AN INTRODUCTION TO

COLLATERALIZED LOAN
OBLIGATIONS

John Martin (Baylor University) and Akin Sayrak (University of Pittsburgh)

Draft—September 2, 2022

We are grateful for the comments of Don Chew (founding editior of the Journal of
Applied Corporate Finance), Marty Fridson (Lehmann, Livian, Fridson Advisors LLC),
Laila Kollmorgen (PineBridge Investments), Jake Pruner (Morgan Stanley), and Don
Puglisi (Puglisi and Associates) to this paper. All errors are our own.

Electronic copy available at: https://ssrn.com/abstract=4207996

Abstract

Collateralized Loan Obligations (CLOs) are now the largest nonbank lender in the U.S. This added
source of financing, which lies outside the purview of banking regulation, is part of the “shadow banking
system.” Lack of transparency and regulatory oversight of CLOs have given rise to the concern that this
growing market might contribute to another financial crisis like the Global Financial Crisis (GFC) which was
triggered by securitizations of sub-prime mortgages. In this paper, we provide a first look at CLOs for
anyone wanting to better understand their distinctive features as a securitization vehicle that allowed
them not only to weather the GFC but thrive in the post-GFC world. Collateral performance of CLOs
suffered during GFC, except their recovery was faster and stronger than that of mortgage-backed
securitizations.
We offer several observations about loan securitization using CLOs that we think will make them much
more accessible to interested readers:
1. CLOs naturally fulfil a need, or they would not exist.
2. CLOs rely on pooling, tranching and overcollateralization to create new securities with investment
risk ranging from AAA to equity risk.
3. CLOs are quite different from the sub-prime mortgage securitizations that triggered the global
financial crisis:
a. The asset being securitized is different. CLOs securitize portfolios of short-term, floating rate
corporate loans made to borrowers from a diverse set of industries compared to sub-prime
mortgages based on a single industry, residential real-estate.
b. CLOs are more transparent. CLOs provide monthly reports to their investors detailing their
loan portfolios and the quality of the collateral pool is monitored regularly.
c. Finally, CLO asset managers actively manage the composition of the collateral portfolio
whereas mortgage-backed securitizations are typically passively managed.
There is much that we do not fully understand about the risks posed to investors and the overall
economy by CLOs. First, CLOs are very reliant on credit ratings, so it is crucial to the health of the market
for CLOs that these ratings be dependable. Second, CLOs securitize “subprime” bank loans which has
dramatically increased the availability of credit to this class of borrowers which may pose unwelcome risk
to the economy. Third, the CLO collateral loan portfolio is diversified across industries; however, this
diversification does not reduce economy-wide or systemic risk. Finally, price and return data for CLO
tranches are not readily available which makes it very difficult to estimate actual investment returns to
the various tranches.

Electronic copy available at: https://ssrn.com/abstract=4207996

Introduction
Federal Reserve Board Chairman Jerome Powell recently stated that “collateralized loan

obligations are now the largest [nonbank] lenders, with about 62% of outstanding leveraged loans.” 1 To

put this into perspective, the leveraged loan market in 2019 was about $1.1 trillion and is used as a source

of funding by more than 70% of U.S. companies!2 Nevertheless, collateralized loan obligations, or CLOs,

which hold the majority of these loans in their investment portfolios, are not well understood by the

public. To further complicate matters, because CLOs are a form of structured finance, they are sometimes

confused with the structured products that were used to securitize the sub-prime real estate mortgages

that triggered the Global Financial Crisis of 2007-2008 (hereafter referred to as “the GFC”).

Our aim in this primer is to provide a first reading for anyone wanting to better understand CLOs

and to point out their distinctive features as a securitization vehicle that have allowed them not only to

weather the GFC but to grow and thrive. For those intent on learning more, there are a host of

technicalities related to accounting, tax, and legal issues that surround the CLO ecosystem that we leave

for more specialized treatments.

After providing a brief taxonomy of structured finance to help identify the unique attributes of a

wide range of structured finance products, we provide some basic facts about the life cycle of a CLO,

including how cash is distributed to investors, who invests in CLO securities, and how CLOs are regulated.

1 Powell, Jerome, “Business Debt and Our Dynamical Financial System,” 2019. Speech at “Mapping the
Financial Frontier: What Does the Next Decade Hold?”, 24th Annual Financial Markets Conference, sponsored by
the Federal Reserve Bank of Atlanta, Amelia Island, Florida. A leveraged loan is a commercial financing provided by
a group of creditors such as a syndication of banks. These loans generally consist of B rated, adjustable-rate,
revolving credit and/or term loans and are traded in the open market.

2 More recent data indicates that new issues of roughly $140 billion in CLOs during 2021 pushed total CLOs
to $1 trillion (https://www.bloomberg.com/news/articles/2021-05-28/structured-weekly-global-clo-outstandings-
approach-1-trillion). In addition, Standard & Poor’s (2020) reported that two-thirds of leveraged loan issuance since
the financial crisis was funded by CLOs.

Electronic copy available at: https://ssrn.com/abstract=4207996

We also reveal the “secret sauce” used by all structured finance products that allows these vehicles to

create AAA credits from portfolios of B-rated investments. Because this is perhaps the least understood

feature of structured finance, we provide a simple illustration of tranching to reveal how over-

collateralization works. The final section of the paper provides an overview of the recent academic

literature on CLOs that focuses on three main issues: (1) the difficulties faced by rating agencies when

evaluating the risks of CLO-issued securities; (2) data limitations encountered by investors wishing to

evaluate the historical performance of the securities CLOs issued; and (3) the potential risks posed by CLOs

to the stability of the financial system.

Background—Securitization, Structured Finance and CLOs

Beginning in the 1970s with mortgage lending and later in the 1990s with commercial bank

lending, financial institutions began a transformation of their business models from originating and

holding loans to origination and distribution (O&D). When in O&D mode, lenders continued originating

loans but instead of holding the loans until they were repaid, they sold them to another type of financial

institution that pooled them and issued claims against the loan pool in a process known as securitization.

