Contents

Rating-implied RelVal

New pricing model identifies opportunities in shifting risk premiums

Historical CDS index risk premiums based on S&P assumptions are considerably higher than they were 12 months ago. A new rating-implied pricing model aims to identify relative value opportunities arising from shifts in risk premiums relative to historical iTraxx/CDX prices.

The new methodology, developed by Dresdner Kleinwort (DK)'s structured credit strategists, identifies opportunities in standard tranches, bespokes and other synthetics based on rating-implied spreads, which can help buy-and-hold investors look beyond current short-term market conditions and focus on historical long-term averages. Rating-implied tranche spreads reflect the long-term credit history and historical pair-wise default correlation of the underlying pool.

"While the base correlation approach has become standard for pricing bespoke tranches and trading correlation, it is not suitable when analysing the historical impact of pair-wise default correlation in investment grade portfolios. Therefore, we have developed a new pricing model based on ratings and pair-wise default correlation," explain DK strategists Andrea Loddo, Priya Shah and Domenico Picone.

Rating-implied spreads can vary according to the set of correlation and default probability assumptions used. The strategists considered a set of assumptions based on S&P's CDO Evaluator model for their analysis. Not all investors believe that these assumptions reflect historical fundamentals, however, so the model also allows alternative inputs to be used.

Under the model, the strategists broke down the differences between market and rating-implied expected losses (ELs) into two components: a correlation component that reflects the impact on the price of the correlation relative to the historical pair-wise default correlation; and a CDS component, which reflects the fluctuations in CDS spreads relative to historical levels. "Given that the market is currently pricing in higher EL than those implied by the ratings, the CDS component is currently positive for all tranches. The correlation component can, however, be negative or positive according to the implied correlation of the particular tranche we consider," they note.

In a scenario where spreads and/or correlation converge towards more fundamental levels, a long position will be beneficial when the overall difference between market EL and implied ratings EL (defined as the risk premium) diminishes – thus resulting in a mark-to-market profit. However, among all the tranches that exhibit this behaviour, relative value opportunities can be identified by looking at the sensitivity relative to a decrease in correlation or spreads.

"In summary, any buy-and-hold investor needs to consider technical impacts not only on the tranche they are concerned with, but also other parts of the capital structure, as these factors can significantly affect the available opportunities and also timing considerations," the strategists add.

In order to calculate synthetic tranche spreads, the strategists generated correlated standard normal random variables and extracted the time of default from the default probability curve, which showed the asset default at year five. Recovery rates are based on market data, as the analysis focuses on the effects of the base correlation and market CDS relative to historicals.

The model relies on historical observations and the variability over time of the input data is thus extremely low. Simulations of default scenarios with these underlying assumptions should provide an indication of a long-term credit trend of generic obligors with a specific credit rating.

On the other hand, the base correlation model captures the current market conditions, which can be heavily affected by factors such as volatility, liquidity, correlation and, eventually, leverage effects. Since comparing model spreads for all tranches on a uniform basis is not easy, tranche PVs of expected loss were calculated and compared in order to identify relative value opportunities.

The analysis demonstrates that relative value opportunities are particularly attractive at the senior end of the capital structure (12%-22% and 15%-30% tranches), especially on the five-year maturity of the iTraxx/CDX indices.

CS

Liquidity boost?

Mixed response to Fed's unprecedented easing

Unprecedented easing by the US Federal Reserve last week led some analysts to call the bottom of the decline in ABX prices. But others believe that the central bank's actions has tarnished its reputation, while still others say that the government hasn't gone far enough to address the country's mortgage crisis.

"In our view, the probability that we have seen a bottom on ABX prices has moved up substantially on the Fed easing and the explicit assistance of financial institutions. The combined Fed commitment across the Term Auction Facility, the Term Securities Lending Facility and term repos now totals US$400bn, and will support mortgage security prices in the near term," note structured credit analysts at JPMorgan. (The Fed's intermediation over Bear Stearns is also expected to act as a backstop for ABS prices – see separate story.)

The Fed's actions initially led to a rally in the Markit ABX index, with short covering driving prices up by as much as six points intraday in ABX.06-2.AAA. The index ended last week 4.52 points higher on the 06-2.AAA, while the closing price yesterday, 25 March, had edged even higher to 77.17.

Although the increased liquidity will support pricing, the mortgage market also requires a significant capital infusion, the JPMorgan analysts say. A steeper yield curve will boost bank earnings and their ability to absorb write-downs, but the steady rise of credit-impaired mortgage assets will offset any such benefits.

"As a result, we believe government assistance is the next leg in the mortgage market's recovery. Given a near crisis of confidence in the banking system in the past week, the chaos in financial markets should incentivise the government to abandon its 'brinksmanship' policy and use public funds to address the mortgage crisis," the analysts add.

Such government assistance is likely to be via a dramatic expansion of the FHA, as per Congressman Barney Frank's recent proposal (see last week's issue), but regardless will require a significant commitment of government funds (around US$300bn-US$400bn). Once a plan gains greater certainty in terms of timing and size of commitment, triple-A ABX prices are expected to rally by 15 or more points.

Analysts at BNP Paribas describe the Fed's actions as an attempt to inch "its way to a full monetisation of housing". But they view the move as negative, with the Fed essentially lowering its lending standards by accepting weaker collateral (investment grade versus only triple-A) from weaker players (primary dealers and brokers versus only commercial banks) and for longer maturities.

"The purpose of this exercise has been solely to benefit the broker-dealer community on several fronts," the BNP Paribas analysts argue. First, it allows prime brokers to force deleveraging onto hedge funds via additional collateral requirements, which can be easily posted with the Fed (see last week's issue).

Second, it is an attempt to reduce the risk premia across asset classes by accepting a wider range of collateral in the belief that buyers will return to risky assets simply because liquidity has been made available on a temporary basis. This has helped to alleviate balance sheet concerns for the financial institutions still holding such risky assets.

Finally, the analysts note that, with the assistance of the rating agencies, the Fed is preventing the correct pricing of triple-A rated CDOs. This comes at the same time that Fannie Mae and Freddie Mac have been permitted to reduce their capital cushion by a third and spend at least US$200bn on new mortgages and RMBS, while serious asset impairment continues across the mortgage sector.

"The Fed has tried various measures to tackle financial problems, which it believes are, and continues to paint as, liquidity concerns, while there remain real solvency issues which it has been unable to overcome. In endeavouring to do so, it has also unleashed inflation and creditworthiness concerns surrounding its own wellbeing, while providing no true assistance to the underlying economy," they conclude.

CS

Backstop tactics

Bear Stearns bail-out could support market liquidity

The US Federal Reserve this week adopted an unusual approach to public crisis management by essentially becoming a backstop for the mortgage market. It is hoped that the central bank's intermediation over the sale of Bear Stearns will help bolster liquidity in the structured credit markets.

Rather than simply recapitalising Bear Stearns, the Fed has established a new company to manage the bank's US$30bn asset portfolio – against which it has lent JPMorgan the funds to purchase Bear Stearns. JPMorgan increased its bid to US$10 per share, valuing Bear Stearns at US$1.2bn rather than US$236m a week ago.

But the move has led to speculation that it is in reality the Fed's deal. "JPMorgan is just the conduit, as the Fed had to be seen to be preventing a financial market meltdown, rather than conducting a specific bank bail-out," analysts at BNP Paribas argue.

The Fed says it has taken this action, with the support of the US Treasury Department, to bolster market liquidity and promote orderly market functioning. JPMorgan will bear the US$1bn first-loss piece of the portfolio, with any realised gains accruing to the Fed. BlackRock has been brought in to manage and liquidate the portfolio under guidelines that are "designed to minimise disruption to financial markets and maximise recovery value".

