Print this
CDOs Under Fire

Chopping up portfolios of bonds, loans and other instruments into different credit tranches has become popular with issuers and investors alike. But what will happen to collateralized debt obligations if credit fears paralyze the financial markets?

By Karen Spinner

The basic idea is simple. Take a pool of bonds, loans or other financial assets and distribute whatever money they generate into different tranches in much the same way that mortgage-backed securities are split apart into IOs, POs and other arcane variations. Investors will be protected from loss by the diversification in the underlying pool of assets and by a variety of credit enhancements built into the structures.

That, anyway, is how collateralized debt obligations (CDOs) are designed to work. Collateralized bond obligations (CBOs), collateralized loan obligations (CLOs) and credit-linked notes (CLNs) have become quite popular in the past year because they allow issuers to sell off assets and free up capital, and allow investors to get a nice yield pickup from a diversified pool of assets.

The latest round of credit downgrades in Asia and Eastern Europe, however, has proven that even the most carefully designed structures can founder if markets are stressed enough. On August 27, Moody’s Investors Service put 11 hybrid CBO/CLO tranches on credit watch, and downgrades have occurred in emerging-market countries such as Venezuela, Brazil, Ecuador, Romania and Ukraine. On October 22, Moody’s downgraded another 15 tranches issued within 12 CBOs, while seven tranches of five CBOs remained under review for downgrade. A day later, it announced it was reviewing for possible downgrade the A1 ratings of three Bank of Tokyo-Mitsubishi-backed CLOs—a sign that the banking crisis in Japan continues to fuel fears of a global economic downturn. The result: most CBO and CLO transactions nowadays seem to be about as attractive to investors as tainted meat. Layoffs are already beginning at some of the major houses, such as Merrill Lynch, that were most active in CBOs and CLOs structured around portfolios containing emerging-market debt.

“Earlier this year there were so many deals that we routinely had to stay late and work on weekends,” says one financial engineer who has since left a major CBO structuring desk. “Now there are literally no deals in the pipeline. Investors won’t touch the stuff.” Standard & Poor’s has rated more than $25 billion in CBO, CLO and similar transactions thus far this year. The question, however, is what will happen if the remaining $20 billion of deals in S&P’s pipeline this year are put on hold.

Even transactions driven by portfolios composed of high-quality credits have become difficult to place in this nervous market. Two major Japanese CLOs—a $2.25 billion deal from Bank of Tokyo Mitsubishi and a $1.7 billion deal from Dai-Ichi Kangyo Bank—have been indefinitely postponed, even though both transactions involved the securitization of high-quality U.S. corporate loans. “We have been getting a lot of calls from the press about CBOs and CLOs,” says one dealer. “Some analysts are trying to draw wholesale parallels between these structures and the mortgage-backed securities market of the early 1990s, in which fundamental shifts in prepayment behavior led to substantial, unpredicted investor losses.” In short, investors are scared.

Investor panic, however, may be an overreaction. Only a small handful of developed-market CBOs and CLOs have been downgraded or shut down over the past few years, and those losses were primarily absorbed by the investors holding unsecured instruments of high-risk tranches. Comparisons of the problems associated with CDOs and those associated with mortgage-backed securities may also be unfair. Mortgage-backed securities were sold to a wide variety of fixed-income investors—and frequently to investors who didn’t understand what they were buying. The CDO investors taking the hits today are hedge funds, proprietary money managers and other market experts who are well aware of the risks and returns of the products. And, in many CLO structures, it is often the issuing bank itself that takes the equity tranche.

The ABCs of CDOs

CBOs, CLOs and CLNs come from the same financial family and are called, collectively, collateralized debt obligations. But, as with most families, some members are more reliable than others, and some run with different crowds. The instruments are particularly perplexing because no two structures are exactly alike, and they may be used for any number of purposes.

