
A Collateralized Debt Obligation (CDO) is backed by portfolios of assets that may include a combination of bonds, loans, securitized receivables, asset-backed securities, tranches of other collateralized debt obligations, or credit derivatives referencing any of the former. Some market practitioners define a CDO as being backed by a portfolio including only bonds and/or loans, but most market practitioners use the former definition. I’ll use collateralized debt obligation as an umbrella term for asset securitizations that may also include collateralized fund obligations (CFOs) and other structured products.
Up to the end of the 1990s, collateralized debt obligations all used special purpose entities (SPEs), also known as special purpose vehicles (SPVs), that purchased the portfolio of assets and issued tranches of debt an equity. The special purpose entity purchased the assets from a bank’s balance sheet and/or trading books. These are known as true sale structures.
Special purpose entities are usually bankruptcy remote, meaning they are delinked from the credit risk of the bank arranger, also known as the originator. The bank arranger can earn servicing fees, administration fees and hedging fees from the SPV, but otherwise has no claim on the cash flows of the assets in the SPV.
Formerly, banks and investment banks underwrote the collateralized debt obligation tranches to provide the funds – often along with bridge financing – for the purchase of the portfolio assets, which backed the tranches. This is no longer always the case. Synthetic securitizations eliminate the need for a special purpose entity, albeit they may also use a special purpose entity to issue limited recourse notes linked to a collateralized debt obligation’s tranched credit risk. These structures rely heavily on credit default swaps and other credit derivatives to transfer risk without transferring assets.
Credit enhancement comes in a variety of forms, and several types of credit enhancement are usually structured into a single transaction. Popular forms of credit enhancement include the following:
The highest rating possible on a wrapped tranche is the rating of the credit wrap provider unless the credit of the wrap provider deteriorates below that of the value of the assets and collateral. The amount of credit enhancement required for a particular rating depends on the deal structure. The amount of enhancement is expressed as a multiple of the expected loss level. For instance, to get an AAA rating, a general rule is that the credit enhancement must equal five times the expected loss level. The amount of enhancement required declines for lower rating requirements.
Overcollateralization is achieved through investment in more collateral than is required to meet the CDO’s static cash flow requirements. The SPE may invest in more physical assets, or equity investors may inject additional cash into the deal. Cash injections are usually invested in virtually risk-free assets such as short-term government assets.
Tranching creates more than one class of debt within a given structure. Junior investors provide credit enhancement for senior investors. Investors in subordinated classes earn a higher coupon but have a higher risk to their initial principal investment. The subordinated debt holders agree to absorb losses before the senior debt holders. Several tranches may exist in one deal, and the payments due to each tranche holder are defined in the prospectus according to the tranche payment priority. The more certain the payment, the higher the credit rating and the lower the nominal return. The less certain the payment, the lower the credit rating and the higher the nominal return. Issuers may deposit cash in a reserve account or a trust account and these funds can be used to meet principal and interest payments as needed. Excess spread may also be trapped in a reserve account.
Excess spread is the current cash flow remaining after payment of investor coupons and fees. The excess spread account may be used for the benefit of all of the tranche investors to offset the loss in value from non-performing assets with remaining funds reverting to the deal manager or the first loss investor at maturity, but this is not necessarily the case. In some structures, the reserve account is for the benefit of the first-loss investor only.
Any potential future stream of payments or future value can be securitized. The following summary shows broad classifications one should be on the alert for in evaluating reported collateralized debt obligation data. Statistics on collateralized debt obligation issuance are often reported according to the type of collateral classification. Many CDOs securitize assets from more than one asset class, so it is difficult to generalize. Different reports may use different definitions and benchmarks, so it is important to read the classification criteria to understand reported data.
One important class is consumer receivables, which include auto loans, auto leases, home equity loans, auto leases, and credit card receivables. Student loans have unique characteristics and are grouped into a separate category. Nonnative English speakers should note that most practitioners refer to the former as asset-backed securities, or ABS and do not lump mortgage-backed securities and other securitizations into this category, even though ABS may sound like a generic term.
