. As the 31 March 2007 fair values of the assets sold were determined on the basis of their sale prices, these sales have not reduced the company's net asset value post year-end and are fully reflected in the announced €7.24 NAV figure.

Proceeds from asset sales have been applied to reduce indebtedness or to increase available cash. The company's leverage has been reduced to approximately 6.6% as at 31 May 2007 (compared to 25.9% as at 31 March 2007, 28.5% as at 31 December 2006 and 17.9% as at 31 March 2006).

According to one structured credit investor, the announcement does have some positives for Queen's Walk. "There are some indications that real efforts are now being made with the company – which maybe could have happened earlier – and these are beginning to bear some fruit. Certainly the diversification of the company's investments appears to have improved and it now has greater flexibility because of the reduced leverage," he says.

However, the investor adds: "Even though it's possible – as some analysts are saying – that Queen's Walk has been cautious with its asset write-downs, the short-term upside is limited. For me, the price is still not yet realistic."

Meanwhile, listed structured credit fund Carador plc – which has so far remained resilient in the face of sub-prime declines – announced last Friday that as at the close of business on 31 May 2007, its unaudited NAV per share was €0.97. A company statement says: "Carador's NAV decreased by 0.45% in May. The May NAV is net of the €0.02 dividend per share declared on 31 May and payable on June 22 2007. This month's calculations include an estimated €293,000 worth of net interest earned in April and May 2007, which equates to €0.0058 per share."

MP

Modern investment management

Andrew Crawford, executive director investment grade credit, and Vivek Jeswani, structured credit products analyst, at Threadneedle answer SCI's questions

Andrew Crawford

Q: When, how and why did you and your firm become involved in the structured credit markets?
A: Managing structured credit products was a natural evolution for Threadneedle. Our fund managers were accustomed to using credit derivatives for relative value assessment and price discovery.

As our fixed income desk grew to over 30 staff, specialisation increased such that in all fixed income asset classes we were equally comfortable in expressing views through derivatives. Indeed, our credit analysts are in the habit of referencing to CDS spreads rather than just bond spreads in making comparisons between companies and their performance is judged, inter alia, on CDS long and short recommendations as well as their cash market calls.

Crucially, our credit selection process is relative value driven in a framework that requires the risk budget for credit to be fully allocated. Under and over-weights must balance out and this zero-sum methodology naturally leads to long and short relative value strategies.

The Threadneedle Credit Crescendo Fund, launched in March, 2003, was the first of our funds to use derivatives and allowed us to pursue strategies such as curve trades, capital structure arbitrage, basis trading and include dipping our toe into the iTraxx tranche market. Today the group manages a range of hedge, total return and other funds that use derivatives.

We launched our credit DPI product, Polestar, in February 2007. This works off the same stock selection process as the rest of our products – the credit selection process and trading strategies are unchanged. Polestar is principal protected by RBS, who therefore also set the leverage constraints and the risk budget.

Q: In your view, what has been the most significant development in the credit markets in recent years?
A: The huge growth in liquidity in credit derivatives, to our minds, is the most significant development in the credit markets in recent years.

Systems and technology are also important and the foremost reason for the rapid innovation in structured credit. Modelling capabilities provide the platforms through which managers, investors and structurers can put derivative liquidity to work.

Q: How has this affected your business?
A: We use derivatives in our funds where mandates allow – and this list continues to grow. Furthermore, our relative value research and price discovery techniques make extensive use of derivatives: we see the ability to separate the rates risk and credit risk elements of a corporate bond as important to decision making.

Our modelling techniques have been aligned with our product development strategy. Arrangers usually provide sophisticated pricing tools but in order to provide best execution to clients, we have to ensure these can be relied upon.

Our quant team stress test the model assumptions and compare pricing tools for scalability, usability and pricing robustness. In most cases, internal models are implemented alongside third party systems as well as the rating agency models. Recent defaults rates are not of any practical use given the benign credit markets, so having the ability to simulate on a more challenging forward-looking basis is a prerequisite.

Q: What are your key areas of focus today?
A: As manager of one of the UK market's largest CPPI products, our structured product area continues to grow and is a marketing focus. We also have other structured credit transactions in their early stages of development and assessment.

Q: What is your strategy going forward?
A: As well as investment products in themselves, Threadneedle sees structured products, including structured credit, as part of the essential toolkit of a modern investment manager.

Applying structuring techniques around long-only and hedge funds and their stock selection process allows us to engineer to alter the risk and return profile of an investment product, solving the investment problem of a more sophisticated investor. To this end, we see alpha generation for structured products as sitting properly along side our other fund management activities with structure or wrapper management as distinct disciplines. Our portfolio managers should be free to focus on our core competency (credit selection) without being overly constrained by the structure so that our management style is not compromised.

Q: What major developments do you need/expect from the market in the future?
A: The anticipated turn in the credit cycle is likely to see a manager shakeout and an investor reassessment.

