Contents

Opportunity knocks

Muni volatility spells good news for new index

An up-tick in trading activity and increased volatility in municipal bond spreads suggests that Markit has picked the right moment to launch its latest CDS index. The new index, called MCDX, will launch on 6 May and reference 50 municipal bonds.

According to one structured credit investor, there has been a spike in muni spreads over the past three to four months due to failures in the auction rate market and the possibility that some municipalities (such as Alabama) may become bankrupt. "It is a good time to trade the credit risk of municipal bonds via a CDS index," he says. "Maybe it will also force municipal bond investors to start analysing the credit risk of their holdings."

Indeed, Markit says that the index is a response to investor demand for a liquid and transparent tool to gain exposure to municipal bonds, at a time when credit pricing has become increasingly important to municipal bond buyers. A number of market participants had also been seeking a way to express a view on the municipal market, the firm adds.

"The Markit MCDX index will attract new investors to the market by increasing tradability and liquidity," explains Niall Cameron, evp and head of indices and equities at Markit. "As the world's first independent synthetic municipal bond index, we expect it to become the benchmark for trading in this asset class."

MCDX Series 10 references 50 equally-weighted CDS contracts on revenue, full faith and credit, and general fund obligations – all of which have US$250m outstanding uninsured debt. Markit stressed in a conference call that monoline-insured bonds, as well as tobacco and healthcare issues would not be included in the index.

Series 10 references 24 revenue bonds, 21 state general obligations and five local obligations. Three tenors will be traded: a three-year, a five-year and a 10-year. It will trade with a fixed coupon.

Like Markit's existing indices, the MCDX will be rolled every six months, with the first roll due on 3 October when MCDX Series 11 will begin trading. The index's criteria states that reference entities must have liquidity in both the CDS and cash markets, while geographic and rating diversification are also taken into account.

The underlying ratings of the bonds in the reference portfolio must be BBB-/Baa3/BBB- or above and not on negative watch. Some 60% of Series 10 is rated double-A and 28% is rated triple-A.

Auctions will be run in case of credit events (failure to pay and restructuring), and cash settlement will be hardwired in the future, according to Markit. The initial participating dealers include Citi, Goldman Sachs, JPMorgan, Lehman Brothers, Merrill Lynch, Morgan Stanley and UBS. MCDX trades are expected to be DTCC-eligible four to six weeks after the launch of the index.

Market participants expect the municipal bond market to continue its strong issuance trend this year. In 2007 new issuance volumes amounted to US$429bn, with US$2.6trn currently outstanding.

AC

Remittance relief?

Mixed response to improvement in ABX credit performance

With the rate of total delinquencies slowing materially across all Markit ABX indices for the second month in a row, some traders expect the index to stage a recovery. However, others suggest that the improvement in credit performance is only temporary and deterioration will accelerate again in the near term.

"ABX witnessed a significant tightening at the triple-A level after the remittance reports were digested, spurring outright long activity as well as short covering," explains one synthetic ABS trader. "But the BBB/BBB- sub-indices continue to decline, reflecting the expectation that the underlying bonds will default over the next one to two years. The market is currently pricing in a value of 15%-20% at this level, but this will eventually reduce to zero as losses build up in the underlying."

Overall, there appear to be two schools of thought about the performance of the index. Some traders believe that ABX triple-As will now recover from their lows, while others reckon the improved remittance data is driven by seasonal factors and is therefore only temporary because US housing continues to decline and there is little hope of immediate government intervention in the sub-prime sector.

The April remittance reports show a continued decrease in the 30- and 60-day delinquency buckets in the majority of trusts across all the indices. The 30-59 and 60-89 day delinquency buckets respectively decreased by 1.1% and 0.6% on average.

But around 45% of bonds showed higher 30- and 60-day delinquency buckets. Analysts at Barclays Capital suggest that this is somewhat foreboding, as the April remittance date has historically yielded improving performance.

"Given the historical seasonal pattern of significant percentage change improvements in 30- and 60- day delinquencies in April, we believe this latest report portends additional collateral performance deterioration over the next several months," they say.

Later stage delinquencies also continue to increase fairly quickly. Monthly aggregate 60+ day delinquencies rose by 125bp, 45bp, 130bp and 166bp (compared with the 163bp, 391bp, 143bp and 207bp rises last month) for Series 06-1, 06-2, 07-1 and 07-2 respectively.

Meanwhile, the significant disparity between ABX prices and the underlying RMBS – respectively estimated to be 90-95 and 70 cents on the dollar – continues to frustrate some investors. Investors complain that it remains difficult to get a sense of where cash bonds are trading because they are bought infrequently and at a discount – and so the index is prone to volatility due to hedge fund short covering as there is no-one buying protection. By way of comparison, CMBX tracks the underlying CMBS much better – albeit the basis has widened from 20bp to around 50bp so far this year.

The trader doesn't expect the US sub-prime securitisation market to return – especially as financing via covered bonds is becoming an increasingly attractive funding option for banks. "The future of the ABX index consequently depends on whether hedge funds continue to trade it. The BBB/BBB- indices are already pretty redundant – and hedge funds are always looking for the next volatile product," he concludes.

CS

Eyes wide shut

Lessons to be learnt from write-down post-mortem

Analysts are hailing the UBS shareholder report on its sub-prime related write-downs as a must-read for all in the structured credit business. The report, which lays bare full details of the bank's US$18.7bn losses in US mortgage securities in 2007, has prompted speculation that other banks with similar exposures will now revisit their risk management practices.

"One can learn a lot from [the UBS report] on how to do – or better, on how not to do – risk management," says Philip Gisdakis, a senior credit strategist at UniCredit Research. The findings of the report, says Gisdakis, appear to be "violations of quite basic risk management principles that credit portfolio managers should always have in mind".

Highlights from UBS' fifty-page report show that trading strategies in the alternative investment management arm of the bank, Dillon Reed Capital Management, contributed approximately 16% of UBS' sub-prime losses for 2007, while the CDO desk contributed approximately 66% of UBS' sub-prime losses during the same period.