The new entity is a Delaware limited liability company established for the purpose of holding the Bear Stearns assets. "Using a single entity (the LLC) will ease administration of the portfolio and will remove constraints on the money manager that might arise from retaining the assets on the books of Bear Stearns," the Fed says. The loan from the Fed and JPMorgan's subordinated note each have a term of 10 years, renewable by the Fed.

Market reaction to the initiative was mixed. One source indicated that it may help to crystallise fairer backstop prices for outstanding securities, thereby facilitating the return of liquidity to the primary ABS market.

But some observers say that the action is reminiscent of the creation of the Resolution Trust Corp that disposed of assets held by insolvent savings and loans banks in the 1980s, suggesting that it has fundamentally altered the role of government in financial markets. Analysts at Moody's confirm that the severity of market dislocations has led the Fed to take unusual decisions.

"Its plan to act as a market maker of last resort by lending up to US$200bn of Treasury notes in exchange for debt including private mortgage-backed bonds – thereby practically accepting illiquid securities as collateral in repos – the creation of a facility for broker-dealers (see separate news story) and its intermediation to avoid a disruptive unwinding of Bear Stearns are signs that the credit crisis may require more direct Treasury involvement. The balance sheet of the Federal government may well have to be deployed," they note.

But the Moody's analysts add that there is no plausible sudden ballooning of the Federal balance sheet that could endanger the US government's triple-A status. "Beyond providing off-balance sheet support in the form of guarantees, the US Treasury, should it decide to do so, would be able to raise the necessary amount to face plausible shortfalls – even in the range of several hundreds of billions of dollars – without impairing in any discernible fashion its ability to repay debt."

BlackRock, meanwhile, has this week – in response to the ongoing dislocation in the US mortgage market – established a separate company that will acquire and restructure distressed residential mortgage loans. Jointly sponsored by Highfields Capital Management, the new company is called Private National Mortgage Acceptance Company (PennyMac) and has a management team of mortgage industry veterans led by Stanford Kurland, chairman and ceo.

PennyMac will raise capital from private investors, acquire loans from financial institutions seeking to reduce their mortgage exposures, and seek to create value for both borrowers and investors through distinctive loan servicing.

CS

Credit and people markets decouple

The latest company and people moves

Credit and people markets decouple
Unlike the market itself, the small number of banks that are still looking to hire for their structured credit businesses are focusing on the junior rather than the senior end of the curve. Headhunters report that, notwithstanding the massive imbalance in favour of CVs over positions, the appearance of good juniors (and that word good cannot be emphasised enough) on their books is very brief indeed before they're snapped up.

Elsewhere – away from the continuing trend of widespread staff cuts – the consensus appears to be that distressed opportunities operations are the way forward. Every firm on the Street is at the very least talking about setting up or enhancing such operations.

At the same time, opportunity funds are providing the best opportunities for former senior bankers, with the buy-side still willing to make room for a select few if a bargain can be struck. One fund that has been in the works for a while and is now reportedly close to reaching its initial US$1bn fund-raising target is former Lehman global head of securitised products David Sherr's One William Street Capital Management.

UBS appoints restructuring head
Matthew French, currently restructuring partner at law firm Lovells, is to join UBS as EMEA head of restructuring. His start date is yet to be confirmed but is expected to be around mid-year.

Shaffran exits Rabo
Alan Shaffran has left his role as head of structuring at Rabobank, which he had joined in 2006 from Citi where he was European head of credit derivatives since relocating from New York in 2001. Shaffran had worked for Citi and its predecessor Salomon Brothers since the late 1980s.

His future destination is not yet known. Donal Galvin, Rabobank's head of global financial markets Asia, will take on Shaffran's role as global head of equity and fund derivatives. A replacement for head of structuring will be announced in due course.

Schwartz joins Winter
Laura Schwartz, who left her role as senior md of ACA Capital Holdings and had been head of CDO asset management at the firm (see SCI issue 60), is understood to have joined the Winter Group.

SPQR partner leaves
Benoit Vincenzi is understood to have left SPQR Capital, where he was a partner. The firm was originally set up by three former members of Deutsche Bank's structured credit group (see SCI issues 35 and 56) – Bertrand des Pallieres, Malik Chaabouni and Moez Ben-Zid.

Vincenzi's future destination is not yet known.

CS completes mispricing review
Credit Suisse Group has completed its review following the CDO mispricing it revealed in February (see SCI issue 76). The bank recorded a total valuation reduction of CHF2.86bn (US$2.65bn), of which CHF1.18bn is related to the fourth quarter of 2007, and CHF1.68bn to the first quarter of 2008. Net income for Credit Suisse for the fourth quarter and full-year 2007 has been revised by CHF789m to CHF540m and CHF7.76bn respectively.

Following its revaluation review, Credit Suisse has determined that the pricing errors were, in part, the result of intentional misconduct by a small number of traders. These employees have been terminated or suspended and are in the process of being disciplined under local employment law. The review also found that the controls put in place to prevent or detect this activity were not effective.

Credit Suisse says its executive board will oversee a series of remedial actions, including reassignment of the trading responsibility for the CDO trading business and enhancement of related control processes, and improvement of the effectiveness of supervisory reviews and formalisation of escalation procedures.

AFP calls for enhanced rating agency oversight
In a letter to the US SEC, the Association for Financial Professionals (AFP) strongly urged SEC chairman Christopher Cox to fully exercise the authority that the SEC was granted in the Credit Rating Agency Reform Act of 2006 and press for the reform of credit rating agencies.

"Today, the value of credit ratings has been significantly diminished and the SEC has the authority to confront the situation head on and hold the credit ratings agencies accountable. It is time for the SEC to act," the letter says.

The membership of AFP includes 16,000 financial executives employed by over 5,000 corporations and other organisations. Members are responsible for issuing short- and long-term debt, and also invest corporate cash and retirement assets for their organisations.

Fitch versus MBIA round 2
Following on from correspondence between Fitch Ratings and MBIA (see SCI issue 79), Fitch has decided to maintain its Insurer Financial Strength (IFS) and debt ratings on MBIA and its subsidiaries for the foreseeable future. Fitch expects to maintain the MBIA ratings as long as it believes that it can maintain a clear, well-supported credit view without access to certain non-public details concerning MBIA's insured portfolio, to which Fitch will no longer have access.

Stephen Joynt, president and ceo of Fitch Ratings, says: "We are disappointed that MBIA has requested that we withdraw our IFS ratings and that they have decided to stop providing us important non-public information about their portfolio. While we respect MBIA's decision not to provide us that information, we trust that they will respect our decision to continue to maintain a rating on MBIA – a company about which many investors are so clearly interested. The approach we are taking with MBIA is consistent with the approach we have taken in other similar situations."

Fitch expects that its review of MBIA will be completed in the course of the next few weeks. Currently, the agency believes that, should the ratings of MBIA be changed as a result of this review, MBIA's IFS ratings will be no lower than the double-A category, which represents very strong capacity to meet policyholder and contract obligations on a timely basis.

In general, Fitch believes that it can rate companies based upon publicly available information. The unique nature of the financial guaranty sector makes maintaining the MBIA IFS and debt ratings more challenging without access to the non-public details on its insured portfolio.

To Fitch's knowledge, the non-public details on MBIA's insured portfolio currently made available to Fitch are not available from any other source. Accordingly, while Fitch has decided to maintain MBIA's ratings at this time, it may become necessary to withdraw those ratings in the future.

MBIA's response was the following statement: "We have previously stated our position on our relationship with Fitch and we completely respect their right to maintain MBIA's ratings for a period of time. While we acknowledge there are rating agencies that rely exclusively on public information to maintain ratings on MBIA, we agree with Fitch that the unique nature of the financial guaranty sector makes maintaining the MBIA IFS and debt ratings more challenging without access to non-public information. Due to market developments, we believe that the non-public information currently in Fitch's possession soon will become out of date, and public information alone will be insufficient to maintain the ratings."