Real volatility, and sometimes distress, at the single name level is likely to benefit managers who focus on stock selection using credit analysis rather than pure quant-driven valuation. Recent sub-prime fears and hedge fund liquidation have kick-started this process.

Investors and asset managers have also begun challenging the ratings agencies. There is likely to be less reliance on resultant structure ratings in the investment decision and perhaps more reliance on ratings on a look through basis – for example, it is possible to hold a triple-B note where the average underlying rating is single-B.

About Threadneedle
Threadneedle is a global asset management company with a distribution reach covering four continents and over fifteen countries.

Founded in 1994, it manages more than £72.2bn (€107,1bn/US$142.2bn) of assets for a wide range of clients including pension schemes, insurance companies, private investors, corporations, mutual funds and affiliate group companies.

Threadneedle is a highly diversified business with investment expertise covering all major asset classes from equities and bonds to property and alternative investments. Its only business is managing clients' money.

Citi hires correlation head

The Latest company and people moves

Citi hires correlation head
Peter Keller has left Deutsche Bank in New York to join Citi as head of global correlation products syndicate. He reports to Nestor Dominguez and Janice Warne, co-heads of global CDOs.

Pawlowski to Primus
Jerzy Pawlowski has joined Primus Guaranty as a senior portfolio manager and head of credit correlation trading. His primary responsibility will be as co-portfolio manager and risk manager for the firm's Harrier Credit Strategies Fund, which commenced operation in 2006 and was funded with capital from Primus Guaranty.

Pawlowski will co-manage the fund with Don Young, who joined Primus in 2006 and serves as senior portfolio manager and head trader, with responsibility for managing the firm's leveraged loan and high-yield bond investment products and vehicles.

Pawlowski, who reports to Charles McLendon, president of Primus Asset Management, previously held positions at leading alternative investment firms and dealer institutions in both correlation trading and structured credit products.

Gile joins BlueMountain
Betsy Gile, currently a member of the Board of Directors of Deutsche Bank Trust Corporation and Deutsche Bank Trust Company Americas, and a former md at Deutsche Bank, has joined BlueMountain Capital Management as a managing director and senior strategic advisor.

The fund says that Gile will leverage more than 25 years in the credit and loan markets to help drive BlueMountain's client and investor relations, internal professional development and talent acquisition efforts.

Calyon hires two in Hong Kong
Calyon has taken on two credit professionals in Hong Kong, both join from BNP Paribas. Wilson Kau joins as head of credit markets and CDO flow sales for Asia excluding Japan, reporting to Pierre Trécourt, head of credit markets and CDOs for Asia. King Yut Quan joins as a credit flow trader, reporting to Anitza Nip, head of credit markets flow trading for Asia excluding Japan.

New synthetic structurer for JPM?
Mike Diyanni is thought to have left Citi in New York. He is understood to be joining JP Morgan as a senior synthetic structurer.

HSBC promotes Cannon
HSBC has appointed Matthew Cannon as head of structured credit products, global markets, Asia-Pacific. In this newly created role, he will be responsible for all trading, structuring and marketing activities in cash and synthetic CDOs, illiquid credit trading, client solutions and other structured credit derivatives in the region.

Cannon will report functionally to Jeff Jakubiak, head of structured credit products, global markets, Europe & Asia-Pacific, and regionally to Monish Tahilramani, head of regional trading, global markets, Asia-Pacific. Cannon was previously responsible for marketing structured credit products across Europe at HSBC.

BNP Paribas adds to advisory
BNP Paribas has hired Nimesh Verma to its financial institutions advisory team, a specialist unit within the bank's global risk solutions unit in its fixed income division. The team gives support to clients on strategic financial issues, including value maximisation, risk management, regulatory developments and capital optimisation structures.

Verma joins BNP Paribas with ten years of experience in advising financial institutions at consultancy Mercer Oliver Wyman. During his consulting career, he focused on commercial and corporate banking, credit portfolio management, credit businesses and the application of risk management to business decision making.

RBS takes on senior structurer
The Royal Bank of Scotland has announced the appointment of Hubert Tissier De Mallerais as an md in its financial institutions securitisation team. He will report to Lee Rochford, head of FI securitisation, and will take on a senior origination and structuring role across Continental and Eastern Europe, covering all financial asset classes. His responsibilities will include developing RBS's principal activities, as well as client business.

Tissier De Mallerais joins from Credit Suisse, where he was head of mortgage and consumer ABS. He is a seasoned securitisation professional with 10 years' experience, having structured and executed a wide range of transactions from UK residential and non-conforming mortgages, RMBS, auto ABS and CLOs in the Netherlands, Italy, Switzerland, Germany and France to working with US and Australian issuers accessing the European markets.

MP

UK fund to bring CDPC

A round up of this week's structured credit news

UK fund to bring CDPC
Hedge fund London Diversified Fund Management (LDFM) is to sponsor and manage a credit derivative product company – Satago Financial Products Ltd. Satago will vie with Channel (see SCI issue 42) to be the first CDPC to launch in Europe.