Some 16% of the CDO desk's losses were located in the CDO warehouse, while 64% of the desk's losses were in the bank's super-senior positions in ABS CDOs. Losses from the FX/cash collateral trading business accounted for a further 10% of the write-downs.

A key admission made by UBS points to monitoring and reporting positions, where the market risk control (MRC) took data feeds from the front-office systems, but in analysing the retained positions, it generally did not 'look through' the CDO structure to analyse the risks of the underlying collateral.

"The CDO desk does not appear to have conducted such 'look through' analysis and the static data maintained in the front-office systems did not capture several important dimensions of the underlying collateral types," states UBS in the report.

For example, the static data did not capture FICO scores, 1st/2nd lien status or collateral vintage, and did not distinguish a CDO from an ABS. Furthermore, UBS acknowledges that the MRC did not examine or analyse such information on a regular or systematic basis.

Nor were there risk factor loss (RFL) limits specifically addressing sub-prime exposure, such as delinquency rates and US house price developments. And overall sub-prime exposure across the investment bank had not been managed using operational limits.

Despite rumours that other banks will now look to strengthen their structured credit risk management practices, it seems doubtful that other banks will be as transparent as UBS regarding their write-downs, according to UniCredit's Gisdakis.

"When you see a report that is so self-critical and so transparent, you have to ask yourself why it is so. The report must be seen as a political tool that will be used to shape the future of the bank. UBS has been one of the hardest hit names for sub-prime write-downs; plus the bank has new management, so will want to appear to be cleaning up its act," he says.

Looking ahead, analysts are not ruling out further-write downs from banks (especially in Europe). The good news, they say, is that the shock effect of bank losses has been priced in, with the market appearing less concerned about the latest round of write-downs because banks have been able to raise capital easily in the public market.

"The question remains who will next invest in troubled banks if write-downs continue to arise, as we believe they will for the next few quarters," says Andrea Cicione, senior credit strategist at BNP Paribas. He stresses that it is difficult to prepare for future write-downs when only net exposures to sub-prime losses, rather than the gross exposure, are made public.

AC

Babson hires technology cio

The latest company and people moves

Babson hires technology cio
Babson Capital has appointed Praveen Kumar as chief information officer. In his new role he will lead a Babson Capital technology team responsible for the operating systems that help manage clients' portfolios.

He reports to coo James Masur and will work with other Babson Capital leaders to develop and implement technology solutions throughout the company. "The strategic importance of our firm's operating platform and systems resources has risen to a new level," says Masur. "Our growth as a global business demands that we more effectively and efficiently develop products and services and acquire and integrate new investment managers, and our technology is instrumental to that effort."

Kumar, who has over 20 years of experience in technology leadership, moves over to Babson from Wellington Management in Boston, where he was director of enterprise systems. Previously, as head of the US technology application development group at JPMorgan Private Bank in New York, he helped develop from scratch new technology applications. At Lehman Brothers in New York, he managed a team that developed new financial services technologies and enhanced client support for trading areas.

ACA prolongs demise
ACA Capital Holdings has entered into a fourth forbearance agreement with its structured credit and other similarly situated counterparties. The agreement runs to 30 May, subject to extension upon certain circumstances, according to the company.

Under the agreement, the counterparties will continue to waive all collateral posting requirements, termination rights and policy claims relating to the rating of ACA Financial Guaranty Corporation, ACA Capital's financial guaranty insurance subsidiary, under their respective transaction documents - including any credit support annexes and similar agreements.

"While its eventual demise seems inevitable, counterparties appear unwilling to pull the plug, forcing it to endure a slow, lingering death. Most counterparties have written off exposure in full and, in our view, ACA's bankruptcy has been fully priced into markets since December," says Michael Cox, an analyst at RBS.

New Asia deputy for Calyon
Arnold Kan has been appointed deputy head of fixed-income markets for Asia ex-Japan at Calyon. He joins from JPMorgan, where he was head of sales for the FX and rates markets.

Kan will be based in Hong Kong and report to the head of Asian fixed-income markets, Frédéric Lainé. He will be responsible for expanding fixed-income sales in the region to new countries and customer bases.

AC

CDO recovery assumptions revised

A round up of this week's structured credit news

CDO recovery assumptions revised
S&P has revised its recovery-upon-default assumptions for CDO securities backed by certain US RMBS collateral, including Alt-A, sub-prime, home equity loan and tax-lien RMBS issued in the US during and after Q405. These revisions follow updates to the agency's recovery assumptions for certain RMBS securities, which were completed earlier this year.

S&P will apply the updated recovery assumptions when rating or surveilling the following securities:

1) ABS CDO tranches for which 40% or more of the collateral consists of the affected US RMBS; and
2) CDOs backed by tranches from other CDOs, which themselves are backed by the affected US RMBS in a proportion of 40% or more of their respective asset balances.

S&P believes the revisions are appropriate, given the observed difference between the actual behavior of the affected US RMBS collateral compared with its original expectations and the corresponding effect on the performance of CDOs holding such RMBS assets. For transactions that are still allowed to reinvest and substitute collateral, the agency will determine the 40% based on the then-current portfolio composition and the concentration limitations stated in the transaction's documentation. For ABS CDO transactions exposed to such RMBS securities to a lesser degree, it will assign recovery assumptions on a case-by-case basis.

To estimate recoveries for the affected CDOs, S&P first applied its newly revised RMBS surveillance criteria to the underlying sub-prime and Alt-A RMBS issued from fourth-quarter 2005 through fourth-quarter 2007. It then ran cashflow projections for the RMBS and used them in its cashflow analysis of the respective affected CDOs.

Next, the agency determined whether, in those circumstances, the affected CDOs would incur losses and estimated their recoveries. When determining the CDO recoveries, it applied the cashflows generated by the underlying collateral throughout the life of the CDO and did not assume any forced liquidation scenarios. S&P says that the changes to its recovery-upon-default assumptions may have a negative impact on the ratings assigned to the affected CDOs because a reduction in expected recoveries typically necessitates more subordination to sustain the ratings on the tranches.

SIFMA forms ratings task force ...
SIFMA has formed an industry-wide, investor-led task force to examine issues surrounding credit ratings. The aim is to pull together a broad, global position and translate it into a set of recommendations.