CME buys CMA
CME Group has acquired Credit Market Analysis (CMA), the credit derivatives market data provider. CME says this offers the potential to leverage CME Group's clearing and trade execution capabilities, enhancing CMA's products to create greater value and efficiencies for its customers.

As a wholly owned subsidiary of CME Group, CMA will remain headquartered in London and will continue to operate under the CMA name. Laurent Paulhac will continue in his role as the company's ceo and the existing CMA management team is expected to remain in place.

MP

EOD haircuts analysed

A round up of this week's structured credit news

EOD haircuts analysed
Events of default (EODs) among US ABS CDOs have been increasing in recent months. These events, which can have significant consequences for investors, have been rare historically due to the limited rating transitions on the RMBS that underlie many CDOs.

However, troubles in the mortgage markets have led to widespread downgrades of US RMBS and, in turn, negative rating actions on certain CDOs that contain these securities. Consequently, 134 ABS CDOs experienced EODs over the past few months, mostly as a result of overcollateralisation (OC) test breaches.

In light of the rise in EODs, S&P is taking a closer look at the effect of rating-based adjustments, or haircuts, to the value of the underlying securities of CDOs when calculating certain EOD test ratios. This feature is now causing some CDOs to fail their OC ratio tests and trigger EODs more often than in the past.

Of the 134 EOD notices so far received by S&P, 109 were caused by failure of OC ratio tests calculated using rating-based haircuts, while 24 were caused by failure of OC ratio tests calculated without rating-based haircuts. One additional EOD was caused by a missed interest payment on a non-payment-in-kind tranche.

A breach of an EOD OC ratio test often gives the controlling noteholders in a CDO the right to accelerate the repayment of the notes. The acceleration of a transaction typically halts interest and principal payments to all but the most senior notes until the senior classes are paid in full. Essentially, the transaction returns to a true sequential waterfall payment structure, in which the notes in the transaction are repaid interest and principal in order of priority.

A breach of an EOD OC ratio test also typically gives the controlling class of noteholders additional rights, including the sole right to liquidate all of the CDO's collateral. The controlling classes of 26 CDO transactions have so far voted to liquidate their collateral, and three liquidations had been completed. Due to the current market conditions, the liquidation proceeds in certain cases have been insufficient to provide full payment even to the senior noteholders.

While the outcome for the other CDOs that have breached their EOD OC ratio tests is still uncertain, S&P believes that the most likely scenario for most is acceleration of the note repayment followed by liquidation of the collateral, if so directed by the controlling class. The noteholders in transactions that are liquidated face market value risk that significantly increases the likelihood that they will suffer losses.

Several of the transactions so far affected by EODs would not yet have failed their EOD OC ratio tests if their OC ratios had been calculated without using rating-based haircuts. For transactions in which the documentation provides for a sequential waterfall after an EOD, S&P may have retained higher ratings on some of the subordinate notes if the transactions had not breached their EOD OC ratio tests and consequently accelerated payment and/or liquidated assets.

Many of the downgrades reflect the acceleration that occurred after the EOD OC ratio test was breached, which caused the payment priority to shift in a manner that was detrimental to the subordinate notes. A liquidation scenario, of course, further impairs the ratings on the notes because the current market value of the underlying securities would likely be insufficient to repay the noteholders' principal and interest when the assets are sold.

Current market circumstances make it difficult to evaluate the impact of rating-based haircuts on the ratings S&P assigns to CDOs. However, the agency maintains its view that rating-based haircuts can offer significant value to CDO transactions because they can provide additional credit protection for the senior notes. Accordingly, S&P has published a request for comment regarding its use of rating-based adjustments in CDO EOD OC ratio tests.

Going forward, S&P will rate CDO transactions that have EOD OC ratio tests only if the tests include the following features:

• If the test is calculated using rating-based haircuts, acceleration cannot be initiated without at least a simple majority vote of each class of notes rated investment-grade at the CDO's inception, voting separately. For such a test, the only other haircuts S&P will recognise under its criteria are those for defaulted securities and equity securities.
• If the EOD OC ratio test is calculated without rating-based haircuts, the test threshold should be low enough that S&P's quantitative analysis shows that all the liabilities affected by either acceleration or liquidation of the transaction would fail the test at their targeted rating levels. In this case, the controlling class may decide to accelerate note repayment and/or liquidate the assets. For such a test, the only haircuts the agency will recognise under its criteria are those for defaulted securities and equity securities. For some CDOs, the definition of "defaulted securities" includes assets that have been paying in kind (PIKing) for a certain period of time. These assets can be included in the definition of defaulted securities and therefore factored into the EOD OC ratio, so long as the PIKable securities are modelled accordingly at the time the transaction closes.

S&P will review any additional haircuts to EOD OC ratio tests that may be included in future transactions to determine whether the haircut is appropriate for the rating assigned. It will also review the definitions used in the transaction documents to classify defaulted obligations to determine whether they are consistent with those used in our default studies.

S&P is also considering the potential effect of OC ratio test EODs on CDOs made up of other CDOs, or those that contain CDOs along with other assets. Analysing CDOs that include underlying assets with their own OC ratio test EOD triggers is complex and transaction-specific, as the post-EOD cashflow performance of the underlying CDO collateral will affect cashflow in the resecuritisation as well. For future CDO-squared transactions and CDOs that contain other CDOs as collateral, the agency will not recognise under its criteria those CDO assets with EOD OC test triggers that include rating-based haircuts unless such CDO assets immediately become the controlling class in the event of an EOD OC test breach.

Request for comment on OC haircuts
S&P is requesting comments from market participants regarding the use of rating- and non-default-based adjustments, or haircuts, to calculate overcollateralisation (OC) ratio tests in CDO transactions. Rating-based haircuts for CDO transactions are used to value assets at less-than-par value, should certain rating concentrations within their collateral pools exceed preset amounts.

For example, a rating-based haircut may value assets at 70% of par when the concentration of triple-C rated collateral exceeds a given deal-specific percentage. Certain CDO transactions can also contain purchase-price haircuts, which are used when an asset is carried in the OC test at the initial purchase price if it had been purchased for less than a predetermined price.

This request for comment only pertains to OC trigger calculations, which are present in the priority of payments of CDO transactions. OC ratio tests have been used extensively as cashflow diversion mechanisms in CDO transactions.

They typically compare the amount of outstanding debt with the net principal amount of the collateral securities backing the debt. The net amount represents a downward adjustment to the principal balance of certain collateral debt securities. The adjustment is typically based on transaction-specific formulas that can be based on the ratings or other credit characteristics of the collateral debt securities.

OC ratio tests typically provide credit support to the rated notes by triggering credit support mechanisms as specified in the transaction documents. One support mechanism diverts cashflow when the credit quality within a CDO transaction deteriorates.

These well-established mechanisms aim to shift the risk burden between classes of notes with different seniorities so that senior notes receive a greater percentage of available cashflow when credit quality deteriorates. Senior noteholders generally prefer rating- and price-based haircuts over other types because they provide an additional cushion in case of downward rating transitions in a transaction's underlying asset pool.

While rating- or price-based haircuts in OC test calculations may give the senior-most tranches more protection in a less-predictable market, the same haircuts may cause rating volatility for the more junior classes, whose repayment of interest could be deferred when their priority of payment is subordinate to the repayment of principal of classes of notes senior to them. The deferral of interest for the junior classes and the faster repayment of the senior classes may increase the structure's weighted average cost of capital.

For classes subject to the deferral of interest, S&P rates to the ultimate repayment of interest and principal. When modelling the return of interest and principal, its cashflow rating methodology considers the OC test levels in conjunction with asset defaults, but does not currently address rating transitions or market price movements, or their impact on OC test calculations.