One important effect of this innovation was to free up lender capital so that the originating financial

institutions could make more loans. Moreover, since many of the buyers of the securities created in the

securitization process were non-financial institutions, this disintermediation brought a new source of

capital into the commercial finance industry.3 Securitization created what amounted to a new market-

based source of liquidity to the banking system that is now commonly known as the “shadow banking

system.” Since this system provides banks with access to a market-based source of capital that is not

3 Interestingly,
the O&D banking model effectively makes depository banks look more like investment banks
that receive commissions in origination of securities or the sell-side.

Electronic copy available at: https://ssrn.com/abstract=4207996

under the control of bank regulators, it has been viewed by many with skepticism as a potential source of

trouble.4

Structured Finance and Corporate Debt

Many types of assets have been used as basis for securitization. However, the general class of

securitizations we want to consider here involve financial contracts, specifically corporate debt contracts.5

The umbrella term used for securitizations of corporate debt is “collateralized debt obligation,” or CDO.

Included under the umbrella are three types of financial entities that have engaged in the securitization

of corporate debt and made up what is commonly referred to as structured finance:

(i) We refer to the first as the “original CDO.” This version of the CDO securitized a variety

of corporate debt contracts but later included mortgages on commercial and residential

real estate as well as securities issued by other CDOs. 6 The remaining types of structured

finance products that securitize corporate debt are more specialized.

(ii) Collateralized Bond Obligations (CBOs) are securities created out of corporate bonds.

(iii) Collateralized Loan Obligations (CLOs) are securities created out of portfolios of corporate

loans. Following the GFC, CLOs became the dominant form of securitization for corporate

debt.7

4
Culp (2013) argues, as did Miller (1998), that the shadow banking system provides a beneficial source of
liquidity to what he characterizes as an inherently “fragile” banking system. The source of the counter argument
derives from the dilution of regulatory control over banks.
http://www.federalreserve.gov/newsevents/press/bcreg/20141107a.htm.
5 Almost every type of debt instrument you can imagine has been securitized at one time or another
including, among other things, student education loans, consumer credit card debt, residential mortgages, and
corporate bonds.
6 CDOs that are based on other primary CDOs may be called CDO-squared, -cubed, etc. These are

fundamentally derivative securities of primary CDO structures.

7 We later address differences in the design of CLOs from other types of CDOs that made the latter less

successful than CLOs.

Electronic copy available at: https://ssrn.com/abstract=4207996

It is important to recognize that all these CDOs and CLOs that are the focus of this paper are themselves

securities. However, they differ from simple debt contracts issued by corporate borrowers in that the

assets that underlie them consist of loans to multiple borrowers.

Collateralized Loan Obligations

The typical CLO is structured as a cash flow or “arbitrage” transaction in which the cash flows

generated by a portfolio of bank loans (interest and principal payments) are used to pay debt service to

the holders of a variety of securities issued by the CLO, commonly referred to as tranches. As shown in

Figure 1, a tranche is one of several related securities marketed to different buyers as part of the same

transaction.

Figure 1. Anatomy of a CLO

This arbitrage is successful if the portfolio of loans throws off a stream of cash flows sufficient to

satisfy the return expectations of the investors that hold the CLO’s debt tranches plus an acceptable

residual value for the equity tranche investors.

In addition to CLOS that repackage loans made by multiple banks, there are also balance sheet

CLOs, which are formed by a single lender as a funding source that securitizes only its own loans. This

Electronic copy available at: https://ssrn.com/abstract=4207996

type of CLO makes up a small percentage (only between 5 and 10%) of all CLOs and operates just like an

arbitrage CLO, with the issuer typically retaining the equity tranche.8

Figure 2. The Life Cycle of the Typical CLO

The CLO Life Cycle

Figure 2 describes the life cycle of a CLO in terms of four overlapping periods that span the six- to

ten-year life of the CLO:

▪ Warehouse period. During this period, which encompasses three to six months prior to the

closing of the CLO, the CLO manager acquires most of the bank loans that will make up the

collateral (assets) portfolio of the CLO.

▪ Ramp-up period. Once the CLO is closed, the manager completes the acquisition of the CLO loan

portfolio during this period.

8https://content.naic.org/sites/default/files/capital-markets-primer-collateralized-loan-obligations.pdf.

Electronic copy available at: https://ssrn.com/abstract=4207996

 ▪ Reinvestment period. Over the next three to five years, the CLO manager can reinvest principal

payments it receives from the loan portfolio. This includes proceeds from loan repayments, sales

of portfolio loans, and the recovery proceeds from loan defaults. The reinvestment process serves

to increase the duration of the CLO, which would otherwise be limited to the one to three year

term of the typical leveraged loans in the loan portfolio. In addition, during the first few quarters

of the reinvestment period, the CLO manager cannot change the terms of the deal and call or

redeem any of the outstanding debt tranches. This no-call period is typically one to two years.

▪ Amortization period. Once the reinvestment period ends, principal payments received by the CLO

(net of CLO administrative expenses and CLO management fees) are distributed to the CLO

tranches in order of their priority.9

Finally, there are three important dates that define the formation of the CLO: pricing, closing and

effective date. The CLO pricing date is the date on which the CLO securities are priced in connection with

the offering. Pricing is described as the spread to a benchmark interest rate. The closing date is the legal

start of the existence of the CLO, and the point at which investors fund the deal. After the closing date

the CLO manager must finish ramping up the CLO by an agreed-upon deadline, which is the date that the

CLO is allowed to go effective.

The Cashflow Waterfall

To illustrate the cash flow waterfall, consider the hypothetical CLO balance sheet found in Table

1.

9 Interest is paid (typically) each quarter to the tranche investors out of interest received from the loan
portfolio in accordance with the CLO indenture. We discuss this in the next section.