Mortgage-backed securities are usually categorized as residential mortgage loans (single-family, multi-family, condominium, cooperatives) or commercial mortgage loans. Manufactured housing loans are often grouped into a separate category due to unique homeowner profiles and collateral characteristics. When CDOs became backed by subprime mortgages, the normal benefits of securitization failed to provide protection due to fundamental flaws in loan origination practices and voluntary due diligence failures, as documented in our real-time crisis analysis.
Receivables such as computer leases, aircraft leases, marine leases and equipment leases are usually grouped separately. “Stranded costs” of utility companies have been securitized for decades and have a separate category. Equipment loans and equipment leases are considered a separate small to medium-sized loan category. Expectations of payments, such as delinquent tax liens have a separate category. Future cash flows deals have another separate category and are sometimes included in emerging market statistics.
A collateralized loan obligation (CLO) is a type of collateralized debt obligation. CLOs are backed by a portfolio of loans. The term CLO is reserved for a securitization that is exclusively backed by loans. In 1990, the first rated CLO backed by U.S. bank loans was brought to market.
Commercial and industrial loans include investment-grade corporate loans and high-yield corporate loans. Small business loans in the United States have their own category due to the government guarantee and unique structural features. In Europe, small to medium-sized enterprise loans are grouped into a multi-country category. These loans are smaller in size than usual commercial and industrial loans, and the obligors are often not publicly rated. Investment-grade corporate bonds and high-yield corporate bonds are often viewed as separate categories.
Bonds backed by receivables such as the right to televise Formula One sporting events, or cable subscriptions are also sometimes classified as CDOs, although they are usually lumped into the broader category of structured finance.
Private equity and hedge fund of fund securitizations are in separate categories. Non-performing loans also have a separate category. Investment-grade sovereign debt may be included in bond obligations, but emerging market debt is usually considered separately, albeit this debt is often found in high-yield bond obligations.
I just mentioned assets that can be used in the CDO portfolio, but even the asset definitions are subject to debate. For instance, asset-backed securities are structured securities in which the underlying collateral is itself a pool of loans or receivables.
A collateralized loan obligation (CLO) is a type of collateralized debt obligation. CLOs are backed by a portfolio of loans. The term CLO is reserved for a securitization that is exclusively backed by loans. In 1990, the first rated CLO backed by U.S. bank loans was brought to market.
I just mentioned assets that can be used in the CDO portfolio, but even the asset definitions are subject to debate. For instance, asset-backed securities are structured securities in which the underlying collateral is itself a pool of loans or receivables.
A collateralized bond obligation (CBO) is another type of collateralized debt obligation. In 1988, the first rated collateralized bond obligation backed by high-yield bonds was brought to market. CBOs are backed by a portfolio of secured or unsecured senior or junior bonds issued by a variety of corporate or sovereign obligors. Often a collateralized bond obligation will include loans in the portfolio, but the majority of the collateral usually consists of bonds. Collateralized bond obligation transactions realize the positive spread between a portfolio of higher-return, higher-risk assets, often rated below ‘BBB’; and lower-cost, highly rated collateralized bond obligation securities issued to purchase the portfolio of bonds backing the collateralized bond obligation.
A collateralized mortgage obligation (CMO) is backed by mortgage-backed securities (MBSs) also called mortgage pass-through securities. CMOs and the individual tranches of CMOs are also called mortgage-backed securities. This terminology is usually reserved for the U.S. market, and we’ll briefly look at this later. Ironically, CMOs, the collateralized debt obligation precursors, are usually excluded from collateralized debt obligation reports and are not considered in the definition of collateralized debt obligations by many market professionals.
Structural differences in the European market did not lend themselves to the same type of cash flow tranching as the U.S. market. Often European mortgages do not enjoy government support. A look at the German market illustrates just some of the differences between the U.S. and European mortgage-backed securities markets. The German Pfandbriefe market has been in existence for more than 200 years. The underlying mortgage loans must have a loan-to-value ratio of less than 60%, so the overall credit quality is deemed to be AAA. In the U.S. market, mortgage-backed securities are a balance sheet management tool. In the German market, however, Pfandbriefe cannot be prepaid and are kept on the bank’s balance sheet, albeit they may be pledged as collateral.
Securitization of other types of mortgage-backed securities as a balance sheet management tool has gained momentum only in the past three years after a very slow start.