"Ensuring an open, competitive and more transparent credit rating system will be an important step towards restoring investor confidence and trust in our markets," says Tim Ryan, president and ceo of SIFMA. "This task force will take an integrated approach to exploring the credit rating system, including key issues such as the current use and quality of ratings, the issuer-pays model and rating agency independence. Given the global nature of the issues and the concerns voiced by regulators, SIFMA, with its worldwide financial services footprint, has a unique opportunity to be a leader in this space and serve as a catalyst for change."

SIFMA's goal is to formalise and advance the dialogue between its members and rating agencies that took place during the past year, through a single group that will focus on this issue. Additionally, the task force will interface with government officials, legislators, regulators, multi-lateral authorities and key credit rating agencies.

The task force will be led by Boyce Greer, president of the fixed income and asset allocation division at Fidelity, and Deborah Cunningham, chief investment officer at Federated Investors. Two dozen firms representing a cross-section of the financial services industry will offer their perspectives in addressing a full range of issues.

Individuals participating on the task force will lend specific expertise in key areas, such as structured finance, municipals and short-term debt ratings. AllianceBernstein, JP Morgan Chase, Citigroup, Vanguard and Schroders are among additional participating firms.

The move came at the same time that the SEC announced that it will publish by end-June its proposed ratings transparency rules, which it hopes will boost rating agency accountability, transparency and competition. The aim is for market participants to be better able to compare one rating agency with another.

One of the proposed requirements is that rating agencies share information on asset pools. Analysts at UniCredit suggest that this would have two advantages: first, disclosing information will improve competition among the agencies; and second, investors can better analyse the portfolios by themselves and form their own ideas about its creditworthiness. Furthermore, the proposals require that different symbols should be used to distinguish different asset classes.

... while Fitch reviews complementary rating scales ...
Fitch is currently engaged in a review to identify and evaluate potential complementary scales or indicators for structured finance transactions. The complementary scales are designed to improve transparency, provide market participants with additional information regarding structured finance securities and help address recent critiques of structured finance rating scales.

Currently Fitch has one ordinal rating scale used throughout its rating activities, including financial institutions, corporates, sovereigns and structured finance. In recent months, there has been increased debate, particularly in the policy-making community, about the appropriateness of this approach for structured finance transactions.

In 2007 Fitch carried out an in-depth verbal consultation process with major issuers, investors and regulators regarding complex structured finance products. At that time, feedback suggested limited interest in adopting a different rating scale for structured finance. However, seasoned market participants also typically indicated an understanding that the current rating scale, which addresses default probability, does not convey all of the risks present in structured finance transactions.

In an effort to add greater transparency and capture additional risks, Fitch has identified three potential additional and complementary rating scales for structured finance securities. These are as follows:

• Loss given default (LGD) ratings: This scale would attempt to quantify the recoveries on a tranche-level basis that a structured finance (SF) creditor would likely receive in the event of default. Such a scale would be similar to the recovery ratings Fitch introduced for corporate finance in 2005. An LGD scale would help address one of the common criticisms of SF ratings compared to corporates. That is, while a SF tranche and a corporate bond that are rated the same may possess similar default characteristics, actual losses given default may differ materially.
• Transition/stability ratings: Such a rating scale would attempt to capture the potential for rating migration – the rating transition or volatility risk. In other words, how likely it is that the rating would change for a given time period. This could be measured in a number of ways, either in isolation or in combination. For example, it could be measured over the short-term, from the perspective of current events (similar to the tranche-level outlooks currently being rolled out by Fitch across its SF ratings) or as a measure over the long-term, based on historical or projected performance of that asset class. It could also be measured as a purely relative measure distinguishing between static and dynamic pools.
• Collateral ratings: These ratings would attempt to measure on a consolidated basis the quality of the underlying collateral or pool of assets for a given transaction. Said differently, such a rating would assess the quality of the underlying assets, independent of all or most of the additional structural features that determine a traditional SF rating. However, such an approach may only be feasible for concentrated loan pools; e.g. CMBS.

Fitch will begin to issue exposure drafts on these proposals by early-June. However, in the interim, it is inviting preliminary feedback on the merits of these initiatives and other potential ratings scales (e.g. liquidity risk). Additionally, the agency will continue to monitor feedback on the idea of a fully separated primary scale for structured finance securities.

... and CMSA warns against separate SF ratings
The Commercial Mortgage Securities Association (CMSA) has responded to a request for comment from the International Organisation of Securities Commissions (IOSCO) regarding its Technical Committee Consultation Report on the Role of Credit Rating Agencies (CRAs) in Structured Financial Markets. The CMSA response includes recommendations for increasing the transparency of CRA methodologies relating to CMBS, while raising concerns about using separate rating symbols for the entire structured finance market.

"CMSA supports IOSCO's recommendations to deal with issues such as how credit rating agencies should avoid or mitigate potential conflicts of interest and improve the transparency of the ratings process," says Leonard Cotton, vice chairman of Centerline Capital Group and president of CMSA. "We urge caution, however, when implementing any recommendations that would create or encourage separate ratings schemes for structured finance vehicles and corporate or municipal debt."

In its response, CMSA commended the IOSCO CRA Task Force for not requiring that structured finance products be rated on a different scale from corporate and municipal bonds. At the same time, CMSA expressed concern about the IOSCO statement in its report that a CRA should disclose whether it uses a separate set of rating symbols for rating structured finance products, and its reasoning for doing so or not doing so. Such a disclosure requirement could result in separate rating schemes becoming a de facto rule or best practice in the industry to the detriment of investors and the capital finance markets.

"CMSA strongly believes that a separate ratings scale could make the structured products market even more volatile by adding to investor confusion," explains Dottie Cunningham, ceo of CMSA. "It would force investors to revise their investment policies to incorporate the new rating structure and develop a new analytical and monitoring infrastructure to interpret the new ratings. These unintended consequences could increase costs for investors and further erode liquidity to the detriment of borrowers and the capital markets."