Electronic copy available at: https://ssrn.com/abstract=4207996

 Table 1. Example CLO Deal Structure

Panel a. Composition of the Collateral Portfolio

Asset Type Senior Secured Corporate Loans 93%

Second Lien Corporate Loans 7%
Number of Loans 201
Smallest Issue $ 300,000
Largest Issue 6,000,000
Average Moody's (S&P) Credit Rating B2 (B)

Panel b. Composition of the Liability Structure

Current Percent of
Spread to three- Percent of Balance (Par Original Capital
Tranche month LIBOR (bps) Capital Value) Ratings (Ranges)*
A 120 63% $ 315,000,000 AAA 60 - 65%
B 175 12% 60,000,000 AA 9 - 13%
C 235 6% 30,000,000 A 5 - 9%
D 335 5% 25,000,000 BBB 4 - 6%
E 620 5% 25,000,000 BB 4 - 6%
F 790 1% 5,000,000 B 4 - 6%
Equity 8% 40,000,000
Totals 100% $ 500,000,000
*Ranges of capital in different tranches that are typically used.

CLO assets, commonly referred to as the CLO collateral portfolio, consist of Ba1/BB+ or lower-

rated floating rate loans, also known as leveraged loans.10 In the hypothetical CLO described in the first

panel of Table 1, the collateral portfolio includes 201 loans, 93% of which are invested in senior secured

corporate loans and 7% are in second-lien loans.

10
CLOs can make very limited investments in some unrated securities and historically held some bonds. For
our purpose we will refer to the collateral portfolio as being comprised wholly of leveraged loans.

Electronic copy available at: https://ssrn.com/abstract=4207996

 To finance the loan portfolio, the CLO entity issues debt tranches that are listed in order of their

priority with respect to the receipt of cash flows. This priority structure determines the order of

preference used to distribute cash flow (referred to as the cash flow “waterfall”). The most senior debt

tranche is rated AAA (Tranche A in Table 1), and these investors get paid first followed by the lower-rated

tranches down to Tranche F which is rated single B. The riskiest tranche is known as the equity tranche.

From the perspective of an investor, a CLO is just a set of securities. The securities or tranches

appeal to a variety of investors, ranging from insurance companies to hedge funds. For example, investors

who are highly risk averse prefer the AAA-rated tranche, while risk-seeking investors are often willing to

invest in lower-rated tranches. And there are even investors willing to accept the risk of a residual

claimant to the cash flows generated by the portfolio of corporate loans. Those investors will acquire the

equity tranche and become the holders of the economic equity in the issuer. By stratifying the cash flow

rights and risks in this way, the issuer makes the CLO securities marketable to a broad set of investors.

The “equity tranche” is technically a subordinated note with a fixed maturity, but such

noteholders are like stockholders in some important ways. For example, their contract does not specify

a targeted rate of return, but notes that, as the residual claimant for CLO cash distributions, they are

entitled to receive the cash that is left over after all other tranches have received their required cash

distributions. In addition, the equity tranche investors possess some important control rights over the

CLO asset manager that are typically associated with an equity security, including the ability to refinance

the higher rated tranches of the CLO and to amend the indenture agreement to extend the life of the

CLO.11

11 If this sounds a bit confusing, it is. In fact, even CLO investors do not have a uniform opinion as to what
the equity tranche represents in their investment portfolios. An informal survey performed by Trivedi and Sageser
(2021) of equity tranche investors asked how they classified their CLO equity investments. The survey revealed that
79% considered it to be opportunistic credit or idiosyncratic investment like specialty finance, distressed investments
or esoteric asset backed securities, 14% placed the CLO equity tranche in a group of alternative investments including

Electronic copy available at: https://ssrn.com/abstract=4207996

 The debt tranches receive a variable rate of return that is specified in the indenture agreement

based on a fixed spread to SOFR.12 In the hypothetical CLO described in Panel b of Table 1, the spread

increases from 120 to 790 basis points for AAA- down to B-rated tranches. As we noted earlier the equity

tranche has no specified rate of return but receives any residual cash flows after paying the debt tranches.

The percentages displayed in the last column of Panel b in Table 1 reflect ranges seen in a typical

CLO. The specific elements of the agreement are negotiated between the asset manager and the investors

providing the capital for the CLO. Nevertheless, there are boundaries to these negotiations related to

achieving the desired credit rating for each CLO tranche. These boundaries reflect the recommendations

of rating agencies and can vary over time in response to changing market conditions. We will have more

to say about this process when we discuss CLO regulation.

The source of the quarterly cash distributions depends upon where the CLO is in its life cycle (see

Figure 2). During the reinvestment period, only interest earned by the collateral portfolio is distributed

to the CLO security holders, while any principal received during this period is reinvested in new loans to

maintain the collateral portfolio. Redemptions may be motivated by realization of the capital gains

accrued in the market value of the loan collateral. The interest owed to each tranche is paid to the AAA

tranche first, AA next, and so forth if interest income is sufficient to cover all promised interest. Interest

rates for the tranches are variable and are defined by using a spread to SOFR. If there is any quarterly

interest income left after paying the lowest rated debt tranche plus management and administrative fees,

investments in private equity, and another 7% thought it simply as a fixed income investment. Although these results
represent an informal survey of a small group of respondents, they highlight the divergence of opinions about what
CLO equity represents in an investor portfolio. Another possible explanation for the observed differences in
classification relates to regulatory considerations and accounting conventions that differ by type of investor.
12 SOFR is the Secured Overnight Financing Rate which began replacing LIBOR in CLO deals in January 2022.
SOFR is a broad measure of the cost of borrowing cash overnight collateralized by Treasury securities. Unlike LIBOR
quotations, SOFR is based on actual transactions which means it is a more accurate means of measuring the cost of
borrowing money. Since these transactions can be observed, it is believed that SOFR is less likely than LIBOR to be
manipulated.