They do, however, share some common characteristics. CDOs are diversified pools of bonds, loans and other assets that are chopped up for investors into different tranches. These multiple tranches may include a senior piece, rated AAA or AA, a mezzanine piece rated BB or B, and an equity piece, which is usually sub-investment grade or unrated. Because rating agencies require a certain amount of over-collateralization to protect senior note-holders, the equity investors usually receive higher returns in return for shouldering the first losses in case of a major credit event. “The first-loss investors credit-enhance the senior and mezzanine tranches,” explains Steve Patterson, a senior vice president in global trust services at Chase Bank of Texas, a trust that administers more than 100 CBOs and CLOs.

“Analyzing and stressing the reference portfolio is critical, because investors may be affected in the event of a reference credit or swap counterparty default.”
Reza Bahar
managing director,
Standard & Poor’s

Approximately twice as many institutional fixed-income management firms use CBOs than use CLOs, says David Farrington, chairman and CEO of Capital Access International, which tracks holders of fixed-income instruments. The firm’s database lists more than 150 institutional investors holding CBOs totaling about $3 billion, and 85 companies holding CLOs totaling $1.75 billion, with insurance companies more active in both instruments. In a survey conducted in August, Capital Access found about half of the 4,500 institutional fixed-income money managers said they were willing to look at both these instruments for their investment portfolios.

CDOs inevitably require a bankruptcy-remote special purpose vehicle or limited liability corporation, both of which are designed to separate the performance of the issuing bank from the performance of issued notes. The performance of CDOs, in other words, depends on the performance of the underlying collateral portfolio, not on the performance of the issuing institution. CDOs also require the services of a portfolio manager to manage the underlying collateral portfolio and a trustee to make sure all the various cash flows go to the appropriate tranches.

CDOs are issued for a variety of purposes. Some originate as a means to create high-yield debt instruments for certain classes of investors, such as hedge funds. Others are issued primarily to obtain medium-term funding that can be reinvested or loaned at favorable rates.

The most growth, however, is in CLOs, CBOs and CLNs structured as balance-sheet management tools that allow issuing banks and insurance companies to transfer the risk associated with certain portfolios of loans or bonds to special purposes vehicles and, through them, to a variety of investors. CLOs require that the issuing bank transfer the asset portfolio to the SPV, thus removing these assets from the bank’s balance sheet. In CLN transactions, credit default swaps are used to transfer the risk associated with the asset portfolio, although the assets remain on the balance sheet of the issuing entity. In both cases, issuers can benefit by minimizing their credit risk and, thus, reducing regulator-mandated capital allocations.

These balance sheet tools have become quite popular in Europe and Asia, where banks have less commercial debt (such as mortgages and credit cards) to securitize. Richard Gugliada, a managing director at Standard & Poor’s, points out that securitizing a large portfolio of loans is a lot more difficult than securitizing a pool of commercial mortgages. “Corporate loans are not interchangeable and include obligor-specific, often idiosyncratic, terms and conditions,” he explains. “But CLOs and CLNs can be an excellent capital management tool in cases where a bank may not hold a great deal of commercial debt and regulators do not make fine distinctions between high-quality corporate credits when determining capital allocations.”

Credit fears, however, are likely to affect the underlying composition of CDOs and dramatically reduce the number of these transactions over the short haul. Dealers have noted that the credit quality of underlying assets built into these transactions has been increasing, allowing issuers to attract more credit risk-conscious investors. Placing the highest-risk equity tranches has become the major impediment to bringing deals to market. “The equity pieces are typically the most challenging classes to place,” says Patterson.

Investors who have purchased higher-rated tranches (other than the riskiest emerging-market variations), however, have little to fear. Most of the credit goes to issuers and ratings agencies, which have worked to build extremely robust structures. In most cases, the documentation in these instruments contains numerous analytic warning bells and fail-safe procedures designed to protect senior and mezzanine investors from losses associated with any downgrades and defaults within the underlying portfolio.

CDOs can be broken down into three general categories: cash flow, market value and credit-linked notes.

Cash Flow CDOs

In cash flow CDOs, the various tranches of the deal are rated according to the likelihood that cash flows generated by the underlying portfolio of assets will cover investors’ principal and scheduled interest payments. In a typical structure, there will be an AAA or AA senior tranche, a BBB or BB mezzanine tranche and an equity tranche. The investment grade tranches are typically overcollateralized. For example, a $500 million pool of collateral could be divided up into a $300 million senior tranche, a $150 million mezzanine tranche and a $50 million equity tranche. The exact ratios of overcollateralization are determined by the rating agencies’ evaluation of the overall credit quality of the underlying portfolio of collateral.