As an alternative to requiring that CRAs justify their decisions regarding rating symbols in disclosures, CMSA recommends that the CRAs issue additional analysis about the potential risk characteristics of rated bond loan pools, as well as new and targeted transparency related to the underlying rating methodology that is being employed in determining rating assessments.

Call for better fair value disclosure
Fitch Ratings says in a special report that companies should not stop accounting for assets at fair value in illiquid markets. However, better disclosure is required as to the rationale, assumptions and sensitivities behind these valuations.

"The most salient issue is not whether fair value per se should be used to report numbers, but how that fair value should be measured," says Bridget Gandy, md in Fitch's credit policy group. "If values are being taken from markets that are not striking a fair balance between buyers and sellers, it is hard to argue that those values are fair."

The report comes at a time when volatile and unstable conditions in the financial markets have caused many reporting financial institutions to call for a relaxation of fair value accounting, allowing issuers the option of when to apply fair value measurement and when to apply historical cost. Fitch argues that the fundamental distortions such unfettered flexibility would permit would not engender greater investor confidence in financial reporting, nor would it foster sound capital markets or sound financial institutions.

In the report, entitled 'Fair Value Accounting: Is It Helpful In Illiquid Markets?', Fitch notes that fair values are helpful to analysts and investors when they represent realistic and reliable indications of the net present values of future cash flows. The disconnect with market prices comes when there is no intention to sell an asset in the short term and a lack of market liquidity means that current values are either much higher or much lower than the amount that will ultimately be received for the asset. However, holding assets in trading books is a clear indication of a company's intent to sell in the short term and market values should be taken.

More extensive disclosure will help investors to understand the limitations around the values reported. These should include indications of market prices versus expected cashflows, amounts companies expect to lose in real cash on assets written down to market values and how such assets will be funded while they are held for longer than originally anticipated.

While Fitch does not think that market prices not directly related to the assets being valued using internal models should be required as inputs, the agency also does not think that they should be ignored. A company that is not using the best observable data available should explain why it is not using this, demonstrate why the alternative measurement is more appropriate and provide an indication of how the value would have differed if the market prices were used as inputs in the notes to the accounts.

SIV capital notes on negative watch
Moody's has placed the ratings of six SIV capital note programmes – totalling approximately US$3bn of securities – on review for downgrade. The affected vehicles include Asscher Finance, Harrier Finance, Kestrel Funding, Links Finance, Nightingale Finance and Premier Asset Collateralized Entity.

SIV capital notes have been the most negatively impacted by the funding and pricing difficulties that have caused the sector to all but disappear in the wake of the ongoing liquidity crisis. Moody's rating actions reflect further deterioration in the market values of SIV portfolios, and the limited benefits to capital notes of the various restructuring proposals implemented by bank sponsors.

The agency's review will focus on the decline in capital net asset value, which incorporates both crystallised losses following asset sales and unrealised losses. Moody's will also review the actual or potential impact on capital notes of any restructuring plans implemented by a vehicle's sponsor.

REIT TRUPs CDOs downgraded
Moody's has downgraded 53 tranches issued by 10 CDOs with significant exposure to residential mortgage REIT Trust Preferred Securities (Trups) and homebuilder securities. The rating actions were prompted by continued credit deterioration and defaults in the residential mortgage REIT and homebuilder sectors.

The affected CDOs include Attentus CDO Series I to III, Kodiak CDO I, Taberna Preferred Funding Series II to VII and Trapeza CDO X. The transactions have significant exposure to these sectors, ranging from approximately 25% to 50% of their aggregate portfolio balances. The rating actions also reflect uncertainties over final workout values, which are expected to be low, for defaulted assets in the underlying collateral pool.

Moody's outlook for REIT TRUP CDOs is negative for 2008. The agency will continue to monitor developments in the specialty mortgage area and will update the ratings of affected CDOs accordingly.

For a more detailed discussion on the sector, see this week's Research Notes.

S&P reports on UK Banking Act
Following the introduction of the UK Banking (Special Provisions) Act 2008 and its assessment of currently available information, S&P considers that at this point the legislation does not have an adverse effect on its ratings on securitisations involving UK deposit-taking institutions.

The legislation gives the UK Treasury powers to make certain orders regarding UK deposit-taking institutions. If an order was made with respect to a UK deposit-taking institution that was a participant in a rated securitisation, it could adversely affect the rights of the noteholders in that transaction.

The wide-ranging powers include various confiscation, appropriation and similar measures regarding the transfer of securities and property of a UK deposit taker and the assets, rights and liabilities of that deposit taker. Orders may have a retrospective effect and may make provision for nullifying the effect of transactions.

If the Treasury were to make an order regarding a UK deposit-taking institution, the securitisation transactions of that deposit taker could be affected. While there are no assurances on how the UK Treasury may use its powers, S&P's assessment of all the information it currently has leads it to believe that it can still rate existing transactions involving UK deposit-taking institutions up to and including the triple-A level.

Moody's affirms Ambac's negative outlook ...
Moody's continues to maintain a negative rating outlook on Ambac Assurance's Aaa insurance financial strength rating, following its parent company's earnings announcement. The agency states that Ambac's incurred losses in its direct RMBS portfolio and ABS CDO portfolio are now somewhat higher than the rating agency's prior expected-case loss estimate, but remain below the stress-case losses used to evaluate the company's capital adequacy.

In Moody's view, the substantial increase in losses reflects both the significant volatility of the company's mortgage-related risk exposures, as well as the challenges inherent in estimating the losses that will ultimately develop from this portfolio over time. The agency will continue to evaluate Ambac's mortgage-related exposures in the context of actual performance, as well as its developing view of the depth of the housing market's decline. In the event that Moody's ongoing evaluation of US mortgage market dynamics leads it to revise upward the stress case assumptions used to evaluate Ambac's capitalisation, the insurer's rating could be placed on review for possible downgrade if it fails to meet Moody's Aaa target capital threshold.

Ambac reported significant incurred losses in its mortgage-related insured portfolio during 1Q08 that contributed to a GAAP pre-tax net loss of US$2.8bn, including nearly US$2bn of combined credit impairment losses on ABS CDOs and loss reserve charges on direct RMBS exposures. Adding these losses to charges taken during prior periods, Ambac's cumulative credit impairment and loss reserve charge for its mortgage-related exposures was approximately US$3.3bn. Moody's prior expected case loss estimate for these exposures was US$2.4bn, with a stress case loss estimate of US$5.4bn (all figures are on a pre-tax basis).