Moody's has assigned a provisional counterparty rating of Aaa to Satago, as well as provisional debt ratings of Aaa to its up to US$250m deferrable interest auction rate notes due 2047 and Aa3 to its up to US$65m deferrable interest preferred perpetual auction rate notes.

According to Moody's, Satago is a bankruptcy remote CDPC incorporated in Ireland. It has been established for the purpose of buying and selling CDS protection on senior tranches of synthetic CDOs.

Moody's provisional ratings are based on eligibility criteria and portfolio parameters for assets purchased, credit enhancement provided by class subordination as well as liquidity tests and eligibility criteria for the investment of proceeds of the rated notes. Satago's capital model, which will be used to monitor the ratings, regularly examines the impact of potential counterparty defaults and potential credit events under the CDS transactions on the computation of required resources.

In addition, Satago's CDS counterparties and noteholders will benefit on an ongoing basis from the oversight of an independent administrator and regular agreed upon procedures performed by an independent auditor on the company. Moody's does not believe that pre-set criteria or risk assessment models can accurately measure or limit all of the possible risks inherent in structures of this type. The maintenance of any given credit rating level is therefore dependent upon the management of the vehicle and the actual portfolio of assets and liabilities, and the observance of the various limits and tests established within the structure.

Fitch launches beta CFXO model
Derivative Fitch has released a beta version of its Vector CFXO model for collateralised foreign exchange obligations (CFXOs) in an effort to increase transparency in the market and the understanding of its ratings. The model can be used to analyse portfolios of FX trigger options and combinations of FX options and sovereign credits.

Fitch notes that the structural risks in a CFXO are similar to those in a synthetic CDO. However, the FX reference assets introduce new risk into the structure. Fitch's quantitative analysis of FX reference assets is based on Monte Carlo simulation using asymmetric GARCH processes with jumps to fit individual FX rates.

Ratings are assigned by a ratings committee and the model results are only one input into the rating process. The ratings committee might want to look at scenario testing of such factors as the historical price movements of the foreign exchange rates in the portfolio, the correlation between these price movements and structural features of the transaction, including the number and level of the FX triggers outside of the model.

S&P explains new PAUG assumptions
In an article published last week, S&P explains its new stressed default assumptions for structured finance assets that are referenced in pay-as-you-go (PAUG) CDS. The revision affects the assumptions underlying S&P's CDO Evaluator modeling tool.

"This change has been made to account for the higher default rates when the CDO transaction synthetically references a structured finance asset," says credit analyst Belinda Ghetti. "CDO transactions that reference PAUG assets have higher default stresses due to these assets having to make loss payments for credit events that could be reversible and are therefore not included in Standard & Poor's Ratings Services' default studies."

Ghetti explains that reversible, or soft, credit events are a unique feature of PAUG CDS contracts, which reference structured finance assets, typically RMBS, CMBS or CDOs.

"The structured nature of these types of assets marks them apart from the reference obligations usually seen in standard CDS contracts, such as corporate or sovereign bonds," she adds. "Of the reversible credit events associated with structured finance assets, write-down is the one that causes the most concern because of its reversibility."

The S&P article examines this further, giving details of how the agency defines and values write-down events in its rating analysis.

RAP CD updated
Derivative Fitch has updated its Risk Analytics Platform for Credit Derivatives. RAP CD v.1.3 incorporates new models for synthetic CDO products, including multiple-maturity CDOs; long/short CDOs; variable subordination CDOs and zero coupon CDOs. The new models complement the existing models for CDOs and CDO-squareds.

Other enhancements in this release include the addition of grid and batch processing and 'what if' hypothetical trade risk analysis. "Development of RAP CD will continue in the third quarter with the addition of models for synthetic CPPI, CPDO, combo notes and forward-starting CDOs," says James Wood, md, Derivative Fitch.

Since RAP CD's launch in mid-July 2006, Derivative Fitch says it has signed up over 30 customers to the service globally and provides daily risk and pricing data for over 150 deals. The agency has also been working closely with a number of arrangers who are using RAP CD as a marketing tool when pricing synthetic CDOs.

Innovation drives European CDOs of ABS
Tranche spreads for CDOs of ABS transactions were slightly down in the first quarter but generally in line with those of the fourth quarter of 2006, according to the latest quarterly report on the European cash CDO market published by S&P.

The move towards new forms of collateral, with higher yields, is one of the key trends in the sector. In addition, structures are increasingly cost-efficient, for example, through the use of hybrid technology.

Transactions rated in the first quarter by S&P included one CDO-squared, one CDO of high-grade ABS and four CDOs of mezzanine ABS tranches, among which was the first cashflow CDO of European CMBS, Glastonbury Finance 2007-1.

European/Asia-Pacific structured finance is robust
Fitch Ratings reports that rating performance in European and Asia-Pacific structured finance remained robust through to the end of the first quarter. In a new report, the agency notes that the overall upgrade/downgrade ratio for European and Asia Pacific structured finance in Q107 was 7.3:1, compared to 3:1 for Q106.