In the past, based on the historical performance of the underlying assets and the cushion between the OC test levels and their respective triggers, S&P generally believed that it would have been unlikely for interest to be deferred due to haircuts in the OC tests and that the possibility of such an event was commensurate with the ratings on the liabilities. However, due to recent market volatility, the lack of predictability of interest deferral and ultimate repayment of principal on the notes subordinate to the OC test cure could be amplified.

In light of increased market volatility, S&P is evaluating whether rating changes and market price adjustments may increase the likelihood of OC tests failures. Similarly, it is evaluating whether, as a consequence of the market volatility, subordinated securities may be subject to a more extensive deferral and accrual of interest, which may negatively affect the ratings on these tranches.

Extensive deferral of interest may also reduce the likelihood that the subordinated notes will receive their ultimate return of interest and principal. These issues have surfaced because of recent market volatility, the amount of downgrades that have affected certain ABS, and the possibility that other asset classes could be similarly affected if there is a serious economic cycle.

Given the current CDO structures, increased ratings migration in underlying collateral may negatively affect the ratings on subordinated securities in a way that may be difficult to predict. Accordingly, S&P is considering changing its criteria to minimise the effect of this potential dynamic on its ratings on future transactions.

The two criteria changes being considered are:

• Only rating securities issued by CDO transactions that employ rating-based haircuts in the OC test so long as the cure for any failure of the OC test is at least subordinate to the return of: a) investment-grade rated noteholders' interest in the interest waterfall and b) investment-grade rated noteholders' interest and principal in the principal waterfall.
• Only rating securities issued by CDO transactions using OC ratio tests that stop payments to notes with investment-grade ratings (at inception) so long as the haircuts used in the OC ratio tests are limited to securities that S&P considers defaulted by its criteria.

Mezzanine investors accept Asscher exchange
HSBC has announced that 24 out of 25 (the equivalent of US$418m out of a total US$458m outstanding) of Asscher Finance mezzanine note investors accepted its voluntary exchange offer. Under the terms of the exchange offer, Asscher mezzanine note investors exchange their notes for notes issued by a new vehicle, Malachite Funding (see last week's issue).

Malachite is a term-funded vehicle with senior funding of about US$6bn provided by HSBC. Malachite also provides for the removal of all market-value triggers.

A voluntary exchange offer was also made to Asscher income note investors on 5 March and the results of that offer are expected to be announced this week.

In addition, HSBC says that Cullinan CP and MTN investors will be invited to indicate their interest in exchanging their senior debt for Mazarin senior debt. Mazarin was established as part of the previously announced reorganisation of Cullinan and provides for senior debt to be 100% backed by a HSBC liquidity facility.

Fitch's monoline methodology explained
In light of recent questions raised about its capital model and the ratings methodology used in the financial guaranty industry, Fitch has provided a Q&A-style commentary in an effort to provide the marketplace with greater clarity on its ratings and process.

Fitch introduced a new capital model (Matrix) in January 2007. The rating agency believed that other models used in the market overstated the risk of many US municipal exposures insured by the financial guarantors and understated the risk of many structured finance exposures.

This was evident when reviewing historical default studies that show the default rate on US municipal debt is a fraction of that on corporate debt, whereas the default rate on structured finance debt is equal to, if not moving higher than, that of corporate debt. Thus, a key goal in developing a new model was to address this perceived imbalance found in existing models.

The following are two examples of assumptions used by Fitch that are not present in the same way in other models and that result in lower overall capital guidelines for many US municipal transactions:

• Fitch's model recognises that default rates implied by each rating agency's municipal ratings scale are generally lower than those implied by its global corporate rating scale. Accordingly, its model converts municipal scale ratings to an approximate "corporate equivalency" basis before capital guidelines are developed. For certain very low-risk classes of municipal debt, such as the general obligation bonds of cities or towns, or certain investor-owned utilities, the default rates can be reduced by more than 50% compared to a similarly rated structured finance or corporate transaction. Even more importantly, our model recognizes the low "loss given default" rates that we believe would likely occur in certain low-risk municipal asset classes.
• While Fitch's capital guidelines for very low-risk municipal asset classes are appropriately low, its model also recognises higher risks in other types of municipal transactions. For example, loss given default assumptions used for corporate-like municipal hospital risks can approximate 60%, and for hospitals corporate-like default rate assumptions are used. The model is granular in distinguishing higher-risk from lower-risk municipal asset classes, and assigns more capital to municipal transactions where it would be warranted.

The modelling assumptions used by Fitch that are not present in other models and that result in higher overall capital guidelines for many structured finance transactions include:

• Fitch's model employs a so-called "regime shift" that assumes stressful, recessionary conditions that occur infrequently, but that the agency believes should be considered in assessing capital. The regime shift impacts both structured finance and municipal securities, but in practice its impact is most pronounced for large, lower-rated structured finance transactions – an example of which might include whole business securitisations rated in the triple-B category. Fitch believes concentrations in large lower-rated structured finance transactions can materially increase a guarantor's portfolio risk and volatility, and its model directly addresses this through prudent incremental capital assessments.
• The model recognises differences in loss given default expectations for senior versus non-senior tranches in the structured finance capital structure through a realistic "waterfall" approach to simulating recoveries. Fitch's approach can result in higher capital assessments for thinner, non-senior tranches that in some cases can reach a 100% loss given default. Some other models employ a single loss given default factor assumption that is used regardless of tranche thickness, which can potentially understate potential losses (and capital) on these transactions.
• In general, Fitch's model assumes slower amortisation schedules for many classes of structured finance assets compared to those assumed in other models. Slower amortisation assumptions result in higher simulated losses and ultimately capital assessments, since the exposure remains in place for a longer period in the simulation, increasing the risk of default. Fitch's assumptions are based on how securities would amortise in a stressful economic environment, which is the basis for an triple-A rated financial guarantor's capital, as opposed to the more optimistic "expected" case typically used by the guarantors themselves for pricing.

It has been suggested that the Matrix model produces results that can change greatly from period to period, and that this was especially true over the past six months. However, Fitch points out that it has not changed any core modelling assumptions since we introduced Matrix in January 2007.

That said, the agency's capital guidelines for the financial guarantors have increased significantly in the past six months, as have capital guidelines produced by S&P and Moody's. The reason for the increase in modelled capital relates to the unprecedented deterioration in the structured finance obligations exposed to sub-prime residential mortgage collateral within a number of financial guarantors' insured portfolios.

This deterioration has resulted in, at times, very sharp downgrades in the underlying ratings of SF CDOs and RMBS transactions, and it is these downgrades that are resulting in higher capital assessments. Simply put, the lower the underlying ratings input into the model, the higher the capital guidelines.

Since underlying ratings migration has been significant, and experienced unprecedented variability, so have financial guarantor capital guidelines. This should be expected in a period of such sharp and sudden deterioration in credit fundamentals, Fitch says.

The agency announced on 5 February a next phase of analysis of sub-prime exposures for financial guarantors. While the agency felt the sub-prime mortgage loss rate assumptions it had been using for its RMBS, SF CDO and, in turn, financial guarantor ratings evaluations were conservative, new data that it received in late January indicated a need to increase its loss assumptions materially.

Negotiations for commutations between various financial guarantors and banks that purchased protections on SF CDOs through CDS have thus far not been successful (see separate news story), since all concerned have such varied views on ultimate losses. Fitch views this variability and uncertainty as a key qualitative ratings consideration, and it is a core driver of its negative outlook or negative rating watch on a number of key players.

As a result, Fitch is carving out SF CDOs insured by the guarantors from Matrix and developing deal-by-deal expected losses to form the basis for its capital assessments on these problem transactions. In developing its 'expected loss' estimates, Fitch is focusing on the underlying credit supporting each SF CDO in order to estimate losses on each collateral pool.