Electronic copy available at: https://ssrn.com/abstract=4207996

the excess is distributed to the holders of the equity tranche. Once the reinvestment period ends,

principal payments received by the CLO are used to pay down the par value of the AAA tranche until it has

been repaid in full, then the AA tranche, and so forth.

Earlier when discussing the life cycle of the CLO, we referenced a six- to ten-year life. Although

technically correct, the owners of the equity (sub-note) tranche have the option to retire the debt

tranches early much like the issuers of a callable bond. Their decision calculus is driven by an assessment

of changes in the cost of refinancing the debt tranches. For example, if the cost of financing declines

because of a reduction in the credit spreads for the debt tranches, such tightening of spreads effectively

reduces the CLO’s cost of financing and increases the residual value that accrues to the equity tranche

investors.13

Who Invests in CLO Securities?

The first thing to note about investors in CLO securities is that not just anyone can buy them. To

own them the investor must be a qualified institutional buyer or QIB (pronounced “Quib”). In the U.S., a

QIB is a purchaser of securities that is deemed financially sophisticated and legally recognized by securities

market regulators as needing less protection from issuers than most public investors. Typically, a QIB is a

company that manages a minimum investment of $100 million in securities on a discretionary basis or is

a registered broker-dealer with at least a $10 million investment in non-affiliated securities.14 The idea is

that buyers of CLOs are very large, sophisticated investors who should not need the usual protections

extended to individual investors by regulators.

13 For example, CLO refinancing jumped to historical highs in the early months of 2021 as the market rallied,
credit spreads tightened, and the CLO managers acted to lower their cost of capital.
14 https://www.investopedia.com/terms/q/qib.asp.

10

Electronic copy available at: https://ssrn.com/abstract=4207996

 Table 2 lists the groups of investors that were invested in each of three major classes of CLO

securities—senior, mezzanine, and equity—at the end of 2018.15 For each type of investor, the data

includes the total dollar value of holdings (in millions), the fraction of the investor’s total investment

devoted to the tranche, and the percentage of the total value of all securities in the investor tranche held

by each investor classification.

Banks and other financial organizations were clearly the largest holders of senior securities, at

38%, while insurance companies were the biggest investors in mezzanine tranches. Both these

institutions face regulatory capital restrictions that make them sensitive to the default risk rating on their

investments, so their preference for the lowest risk category tranches is no surprise. Banks (and other

depository institutions) placed 95.4% of their CLO investments in the senior tranche, which represented

27.9% of the entire tranche. Insurance companies invested 49.1% of their total investment in the senior

tranche, which represents 26.5% of the total holdings of senior notes across all investor groups. Finally,

the largest investors in the equity tranche were mutual funds, accounting for 28.6%. Other investors in

the equity tranche include the “Fund or Other Investment Vehicles” category, which includes hedge funds

and private equity investors, with 27%, and “Nonfinancial Organizations,” with 20.3%.

15 We report only one year of data for illustrative purposes, but this data is reported annually by the Federal
Reserve.

11

Electronic copy available at: https://ssrn.com/abstract=4207996

Table 2. Domestic Holdings of U.S. CLO Securities by Investor Type and Tranche (12/2018)

Definitions of investor classifications:

Insurance Company: A financial institution which underwrites the risk of loss of, or damage to,
personal and business assets (general insurance) and life and limb (life and accident insurance).

Mutual Fund: Investment companies that are registered with the Securities and Exchange
Commission and regulated under the Investment Company Act of 1940.

Depository Institution: Any financial institution that accepts deposits. Examples include banks
(commercial, industrial, mutual and stock), credit unions, and savings and loan associations.

Other Financial Organizations (incl. BHCs): Businesses and institutions other than banks and
other depository institutions, that are primarily engaged in proprietary investments and/or in the
provision of financial services to other organizations and households.

Fund or Other Investment Vehicle: Hedge funds, private equity funds, and other investment
partnerships.

Pension Fund: Investment companies that engage in the management of retirement assets.

Nonfinancial Organizations (incl. household): Non-financial corporations and private citizens.

12

Electronic copy available at: https://ssrn.com/abstract=4207996

 How are CLOs Regulated?

Since CLOs issue securities, you might think they would face the same kinds of reporting and

disclosure requirements as companies that issue stock or bonds. However, this is not the case. As we just

discussed, only large institutional investors (QIBs) can own CLO securities. Since this group is more

sophisticated than ordinary investors, they are assumed to be capable of assessing the risks of investing

in CLOs and are not subject to the same type of oversight and supervision as stocks, bonds, and other

securities provided less sophisticated investors. Even so, CLOs are subject to both statutory constraints

and market disciplinary forces that work to constrain their structures as well as their day-to-day

operations:

Statutory regulations. The global financial crisis led to the implementation of two new regulations

on CLOs that were designed to make them less risky investments. First, in 2014, the Volcker Rule was

applied to CLOs, which had the effect of limiting the use of non-loan assets in CLO portfolios. Prior to the

imposition of this rule, CLOs regularly held 5 to 10% corporate bonds in their asset portfolios. Following

the application of the Volcker rule, many older vintage CLOs eliminated all non-loan assets from their

collateral such that compliance was no longer necessary.16

A second effort to reduce the riskiness of CLOs was made by applying Section 941 of the Dodd-

Frank Act 17 to impose a risk retention rule on CLOs requiring the CLO managers to retain at least 5%

ownership of the original value of the portfolio. The risk being addressed here related to balance sheet

CLOs enabling the loan originators to remove the loans originated from their balance sheets. The idea

was to better align CLO manager interests with those of the investors by forcing them to invest alongside

16 The Volcker Rule was later amended in 2020 to allow for bonds up to 10% in CLOs.
17 Section 941(b) of the Dodd-Frank Wall Street Reform and Consumer Protection Act authorized
regulation of the asset-backed securitization market.