... and rates BHAC triple-A
Moody's has assigned Aaa insurance financial strength ratings to Columbia Insurance Company (Columbia) and Berkshire Hathaway Assurance Corporation (BHAC). Columbia and National Indemnity Company (NICO) are the lead companies within the Berkshire Hathaway Reinsurance Group (BHRG), one of the core (re)insurance operations of Berkshire Hathaway.

BHAC is a new municipal bond insurer established by Berkshire in late 2007. All financial guaranty policies written by BHAC on or after 31 March 2008 are unconditionally guaranteed by Columbia. The rating outlook for these companies is stable.

Columbia's rating is based on its extremely strong capitalisation, moderate risk exposures and healthy profitability, as well as the implicit support from the broader BHRG and Berkshire, according to Moody's. Columbia reported statutory surplus of US$8.7bn at year-end 2007, well in excess of its loss and loss adjustment expense reserves of US$2.bn. However, Columbia is exposed to significant earnings volatility from year to year, adds Moody's. The vast majority of its premiums consist of reinsurance assumed from NICO and other Berkshire affiliates, including exposure to large-limit transactions written by NICO.

Moody's says that BHAC's rating reflects its strong capital position and its high-quality insured portfolio of public finance credits, as well as the guaranty from Columbia and implicit support from Berkshire. Moody's notes that the guaranty from Columbia ranks junior to policyholder claims at Columbia. Nevertheless, the guaranty in combination with Berkshire ownership and oversight represents robust support for BHAC.

Canadian ABCP restructuring approved
The Pan-Canadian Investors Committee for Third-Party Structured ABCP has announced that noteholders have voted overwhelmingly in favour of its restructuring plan. Approximately 96% of noteholders, by number and value, represented in person or by proxy at the meeting, endorsed the plan. More than 90% of noteholders by value cast votes on the plan.

The favourable vote follows eight months of work to restructure the remaining 20 trusts covered by last summer's Montreal Accord, affecting US$32bn of notes. The results of the vote will now be brought forward for consideration by the Ontario Superior Court of Justice for final sanction at a hearing currently scheduled for 2 May.

"We are delighted by the results of today's vote and will now proceed to a final hearing before the Court," says Purdy Crawford, chair of the Investors Committee. "If the Court sanctions the plan as approved by noteholders, it should be fully implemented before the end of May."

Rating Change index continues decline
Fitch says that its ESF Index of Rating Change has declined for the last three consecutive quarters, following 11 quarters of improvements. The ABS, CMBS and RMBS indices deteriorated as a result of the downgrade of the monoline insurers, Ambac Assurance UK Limited, FGIC Corporation and XL Capital Ltd.

The just-published ESF Index of Rating Change shows a decline of 1.1% in the quarter to March 2008. Upgrades marginally outweighed the downgrades by 43 to 42, a ratio of 1.02.

"The ABS index was the weakest this quarter, with a decline of 2.9%," says Charlotte Eady, associate director in Fitch's CMBS team. "Although the RMBS index was almost flat with a 0.01% fall, all asset classes experienced negative movement. This hasn't occurred since June 2004."

"Of the 42 downgrades, eight were CDO tranches, while the remaining 34 were limited to triple-A minus rated credit-wrapped tranches," adds Rodney Pelletier, md in Fitch's CMBS team. "The value of non-CDO securities downgraded amounted to €3.5bn, while the value of CDO tranches downgraded was €2.1bn. Overall, the value of the downgrades was considerably higher than that of the upgrades."

The CDO index declined by 2.3% in Q108, which is a marked difference from last quarter when it declined by 22.3% The CMBS index was down 1.2% versus a positive change last quarter of 0.4%.

Italian structured finance bonds were the weakest performers by country, declining 4.7% due to nine downgrades – all credit-linked to monoline insurers. The UK index declined marginally (0.6%) as a result of 25 downgrades.

The ESF Index of Rating Change measures the weighted average change in the Fitch rating factor across all ESF transactions with a public rating by Fitch.

Synthetic CDOs hit
S&P has taken credit rating actions on 130 European, 64 US, 18 Asia Pacific and 15 Japanese synthetic CDO tranches.

Of the US transactions, S&P lowered 116 ratings, one of which was also placed on credit watch negative, while 14 others remain on watch negative. S&P also placed four other ratings on watch negative; withdrew 29 ratings after the classes were terminated; and affirmed 15 ratings and removed them from watch negative.

Regarding the affected European transactions, the ratings on 93 tranches were removed from credit watch with negative implications and lowered; 20 tranches were lowered and remain on credit watch negative; five tranches were lowered; two tranches were removed from credit watch positive and placed on watch negative; one tranche was removed from watch negative and affirmed; two tranches were removed from watch negative and raised; two tranches were removed from watch positive and raised; and five tranches were raised.

Of the 118 tranches lowered 51 reference US RMBS and US CDOs that are exposed to US RMBS, which have experienced recent negative rating actions; and 67 have experienced corporate downgrades in their portfolios.

Of the affected Asia-Pacific transactions 14 were downgraded. Four of these CDOs remain on watch with negative implications. Additionally, S&P raised the ratings on two CDOs and affirmed the rating on one CDO. Another was revised from watch with negative implications to developing implications.

These rating actions follow a review and watch placements on 14 April. The CDOs that had their ratings lowered have SROC levels that are less than 100%.

This indicates that they have insufficient credit support to retain their former ratings following negative rating migration within the reference portfolios. It should be noted that these deals do not have exposure to US sub-prime RMBS transactions.

For the transactions that had their ratings raised, their SROCs are maintained above 100% at the higher rating level. For the transactions that had their ratings affirmed and removed from watch with positive implications, the review indicated that they have insufficient credit support to be raised to the higher level.