Fitch's SF CDO loss estimates build off the loss expectations reflected in Fitch's recent RMBS and CDO rating actions. The loss estimates vary by asset type, vintage and original credit quality.

These projected losses on each pool are then applied against each specific transaction, giving credit to subordinate tranches within the SF CDO capital structure, expected default timing, control rights and other structural features. From there, the expected loss on the insured tranche of each SF CDO transaction is aggregated, and discounted to a present value.

Further rating action for EVVLF
Moody's has lowered the ratings on the capital notes and MTNs issued by Eaton Vance Variable Leverage Fund (EVVLF) from Ba3 to Caa3 and from Aaa to Aa3 respectively. The MTN rating remains on review for possible downgrade.

Despite recent deleveraging of the vehicle, the risk of an enforcement event remains high because of the deteriorated market value of its asset portfolio. An enforcement event may lead to the liquidation of the portfolio over a six-month period.

As a result, Moody's believes that the expected losses associated with the capital notes and MTNs are no longer consistent with their prior ratings. In addition, Moody's has put on review for downgrade the short-term Prime-1 ratings assigned to both the commercial paper programme and to medium term notes with a maturity of less than 365 days – despite there being no such short maturity medium term notes or commercial paper outstanding. This rating action was prompted by concerns about EVVLF's ability to continue making timely payments if the vehicle were to enter enforcement.

Moody's will continue to monitor the situation and evaluate any alternative restructuring proposals that may be presented by EVVLF's management.

Agreement reached for Apex and Sitka trusts
DBRS has been informed, and BMO has publicly announced, that an agreement has been reached with all four swap counterparties to the Apex and Sitka Trusts and certain investors for a restructuring of the conduits.

DBRS has confirmed that the credit quality of each of the underlying assets held by the Trusts meets the minimum requirement for a triple-A rating. This assessment speaks to the credit risk of the underlying assets from a probability of default perspective only and does not consider mark-to-market, collateral call or funding risk faced by the conduits.

FGIC on negative watch
S&P has revised its credit watch with developing implications on FGIC to credit watch with negative implications. FGIC's principal owner, The PMI Group, recently announced that it no longer views the monoline as a strategic investment and will not be contributing capital as part of any recapitalisation plan.

In addition, while FGIC's ability to write new business was already severely constrained, the company formally announced a suspension of new business writings in order to preserve capital. In S&P's opinion, these announcements negatively affect the monoline's ability to implement plans to raise additional capital and resume writing business in the future.

With respect to business position and the company's ability to write new business, in February 2008, FGIC applied to the New York State Insurance Department for a new license. FGIC has said that this new license could be used in connection with the establishment of an affiliated new company that would principally be involved in the insurance of municipal business.

According to FGIC's plan, structured finance business and possibly some other sectors would remain as insured obligations of FGIC. Management believes that the creation of this municipally focused company is central to its ability to raise capital to generate new business.

The revision to credit watch negative reflects S&P's belief that, in terms of both capitalisation and ability to write new business, prospects for the currently existing legal entity (FGIC) under this split-company plan are negative. Should the split-company plan not succeed, FGIC's ability to raise capital and resume writing new business would be less likely, in the agency's view. Under either scenario, prospects for a rating upgrade for FGIC are difficult to envision.

FAS 161 issued
The Financial Accounting Standards Board (FASB) has issued FASB Statement No. 161, Disclosures about Derivative Instruments and Hedging Activities. The new standard is intended to improve financial reporting about derivative instruments and hedging activities by requiring enhanced disclosures to enable investors to better understand their effects on an entity's financial position, financial performance and cash flows. It is effective for financial statements issued for fiscal years and interim periods beginning after 15 November 2008, with early application encouraged.

"Use and complexity of derivative instruments and hedging activities have increased significantly over the past several years. This has led to concerns among investors that the existing disclosure requirements in FASB Statement No. 133, Accounting for Derivative Instruments and Hedging Activities, do not provide enough information about how these instruments and activities affect the entity's financial position and performance," explains Kevin Stoklosa, FASB project manager. "By requiring additional information about how and why derivative instruments are being used, the new standard gives investors better information upon which to base their decisions."

The new standard also improves transparency about the location and amounts of derivative instruments in an entity's financial statements; how derivative instruments and related hedged items are accounted for under Statement 133; and how derivative instruments and related hedged items affect its financial position, financial performance, and cash flows.

FASB Statement No. 161 achieves these improvements by requiring disclosure of the fair values of derivative instruments and their gains and losses in a tabular format. It also provides more information about an entity's liquidity by requiring disclosure of derivative features that are credit risk-related. Finally, it requires cross-referencing within footnotes to enable financial statement users to locate important information about derivative instruments.

IASB issues financial instruments discussion paper
The International Accounting Standards Board (IASB) has published for public comment a discussion paper entitled 'Reducing Complexity in Reporting Financial Instruments'. The existing requirements for the reporting of financial instruments are widely regarded as difficult to understand, interpret and apply, and constituents have urged the IASB to develop standards that are principle-based and less complex. The document is the first stage in a project which aims to replace IAS 39 Financial Instruments: Recognition and Measurement.

The discussion paper analyses the main causes of complexity in reporting financial instruments and proposes possible intermediate approaches to address some of them. Those approaches seek to improve and simplify measurement and hedge accounting by amending or replacing the existing requirements.

Furthermore, the discussion paper sets out the arguments for and against a possible long-term approach that would use one measurement method for all types of financial instruments in the scope of a financial instruments standard.

The IASB is seeking views on both the possible long-term and intermediate approaches, and is interested in hearing about possible alternatives on how it should proceed in developing new standards for reporting financial instruments that are principle-based and less complex.

The publication of the discussion paper also fulfils the commitment set out in the Memorandum of Understanding between the IASB and the US Financial Accounting Standards Board (FASB) on a roadmap for convergence between IFRS and US GAAP. The discussion paper will be considered for publication by the FASB for comment by its constituents.

David Tweedie, IASB chairman, comments: "IAS39, which the IASB inherited from its predecessor body, is far too complex. We are determined to simplify and improve IAS 39 by creating a principle-based standard. Those who believe in reducing complexity in accounting standards have now the opportunity to shape the way ahead."

The IASB invites comments on the discussion paper by 19 September 2008.

Remittance deterioration slowing?
The March remittance data for the 80 sub-prime RMBS deals referenced by the ABX indices is in. The report shows that severe delinquency rates (loans that are 60 days delinquent or worse) rose across all four ABX series during March, reaching 34.1%, 32.5%, 28.1% and 23.1% for the 06-1, 06-2, 07-1 and 07-2 series respectively. Severe delinquency rates for several deals in the 06-1 series are now in the mid to high-40s.

However, the March data indicates that the pace of deterioration is finally slowing, according to analysts at Deutsche Bank. In particular, the increase in the severe delinquency rate declined from 2.32% to 1.60% for the 06-1 index, 2.63% to 2.21% for the 06-2, 1.98% to 1.40% for the 07-1 and 2.61% to 2.04% for the 07-2.

Voluntary prepayment speeds have collapsed. Those for the 06-2 series declined from 18.1% to 14.9% during March, in spite of the fact that most of the underlying loans are now approaching their rate resets. Prepayment speed for 06-1 slowed from 14.1% to 10.9%, while the speeds for 07-1 and 07-2 slowed from 8.3 to 7.1 and 7.6 to 7.2 respectively.

Cumulative loss rates continue to march northwards, however, with 06-2 now at 2.21% and 06-1 not far behind at 1.95%.