Q107 rating activity in European and Asia-Pacific structured finance increased compared to the same period in 2006. An aggregate of 158 upgrades and downgrades occurred in the first three months of 2007; almost half of those rating actions came from the European RMBS sector, all of which were upgrades.

2006 CDO rating activity in Europe and Asia Pacific was similar to 2005, with about 300 rating actions. However, the upgrade/downgrade ratio weakened to 2.1:1 from the strong 17:1 in 2005. This was despite significant upgrade activity in Q406 from European SME CDOs.

Sun and CDO2 offer new service
Sun Microsystems has announced the availability of a dynamic on-demand financial risk simulation and pricing service from Network.com offered by CDO2, the structured credit pricing and risk technology provider.

The companies say that the new version of their CDOSheet application, available from Network.com, delivers the latest pricing models and risk analysis technology to customers. Financial modellers can now run simulations directly from their desktops without any awareness of the advanced grid computing technologies powering the service.

CDOSheet application users capture the necessary deal indicatives and model parameters from within a spreadsheet interface. The Sun Grid Compute Utility, powered by the Solaris 10 OS and Sun Grid Engine running on Sun's x64 hardware, allows a single risk analysis from the CDOSheet application to be performed in parallel on the grid, thereby minimising the time necessary to complete the simulation.

"As an active manager of our CDO portfolios, we have a varied and unpredictable demand on our systems depending on the market conditions. Network.com meets that demand for our utilisation spikes with an efficient and cost-effective infrastructure," observes Jana Becher, md of AC Capital Partners.

OpenLink upgrades Findur
OpenLink – provider of cross-asset trading, risk management and operations processing software solutions – has announced the release of a new version of Findur's credit derivatives solution.

The firm says that Findur version 8.1 now features the latest developments in front-to-back straight-through processing for credit derivatives. Several major financial institutions are in various stages of implementing the latest features across their credit derivatives desks.

Highlights of Findur's credit derivative module in version 8.1 include: the latest credit derivative products, including credit indexes, tranched indexes, synthetic CDOs, default baskets and CDS options; state-of-the-art valuation methods, including copula based semi-analytical and Monte Carlo pricing models; advanced real-time analytics and stress testing capabilities; credit event management and settlement modules; and new interfaces to data services.

Markit launches electronic affirmation matching service
Markit Group has launched a new electronic affirmation service within Markit Trade Processing (MTP). The first dealers to sign up are Bear Stearns, Goldman Sachs and Lehman Brothers.

The new service, which builds on Markit's existing affirmation model, is – the company says – unique in its ability to electronically match trades between counterparties prior to confirmation. Alternative vendor affirmation solutions simply provide a connectivity mechanism that routes information between sell-side and buy-side firms, Markit states.

MTP aims to allow clients to reduce risk by automating the post-trade lifecycle for all OTC derivative instruments. Last month alone, the service processed in excess of 60,000 trades.

MTP provides trade date affirmation and trade confirmation through industry utilities such as the DTCC.

MP

Trading ideas - creeping steeper

Tim Backshall, chief credit derivatives strategist at Credit Derivatives Research, looks at an IG-HY Homebuilders curve steepener

The complex dynamics of HY and IG curve levels for US Homebuilders has been playing on our mind for a while. Thanks to a new curve score model, we are able to effectively judge the relative trajectories of these similar credits as they widen and tighten.

With prices falling, inventories rising, foreclosures rising to their highest levels as a percentage of loans underwritten since 2002, tightening lending standards, higher rates, and a seeming deterioration in consumer credit generally (Best Buy and Circuit City), the future does not look bright. Add to that the wall of sub-prime mortgage first resets we face in the next 12 months and our expectation is that things get worse before they get better (for builders as well as lenders).

Exhibit 1 shows the relative steepness of the IG and HY curves over the past two years or so. It is important to note that spreads have been moving around a lot here but it is clear that the dramatic flattening of HY curves (around the BZH investigation) pushed them back in line with IG even though spreads were 200+bp wider.

Exhibit 1

One critical point here is that if IG spreads widen (as per their HY cohorts) then there is no good reason to not expect curves to steepen considerably (from current 50-60bp to 90-100bp in 5s-10s). Multiple outlook changes from the agencies seem to signal potential downgrades and our view is that if one goes many will go as although they vary by sector, geography, and liquidity, a housing slump this large with the inventory they have will require market remediation in terms of some companies disappearing.

The relationship between IG and HY spreads and curves in general is tricky to model, largely due to the idiosyncratic nature of HY credits versus the more systemic curves in IG, but Exhibit 2 provides an interesting insight into the relative changes in the two Homebuilder curves over time. This scatter plot shows the movements in IG and HY curves through time with the red marks indicating the current level. The dotted lines show that in general curves have steepened as spreads widened (especially in IG) but the green arrow indicates the recent flattening that we would expect to see as credits deteriorate past our inflection point.