13

Electronic copy available at: https://ssrn.com/abstract=4207996

of the outside investors. The rule remained in effect from December 2016 until May 2018, when the

courts dropped the requirement for those CLO sponsors whose investments involved securitizing only

“open market” purchases of loans they did not originate.18 This court ruling constituted a win for the Loan

Syndications & Trading Association (or LSTA), which maintained that an open-market CLO manager is not

a “sponsor” because it does not sell or transfer assets to the issuing entity negating the need for risk

retention.

Securities Laws. CLO Collateral managers are typically are required to register with the SEC as

investment advisers under the Investment Advisers Act of 1940. However, when CLOs issue their

securities, these are private transactions and, under the Securities Act of 1933 are exempt from

registration. Consequently, offering materials are neither filed with nor reviewed by the SEC.19

Rating Agencies. The very nature of the intermediation model underlying CLOs requires that they

meet rating agency requirements in the design and management of both their assets—that is, their

portfolios of leveraged loans—and their capital structures, or mixture of debt and equity tranches.

Specifically, rating agencies provide guidance to CLOs in the construction of their loan portfolios related

to the following elements:

1. Concentration limits. These limits constrain the ability of the CLO to invest in some types of

riskier investments that offer the promise of higher rates of return than leveraged loans. For

18 Open market purchases refer to loans acquired in either the primary (new issue) market or the secondary

market for loans. Open market purchases by their nature exclude CLO transactions involving loans originated by a
lender and then transferred to a balance sheet CLO it forms.
19 SEC Chairman Jay Clayton provided the following response to a question by Senator Elizabeth Warren

about the SEC’s role in monitoring CLOs, “Collateral managers of CLOs typically are required to register with the
SEC as investment advisers under the Investment Advisers Act of 1940. It is important to note that beyond that
requirement, the SEC has limited jurisdiction with respect to origination and distribution in CLO markets, and
generally speaking, even less authority with respect to the origination of the underlying leveraged loans.”
https://www.forbes.com/sites/mayrarodriguezvalladares/2019/05/06/the-sec-tells-senator-elizabeth-warren-that-
it-is-monitoring-clo-markets/?sh=1ab1e7702124.

14

Electronic copy available at: https://ssrn.com/abstract=4207996

 example, a mandate that at least 90% of the CLO investment portfolio be invested in senior

secured loans.20

2. Borrower diversification. There are several tests that the CLO asset portfolio must meet,

including maximum exposure to any one issuer or industry. This typically results in the loan

portfolios being diversified across 150-300 distinct borrowers operating in 20-30 industries.

Moreover, the proportion of the loan portfolio composed of a single borrower is typically

limited to less than 2% of the total portfolio.

3. Loan liquidity (minimum issue size). CLO contracts often restrict managers from purchasing

loans to smaller firms because of concerns about limited liquidity should the CLO manager

try to sell its position in the secondary market prior to the loan’s maturity.

The rating agencies also provide guidance to CLOs about the structure of their liabilities, such as

the relative sizes of the debt tranches as well as the minimum size of the equity tranche,21 and the

minimum level of over-collateralization. For example, if the CLO plans to issue debt tranches with a par

value of $500 million and the rating agency requires 20% over-collateralization of the debt, the CLO would

have to purchase a loan portfolio with a par value of $600 million to provide the needed level of over-

collateralization. Historically, however, the over-collateralization requirement has been less than 10%

over the total size of all the CLO tranches rated AAA thru BB.

In addition to the guidance that rating agencies provide CLO managers concerning the CLO

investment portfolio and the design of their capital structure, they also monitor the quality of CLO

collateral over its life using interest coverage (I/C) and over-capitalization (O/C) metrics. Data related to

20 Senior secured typically refers to loans that have a first priority lien on company assets.
21 Panel b. of Table 1 provided information related to the typical structure of CLO liabilities in each rating
classification.

15

Electronic copy available at: https://ssrn.com/abstract=4207996

interest coverage and over-collateralization metrics are available to the rating agencies on a regular basis

throughout the life of the CLO.22

Financial Markets. Although market forces are not generally seen as sources of “regulation” for

CLO structures, the CLO’s indenture contains provisions that reflect direct negotiations between the

investors and the CLO management that constrain the actions of the CLO manager. Specifically, the CLO

manager negotiates a variety of the terms in the CLO indenture directly with investors in the different

tranches. Examples include restrictions on the attributes of the loans and securities that will be included

in the collateral portfolio as well as limits for over-collateralization and interest coverage tests. These

negotiations are particularly important when seeking buyers for the equity tranche—the investors who

hold the residual rights to the CLO cash flows and hence face the greatest risk of loss.

Finally, CLOs are finite-life investment vehicles. This means that the managers of CLOs, like private

equity or hedge fund managers, continually face the pressure to perform so that they will have the

opportunity to raise their next CLO or fund.23 This design feature encourages CLO managers to be mindful

of any actions that could damage their reputational capital with current and potential future investors.

Who “Owns” the CLO?

We earlier pointed out that the “equity tranche” of the CLO is technically a subordinated note,

although holders of the equity tranche have some rights that are commonly associated with equity. For

example, unlike the debt tranches, the equity tranche is not rated and does not have a stated rate of

return. Moreover, the holders of the equity tranche have rights to the residual cash flows from the CLO

22 CLO investors can monitor the CLO’s performance using the CLO’s monthly trustee reports (except for
the current market value of the underlying CLO loan assets). Bloomberg reports prices for individual leveraged loans
which provides the information necessary to estimate the total value of the loan portfolio.
23 Technically, the life of the CLO can be extended through a refinancing of the CLO debt tranches. However,

this can only be done with the support of the owners of the equity tranche. Consequently, the CLO manager’s
reputation with the equity holders is critical to getting the extension done.

16

Electronic copy available at: https://ssrn.com/abstract=4207996

as well as some control rights normally associated with an equity security.24 Finally, the holders of the

equity tranche choose the collateral manager and make decisions about extending the life of the CLO and

refinancing one or more of the CLO’s debt tranches.