CS

The new and improved LevX

The different types of trades that can be implemented using the new LevX index are discussed by Mahesh Bhimalingam and Eugene Regis of Barclays Capital's high yield and leveraged finance strategy team

The introduction of the new LevX is a further evolution in the derivatisation of loan markets. Given that this is not just a simple roll on changes in constituents like the other iTraxx indices, it may seem in the current climate to be adding another layer of complexity or yet another instrument to short. However, Series 2 offers key changes from Series 1, such that it will arguably have a bigger impact than the original LevX indices in loan market trading flows.

The key changes in Series 2 are the advent of non-cancellability of individual contracts and the substitution of reference obligations. We expect them to have a significant impact on the European LCDS market by improving liquidity and allowing investors to express a view by taking a direct position in the most liquid LCDS names, with a much reduced risk of cancellation of the underlying loan due to the continuity feature built in.

This would reform the benchmark for a relatively new market and allow pure credit risk trades without the threat of a disappearance of P&L due to cancellation. As such, this could eventually allow for more sophisticated structured products such a tranches on the LevX. This contrasts with the initial concerns about cancellation risk and uncertainty of duration on contracts in Series 1.

Hedge funds have been able to take advantage of relative value and capital arbitrage trading by taking positions in different iTraxx CDS indices or by trading the index intrinsic against the market levels of the indices. Since hedge funds have shown an appetite for LCDS, the new index will also make it easier to implement these trades in the leveraged loan space by increasing transactability on valuations, given that duration can be much more accurately estimated. For example, hedge funds will now find it easier to implement relative value trades between the Crossover and the LevX indices.

Loan/credit portfolio managers will have greater ability to hedge their loan portfolio exposure without compromising their client relationships. Structured products (tranches and first-to-default products) are easier to create and hedge, given that the price and risk of these products can now be easily calculated and appropriate delta hedging can be employed on the new LevX index because cancellation risk is minimal.

This has already been seen in evidence on the US LCDX, where there are tranches trading as well as active hedging using the index. In addition, the new LevX could be actively used in the ramping up of cash CLO portfolios, as it is more efficient from a duration perspective.

That said, we believe that the current stasis of the CLO market, with very few deals being ramped up in this climate, means that LevX will be used more as an index to trade relative value in expressing a view on capital structure (senior-sub) and index-intrinsic type trades between the index and underlying LCDS contracts. In essence, the new indices will enable investors to trade this market more efficiently.

All these functions are implemented at a transaction cost which will be decreased substantially with the launch of the index. The increased transparency in the LCDS market from the new LevX will mean market participants are able to price their positions better across a much larger universe of names. The reduction in the uncertainty of cancellations should help cut transaction costs and reduce the bid/offer spread (market conditions dependent).

In the current market environment, we expect a choppy start for the new LevX. While it will be easier to structure more products from the index, the current environment and lack of a bid for structured credit will not transform the leveraged loan space overnight. However, the foundations for a more liquid, transparent and easily risk-manageable benchmark have been laid.

Index tranche synthetic CLOs
As the new index starts to trade on a nearly non-cancellable basis, the contract will start to have a near-bullet maturity. Thus, we expect to see further development in synthetic CLOs structured with index tranches. They are implemented in the same way as index tranches are implemented on the iTraxx indices.

The cash CLO market was extremely popular with investors, but as was seen with cash CDOs, the synthetic CLO market could develop into a sizeable alternative to the cash CLO market. It should be noted that the CLO market in general is having difficulties in terms of funding and placing the triple-A tranche, but once the market recovers we expect tranches on LevX to be offered to investors.

Index tranches
Figure 1 shows the construction of unfunded synthetic index CLOs along with typical attachment points on the tranches. This market is already in vogue in the US, with tranches on the LCDX having been actively traded before the current illiquidity reduced trading flows in structured products in general.

Figure 1

The new LevX will be the underlying portfolio of names for the synthetic CLO. The change in documentation would make it easier for dealers to delta hedge these tranches using the index, given improved and greater certainty of duration of the index as well as the tranche.

In addition, the existing CDO pricing technology can be applied without the need to worry about cancellation risk, as was the case with the old LevX. With the new documentation, the new LevX would be the primary hedging vehicle for synthetic CLO traders. Greater liquidity in the index, as well as single name CDS, is required for structured products to take off.

Trade examples using LevX

Selling/buying the index and unwinding
There is no major change in the mechanics of trading Series 2 of the LevX with the introduction of nearly non-cancellable contracts. Suppose the client wants to take a long risk position. They would buy €10m iTraxx LevX Senior Series 2 of five-year maturity in unfunded form, which is analogous to selling protection/lifting the offer.

If the deal spread for iTraxx LevX Senior is 525bp and the price quoted currently is 100.1/100.2, this implies that at that point – given that the offer price is above par – the current implied spread for the series would be lower than the deal spread of 525bp. The client only pays the 20 cent premium (in addition to accrued interest) when buying the LevX at 100.2, as the index trades in unfunded form.

Let us also assume now that after two days the market is 100.4-100.5 and the client wants to unwind. The client will sell the index (buy protection) to a market maker at a higher price (lower spread) and therefore lock in profits. The profit will be the difference in prices, which is 20 cents along with accrued interest over the holding period.

The market maker therefore buys the index (sells protection) from the client. The current spread must be less than 525bp, even though the client will be paying a deal spread of 525bp.

In order to compensate for that, the market maker pays the premium of 40 cents (as the index is trading above 100; i.e., premium * notional) and accrual for two days. If the client does not unwind, the market maker pays deal spread (525bp) quarterly to the client on a notional amount of €10m, unless there are defaults or refinancings (see Figure 2).

Figure 2

Note that trading works in a very similar fashion to a bond, since quotation and trading are based on price rather than spread (i.e., no recovery rate assumptions and curves are necessary for the spread to price conversion). In essence, the marking to market of LevX is like a bond but without the initial par payment.

Index versus intrinsic arbitrage
Once the index comes into play it is highly likely that the index would trade away from its 'fair-value' spread implied by the single name contracts embedded in it, depending on supply (loan portfolio hedgers) and demand (CSO players) for the index risk. If the market value is significantly different from the fair value, even after accounting for bid-ask, then the difference could be arbitraged out by buying LevX and buying LCDS protection on the single names or vice versa.

Let's assume that the duration of each name is between two and 2.5 – by using the fair spread calculation formula, we arrive at a fair spread of 500bp for the index. If the market spread arrived at by backing out from the price using option pricing model and flat curve and recovery rate assumptions is 510bp and there were no transaction costs related to bid/ask for each name, then the investor could easily buy the index and buy protection on each of the single names to make an arbitrage profit (see Figure 3).

Figure 3

Because of the bid/ask transaction costs, this difference (20 in this example) has to be significantly large, typically in the 20+ range for index-intrinsic arbitrage to be profitable. The trade can be held until maturity (pure arbitrage) or unwound when the index intrinsic skew tightens (mark to market).

With the use of nearly non-cancellable LCDS in the LevX, it is now easier to calculate the duration on such instruments and make these trades accordingly. We would expect a more liquid market in this space after the roll and once conditions normalise.

Index versus single-name views
The index can be used as a tool for putting on overweight and underweight trades on particular names. For example, if an investor is comfortable/constructive on market conditions but has a negative credit view on a particular name, they can buy LCDS protection on that name but hedge that market risk by buying LevX of the appropriate subordination (on a beta-adjusted duration-neutral basis) or vice versa if he has a positive view (see Figure 4). The issue of cancellability on the index has stymied such trades before, but we expect such trades to pick up now.

Figure 4

Suppose the investor has a negative view on Company XYZ and believes that spreads will widen in the future. The investor will buy LCDS protection on Company XYZ at say 625bp.

The investor needs to hedge the market risk on this trade. So they will then buy LevX at, say, 100.75 and the amount of index that the investor will sell will be a beta- and duration-based ratio with Company XYZ. This means that if Company XYZ spreads widen relative to the market, they will make more profit on the single name Company XYZ LCDS protection than the loss on the LevX index.