Collateralized debt obligations are further classified as synthetic (credit derivatives) or cash. Some collateralized debt obligations are backed by combinations of cash and synthetic securities. Collateralized debt obligations are either arbitrage deals, balance sheet deals, or both. Both can have any combination of cash and synthetic underlying collateral.
Balance sheet CDOs are normally used for regulatory capital relief, albeit funding may be another motive for a balance sheet deal.
Arbitrage CDOs take advantage of greater market prices for the underlying obligations versus the price at which the combined liabilities can be sold.
Arbitrage CDOs are either market value CDOs or cash flow CDOs. They can be backed by either cash assets, synthetic assets, or a combination of both. Cash flow CDOs make up 85–90% of the arbitrage market, and are growing at a much faster rate than market value CDOs.
Market value collateralized debt obligations make up approximately 10-15% of the arbitrage CDO market. Market Value deals, like most CDOs, will either 1) pledge assets to a trustee to back the debt for the benefit of the investors, or 2) a bankruptcy remote SPV buys the assets and pledges the collateral to back debt or equity (or preferred stock) for the benefit of the investors. Market Value deals may also use a third, less common, structure: investors have a claim on all of the assets in the fund and the seniority of their claim depends on the class of debt or equity (preference shares or preferred stock) they own.
Market value structures meet principal (if applicable) and interest liabilities by generating cash from trading assets, and from interest on invested assets. If a collateral pool consists of defaulted bonds or loans, the market value structure is usually used, since these assets do not generate predictable cash flow streams, but have significant market value upside potential. Managers who have good track records trading investment-grade debt and high-yield debt often prefer market value deals.
Market Value deal managers trade actively and aggressively and usually employ leverage. Of course, not every trade results in a gain. Both credit and market considerations are important to market value deal managers. Investors who must market their assets to market might prefer market value structures, since market value CDO managers must mark the CDO’s portfolio to market.
The ratio of the market value of assets to the face value of liabilities is the focus of a market value collateralized loan obligation. Market value deals require overcollateralization or a “haircut”. The overcollateralization protects investors from asset price volatility due to changes in general interest rate movements, general credit spread movements, or other general market movements.
Market value CDOs require maintenance of a minimum overcollateralization level. If this level is breached, the assets must be sold to pay down liabilities or the assets must be sold and exchanged for highly rated liquid instruments. This is known as a trigger event. The rating agencies monitor the debt coverage ratios. The performance of a market value collateralized debt obligation depends upon the manager’s trading ability.
Non-synthetic cash flow collateralized debt obligations invest in cash assets as collateral. The structure passes principal and interest payments generated by cash flows of the underlying collateral through to the investors. For a deal to be successful, the cash flow from the collateral pool must meet all deal liabilities including the interest and principal obligations from the notes issued by the CDO.
The cash flows are hedged, reinvested, and tranched by time and seniority. The rating of a note depends on the probability of cash flow sufficiency to service that particular note. If the cash flow deal is properly structured, investors only experience a loss if there are defaults in the collateral pool.
Cash flow CDOs backed by a pool of cash obligations require a manager, whereas synthetic cash flow deals may or may not have a manager. For the cash asset-backed CDO, the manager’s role is more limited than that of a market value CDO manager. The manager is carefully chosen for credit expertise rather than trading expertise. The cash flow CDO manager of a cash instrument backed deal chooses the initial diversified portfolio of cash instruments and chooses credits to purchase during the reinvestment period. The manager may also participate in a workout in the event of default to maximize recovery. We’ll examine these structural details when we compare cash and synthetic CDOs.
Deals can be managed or unmanaged. Only CDOs that reference a static pool of assets can dispense with a manager. This is most common with arbitrage synthetic CDOs, or cash CDOs where a fixed pool of assets is used in a pass-through-type structure.
Cash CDOs are usually broken into broad categories that represent the bulk of the issuance: high-yield loans, multisector, investment grade corporate (which normally includes one-quarter to one-third high-yield collateral), and high-yield bonds.