Typically, the underlying portfolio composition of cash flow CDOs is relatively fixed according to industry sector, geographical and credit quality concentrations; the overall portfolio must also meet a predetermined weighted average reflecting overall credit quality. For example, an extremely simplified portfolio might include 50 percent technology debt, 25 percent financial sector debt and 25 percent manufacturing debt, distributed across an 80 percent allocation to the U.S. market and a 20 percent allocation to the U.K. market. The typical collateral pool, however, is much more diverse.

Because the cash flows associated with the underlying portfolio are so critical for covering investors’ principal and interest payments, ratings agencies do not allow much room for deviation from the portfolio composition and overall ratings benchmarks outlined in the original agreement. The collateral manager may, according to the terms of the contract, engage in limited trading, particularly if certain assets appear to be in danger of a downgrade or other credit event. To hedge cash flows, the collateral manager may also be permitted to purchase caps and swaptions; these hedge transactions are often specified in the contract documentation.

“If you believe that the market is now overstating credit risks, this may be a good time to invest in an equity or subordinated tranche—if your risk profile permits it.”
Bidyut Sen
Lexam Capital

Basically, the collateral manager’s role in cash flow CBOs is passive; the manager monitors the portfolio and related hedge transactions to ensure compliance with contractual and rating agency guidelines, and takes corrective action to ensure that portfolio-wide cash flows are sufficient to cover regular payments to investors.

Cash flow CDOs are intended to provide regular returns to a spectrum of investors, and their relatively static portfolio structure is designed to minimize risk. Typically in CDOs, available cash flow from the collateral—after payment of scheduled interest on the liabilities, reinvestment purchases and ongoing expenses—is paid out as dividends to the subordinated note-holders. This dividend is paid to investors at predetermined intervals, although some CDOs specify for dividends to be reinvested for a period of time.

Cash flow CDOs have more restrictions in terms of portfolio composition and trading flexibility. As a result, they tend to be less volatile compared to market value CDOs, says Bidyut Sen, president of Lexam Capital. The ultimate return on cash flow CDOs depends on actual defaults, whereas in market value CDOs the ultimate returns depend on the market value of the portfolio at maturity. According to Sen, equity investors generally represent 10 percent to 20 percent of the total capital structure in cash flow CDOs, and equity returns range from 15 percent to 20 percent.

“Lower-rated credits will affect the overall portfolio rating in a manner that is disproportionate to their representation.”
Robert Grossman
group managing director
of the loan products unit,
Fitch IBCA

Credit-rating agencies evaluate cash flow CDOs by carefully reviewing the terms of each contract, the notional amount of collateral to be allocated for each class of investor and the composition of the underlying portfolio, which may include investment-grade debt, non-investment-grade debt or a combination of the two. According to Fitch IBCA’s Robert Grossman, group managing director of the loan products unit, part of this evaluation process involves developing a base default rate for the entire portfolio, in which a nonlinear relationship between credit ratings and default probabilities is assumed; in other words, a BBB instrument will have a much more significant impact on the base default rate than an AA instrument. “Lower-rated credits will affect the overall portfolio rating in a manner that is disproportionate to their representation,” explains Grossman. Similarly, this overall rating may be adjusted upward by a “stress multiple” based on the maturity of the portfolio, the asset manager’s track record and other factors.

Market value CDOs

Although market value CDOs are similar in structure to cash flow CDOs, there are striking differences in how the underlying pool of collateral assets is managed. These instruments are called market value CDOs because the market value of the underlying portfolio of traded instruments determines the level of low-risk collateral that must be held to protect senior tranche investors.

Essentially, the collateral pool underlying market value CDOs behaves like a margin account. Assets are dynamically reallocated between a trading account and a risk-free holding account designed to protect senior bond-holders. When the market value of the underlying pool of collateral drops to unfavorable levels, the portfolio manager must respond to a “margin call” by liquidating assets from the trading account and transferring the value to a holding account to protect the senior bond-holders from losses.