Assuming a beta of 1, given that the durations are nearly the same, the hedge ratio would be 1. In a market tightening environment, if Company XYZ widens to 650bp and the index has moved to 100.5, the investor makes a profit of about 0.5 (assuming duration of 2 on the LCDS) and a loss of 0.25 on the index leg – resulting in a net profit of 0.25.

Senior versus sub-loan plays
Investors who believe that sub valuations (second and third lien) are too wide relative to senior (first lien) valuations in the market can make use of these indices to take advantage of putting the senior versus sub trade (buy sub LevX and sell senior LevX on a duration-neutral basis) on many names at once with minimal bid-ask spread (see Figure 5). Investors who feel the differentials are too tight can put the trade the other way round. That said, if they believe that the differential is odd only in a particular name, it is more effective to put the trade on using the single name LCDS than the index.

Figure 5

If an investor believes that sub levels are trading too wide relative to the senior level in the whole market, they can then buy sub LevX at, say, 100.1 (implying a spread of, say, 800bp) and sell senior LevX at 100.2 (spread of, say, 480bp) on a duration-neutral basis (hedge ratio 1.25 sub to 1 senior). Assuming the investor is correct and sub spreads have tightened relative to senior in an overall tightening market environment, this could make sub LevX trade at 100.4 (spread say of 750bp) and senior 100.3 (spread say of 460bp). The investor unwinds both legs making a profit on the sub leg (100.4-100.1 = 0.3) and a loss on the senior leg (100.3-100.4 = -0.1), resulting in a net profit of 0.275 (0.3*1.25-0.1 = 0.275).

Secured versus unsecured plays – duration-neutral basis
Investors should find it easier to play on secured versus unsecured market differentials now that the LCDS contracts in the LevX are nearly non-cancellable, giving greater spread accuracy when implying a spread on the LevX. For example, if investors believe that unsecured spreads are trading too wide to secure at the whole market level, they can sell protection on the iTraxx Crossover and sell LevX on a duration-neutral basis (see Figure 6). This trade has the risk that the constituents of the Crossover and the LevX are not exactly the same and hence any idiosyncratic events in names that are not common can affect the trade performance.

Figure 6

If an investor believes that unsecured levels are trading too wide relative to senior secured levels in the whole market, hey can then sell Crossover protection at, say, 600bp and sell senior LevX at 100.2 (say a spread of 480bp) on a duration-neutral basis (hedge ratio of 1.6 LevX:1 Crossover). Assuming that the investor is correct and unsecured spreads have tightened relative to senior in an overall tightening market environment, this could make unsecured Crossover spreads trade at 550bp and senior LevX at price 100.4 (say spread of 460bp). The investor unwinds both legs making a profit on the Crossover leg of about 1.6 (50bp times duration of 3.2) and a loss on the senior LevX leg (100.4-100.2 = -0.2), resulting in a net profit of 1.28 (1.6-0.2 *1.6 = 1.28).

Secured versus unsecured plays – default-neutral basis
For investors who believe that senior risk is more attractive than subordinated risk, another way of trading this differential is to imply a spread on the LevX and adjust it for the recovery rate (see Figure 7). An investor would buy LevX and hedge themselves by buying protection on the Crossover index on a recovery-adjusted ratio of 2:1 (60% loss on the Crossover index versus a 30% loss on LevX). As with the trade above, this trade has the risk that the constituents of the Crossover and the LevX are not exactly the same and hence any idiosyncratic events in names that are not common can affect the trade performance.

Figure 7

An investor feels there is value in the loan market after a big sell-off. He buys LevX at 100.3, implying a spread of 465bp and hedges himself by buying Crossover protection at 600bp. This is done at a hedge ratio of 2:1 after adjusting for the recoveries, implying a spread differential of 330bp.

Assume the investor is again correct and LevX outperforms Crossover. The investor then unwinds his LevX trade at 103 and the Crossover tightens to 500bp.

The implied spread on the LevX is now 350bp leading to a recovery-adjusted differential of 200bp, a tightening of 130bp. In cash terms, the loss on the Crossover position is 3.2 (100bp times duration of 3.2) but the recovery adjusted gain on LevX is 5.4 ((103-100.3)*2), leading to a gain of 2.2 (5.4-3.2).

© 2008 Barclays Capital. This Research Note is an extract from an article first published by Barclays Capital's high yield and leveraged finance strategy team on 18 March 2008.

Trading ideas: home unimprovers

Tim Backshall, chief credit derivatives strategist at Credit Derivatives Research, looks at a pairs trade on Lowe's Cos Inc. and Home Depot Inc.

With spreads galloping tighter across much of investment grade land (more technically than fundamentally driven) and the herald call from Wall Street that the bottom is in, we are tempted (again) to jump back into some retailer relative value. The building products-related end of the retailing sector has seen significant volatility but has remained less impacted than many of the more mainstream retailers so far.

The rally of the last few days, and dramatic rise in the RTH ETF seems a little over-optimistic to us and we think a relative value position between HD and LOW has room to outperform if a consumer-led recession comes to pass (as many still suspect it will). We have recently taken advantage of our recessionary view through pairs trades across cyclical and non-cyclical sectors, but this pair focuses on the retailer space specifically and picks two recently weak credits that offer affordable carry and idiosyncratic divergence potential if the consumer continues to pull back.

Just to be clear, we see at least three major signs pointing to a consumer-led recession in the US. First, year-on-year growth of consumer spending is in a downward trend.

The American homeowner used easy credit from home equity loans, car loans and credit cards to fight off a recession in early 2002. YOY Growth of consumer spending rebounded in mid-2001 and since late-2003 has been on a steady decline.

Second, over the next few years, the American consumer will be hit with over US$800bn worth of ARM resets. If homeowners do not default, then they must pay additional money to finance their homes which must have a substantial decrease on overall consumer spending in the coming years.

Last, due to the ongoing credit crunch, tighter lending standards will become the norm, causing refinancing and new loans to be more difficult to acquire than in the recent past. Unfortunately, compared to the 2001-2002 recessions, the American consumer is highly leveraged and does not have the capacity to take on further debt to increase their spending. Continued stress in the financial system (which it seems many believe is over) will only exacerbate this, as slowing GDP and home price drops have a much more direct and significant pro-cyclical impact on aggregate bank credit than interest rates.

We believe the probability of a consumer-led recession has increased as the consumer has become severely impacted. The retailers are a perfect place to play this perspective and the building products-related end appears an even more perfect storm waiting to happen.

LOW and HD have recently reported but we await their 10-Ks for further surprises that we suspect will cause HD to further underperform LOW. The recent rallies in this sector have helped reduce carry and reopen the opportunity to pick value here, as well as lower the relative multiple and reduce the beta from its highs – all of which provide fertile ground for execution over and above our fundamentally weak view.

Building our case
LOW and HD reported disturbingly similar (i.e., similarly weak) results for their fiscal fourth quarters, and introduced equally grim 2008 outlooks. Facing the same business risk, the nation's two biggest home improvement retailers are both being hit by a slumping housing market, tighter credit conditions (to borrow corporately as well as for the consumer) and the overall weak economic environment.

Although we have yet to see the complete gory details of the 10-Ks, both companies saw earnings plummet and same-store sales fall considerably. However, according to Carol Levenson, Gimme Credit's retail sector expert, LOW claims to have won the market share race, but she does not see the evidence in the results.

Carol's fundamental outlook for LOW is stable and for HD is deteriorating, with both companies saying they are planning conservatively for whatever fiscal 2008 holds, and are cutting back on new store openings and supposedly stock buybacks. Both see same-store-sales in the red in 2008 (with HD slightly outperforming LOW) and both see EPS plunging, with HD's down significantly more than LOW.