Synthetic CDOs are backed by credit default swaps on a diversified pool of reference obligors. The credit default swaps can reference any type of cash obligation. They can be either balance sheet deals or arbitrage deals and are often classified into three categories: 1) managed CDOs; 2) CDOs with the right of substitution; and 3) static CDOs. However, the flexibility of synthetic instruments created unique challenges during the financial crisis, as detailed in our comprehensive analysis of synthetic CDO risks and market practices.
Managed CDOs will have an actively involved manager and may look similar to a market value CDO in that the underlying pool of assets is actively traded in an attempt to add value. Some managers are more passive and trade only to avoid losses. CDOs with the right of substitution have managers who may not be “arms-length” managers. The bank arranger’s structuring group may handle management activities that allow them to trade out of credits according to pre-set rules. The object is to trade out of deteriorating credit before default and trade into a higher credit quality obligation.
Static CDOs can be cash or synthetic. Static pools of credit default swaps back a static synthetic CDO. No trading is allowed. By definition, this is a cash flow CDO. Synthetic static arbitrage CDOs differ from cash CDOs that are also cash flow CDOs because the synthetic deal normally has no pre-planned reinvestment period. Since the CDS market allows cash flow maturities to be defined at the deal inception, there is no need for reinvestment, other than possibly to reinvest recovered amounts after a default, if any.
Understanding the risks inherent in collateralized debt obligations requires careful attention to both structural features and market dynamics that can lead to significant losses. Historical analysis reveals several critical warning signs that investors and risk managers must evaluate when assessing CDOs.
Moral Hazards in Management: CDO structures can create dangerous conflicts of interest when managers have claims on equity cash flows while making decisions affecting senior noteholders. Four structural conditions pose particular danger: losses allocated in reverse order of seniority with limited loss deductions, excess spread that doesn’t benefit noteholders or absorb losses, inadequate restraints on managers’ ability to deteriorate portfolio quality, and managers with claims on excess spread. These conditions can incentivize managers to trade out of good credits into lower-rated, higher-spread credits, potentially compromising even single-A tranches.
Transparency Challenges: Many cash assets underlying CDOs, particularly distressed loans, are difficult to mark-to-market, forcing investors to rely on manager pricing without verification mechanisms. Market-value CDO structures face particular scrutiny given historical scandals where managers allegedly marked up asset prices shortly after increasing performance fees while actual market prices declined.
Single-Tranche CDO Risks: In synthetic CDO trading, investment banks often sell only mezzanine tranches while retaining risk equivalent to the remaining tranches of the full notional amount. This creates concentration risk that is frequently underreported when banks discuss only sold tranche amounts rather than total risk exposure.
Correlation and Concentration: When multiple CDOs reference similar underlying names or use common indexes as portfolio reference entities, investors face significant concentration risks. This became apparent during credit events when supposedly diversified portfolios experienced correlated defaults.
The subprime mortgage crisis demonstrated how CDO structures could amplify rather than mitigate underlying asset risks. Our prescient analysis during the 2007 crisis identified predatory lending practices that fundamentally undermined CDO collateral quality, with loss projections that proved six times more accurate than Federal Reserve estimates.
Key warning signs that emerged included aggressive “reaging” of mortgages by servicers, recovery rates falling to 3-6 cents on the dollar, and widespread use of stated-income loans with inflated borrower qualifications. These practices, combined with CDO structures that obscured rather than revealed underlying risks, contributed to massive losses across multiple investor classes.
The evolution of collateralized debt obligations provides crucial lessons for understanding both the potential and the pitfalls of structured finance. Our real-time analysis during the 2007-2008 financial crisis offers unique insights into how CDO structures performed under stress.
Early Warning Indicators: In 2005, we identified mounting risks in CDO structures, particularly those backed by subprime mortgages, documenting specific warning signs that investors should consider when evaluating these complex instruments.
Accurate Loss Projections: Our 2007 forecast of approximately $300 billion in subprime losses—six times higher than Federal Reserve estimates—proved remarkably accurate, based on realistic assumptions about default rates and recovery values that market participants were reluctant to acknowledge.
Predatory Lending Impact: The crisis revealed how predatory lending practices by mortgage originators fundamentally compromised CDO collateral quality, with widespread income inflation on loan applications and unsuitable loan products that set borrowers up for failure.
