Hunting for paper

The landscape for CDOs has changed dramatically over the past year, when the securities fell out of favor in the midst of rampant credit downgrades and defaults. Saskia Scholtes looks at the turnaround

In late 2003, investment bank UBS surprised the fixed-income markets by abandoning plans to issue a collateralized debt obligation backed by other CDOs, and liquidating the portfolio of underlying assets which the bank had accumulated for the deal.

Collateralized debt obligations are tranched securities issued by a special-purpose vehicle against a diversified pool of cash or synthetic credit exposures, such as asset-backed securities, high-yield bonds, leveraged loans, or even other CDOs. In the US, the repackaging of ABSs, including multi-sector ABS and CDO collateral, accounted for almost a third of issuance volume in 2003, and was followed closely by high yield or leveraged loan collateral. In Europe, hybrid or partially funded synthetics backed by investment-grade loans were the preferred issuance.

UBS had initially planned to market a $1 billion collateralized debt obligation backed by super senior CDO tranches, ABSs and loans. Among the CDO pieces purchased for the structure were Pimco’s Pacific Coast deal and American Capital Access’s ABS 2002. UBS’s so-called Cherry Valley CDO was to have been a cash deal with a credit default swap referencing the underlying portfolio.

At a time when investor demand for such products is high, liquidating the Cherry Valley portfolio looks like an unusual move for UBS – you would be forgiven for thinking that something was awry. But in reality, tightening spreads in the markets for CDOs and CDO collateral reduced the arbitrage opportunities of the Cherry Valley structure. And since UBS had already accrued a profit on the underlying collateral, by liquidating the portfolio the bank was simply realizing its gains.

The decision to pull the deal for these reasons marks a complete turnaround in the market for CDOs. In 2002, these products were the pariah of the structured finance universe: widespread downward ratings migration and some substantial defaults produced a raft of multiple-notch downgrades for CDO tranches.

Investors who had bought CDOs on the merits of diversification suddenly found that every portfolio had exposure to say Enron or WorldCom. What’s more, they found that the structural characteristics of the deals were insufficient to protect them from the loss. Investor confidence in collateralized debt obligations seemed all but lost.

In 2003, however, that confidence returned in force. In today’s market, diversification is no longer regarded as a poisoned chalice: improvements in corporate credit quality have reduced the likelihood that several names in the same pool will default, and the strong bid for credit has dramatically reduced spread volatility on CDO collateral. At the same time, increased diversification is now possible in synthetic CDO deals as the number of reference entities in the credit default swap market continues to grow apace.

Second, issuers and arrangers of CDOs have recognized the need for improved structural features: investors are now better protected from the effects of collateral deterioration, and the interests of the asset manager are now more closely aligned with those of the investors in each tier of the capital structure.

Third, the CDO market has become increasingly transparent. The Bond Market Association has made efforts to demystify the complex structures of CDO deals with its CDO Transaction Library. And the rating agencies have developed methodologies to allow investors to improve their analysis of a deal’s underlying collateral. One recent innovation in this field is Standard & Poor’s ‘drill-down’ methodology to assess the risk of loss in synthetic CDOs that reference other CDO deals.

However, one of the most important factors in bringing investors back to the CDO market has been the dramatic tightening in corporate bond spreads throughout the course of 2003. The resultant lack of arbitrage opportunities available on single-name credits has forced investors to look for yield in alternative products and strategies, and the CDO market has benefited from the trend.

The long view

According to market participants, demand for CDOs in both the primary and secondary markets is now at unprecedented levels. The most active players – banks, reinsurers, and long-term money – have become increasingly committed. In the primary market, long-term accounts such as pension and life funds have continued to use CDOs to access additional yield with a similar or lower level of risk than in the straight corporate bond market.

But increasingly, says Dominic Powell, global head of structured products and loans at UK fund manager Henderson Global Investors, these accounts are using synthetic CDOs for liability-matching purposes. By referencing the underlying portfolio with credit default swaps, single tranches of CDOs can be structured to match the fund’s maturity requirements and risk tolerance levels.

But there are also new accounts joining the CDO market. For one, the creation of short-term CDO debt with maturities of one year or less has brought some money-market accounts to the market. And according to David Wilmot, board director at fund manager Duke Street Capital Debt Management, there is also a nascent market for CDOs among retail investors. “A lot of high net worth retail investors are looking closely at the potential benefits offered by alternative asset classes,” he says. “For these investors, collateralized debt obligations are a very attractive risk-return proposition compared with other investments like public equity, and we expect to see retail interest in CDOs become more pronounced as the market moves into 2004.”

For Klaas Smits, portfolio manager at Dutch firm Robeco, this trend is quite logical. “While most high net worth individuals have been accustomed to returns of between 2% and 3%, they can now enter the CDO market and receive between 9% and 10% on their investment, depending on the structure and their risk appetite.”

At the same time, the burgeoning secondary market for CDO tranches is seeing growing participation from managers of CDO-squared deals, hedge funds and investment boutiques. The unprecedented credit correction of 2001–2 produced a spate of CDO downgrades, forcing investors prevented from holding junk to sell and creating a raft of arbitrage opportunities for yield-hungry funds.