One point of critical note in Exhibit 2 is that the IG curve is dramatically flatter than where the HY curve was at approximately the same levels. The red arrow indicates that at around 150bp in 5Y, the HY names were trading up near 90bp after sitting at 60bp for some time (follow the light blue line).

Exhibit 2

It is not a stretch to believe that 5Y CDS (being the most liquid) is a good anchor and that given how flat IG curves are, we expect curves to steepen considerably (driven by 10Y widening more than 5Y tightening). This 5Y tightening (bull steepening) is what we saw a few weeks back when the technical bid overwhelmed even the biggest bears.

Our summary curve view is that while IG and HY curves are close in absolute steepness, IG is too flat in 5s-10s. HY remains in line with our models and our belief is that we will not likely see a jump from current IG to HY levels without going through the mid-range of curves and levels, i.e. although IG is trading as flat as its HY cohorts, it must go steeper before it can flatten into distress (in the worst case scenario).

Building builder curves
Analysing the curves through time gives us an indication of where we might expect levels and curves to be but an investigation of current levels shows some interesting curve shapes and relative-values. Exhibit 3 shows the current IG and HY builders' curve (dotted lines) compared to the market (solid lines).

Exhibit 3

The 5s-10s curve is clearly hump-shaped as we have discussed before with peak steepness seemingly around 85bp at around 225bp in 5Y. A quick scan across the credits (the red and blue dots) indicates that at the wides, HOV (light blue arrow) is trading in line with the curve's fair-value for 5s-10s but at the IG end, the dark blue arrow indicates that PHM is trading significantly flat to both the builder curve and the market curve.

This simple analysis provides further reason to expect IG curves to steepen as they deteriorate as well as positioning PHM as one of the flattest (and potentially most likely to steepen). As we previously mentioned, we would expect 10Y widening to be the driver of steepening as opposed to 5Y tightening (especially if builders get downgraded to junk and any technical bid disappears).

Judging the relative steepness of curves is hard as credits tend to trade on different trajectories as they widen and tighten. Given that these are all builders we would expect that the trajectories would be similar and a new model is illustrated in Exhibit 4 to show these potential trajectories. The behaviour of the 3s-s segment of the curve is very sensitive to liquidity, leverage, and potential buyout rumours, as well as investors' somewhat myopic view of a short credit-cycle.

Exhibit 4

The curves in Exhibit 4 represent a series of trajectories regressed across our CDS universe over the past four years. This view is of all liquid credits across all trading and the solid blue curve represents the average movement.

The dark red curve indicates the trajectory that many LBOs have taken and the green curves show the trajectories that many weak-performing credits have taken in and out of distress. The blue line has an r-squared of over 65%, which is very high given the large number of data points, and clearly represents the hump-shaped behaviour we have been discussing with clients for a number of months.

We can use these curves to generate a more consistent view of the relative steepness or flatness of curves – 5 being the steepest and 0 being the flattest. As one would expect, many of the LBOs and LBO-prone names score 5 and many of the weakest performers score 1 or 0 (as do many Financials surprisingly in our view).

Notably, PHM is among the steepest curves in the 3s-5s segment of the CDS market (Curve Score 5) as opposed to its 5s-10s segment where it is very flat (Curve Score 1) – this is among the widest divergences we have seen and given the fact that short-term liquidity may be fine we prefer to play the long-end of the PHM curve with a 5s-10s steepener – or in our case buying 10Y protection.

On the other hand, although HOV's curve is flat to the average level, we see it as still steep to the weaker-performing credits that make up the lower end of our range. The curve score for 5s-10s is 1 (flat as we would expect for a distressed credit) and in 3s-5s the score is 2 which is somewhat surprisingly flat. Given this view, we might look to 3s-5s steepeners in HOV as 3Y keeps anchored while 5Y widens. In our case we prefer to sell 3Y protection based on our view of short-term liquidity.

Picking the pair
Technically, we see PHM and HOV as having divergent curves and as we expect IG curves to continue to steepen and HY levels to widen but short-term liquidity to hold in. We believe that the housing woes will spread from lower-end to mid-tier and interestingly both HOV and PHM serve about the same price point on average (at current levels) – PHM US$337k and HOV US$329k. Potentially we could choose TOL (as a higher end home builder) or DHI (as more entry-level) for our short leg but the PHM curve and rating agency outlooks steer us towards PHM.

The historical relationship between 3Y HOV and 10Y PHM is interesting in terms of its cyclicality. Exhibit 5 shows that HOV has pulled away from PHM a few times, only to pull back. The jump in HOV's spreads in Mar 2007 (thanks to BZH revelations), kicked the relationship out of line and it is clear that HOV has a higher beta than PHM. While it is too easy to remark that the relationship will converge once more (given the idiosyncratic issues at hand), we are comfortable selling the spread at these levels of differential with the view that PHM underperforms HOV as builders deteriorate.