Nevertheless, there is another equity-like component in the capitalization of the CLO. It resides

in the Special Purpose Vehicle (SPV) that is created for the sole purpose of issuing the CLO securities and

holding the collateral portfolio. 25 To illustrate consider the capitalization of the SPV for the Apidos CLO

XXXI shown in Table 3.26 The capitalization of the SPV differs from the CLO capital structure example

presented earlier in Panel b. of Table 1 in one important respect. The CLO is issued using an SPV that

incorporates “Issuer Ordinary Shares” or common equity of $250 for the “co-issuers.” Although the co-

issuers own the common equity in the SPV, the “issuer shares” entitle the issuers to receive fees only for

services related to facilitating the issuance of the SPV tranches and other services rendered in support of

the operations of the CLO. Specifically, unlike corporate shareholders, the owners of the SPV co-issuer

shares have no claim to any of the CLO residual income; this right resides exclusively with the investors in

the equity tranche.27

It is important to note that CLOs are typically incorporated offshore to avoid double taxation.

Specifically, the CLO must not engage in a trade or business within the United States if it is to avoid

24 Moreover, for U. S. Federal income tax purposes the holders of the Subordinated Notes treat them as
equity. The note holder must generally choose either (1) to elect to treat the Issuer as a "qualified electing fund"
as a result of which such U.S. Subordinated Noteholder must include its pro rata share of the Issuer's ordinary income
and net capital gains on a current basis without regard to cash distributions or (2) to pay income taxes only when
cash distributions are actually received or gains realized upon disposition of equity, but subject to potentially
significant additional taxes.
25 The SPV is important because it provides legal isolation of the CLO assets. Legal isolation refers to the

independence of the special purpose vehicle used to issue the CLO securities and hold the collateral portfolio from
the issuer of the loans in the CLO collateral portfolio. This allows the risk rating of the CLO to depend solely on the
quality of the underlying assets of the CLO and to be independent of the issuer.
26 Information based on Offering Memorandum dated May 27, 2019. https://ise-prodnr-eu-west-1-data-

integration.s3-eu-west-1.amazonaws.com/legacy/ListingParticulars_44585372-5f3b-4eac-9d45-
b481b4e12de6.PDF
27 The specific rights and responsibilities of the co-issuers are laid out in the Issuers Memorandum of

Understanding.

17

Electronic copy available at: https://ssrn.com/abstract=4207996

becoming subject to U.S. corporate income taxes. And, this means that the SPV must have a domestic co-

issuer since some U.S. institutional investors cannot hold foreign securities. The domestic or on-shore co-

issuer is a U.S. domiciled corporation or limited liability company. It issues the CLOs debt securities along

with the offshore issuer. But, importantly for tax purposes, it does not have an ownership interest in any

of the CLOs assets.

Table 3. Capitalization of the Apidos CLO XXXI issue April 19, 2019

How do CLOs Differ from MBSs?

As we discussed earlier, structured finance products come in many forms. They differ first and

foremost in terms of the type of instruments (assets) that make up the collateral portfolio. But the

differences do not stop with the assets in the collateral base. There are other important structural and

operational differences among types of structured finance products.

CLOs are technically a form of collateralized debt obligation (CDO), as are the variety of

mortgage-backed securities (MBSs). CDOs based on nonconforming residential mortgages became

synonymous with the subprime mortgages that triggered the global financial crisis. However, there are

18

Electronic copy available at: https://ssrn.com/abstract=4207996

 several attributes of CLOs that distinguish them from their now infamous cousins. Table 4 highlights

some of the more important differences in the CDOs used to collateralize subprime real estate loans and

CLOs.28 Specifically,

1. Differences in the collateral pool. The collateral pool of a CLO comprises short-term senior

secured corporate loans that pay variable rates of interest based on LIBOR or SOFR.

Moreover, the CLO liabilities issued to finance the loan portfolio are also variable rate

securities. By contrast, the CDOs that triggered the GFC relied heavily on long-term real estate

mortgages with longer duration and fixed coupon rates. Moreover, many of the mortgages

were exclusively issued for purchase by government-sponsored entities such as Fannie Mae

and Freddie Mac, which had Congressionally imposed quotas. The imposed quotas created

the de facto impact of debasing underwriting standards such as originating residentials

mortgages with very little to no down payment and treating them as safe securities.

2. Transparency of the securitization. CLOs provide monthly reports detailing their loan

portfolios including purchases and sales for the period as well as current loan prices and

ratings. The quality of the collateral pool is monitored regularly by rating agencies and

investors using both over-collateralization and interest coverage tests. Failure to correct

breaches of indenture covenants can trigger purchase of additional collateral using interest

proceeds. CDO reporting typically was rarely, if ever linked to the asset quality of the

underlying portfolio, which can lead to potential problems in the long run especially

considering the longer duration of CDO assets based on mortgages.

28CDOs are the securitization vehicle featured in Michael Lewis’s book The Big Short. For a nice account of
the more fundamental problem with the $4.6 trillion of U.S.-originated non-traditional mortgages that Lewis largely
overlooked, see Don Chew’s review of Lewis’s book in this journal, “The Economic (Not Literary) Offenses of Michael
Lewis: The Case of The Big Short,” Journal of Applied Corporate Finance, Vol. 32. No. 4 (Winter 2020).

19

Electronic copy available at: https://ssrn.com/abstract=4207996

3. Management. The CLO asset manager is typically actively engaged in managing the contents

of the collateral assets (loans) not only during the building of the portfolio but afterwards

through the end of the reinvestment period. Although some CDOs had managed collateral,

most mortgage portfolios were static. Active management of collateral has the potential to

mitigate emerging risks as the managers can respond to changing market conditions and take

evasive action on behalf of CLO investors. Active trading of collateral may also help generate

return enhancement for the equity tranche.