Paul Hollowday, London-based asset manager at hedge fund Cognis Capital, says: “While hedge funds have been exploiting the arbitrage opportunities on all CDO tranches in the secondary market, from the senior non-PIKable all the way through to the mezzanine and subordinated tranches, many hedge funds have chosen to go long the most junior CDO tranches and bought CDS protection on a basket of the assets in the portfolio.”

As a result of this increased interest, the most active dealers in the secondary market are now reporting weekly volumes of between $150 million and $300 million.

In 2004, analysts and structurers expect improving credit quality and tight corporate bond spreads to continue to translate into incrementally stronger demand for CDO paper in both the primary and secondary markets. But at the same time, they warn that opportunities in the sector may be increasingly difficult to find.

In the primary market, the problem is that new CDO issuance is necessarily contingent on the spreads for CDO collateral. The strong bid for credit generally, and for CDO collateral specifically, together with an absence of collateral issuance and improving credit quality, has brought about significant tightening in these markets and created a major hurdle for CDO issuance. According to Moody’s, US CDO issuance for 2003 was down by roughly 10% from 2002 in terms of the number of deals rated, and down by 20% in terms of volume.

In the synthetic market, the steady decline of single-name credit default swap spreads has had the same impact on issuance, particularly since the CDS market is dominated by investment-grade names with the tightest spreads in the cash bond market. From a peak of 192bp in October 2002, CDS premiums on the Trac-x credit default swap index have declined to around 50bp. Douglas Lucas, CDO analyst at UBS, says: “This has made synthetic CDOs almost impossible to do for the last two months.”

The tightening in spreads in both the cash and CDS markets has quite simply reduced the arbitrage opportunities available to investors in CDOs – as was the case with UBS’s Cherry Valley. And in some cases, tightening spreads have reduced issuer incentive to structure deals.

One recent example is MBIA’s decision to cancel a $2.5 billion CDO of synthetic CDOs. According to market sources, the insurer had been working with Deutsche Bank to securitize part of its CDO portfolio as a means of managing its mark-to-market volatility risk. But the rally in spreads on the underlying collateral will have lessened MBIA’s need to manage that exposure, and the deal was pulled.

So in this tight spread environment, where can investors still find value? According to Russell Hurst, CDO analyst at Banc One, above-average returns continue to be available for multi-sector ABS repackagings and CDO-squared transactions. For CDOs backed by high-yield bonds, high-yield loans and investment-grade obligations, Banc One’s CDO arbitrage index shows returns below those of the other collateral classes, but for the moment at least, execution is still possible. However, Douglas Lucas at UBS warns that investors should be aware that lower collateral spreads could mean a slight shift into lower-rated collateral to gain the same yield.

In 2004, rating agency Moody’s actually anticipates a recovery in issuance volume, believing that growth in newer CDO market sectors, such as middle market collateralized loan obligations, will make up for the relative absence of high-yield cash CDOs. Jeremy Gluck, a CDO analyst at Moody’s, says: “Credit spreads in the market can’t get much tighter, so the CDO market is destined to turn around at some point in the future.”

In the short term, Banc One’s Hurst expects the ABS repackaging growth trend to continue and to be limited only by the supply of repackageable new-issue mezzanine and subordinated ABSs. “We expect record issuance of residential real-estate ABSs and a significant increase in commercial mortgage-backed securities issuance as we enter the first stages of economic expansion in the US and hopefully a recovery in Europe,” he says. “This asset class is currently a favorite for ABS CDOs, and the CDO bid has driven mezzanine and subordinated spreads for residential ABSs to their tightest level in over two years, so the increase in supply will bode well for the CDO market.”

The pipeline for CDO issuance in the months to come currently has 12 cash transactions planned in the US, representing $3 billion. This includes seven high-yield loan transactions, four CDOs backed by ABSs, and one hybrid partially funded investment-grade synthetic CDO, all in various stages of marketing. And according to Hurst at Banc One, it is also rumored that three high-yield bond CDOs are about to begin pre-marketing, after only one or two such deals closing in all of 2003.

In the secondary market, the sharp rally and compression of spreads in the underlying credit markets during 2003, combined with the strong bid for CDO paper, have had a resounding impact on CDO performance. In general, the market across all collateral classes is experiencing a firmness that it did not have for most of 2003 due to the increased credit quality of the underlying collateral.

For CDOs backed by corporate bonds, the prevailing secondary market trend in 2003 was for spreads to become more congruent with the primary market. According to Sivan Mahadevan, credit derivatives analyst at Morgan Stanley in New York, many floating-rate senior notes now trade at or near par – a situation that was simply unthinkable a year ago. What’s more, Douglas Lucas says that UBS recently received its first ever bid above par for a floating-rate CDO. “The most compelling relative-value opportunities now are in structured finance-backed CDOs,” he says.

But Paul Hollowday at Cognis Capital says that many investors have already missed the boat. “Many CDOs, particularly collateralized loan obligations, were stupid cheap back in March,” he says. “But now spreads have tightened so much that the asset class is quite expensive. For the moment at least, further opportunities will be few and far between.”