Often, investors in CDOs do not receive coupon or dividend payments in the early years. Instead, trading profits or net positive cash flows are reinvested for a preset period of time in order to enhance leverage and yield for the equity tranche. In market value CDOs, where the bank issuing the structure has retained a considerable portion of the equity tranche for its own account, the extent of reinvestment and leverage is often higher. Equity investors in such market value CDOs expect to make 25 percent to 35 percent returns, but are also taking more risk in terms of portfolio composition and active trading.

Because the portfolios underlying market value CDOs are actively traded across a wide selection of asset classes and credit ratings—and because overall portfolio composition may change over time—credit rating agencies carefully analyze the margin formulas associated with these deals. For example, market value CDOs may include bonds, high-yield loans or even pure equity, or a combination of these assets. Each type of asset that the portfolio manager is allowed to trade receives a “haircut” that is approved by the rating agency. This haircut refers to how much collateral or margin must be drawn from trading accounts and transferred to the holding account to support an unfavorable change in market value.

According to Nels Anderson, a vice president at Moody’s in London, the vast majority of market value CDOs are based on pools of liquid U.S. assets, such as bonds, traded loans and so on. The reason is straightforward: There are usually a great deal of historical data available on these assets, which, in turn, make it possible to model the price variation of the various asset types under a wide variety of market conditions. “Emerging market assets are extremely difficult to model, as are many European and Asian assets, simply because there are not enough data,” says Anderson, “As a result, few market value CDOs involve substantial portions of non-U.S. assets.”

Chase’s Patterson adds that market value CDOs are monitored constantly by collateral managers and their trustees. “The structure of these deals demand that the collateral manager closely monitor the portfolio to ensure that corrective action is taken in the event of a significant deterioration in the market value of the collateral,” he explains.

Credit-linked notes

It is not necessary to set up a special purpose vehicle to get the benefits of a CDO. Issuers looking to move risk off their balance sheets can mimic the behavior of common cash flow CDOs with credit derivatives. Instead of transferring assets from the issuer’s balance sheet to an SPV, issuers can use a credit-linked note to shift the risk associated with a portfolio of assets to the SPV while the assets themselves remain on the issuer’s balance sheet.

The standard instrument used for this purpose is a default swap, a transaction in which the issuer pays the SPV a premium in return for contingent payments should one or more credits in the reference portfolio default. The reference portfolio is designed to mirror the mix of assets on the issuer’s balance sheet. Once the risk has been transferred to the SPV, it can issue notes to seek new funding for reinvestment or to support the issuing bank’s lending and other business activities.

Rx for the Credit Crunch?

Perhaps the most important trends in CLOs is their use as a balance-sheet management tool. They have become an increasingly important technique for banks that want to manage their portfolios to reduce regulatory capital requirements. S&P’s Richard Gugliada believes that balance sheet CLOs may even be an important, make-or-break strategy for the many Japanese banks now under siege: “During the U.S. banking difficulties of the early 1990s, many U.S. banks survived by securitizing portfolios of assets, which resulted in continued access to liquidity,” he explains. “The Japanese banks that survive the current crisis will be those that can successfully securitize assets. CLOs are continuing to provide access to liquidity during the current difficulties.”

While some of the mortgage, bond and consumer loan securitization deals in the 1980s and early 1990s were criticized as ways for investment banks to dump equity tranches on unsuspecting investors, this caveat does not appear to apply to the current balance-sheet CDO market. “In most CLOs, the issuing bank will retain the equity portion of the deal themselves,” notes Lehman’s Carter.

Credit-linked notes may be particularly helpful to insurance companies that face hold-to-maturity investment restrictions and banks that want to maintain a certain-sized balance sheet for investor relations purposes. “We wanted to reduce our capital allocations without doing anything that might appear to shareholders as the ‘incredible shrinking balance sheet,’” explains one Midwestern CLO issuer.

Credit derivative structures that transfer risk from one entity to another but do not issue notes are likely to become increasingly important in today’s market climate, in which even highly rated CLO tranches are difficult to place.