Our analysis documented extensive fraud across multiple major institutions:
The aftermath of CDO-related losses extended far beyond financial markets. Our post-crisis analysis documented how predatory mortgage practices led to urban blight, with entire neighborhoods abandoned as banks walked away from properties worth less than foreclosure costs. This created lasting economic damage in communities while institutions responsible for the crisis received taxpayer bailouts without meaningful accountability.
The CDO crisis exposed fundamental flaws in the credit rating process, leading to our decision to revoke support for the credit rating agency system and advocate for structural reforms in how complex securities are evaluated and regulated.
These case studies demonstrate that while CDO structures can provide legitimate economic benefits through risk transfer and capital efficiency, their complexity requires heightened due diligence, robust regulatory oversight, and careful attention to the incentive structures that govern all market participants.
This menu of different products may seem confusing at first. The structural details are important. The place to start evaluating the fundamental value of any securitization is with the pool of assets or portfolio. If the investor likes the composition of the initial portfolio, then the negotiation begins for the best possible structural protections.
Adapted from: Structured Finance and Collaterlaized Debt Obligations, by Janet Tavakoli, John Wiley & Sons, 2008.
Janet Tavakoli is the president of Tavakoli Structured Finance, founded in 2003. She is a globally recognized expert in structured finance and derivatives who has advised financial regulators, served as an expert witness in major financial litigation, and authored definitive works on credit derivatives and securitization. Her expertise spans complex financial instruments and risk management, making her a sought-after authority in the field of structured finance.
Tavakoli, Janet. “Bear Stearns Bailout: 2007 Warning Predicts 2008 Crisis.” Tavakoli Structured Finance, June 27, 2007. Updated with Epilogue 2025.
Tavakoli, Janet.”Cash Flows from Synthetic CDOs,” TSF. Based on “The Elusive Income of Synthetic CDOs” – Journal of Structured Finance – Winter 2006 Volume 11, Number 4. – Janet Tavakoli Updated 2025.
Tavakoli, Janet. “Comments on SEC Proposed Rules and Oversight of NRSROs” Letter to Securities and Exchange Commission, February 13, 2007.
Tavakoli, Janet. Dear Mr. Buffett, What an Investor Learns 1,269 Miles from Wall Street. Hoboken, New Jersey: Wiley, 2009.
Tavakoli, Janet. “Dicey Deals Done Dirt Cheap: (MBIA and Ambac Face Massive Downgrades: Monoline Crisis) Tavakoli Structured Finance. Originally published January 3, 2008 | Updated July 2025.
Tavakoli, Janet. “Goldman Sachs and the AIG: The Hidden Financial Crisis Story,” TSF. November 2009. Updated and Expanded 2025.
Tavakoli, Janet. “Letter to the Editor: Cynical use of derivatives has market in a pickle.” Financial Times, January 31, 2008.
Tavakoli, Janet. “Letter to the Editor: Greater Global Risk Now Than At Time of LTCM.” Financial Times, May 7, 2007.
Tavakoli, Janet. “Letter to the Editor: Subprime lending excesses have damaged US’s standing as global leader in finance.” Financial Times, March 19, 2007.
Tavakoli, Janet. “Merrill Lynch CDO Crisis: Analysis & Aftermath (2008-2025),” TSF. July 30, 2008. Updated 2025.
Tavakoli, Janet “Michael Lewis vs Wall Street Reality: Expert Analysis,” TSF. March 10, 2010. Updated 2025.
Tavakoli, Janet. “Repairing the Damage of Fraud As a Business Model,” Presentation to the Federal Housing Agency’s Supervision Summit, December 8, 2010.
Tavakoli, Janet “Subprime Mortgages: The Predators’ Fall,” GARP Risk Review, 2007.
Tavakoli, Janet. “Synthetic Collateralized Debt Obligations: Risk, Valuation, and Market Practices.” Tavakoli Structured Finance.
Collateralized Fund Obligations (CFOs): Complete Guide
Credit Derivatives and Credit Default Swaps
Credit Rating Agencies: How Moody’s, S&P and Fitch Failed
Special Purpose Entities: Guide to SPEs in Finance
Total Return Swaps (TRS): Benefits, Risks, and Applications