20

Electronic copy available at: https://ssrn.com/abstract=4207996

 Table 4. Comparison of the Attributes of CLOs and CDOs

ATTRIBUTES COLLATERALIZED LOAN ASSET BACKED SECURITIES

OBLIGATIONS (CLOS) (ABS) COLLATERALIZED
DEBT OBLIGATIONS (CDOS)

UNDERLYING COLLATERAL Senior secured loans Mezzanine tranches,

ASSETS high grade ABS
tranches, subprime
mortgages

TRANSPARENCY Detailed monthly reporting Reporting generally did

including all not link to the
loans/purchases and sales, underlying assets
current ratings, and loan
prices

MANAGEMENT Actively managed by some Static and managed

of the largest US asset portfolio
management firms securitizations

INDUSTRY Low—CLOs are required to Highly correlated to

CONCENTRATION OF have a diversified portfolio housing prices
UNDERLYING ASSETS of loans spanning multiple
industries

HISTORICAL LOSSES ON Negligible High

DEBT TRANCHE
INVESTMENTS

4. Industry concentration of the collateral portfolio. CLOs are required to distribute their loan

portfolio across multiple industries whereas the CDOs that led to the financial crisis were

linked to the housing market alone.29

5. Historical losses. Historical losses to investors in the debt tranches of CLOs have been

negligible while the same cannot be said about CDOs that suffered substantial losses during

the global financial crisis. (Indeed, we still do not know what the losses on nontraditional

mortgages have been, though current estimates continue to run above 10%.)

29 One could argue that CDOs were not completely undiversified as they typically represented properties in
a variety of geographic locations.

21

Electronic copy available at: https://ssrn.com/abstract=4207996

The Alchemy of CLOs—Making AAAs out of Single Bs

CLOs are financed by securities they collateralize with a portfolio of below investment grade

leveraged loans. The alchemy of the CLO involves transforming the cash flows from a portfolio of below

investment grade loans into a mix of investment grade financial securities plus the residual claim.30 The

biggest component of the investment grade securities has a AAA rating.31 How do they do it? Technically,

they accomplish this in a three-step procedure that involves pooling, tranching, and over-collateralization.

Pooling and Tranching

Pooling refers to the process of forming a portfolio of leveraged loans from a wide variety of

companies operating in a diverse set of industries. The loan portfolio is assembled in an SPV, which

effectively removes these loans from the balance sheets of the lenders who originated the loans. The CLO

uses the cash flows received from the collateral portfolio to pay the investors that purchase the newly

issued securities it sells to finance the acquisition of this portfolio.

These securities that are created by the CLO comprise tranches defined by the priority of their

claims on the collateral portfolio’s cash flows. The most senior credit tranche has the first claim, the next

lower priority claim has the second claim, and so forth, ending with the equity tranche which only gets

paid if all other financial claims have been satisfied. 32 In this way, the tranche with the highest priority of

claim has the lowest risk and highest credit rating, and so forth down to the equity tranche, which is not

30 This segment relies on the discussion in Coval, Jurek, and Stafford (2008).
31 Dang et al. (2020) argue that CLOs create “privately produced safe debt” which is indeed safe up to the
point where some bad public news about the fundamental value of the backing of the debt prompts sophisticated
investors to acquire private information thereby creating adverse selection.
32 If the SPV issues claims were not prioritized and were simply fractional claims that were not prioritized,

the structure would be called a pass-through securitization. This structure is interesting since the expected risk of
portfolio loss would be the same as the average risk of loss on the underlying leveraged loans in the pool.
Consequently, the portfolio credit rating in a pass-through securitization would be the same as the average rating of
the securities in the loan pool and hence, there is no credit enhancement.

22

Electronic copy available at: https://ssrn.com/abstract=4207996

rated. Overcollateralization and differentiated cash flow rights via tranching provide the basis for

achieving AAA-rated securities using B-rated loans.

Making AAA Securities—A Simple Example

The challenge faced when structuring a CLO is to use the collateral and cash flows from a portfolio

of loans with an average B rating to create tranches of securities worthy of ratings ranging from AAA to

BB. The key to this transformation is overcollateralization. To illustrate the process, we consider a very

simple example in which the CLO has only two tranches of securities, a $100 million AAA debt tranche and

a $25 million equity tranche. To further simplify the example, we assume the CLO is formed at the

beginning of the year and liquidated in one year. The balance sheet for our example CLO is found in Panel

a. of Table 5.

Suppose we have a pool of one-year maturity B-rated loans with a one-year default probability of

2.5% and an interest rate of SOFR plus 2.75% or 3%, assuming SOFR is 25bps. With a sufficiently large

number of loans in the pool, we assume that 2.5% of the loans will default within one-year due to chance.

And let’s further assume that the recovery rate in the event of default is 70% on average.

For illustration purposes, we assume that to achieve a AAA rating on the senior tranche of the

CLO requires overcollateralization of 25%, or total collateral of 125%. Neglecting transaction costs, this

implies that our sample CLO has $125 million in capital raised using $25 million in equity tranche capital

and $100 million of debt from the sale of the AAA debt tranche that pays a floating rate equal to SOFR

plus 125 bps. In this case, the investors in the AAA tranche are promised an end-of-year cash flow

$101,500,000 –which amounts to a 1.5% return on their $100 million investment.

Panel b. of Table 5 describes the end-of-year expected cashflows to the entire loan portfolio, the

promised return to the AAA tranche holders and the residual earned by the equity tranche. The CLO

manager projects the rate of return on the $125 million loan portfolio to be 3%. However, 2.5% of the

23

Electronic copy available at: https://ssrn.com/abstract=4207996

loans will default and pay no interest for the year with a recovery rate of only 70 cents on the dollar. For

the 97.5% of the portfolio that doesn’t default, the CLO will recover full promised interest of 3% plus 100%

of the face value of the securities. Given the default risk and recovery rate assumptions, the expected

end-of-year cash flow from the $125 million portfolio is $127,718,750. Netting out the $101,500,000

commitment to the AAA tranche leaves $26,218,750 for the residual claimant equity tranche, which

represents a 4.875% return.