Meanwhile, the action on the Japanese CDOs relates to the lowering of ratings on 18 tranches from 13 transactions. At the same time, S&P raised its ratings on two Japanese synthetic CDO tranches relating to two different transactions (also listed below).

Regarding the downgrades, S&P removed its ratings on five tranches from credit watch. The ratings on the other 13 downgraded tranches remain on watch with negative implications.

Fitch CDS pricing services launched
Fitch Solutions has announce that Catlin Group Limited, a Bermuda-based international specialist property/casualty insurer and reinsurer, has chosen its CDS Pricing and Market Implied Ratings services to help improve the pricing of its emerging market credit insurance products. Having commenced on 21 April, the services combine Fitch Solutions' CDS data with its proprietary CDS and equity market implied ratings, and Fitch Ratings' fundamental credit ratings.

Furthermore, these services also now benefit from the addition of Fitch CDS Benchmarking for highly illiquid entities. This new pricing tool provides indicative spread values for CDS where there is insufficient market information to determine a consensus price based on market-maker contributions.

"Knowing what other markets price securities at is an increasingly important benchmark for us when pricing our own credit insurance products," says John Lentaigne, credit and political risk underwriter at Catlin Bermuda. "Fitch has a strong track record in assessing credit risk and the combination of its CDS Benchmarking, CDS Pricing and Market Implied Ratings tools will give us unique coverage of emerging market entities and a vital insight into how the market is pricing risk in emerging markets."

Thomas Aubrey, md at Fitch Solutions, adds: "We are delighted to be working with Catlin Group. Assessing credit risk is central to the investment and risk management process – and the combination of our CDS Pricing, CDS Benchmarking and Market Implied Ratings services provides users with a single platform to derive key insights into the direction of credit risk and ultimately make better investment decisions."

The indicative spread value for Fitch CDS Benchmarking is determined by looking for proxies in the liquid CDS universe, where a consensus price is available. It is available for both single name CDS and ABS CDS, either as additional data coverage within the CDS consensus data pricing feed or as a stand alone data feed.

CS

TRUPs CDO update

With TRUPs CDOs having been more negatively impacted than CLOs by the market's liquidity and risk re-pricing, JPMorgan's structured finance research team reviews the sector's credit performance and outlook

Trust preferred (TRUP) CDOs are backed by what is essentially preferred stock, surplus notes, or subordinated debt from bank, insurance and REIT companies. The market began in 2000 with 100% bank transactions and gradually included more insurance or REIT collateral as a means of additional spread or diversification. In the second half of 2007, investors switched back to bank and insurance portfolios, given credit concerns in housing (see Chart 1).

Transactions with REIT collateral have experienced a number of defaults and ratings downgrades; subordinate tranches likely face large losses, depending on the proportion of mortgage REIT and homebuilder exposure. Bank and insurance transactions have experienced relatively few deferrals, but nonetheless – as with other structured products – the new issue market has dried up.

Transactions from 2002-2004 are unlikely to be called anytime soon. Thus, secondary is the largest potential source of paper, but this is even more shut down given the 'niche' nature of TRUPs and lack of transparency and liquidity. Still, SIVs were large buyers of triple-A TRUPs CDOs and we expect distressed purchase opportunities.

In this note we review the credit outlook and explore yield stability under various default scenarios. We establish reasonable price ranges, assuming an illiquidity premium to CLOs and in the context of significant extension risk (e.g. to 20-30 years).

Bank and insurance TRUPs CDOs
The key performance variables for TRUPs investments are going to be (not surprisingly), defaults, recoveries, prepayments and call prospects.

Defaults
Bank defaults have historically averaged about 0.25% a year, although in most years there is only a handful and during bank crises the fail rate has spiked to 1%-2% (broad commercial banks) or 5%-10% (savings and loans during the S&L crisis) (Chart 2). There were no failures in 2005 or 2006, according to the FDIC, and three in 2007; these three were not driven by sub-prime (according to Fitch). Broadly, we do not anticipate a widespread sub-prime crisis in the regional bank sector; what could become an issue is direct commercial or real estate-based lending.

However, that said, the banks included in TRUPs portfolios are often not covered by industry analysts (market cap US$25m-US$250m), so the manager's due diligence and credit skills are imperative. For purposes of modelling TRUPs CDOs, we feel like something between 0.75%-1% is a reasonable base case default rate going into a weak real estate environment – although we note that Fed rate cuts and a lessening of competitive pressures may actually bring opportunities for some smaller banks. We also consider stress cases extending to 4%.

Insurance impairment rates have been higher historically (roughly 0.80% on average), although the series has been less cyclical (Chart 3) and the market environment is solid currently. Below we run stress cases on a 100% bank transaction, but would probably use a 0.75% base case assumption for the insurance bucket of any hybrid transactions, with stress defaults up to 2%-3% annually. Insurance companies/DPCs in the financial guaranty business, even away from sub-prime or ABS, should be stressed more severely if they are required to post collateral on mark-to-market declines.

Recoveries
There is little applicable recovery data, except for an earlier Moody's study on corporate preferred stock recoveries, and the last seven years' experience of TRUPs collateral. However, recoveries are expected to be low, given TRUPs' deep subordination.

Recoveries are also likely to be delayed – the ability to defer interest is a condition of equity capital treatment for bank and insurance issuers. A large percentage of deferrals have eventually cured (Table 1). This can be because the issuer recovered or was acquired (more common in bank than insurance credits).

To be conservative we assume a bank recovery rate of 25% in our best case environment, down to 5% in severely stressed environments. This can be thought of as a 0% recovery on non-cured assets and a full recovery on the rest.

We also assume a 24-month recovery lag. For insurance buckets, we would probably cut the bank recovery assumption in half.

Prepayments
Bank and insurance TRUPs usually have a 5- to 10-year non-call period, again for regulatory capital reasons. Some issuers are likely to call their transactions in the future, either because of organic credit improvement, acquisition by a higher-rated entity or a pending switch to floating rates for hybrid fixed/floating issues.

We are not so bold as to forecast prepayment incentives 3-5 years out, but it seems likely that calls would be less attractive to issuers now, given the credit crunch, diminished CDO issuance and lower floating interest rates. Thus we run three prepayment assumptions, ranging from what was a common base case scenario earlier (15% CPR at year five and 3% thereafter), half of this, and no prepayments.