Exhibit 5

Construction of the trade requires an analysis of DV01s and Beta. A first cut at the two CDS curves supports our view that we should be selling 3Y HOV protection and buying 10Y PHM protection as the relative spread/DV01s indicate these as the most attractive positions (HOV 3Y being highest and PHM 10Y being lowest). The DV01 differential between the two positions is 2.63:1 with HOV 3Y at US$2,640 and PHM 10Y at US$6,947 for each US$10m notional.

Building the trade with this ratio would be DV01 neutral and more a bet on the divergence of the pair from Exhibit 5, but we have already remarked that HOV is clearly more volatile than PHM and Exhibit 6 indicates that the Beta is approximately 0.38 (for each 1bp move in HOV, PHM moves 0.38bp). The r-squared is reassuring at 75% and there are few outliers to worry about in these liquid credits.

Exhibit 6

We prefer to construct the trade as relatively neutral as possible and so combine our DV01 ratio with the Beta. This, rather surprisingly, leaves us with 2.63 times 0.38 or 0.999947 as our hedge ratio which we obviously round to 1. This DV01/Beta adjusted trade is best positioned in equal notionals.

Therefore, based on the relative curve differentials, absolute curve and CDS level changes, short-term liquidity versus long-term weakness, and fundamental idiosyncrasies, we prefer the well hedged equal notional short 10Y PHM and long 3Y HOV. This trade takes advantage of the continued steepening trend in IG builders and relatively flat curves in HY and across a number of scenarios we feel it is an attractive trade.

Scenario analysis
We see four possible scenarios for this trade, ranging from a housing market recovery to a near-term liquidity crisis. Only in the case of a significant improvement in housing and therefore homebuilders do we have cause for concern and even in this case we feel that there is some protection.

The most likely scenario is that pressure continues in the housing market. This scenario should entail both IG and HY spreads moving wider and while the size of the change is hard to quantify and is clearly market driven. The carry and roll-down on the position provide ample MTM volatility cover and we would expect (based on the last few weeks) that 10Y PHM would underperform 3Y HOV making this scenario profitable for us.

A maintenance scenario in which credit trading slows in the doldrums of the summer and spreads drift from results to results would likely prove profitable for our position also as roll-down and carry is well in our favour. While we do not expect this scenario (given current volatility), we are pleased that we would expect profitability in nothing changes.

Less likely but nevertheless possible is that the housing market recovers and we all skip back to Neverland (sorry for the sarcasm). In this case we would expect the 10Y PHM to rally considerably but we are somewhat comforted by the fact that in the recent rally (technical-bid driven), we saw HY rallying with it as the relationships held (and actually 3Y HOV outperformed). This scenario is likely to prove the worst for our position but the short-term HOV leg combined with carry should help us considerably.

Finally, the least likely scenario, given current levels of liquidity, is a jump-to-default crisis or extremely bearish scenario. If this occurs and we see HY curves flatten further and potentially invert we would be hurt on our HOV leg as the short-end rose to around 500bp levels. However, if this were to occur, we feel comfortable that the IG curves would have steepened and widened considerably as this would be an extremely bearish scenario indeed.

Our best performance is a continued drift wider and steeper in IG as HY remains flat, which we feel is best reflected by our expectation that the housing market is at least a year away from a bottom as sub-prime foreclosures hit the lower-end of the property ladder, higher-rates hit the mid-tier, and excess inventory forces prices down further.

Risk analysis
The pair are DV01/Beta hedged to equal notional which implied that we are more exposed to PHM tightening losses than HOV widening losses but the Beta-adjustment should help with that.

See scenario analysis above for more details. The roll-down and carry on the trade should further manage the expected short-term volatility of the homebuilders.

Liquidity
Both names offer good liquidity in the CDS market. As members of the CDX IG8 and XO8 indices they consistently rank in the top 200 issuers by quoted volume on a daily basis. Bid-offer spreads could be better but HOV 3Y regularly trades at 10-15bp and PHM 10Y less than 10bp.

Fundamentals
This trade is based on our view that the US housing market will continue to weaken (even with rate cuts) as sub-prime mortgage issues, tightening lending standards, high inventory, prices falling, and rising foreclosures. That's a quick look at our macro view (on which we could expound but many have heard the story too many times) and note that each credit has its own idiosyncratic story to tell.

For more details on the fundamental outlook for each of the credits, please refer to Gimme Credit. Kathleen Shanley, Gimme Credit's IG Homebuilders Sector expert and Vicki Bryan, Gimme Credit's HY Homebuilders Sector expert both maintain fundamentally deteriorating views on the two credits, but their outlooks for performance are slightly different and fit with our trade framework.

While both remain negative fundamentally, Vicki's view is that HOV may offer some upside in senior notes (CDS references) while Kathleen sees PHM as likely to get the downgrades it (from our view) deserves.

Summary and trade recommendation
Considering the complete homebuilder curve through time offers us insight into the relative attractiveness of curve trades, and belies our expectation of continued steepening in IG and stable to flatter curves in HY. Idiosyncratically, PHM and HOV stand out in different segments of the curve as fundamentally we see long-term weakness in PHM and solid short-term liquidity in HOV.