Capital tools

While credit-derivative-oriented structures can be used to build investor-driven, high-yield portfolios, they are most often used to reduce regulatory capital allocation requirements without the expense and complexity of traditional CLO and CBO structures. Because traditional balance sheet CLO and CBO structures require that assets be legally transferred between the issuer and the SPV, a trustee must be engaged to monitor these assets, and the terms of the transfer must be carefully documented for each asset. If these assets include bank loans, the client must be informed that its loan is being legally transferred to another entity—although the issuer will continue to manage the relationship. With credit derivative structures, however, assets remain on the issuer’s books, thus eliminating hassles associated with the legal transfer of ownership.

Indeed, issuers can reap the capital management rewards of balance sheet management through credit derivative structures without actually issuing notes to investors. “Credit-derivative-based structures essentially separate the regulatory capital and credit management decision from the funding,” says Michael Carter, a director in Lehman Brothers’ asset-backed securities finance group. This means that banks can reduce capital allocation requirements associated with even those portfolios that do not lend themselves easily to the development of an attractive investment vehicle.

Credit-derivative-driven structures, then, represent a growing opportunity for banks and insurance companies to transfer balance sheet risk (and, thus, capital allocation requirements) without having to place notes. Essentially, credit derivatives offer issuers a form of virtual securitization, and the flexibility to offer investors a wide spectrum of rated notes, a limited issuance or nothing at all. This is an important consideration in a market in which investors are more reluctant to accept lower-graded securities, particularly securities rated BBB or lower.

When analyzing credit-linked notes before issuance, credit-rating agencies focus on the credit-worthiness, diversification and other attributes of the reference portfolio—which remains relatively static—in order to determine the support requirements for each class of investors. “Analyzing and stressing the reference portfolio is critical, because investors may be affected in the event of a reference credit or swap counterparty default,” says Reza Bahar, a managing director at Standard & Poor’s. “Therefore, determining default probabilities is important to determining the credit ratings associated with each tranche.”

Seasonal high yield?

One of the major reasons that CDOs have been under the spotlight lately is that credit spreads have widened considerably, and downgrades across the emerging markets have resulted in downgrades and losses in a small number of high-yield CDO structures. Usually, these high-yield structures are known as arbitrage CDOs; in other words, these are investor-driven structures in which the underlying portfolio of assets is purchased on the open market rather than transferred from the issuer’s balance sheet. Arbitrage CDOs tend to be higher-yield structures.

The investors most affected by the downgrade incidents have been the equity tranche note-holders. Typically, the investors holding the equity portion of any deal are highly qualified portfolio managers, hedge funds or even the issuing institutions themselves.

According to S&P’s Gugliada, high-yield CBOs, in particular, are likely to be a cyclical phenomenon. When credit spreads were lower over the past year or so, yield-hungry investors were attracted to these structures. “We were interested in market value CBOs, because we were looking to outperform our benchmark in a relatively tight spread environment, in which defaults and downgrades were comparatively rare,” explains one U.S.-based end-user. “We still have some CBOs on the books now, but, today, we would only consider AAA or AA tranches, based on our analysis of the global economy.”

Now that default rates and credit risk in general appear to have increased on a global basis, arbitrage CBO and CLO structures tend to be backed by higher-quality assets. “In balance sheet CLOs, banks are targeting portfolios that have traditionally produced lower returns on regulatory capital and that will require the smallest amount of subordination,” says Lehman’s Carter. In other words, banks are trying to maximize regulatory capital benefits while minimizing the first-loss tranche. Moody’s Anderson adds that a number of proposed deals have simply “disappeared” because the issuers decided they could not easily place the equity or first-loss tranches. Even balance sheet CLOs backed by high-quality corporate loans are harder to place, because they have been tarred by losses associated with high-yield structures.

Depending on your view of the markets, this could be a good time to invest in arbitrage CBOs and CLOs. Lexam Capital’s Sen, for example, notes that if you believe that the market is now overstating credit risks, this may be a good time to invest in an equity or subordinated tranche—if your risk profile permits it.