Default risks implied by the equity market remain quite similar, with the most recent rally in stocks pushing HD to modestly oversold (in credit) versus LOW (although we suspect this is temporary, as the broad technical rally in spreads – as financials appear safe again – is pulling the higher beta credits in more aggressively than lower-beta, as one would expect). LOW has a modestly better debt/EBITDA rank in our models, but aside from that they are very similar in ranking terms.

Both companies view 2008 as forcing further operating margin contraction and, as Carol mentions, the only ray of hope for credit investors is that the firms give a nod to any commitment to preserve cash. Stock buybacks in 2007 were funded via borrowing in both companies, but LOW has the better credit profile and its slightly more diverse mix lowers business risk.

Spread differentials are wide and reflective of this differentiation, but we feel the recent pullback gives us an opportunity to benefit once the technicals have abated and the differential pushes back out to its wides again. With LOW emphatically stringer than HD (especially ex-HD supply), any resuming of HD's (postponed for now) recap will cause a much larger dislocation.

Building an RV
LOW has a slightly higher MFCI Rank of 5.9/10 versus HD's 5.7/10 as most of our factors (ex debt/EBITDA and FCF/Debt) are very similar. This is also the case in the market factors, such as implied volatility and default risk.

Exhibit 1 shows that until late-2006, these two credits traded within 5-10bp of one another. HD started to become in play for LBO investors and eventually went the levered recap route itself, pushing the differential into the 50bp range.

Exhibit 1

The sell-off since January 2008 has driven the higher beta HD much wider comparatively and we have seen the differential as wide as 87bp in late February. The last few days have seen that differential pull back rapidly to almost 50bp (which seems to offer a reasonable floor at around 30-35bp and upside of at least 87bp on the next wider swing).

As we would expect, Exhibit 2 shows the rolling three-month beta holds relatively steady through much of mid-2003 to early-2006 before LOW started to under-perform quite dramatically (albeit almost invisibly on Exhibit 1). Once HD became in play for an LBO (rumours) or inevitably an internal recapitalisation, the beta fell back as HD under-performed.

Exhibit 2

The beta has gently risen from almost zero as the market has become more systemically driven that idiosyncratic and the most recent rally has pulled that back in to almost 1 as HD has out-performed LOW on the way tighter. We will take advantage of this falling beta to position our pair at a slight disadvantage to LOW, as we feel that systemic risk concerns will rise again and this will exaggerate the push into HD (as a weaker credit).

With spreads having shifted so dramatically over the last few years, from less than 10bp to over 200bp, Exhibit 2 also shows the relative multiple of the two credits to give us some more relativistic perspective of the most recent moves. Until late-2006, LOW traded wider than HD, but that differential leaped as event risk concerns priced into HD.

The two credits spike to 2x multiple as HD below out before rumours abated and HD pulled back, but the sector remained wider. In the middle of last year, HD pushed to almost 2.5x the spread of LOW as both recap and systemic risk issues hit.

Since then the multiple has drifted lower. We do not see much risk of the multiple dropping below 1.25x – post-event risk concerns don't easily get forgotten – and perhaps if we see an emergence of the HD recap (if times improve) then we could once again visit 2.5x. Our view again is that there is more divergence than convergence opportunity here.

So from a fundamental perspective, our factors are similarly ranked across the pair but moreover the analyst perspective is that of HD deteriorating worse/faster than LOW (especially with the postponed for now recap hanging out there). These two credits have seen some very large swings over one-week periods and, as we have seen in Exhibits 1 and 2, we are well off of any extremes (and in our view likely to go back there).

Exhibit 3 illustrates just how volatile the pair has been. The scales are adjusted but notice the far larger moves in HD (y-axis) than LOW (x-axis).

Exhibit 3

The complete history shows an average beta of around 0.45x with a low r-squared, but if we look only at the most recent history (since mid-2007), then the r-squared is much more significant and the beta is around 0.5x (or HD has been twice as volatile as LOW). So our longer-term rolling beta is around 1x and the shorter-term beta is around 2x. Given this and the similarity of the DV01s, we are entering the trade at 1.5x LOW versus 1x HD in an effort to calm the P&L swings and potentially maximise profitability in our expected (idiosyncratic widening) scenario.

The shift in beta brings our position almost carry-neutral and set against the idiosyncratic fundamental view of HD under-performing LOW, a continued macro-economic slow-down, a housing slump which will directly impact these businesses, and a continued reduction in bank aggregate credit leaving consumers and corporates with much stricter lending standards and higher funding costs, we feel this pair warrant attention. Technically, the tightening of the last few days helps us dramatically, as we once again fade the rally and choose a pair to benefit from the systemic sell-off we expect in a more conservative relative-value manner.

Risk analysis
This pairs trade carries a direct risk of non-divergence (convergence should we see the economy shift back into high gear). In other words, there is the possibility that the names will not tighten and widen as expected. However, based on historical performance of the technical indicators, modest carry cushion and roll-down and the conviction of the fundamental analysts, we believe these risks are well mitigated.

Liquidity
Both HD and LOW offer good liquidity in the CDS market. HD is a member of the CDX IG9 and 10. They both consistently rank in the top 200 issuers by quoted volume on a daily basis.

They also have (relatively) tight bid-offer spreads: regularly less than 10bp and sometimes down to 4bp. We see no concerns with execution of this trade.

Fundamentals
This trade is significantly impacted by the fundamentals. For more details on the fundamental outlook for each of HD and LOW, please refer to Gimme Credit.

Gimme Credit's retailer industry expert, Carol Levenson, holds a negative view on HD, citing the following:

Home Depot Inc.: [Deteriorating: -1] Despite weakening business conditions (which have only become worse subsequently), HD decided to take the credit quality plunge with a 'recapitalisation' involving US$22.5bn in stock buybacks using the proceeds from the sale of HD Supply and borrowed funds. Although this leveraging plan is "on pause" for the moment, the new leverage target is an aggressive one and management will have a lot of repair work to do both with the balance sheet and with its bondholder relations.

Carol downgraded LOW from improving to stable last September but remains somewhat positive on LOW's relative outlook as an outperformer in the retailer sector, citing the following:

Lowe's Cos Inc.: [Stable: 0] It may not be the biggest but it remains the strongest in its niche by many measures, and we continue to be impressed by its ability to fund internally its ambitious capital plans – although recently more aggressive share repurchases have been funded with a modest increase in borrowing. With the outlook for home improvement spending dimming, we were hoping for a more conservative approach to cash returns to shareholders, and we think management has come around to this way of thinking with its plans for 2008.

These fundamental views reinforce our model-based views of LOW outperforming HD as the consumer cuts back.

Summary and trade recommendation
The rally of the last few days has driven retailers back to almost two-month tights, but seems far more technically-driven (roll and covering) than fundamentally-prone. The dismal same-store sales numbers from the major retailers remains confirmation of our view of an increasingly likely consumer-led recession biting into the credit profile of many 'over-levered' US corporates.

The home improvers are particularly at risk with the housing slump continuing (despite the blip in home sales) and with their most recent results showing huge drops in earnings and reduced outlooks before we've even started this recession. Fundamentally, LOW (stable) remains the better credit, and the less diverse business of HD (deteriorating) combined with the recap risk (postponed for now) remains somewhat of a limiter to downside in this pair.

Margin contraction continues as the results of both credits fall somewhat similarly and the spread differential and relative multiples have pulled off recent highs. Our models shade it to LOW as an outperformer here and, combined with our detailed beta-analysis which shifts us to carry-neutral, we feel the short-term rally will fade and the most risk-reward appropriate way to play the systemic sell-off in retailers is the idiosyncratic divergence of LOW and HD.

Sell US$15m notional Lowe's Cos Inc. 5 Year CDS protection at 110bp.

Buy US$10m notional Home Depot Inc. 5 Year CDS protection at 162bp to gain 3bp of positive 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).