Call prospects
CLOs usually have optional calls occurring after year three or four; base case cashflows are run to a natural amortisation cycle, which is much shorter than in TRUPs CDOs (8-12 years). TRUPs CDOs also have optional calls, but here we look strictly at auction calls, which typically kick in at year 10 (e.g. beginning 2016-2017). These are structured in order to prevent average life extension to 20-30 years, although the mechanism has not been tested (the market is too young).

Auction calls are mandatory: if market value proceeds are sufficient to pay down debt and any due fees, the manager is required to execute. Although the liquidity crunch is severe, 10 years out is a long time; we would assign at least some probability (perhaps 1/4-1/2) to auction calls going through in a non-catastrophic bank environment.

Cashflow stability
First we look at the cushions to default rates for a typical 100% bank transaction issued in 2007. Although different transactions may have different capital structures, the triple-B tranche is at risk here by around 1.2%-1.5% annual defaults, and single-As are at risk near 1.8%-2.2% (Charts 4-5).

Sensitivity to both recovery rates and prepayment speeds increases up the capital structure, in part because the most subordinated tranches are extremely extended and do not amortise early, so a small change in the default rate is applied over a longer time period. This analysis suggests that the most junior tranches are safe under a base case, but have little cushion to a stressed environment; senior tranches appear adequately supported in general – although in a stressed case they would experience ratings and MTM volatility.

Their breakevens are similar to CLOs. However, at current steeply discounted prices, even junior tranches may offer attractive returns. We thus move on to a yield stability analysis.

Yield prospects
In a market that is not trading, it is difficult to guess at clearing levels, so instead we focus on what combination of defaults, prepayments and prices will produce required discount margins (DM). In our view, TRUPs CDOs should be marked to yield several hundred basis points back of CLOs (more at subordinate levels), in line with middle market illiquidity premia. That said, at these levels they would offer attractive buy-and-hold returns.

However, there are a few points to note. Triple-As and double-As are generally insensitive to our default scenarios. The triple-A return is highly sensitive to prepayment speed since it amortises first – many TRUPs CDOs have a senior 'turbo' feature past the auction call, which further speeds amortisation.

In fact, higher default rates improve triple-A returns (e.g. 50bp-100bp from 1% CDR to 4% CDR), since they increase OC-related pay-downs. Past the single-A level, credit becomes the main (and negative) driver; single-A returns should steady until the 2.5% default scenario. The triple-B tranche looks particularly attractive in mid-range default scenarios (from 0.75%-1.25%) because it benefits from an additional OC test, which is dedicated to paying down the triple-B alone. (Other transactions may achieve this through a reverse 'turbo' feature, which pays down triple-B or double-Bs prior to the auction call date using excess equity proceeds).

However, as we mentioned earlier, there is little cushion to stressed defaults, and at 1.25% CDR returns start to decline quickly. Similarly, equity has a considerable amount of upside but is extremely sensitive.

The long-dated nature of TRUPs CDO cashflows implies significantly lower prices versus CLOs (in order to achieve a similar DM). Thus, TRUPs CDOs have been even more negatively impacted from the general risk and liquidity re-pricing than CLOs. But the flip side of this is that the upside to faster bank prepayments or successful deal auction calls is high (Chart 6). Between 100bp-500bp spread upside is possible, less when prepayments are near the base case, and in general only up to a mid level of defaults (1.25% CDR), because after that collateral loss becomes too extensive for the auction call to go through.

One exception might be the 5%-10% of mezzanine liabilities (single-A to double-B), which were structured as fixed-rate or hybrid securities, yielding a spread over Treasuries. These may out-earn Libor-based coupons by approximately 2%-3% over the next several years (rather than gradually). Especially at the equity level, default timing may be an important component of yield.

Lastly, we make the technical caveat that TRUPs CDOs often have large position sizes. It is not uncommon for individual exposures to tally 3%-4% in size. This means that a flat 1% CDR, run in Intex, is standing in for a large, concentrated default occurring at random times..

Furthermore, with regard to hybrid transactions that have a handful of REIT exposures, these may eat up all available debt subordination and leave little cushion for any bank or insurance losses. We consider REIT performance below; default assumptions should be input on a case-by-case basis for these names.

REIT performance
Most of the deferrals on 2005-2007 vintage TRUPs CDOs in Table 1, according to Moody's, are due to REITs or homebuilders. By our estimate, roughly half the TRUPs CDO issuance from that period had REIT exposure; that implies deferral rates on REIT-exposed pools more like 7+%. There is little publicly tracked data on lower-rated REIT performance, but as of 30 September there was an average deferral rate of 3.6% across REIT focused pools, with nearly 20% rated triple-C or less (Chart 7).

Although REIT transactions tend to have more subordination, the breakevens in Chart 4 indicate that BBB/BB tranches could be wiped out, with single-As and double-As at risk (capital structure probably extending from 15%-40%). Collateral deterioration has led to large and widespread downgrades in REIT TRUPs CDOs (Table 2).

Government intervention could relieve some homebuilder prospects, whether through a sub-prime bailout that limits foreclosed supply or retroactive tax relief, but we are not convinced this would make its way through the issuer capital structure in the form of actual dividend payments or higher recoveries. Even apart from housing concerns, hotel or retail REITs may see earnings shrink in the face of a consumer-led recession.

Ultimately, given size concentrations, REIT and CMBS exposures will need name-by-name analysis for cashflow or valuation purposes (across a pool a default assumption of 15%-25% seems reasonable, given the triple-C exposure). For predominantly bank/insurance transactions with modest REIT exposures, this means a binary name-by-name stress (100% CDR, 0% CDR and so on).

© 2008 JPMorgan. All rights reserved. This Research Note is an excerpt from JPMorgan's Global CDO Weekly Market Snapshot, first published on 18 April 2008.