Some CDO structures have been attracting considerable investor interest despite the current cautious market environment. One particularly successful structure is Chase’s Secured Loan Trust (CSLT) note. These transactions involve an underlying portfolio of leveraged bank loans, the economics of which are transferred to a trust via a total return swap. The trust then issues notes that are typically rated BBB. “These are investor-driven transactions, with specific loan portfolios created in response to client needs,” explains Joyce Frost, head of marketing for Chase Securities’ credit derivatives group.

Since the note’s inception in September 1996, Chase has structured $8.8 billion in CSLT deals. Because leveraged bank loans have higher recovery rates and lower volatility than high-yield bonds, Frost argues, they provide an attractive asset class to leverage via the total return swap. Additionally, in the current environment, leveraged bank loans have held their value extremely well relative to high-yield bonds.

Buyer Beware

Although CDOs are no more intrinsically risky than any other complex structure, both issuers and end users should proceed with care to ensure that potential risks are identified up front, and monitored throughout the life of the deal.

Because CBOs, CLOs and CLNs are complex structures, there is always some risk that negligent investors will purchase something they’re not ready for. However, this so-called complexity risk is widely regarded as a case of buyer beware. “It is important for investors to carefully review contracts and terms,” says one U.K.-based fund manager. “It is obvious, but investors need to know what they are doing.” Paradoxically, he added, it is often the higher tranche investors who are less diligent than those who purchase the equity tranche. In any case, complex documentation must be accurately captured within all appropriate compliance systems once the deal is closed.

Some CDO structures can present liquidity risks for both issuers and investors. Once a CDO structure is closed, it can be difficult for the issuer to unwind the transaction. For investors, the liquidity of a particular note depends on the current market climate and the relative level of issuance at the time. For example, explains Lehman’s Carter, in a market environment in which numerous CBOs and CLOs have been issued, these notes may be more liquid than similarly rated single-corporate bonds.

For both issuers and investors, it is important to be able to incorporate CBOs, CLOs and CLNs within portfolio-wide credit- and market-risk management analyses. It is also important to understand how these instruments can be broken down into cash flows in order to incorporate them into firm-wide value-at-risk and P&L models most effectively. “One of the most effective ways to model the behavior of an entire bank-wide portfolio that may include everything from traded instruments to credit card loans to asset-backed securities is to slice and dice each asset into its component cash flows,” says Jefferson Braswell, who, as CTO of Risk Management Technologies, has specialized in the design and implementation of bank risk management and technology strategies. “The ability to model CBO, CLO and CLN cash flows is an important first step to including these instruments within bank-wide risk management and other analyses.”

CLOs, in particular, may involve certain legal risks. The loan documentation tends to be customized, introducing the potential for unique risks, and necessitating substantial legal review of a loan in a portfolio. This is one of the major impediments to the securitization of corporate loans. “Currently, the trend in bank loans is to standardize documentation in order to facilitate the securitization of loans,” notes S&P’s Richard Gugliada.

Most of the rating agencies maintain rigorous models designed to stress-test proposed CDOs and ensure notes’ advertised performance under numerous market conditions. Investors must consider the possibility, however, that rating agencies’ models may not be 100 percent accurate.
For example, while correlations between stock price and credit quality are quite well-accepted, there are few data surrounding correlations between interest rates and credit quality. Certainly, this dearth of data has considerable implications for CBO and CLO portfolios. The usual separation of credit and market risk reporting can also exacerbate this uncertainty. “Although credit and interest rate models have been developed separately, the challenge for CDO users and issuers is to integrate these models in a consistent way across the enterprise, that will help managers capture increased market and credit risk in tandem,” says Braswell.

For issuers of balance sheet CLNs, it is important to ensure that your local regulators will accept credit-derivative-driven structures as a legitimate way of reducing capital allocation requirements. For issuers of conventional CLOs, it is important to monitor carefully any trading activity following the close of the deal. Says one investment banker, “If a CLO issuer consistently buys back failing loans to support the SPV, the CLO is no longer remote from the bank’s balance sheet, and regulators may then disallow capital allocation relief.”


Was this information valuable?
Subscribe to Derivatives Strategy by clicking here!