Adjusting our weights for DV01- and Beta-neutrality leaves us with an equal notional long short-dated HOV against short long-dated PHM, which provides us with ample carry and roll-down benefits while our detailed scenario analysis points to profitability in most market outcomes.

Buy US$10m notional Pulte Homes Inc. 10 Year CDS protection at 220bp and

Sell US$10m notional K Hovnanian Enterprises 3 Year CDS protection at 340bp to gain 120bp 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).

ECMBX - a primer

Chris Nias, quantitative securitisation research at Barclays Capital, looks at the soon-to-be-launched European CMBS index, how it will trade and its likely constituents

The soon-to-be-launched ECMBX index comprises an equally weighted basket of 20 pay-as-you-go (PAUG) CDS referencing European CMBS selected according to a defined set of rules. The index coupon will be fixed at roll and the index will be quoted on a price basis rather than spread due to the prepayment uncertainty.

There will be indices based on triple-A and triple-B, euro and sterling denominated tranches. Similar to the CMBX and ABX index families, the key advantage this index has over more traditional bond indices is that it can be traded in CDS form; traditional bond indices could only be traded in TRS form, there was less liquidity and therefore generally higher trading costs were incurred.

As with any CDS, trading the index enables transfer of credit risk, which can be used either for hedging or speculating. The additional benefit of an index is that it does this on a portfolio basis. The increased liquidity lowers costs and it can be used as a proxy hedge where no market exists in a particular underlying; it also allows the separation of market and idiosyncratic risk. Again, the contract has an unfunded notional and, in addition to making it a cheaper hedging tool, this makes it popular with investors employing leverage.

A trade on the index is just a portfolio trade of the 20 underlyings in single-name CDS format. The cash flows on the index would exactly match the cash flows that the investor would receive were they to have done the individual trades at the same coupon (with the appropriate selections made in the documentation – for further details see the later section "Documentation"). Being based on the PAUG CDS, the cash flows on the ECMBX contract are designed to closely mirror the cash flow of the portfolio of cash CMBS bonds.

Currently, there are four indices planned within the ECMBX family. These will be at the top and bottom of the investment grade spectrum and cover the sterling and euro markets as follows:

The index has gone through several iterations since work first began back in 2005. The original plan was for a futures-style contract, but more recently this has been dropped in favour of a format that closely apes the US ABX and CMBX indices. This was primarily due to the wide scale take up of the PAUG contract in Europe for trading on single names and also the demonstrated success of this format in North America for use in indices. It is currently expected that the first series will roll in September of this year.

The index will roll every six months, giving each index unique vintage characteristics. The constituents for each new series will be selected through a set of rules (see the following section "Eligibility criteria and generating the constituent list") this brings transparency, confidence and therefore liquidity to the index as it minimises the need to use less predictable dealer votes. Additionally, trading of the standardised contract that has now been widely accepted within Europe for single-name trading will reduce basis risk between index and single names and will benefit liquidity.

Each series will represent exposure to CMBS originated within a certain period. However, based on our preliminary analysis, only approximately 20% of the index will change between rolls due to the limited number of qualifying deals and the large (two years) look back period. This will not provide as clear vintage separation as is seen in North America where 100% of constituents change between indices. Whether there is sufficient differentiation in constituents or whether there will be sufficient vintage differentiation will influence whether off-the-run series retain liquidity, or whether investors roll positions into the new series.

The corporate indices (iTraxx and CDX) have few changes between rolls and, as a result, liquidity is focussed on the on-the-run index and, at roll, investors will tend to roll their positions into the new contract so as not to be stuck with an illiquid position. This is in contrast to ABX and CMBX where it is expected that, due to the very different characteristics of the different series, there will be liquidity across all series.

Roll process and rolling trades
The index roll is the process by which a new series of the index is created. The primary difference will be that the new index will incorporate eligible deals from the past six months and will have removed deals from the earliest six months or those that have refinanced. Also, when the index coupon is fixed, this will reflect current market risk premia. Figure 3 illustrates the countdown to the index roll date and the tasks that must be completed by the index administrator on those days.

Index roll tasks

• Proposed offerings list (T-10): The index administrator will generate an initial list of index constituents using the eligibility criteria discussed in the following section ("Eligibility criteria and generating the constituent list"); these will be the most recent 20 deals that meet the criteria. This process of a rules-based index creates greater confidence with investors due to the transparency of the process.
• List finalised (T-10 to T-6): Index members have the opportunity to nominate for rejection any deal in the list. Once a deal receives three nominations for removal, the index administrator will solicit a vote from all members for the security to be rejected. Once a deal has been removed from the list, the administrator will replace it with the next eligible deal. Again, members will have the opportunity to nominate deals for removal. This process is continued until the members are happy with the constituents.
• Index composition announced (T-4): Index members and the public are informed of the composition of the index.
• Draft annex published (T-3): The draft annex for each index is published to members and the public. This is the legal document listing the index constituents. The trade confirmation will reference this.
• Fixed rate polling process (T-1): The process of setting the index coupon occurs. This is fixed in the same manner used by the British Bankers Association when fixing LIBOR. Index members submit a spread for each index and the published spread will be the inter-quartile mean rounded to the nearest basis point.