Trading ideas: boxed in

Dave Klein, senior research analyst at Credit Derivatives Research, looks at a capital structure arbitrage trade on Pactiv Corp

Credit and equity risk are unambiguously linked, as the risk of debt holders not receiving their claims is akin to the risk of equity prices falling to zero. Both credit and equity risk are directly tradable with liquid instruments, such as CDS and equity puts. In this trade, we analyse hedging CDS directly with equity.

The trade exploits an empirical relationship between CDS and equity and an expectation that equity drops precipitously in the case of default. For certain names, the payout from buying CDS protection and buying equity behaves like a straddle.

If equity sells off, we expect CDS to sell-off more in dollar terms. If CDS rallies, we expect equity to rally more (again in dollar terms). This is also the basis for the so-called 'wings' trade, where CDS is financed using equity dividends.

For this trade, we choose our hedging ratios based on a fair-value model that derives CDS levels from equity prices and implied volatility. Given the straddle-like payout, we are going long volatility and taking advantage of a non-linear relationship between CDS and equity. This trade on Pactiv Corp. (PTV) takes advantage of this relationship by buying equity shares and buying CDS protection.

Delving into the data
When considering market pricing across the capital structure, we compare equity prices and equity-implied volatilities to credit market spreads. There are a number of ways to accomplish this, including the use of structural models that imply credit spreads (through an option-theoretic relationship) from equity prices and the analysis of empirical (historical) relationships between the two markets. We refer the reader to a CDR Trading Technique article – Capital Structure Arbitrage – for more detail.

The first step when screening names for potential trades is to look at where equity and credit spreads stand in comparison to their historic levels. Recently, PTV's CDS has outperformed its equity and this trade is a bet on a return to fair value.

In order to judge actual richness or cheapness, we rely on a fair-value model and consider the empirical relationship between CDS, implied volatility and share price. Exhibit 1 plots five-year CDS premia versus an equity-implied fair value over time.

Exhibit 1

If the current levels fall below the fair value level, then we view CDS as too rich and/or equity as too cheap. Above the trend line, the opposite relationship holds.

At current levels, PTV CDS is rich (tight) to fair value. This bolsters our view of buying protection on PTV.

Exhibit 2 charts market and fair CDS levels (y-axis) versus equity prices (x-axis). With CDS too tight when compared to equity (with volatility set to our expected fair value), we expect a combination of shares rallying and CDS widening.

Exhibit 2

The green square indicates our expected fair value for both CDS and equity when implied volatility is also considered. Note that these levels fall below the green line, as we expect a combination of spread widening, equity rally and implied vol movement to bring PTV back to fair value.

Hedging CDS with equity
Our analysis so far has pointed to a potential misalignment between the equity markets and credit spreads of PTV. It would appear that we should buy protection (sell credit) against a long equity position.

As default approaches, we see CDS rates increase (to points upfront) and equity prices fall close to zero. In this situation, our equity position will drop in value (bad for us), but this loss should be more than offset by our gain due to the CDS sell-off. If equity rallies, we expect CDS to rally as well.

Exhibit 3 charts the P&L for the trade after two months for various CDS-equity price pairs. The green square shows the expected P&L for a return to fair value.

Exhibit 3

The longer we hold the trade, the more difficult it will be to make money, given the negative carry/negative roll-down we face. However, we believe we can exit profitably in a reasonably short time period.

The main trade risks are that PTV volatility drops and we are unable to unwind the trade profitably or that PTV begins trading under a different regime and the current vol-equity-CDS relationship no longer holds.

Risk analysis
This position does carry a number of very specific risks.

Recovery and 'default' stock price assumption: In the default scenario the cheapest-to-deliver CDS obligation may have a higher than expected market value and the stock price might not fall as assumed.

CDS present value (PV): The CDS PV is an expected value, but not necessarily a realised outcome. In practice, the CDS may trade on an up-front and/or running basis.

Corporate actions: Spin-offs and private equity buyouts, for instance, could radically change the equity-CDS relationship, leaving investors with a mis-hedged position. It is our expectation that an LBO (however unlikely) would be a positive event for this trade as we would expect CDS to sell-off and equity to rally. A company bail-out (a la Bear Stearns) would have a massively negative impact on the trade, as credit would rally and equity value would be destroyed.

Mark-to-market: In our view, credit and equity markets operate largely independently and this can lead to trade opportunities. At the same time, however, any relative mis-pricing may persist and even further increase, which could lead to substantial return fluctuations. Additionally, the trade faces a fairly substantial bid-offer to cross in CDS. The negative carry and rolldown hurt us the longer we hold the trade.

Negative carry: As constructed, this is a negative carry trade. We go long equity and buy protection, both of which cost us. The longer we hold the trade, the more difficult it becomes to recoup our costs.

Overall, frequent re-hedging of this position is not critical, but the investor must be aware of the risks above and balance that with the negative carry. If dynamic hedging is desired, this is best achieved by adjusting the equity position, given transaction costs.

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. Our data on liquidity, created from the volume of bids, offers and trades we see each day, provide us with significant comfort in both the ability to enter a trade in PTV and the bid-offer spread costs.

PTV is a reasonably liquid name and bid-offer spreads are around 10bp. PTV is also liquid in the equities market.

Fundamentals
This trade is based on the relative value of PTV's CDS and equity, and is not motivated by fundamentals.

CDR's MFCI model assigns a negative score to PTV and indicates the company is trading too tight when compared to similarly ranked peers. PTV has middle ranks for most MFCI factors except for its high sales-to-debt and FCF ranks. The MFCI-expected fair value for PTV is 153, inline with our CSA-implied fair value of 135bp.

Summary and trade recommendation
Is the credit rally over? We hesitate to venture a guess. Given how quickly credit has rallied over the past month, it is no surprise that most names' CDS are now trading tight to fair value.

In our recent CSA strategy piece, we highlighted PTV as especially disconnected from its past credit-equity relationship. By buying CDS protection and buying stock, we are betting on a return to fair value and, given PTV's negative MFCI score and high expected CDS spread, we anticipate that the company's equity will outperform its CDS over the next few months.

Buy US$10m notional Pactiv Corp. 5-Year CDS at 100bp.

Buy 17,700 Pactiv Corp. shares at a price of US$24.02/share to pay 100bp of carry.

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

Copyright © 2008 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).