As discussed, where there is a large overlap in reference obligations, investors may prefer to roll from one series into the next. This allows them to retain exposure to the most liquid series of the index. This is achieved by selling (buying) a long (short) index position in series 1 and buying (selling) the same size position in series 2. The Street convention for this is for the dealer to bid or offer the roll as follows:

Eligibility criteria and generating the constituent list
The specification of eligibility criteria ensures that the index is representative of the market as a whole. Specifying these rules, a priori, makes the index generation process transparent and predictable. This is preferable to the use of votes and other more qualitative means.

As previously mentioned, there will be triple-A and triple-B indices for euro and sterling denominated CMBS. For a transaction to be eligible it must have a triple-A tranche that fulfils the criteria set out in Figure 6 and also have a triple-B tranche; that means the triple-A and triple-B constituents will both reference the same 20 deals. As we see in Figure 5, the indices transect the capital structures of the 20 deals; this is the same approach used in CMBX. This should provide great stability across the rating curve.

When generating the proposed offering list, in order to limit the list to the 20 deals needed for the index, we simply select the most recent 20 deals. Where there is a "last place tie" (ie, two deals in 20th place), the larger deal will take precedence. It is at this point possible, by a vote of index members, to have a deal removed. In this case the administrator will replace that deal with the next eligible deal.

Finally, for tranche selection, where there are multiple tranches that meet the criteria, the following rules applied:
1) Tranche rated by all 3 agencies takes precedence.
2) If there are still competing tranches then the most senior takes precedence.
3) If there are still competing tranches then the largest issue size takes precedence.
This is slightly different to CMBX, which operates as follows:
1) If there are competing tranches then the most senior takes precedence.
2) If there are still competing tranches then the longest WAL takes precedence.

Possible constituent list for an ECMBX 2007-H1/2
There may still be some minor alterations to the eligibility criteria to ensure an adequate and representative sample of bonds are available on index roll dates. As such we did not perform a detailed analysis of the index constituents. However, based on the draft criteria, we looked at what bonds will likely be in the September index:

Under the current criteria, we do not find 20 eligible bonds for the GBP index. However, we expect to see some alterations to the rules that will rectify this.

We looked at instances of deals being voted out of CMBX in order to gauge the effectiveness of the rules-based index approach. We found that, in the three series to date, there have been no bonds voted out by the index members. In light of this, we expect that most of the above will make it into the index. We can calculate a few simple summary statistics off this.

We see that there is very little difference between the GBP and EUR indices, only a marginally longer WAL on the GBP triple-B than EUR triple-B. We also note that none of the triple-B constituents of the sterling index are rated by Moody's.

Documentation
The index contract will be based on the recently released April 2007 contract, which has become the standard for trading within Europe. As we covered when discussing single-name CDS of ABS, there are a number of options within the contract terms; these pertain to such things as step-up provisions, interest shortfall caps, etc. To ensure that the contracts are fungible, these options are all preset for the index. The following are the terms under which the index will trade:

• No physical settlement: It will not be possible to deliver bonds into the index contract.
• No step-up provision: Step-ups in bond coupons will not reflect in the index coupon, nor will there be any option to cancel as a result of those step-ups.
• Fixed cap applies: The seller's exposure to floating payments due to interest shortfalls is capped at the fixed rate on the index.
• The WAC cap provision applies. This means that, if a bond's coupon is curtailed due to a WAC Cap, this will not be counted as a shortfall.
• No implied writedown: Implied writedowns will not be treated as floating events.

Fixing process and rules
The index Administrator (currently Markit) will collect closing mid-market prices daily and publish the results on their website. The day's closing price is calculated using the same methodology used by the British Bankers Association when calculating Libor rates. That is to say, it is the simple average of the inter-quartile range. The following table shows how many prices will be discarded depending on the number of contributors.

Dealers will mark, and Markit will publish, daily closing mid-market levels on the two most recent series. Older series will be marked monthly unless they retain significant liquidity.

Index administration
Markit, as the Administration, Calculation and Marketing Agent for the index will provide the follow services in support of the index:

• Markit will provide a monthly settlement instruction file on their website as is done for ABX and CMBX. This will detail all fixed, floating and additional floating payment amounts due.
• Markit will provide an index calculator on their website similar to that provided for ABX and CMBX, this should provide investors with a simple tool for running scenarios on the index.
• Markit will poll for, and publish, daily prices for the on-the-run index and monthly prices for the off-the-run indices.
• Markit will manage the semi-annual index roll process.

© 2007 Barclays bank PLC. All rights reserved. This Research Note was first published by Barclays Capital on 21 June 2007.