Panel c. of Table 5 contains the current values of the debt and equity tranches when we assume

a 1.5% investor required rate of return on the AAA tranche and a 4.875% rate of return on the equity

tranche. These valuations match up with the initial capital contributions of both the debt and equity

tranche investors by design.

If the loan portfolio performs better than expected, with lower defaults or higher recovery rates,

then the additional income would flow through to the equity tranche investors. Conversely, if the loans

perform worse than expected, realizing higher default rates or lower recovery rates, the equity tranche

must absorb the losses to protect the debt tranche. Panel d. of Table 5 illustrates the effect of both worse

and better than expected loan portfolio performance. In the worse than expected case, the AAA debt

tranche cash flow is the same, but the equity tranche investor suffers. In the better-than-expected

performance illustration, the equity tranche realizes the benefits of the improved loan portfolio cash flows

while the debt tranche payout is fixed.

24

Electronic copy available at: https://ssrn.com/abstract=4207996

Table 5. Hypothetical CLO

Panel a. The CLO Balance Sheet

25

Electronic copy available at: https://ssrn.com/abstract=4207996

 Improving the Risk-Return Profile of CLO Tranches

Thanks to overcollateralization, the par value of the CLO investment portfolio of leveraged loans

will be greater than that of the debt tranches issued by the CLO. Note that, in Table 5, the CLO had a

portfolio of leveraged loans that was 25% greater than its debt tranche. Typically, the rating agencies

provide guidance to CLOs about market expectations for the amount of overcollateralization required to

meet investors’ expectations.

The risk-return profile of CLO tranches can be improved either by reducing risk or increasing

returns. Risk reduction is accomplished by choosing a portfolio of loans with lower default risk than

corporate debt with similar ratings. Returns are generally increased by paying lower prices when acquiring

the collateral loans. Both actions are measures of the effectiveness and success of the collateral

managers.

From the investors’ perspective, CLOs can enhance their returns by offering larger yield spreads

than other fixed income securities with similar ratings. The wider spreads represent investors expected

compensation for the greater complexity, reduced regulatory oversight, and illiquidity of CLO debt

tranches. Moody’s Investor Service (2019) provides a comparison of representative spreads of CLO debt

tranches to the option-adjusted spreads of comparably rated high yield bonds.33 On August 5, 2019, the

spread to Libor for the AAA CLO tranche was 101 basis points higher than a comparable priced corporate

bond with the same rating, while the spread differential for a single B rated tranche was 604 basis points.

33 The corporate option-adjusted spreads (OAS) are based on LIBOR and represent the
measurement of the spread of corporate bond yields over the risk-free rate, which is then adjusted to
account for the embedded options on corporate bonds. This makes the optionable corporate bond yields
comparable to the returns on CLOs.

26

Electronic copy available at: https://ssrn.com/abstract=4207996

Performance of CLO Tranches

Have the returns earned on the various CLO tranches been better or worse than those earned on

similar risk investments? In a recent study of CLO performance, Cordell et al. (2020) began by analyzing

the returns earned on CLO asset (collateral) portfolios and found them to be comparable to the returns

on a broad-based index of leveraged loans. They interpret this result as an indication that CLO portfolio

managers do not exhibit skill in selecting better performing loans. At the same time, however, CLO debt

was found to outperform similarly rated and duration-matched corporate bonds. Annualized return

differences ranged from 0.70% for AAA and AA rated tranches to 1.90% for non-investment-grade

tranches. Moreover, after management fees, the risk-adjusted returns to the equity tranches were not

statistically different from zero, suggesting that equity tranche investors earned an equity rate of return

commensurate with the risks of the investment.

The returns to the senior tranches sold by CLOs exhibited extremely low risk to investors during

the global financial crisis. For example, a report from S&P citing data attributed to Moody’s stated that34

no AAA or AA rated CLO has lost principal so far, with only 0.1% of single-A tranches, 3% of BBB tranches
and 6% of all BB rated tranches taking losses, according to Moody’s. In comparison, 43% of
collateralized debt obligation (CDO) tranches [a form of MBS] originally rated AAA ended up taking
losses, a number that would jump to nearly 63% for tranches originally rated BB.

Moody’s Investor Service (2019) provides added insight into the default risk of CLO tranches.

Specifically, it contains a comparison of default and loss rates for CLO debt tranches and comparably rated

corporate bonds. In every comparison both the incidence of and the losses suffered in default was lower

for CLO debt than comparable bonds.

34
https://www.spglobal.com/marketintelligence/en/news-insights/latest-news-headlines/leveraged-loan-
news/those-700b-in-us-clos-who-holds-them-what-risk-they-pose.

27

Electronic copy available at: https://ssrn.com/abstract=4207996

 The comparison of losses in CLOs versus CDOs is telling because CLOs are often linked in the minds

of the investment community as well as the broader the public to the CDOs that invested in sub-prime

mortgages. The mortgage-backed securitizations suffered from several flaws, not the least of which was

the securitization of subprime mortgages, many with low down-payments whose probability of default

went up dramatically when real estate prices plummeted. CLOs experienced some challenges during the

crisis that led to revisions in the type of collateral assets they could hold and a reduction in the size of the

AAA tranches. Nevertheless, in the wake of the 2007-2008 financial crisis, CLOs continued to grow

approaching a trillion dollars and in fact providing the majority source of funding for leveraged lending in

the U.S. economy.

Open Questions about the Risk of CLOs

Two fundamental questions have been raised about the risks posed by CLOs to investors: the first

concerns the reliability of credit ratings provided by the major rating agencies, the second is about the

risks that CLOs might pose to the economy. Stated more directly, could CLOs be the trigger that sets off

the next financial crisis?

How Well do Rating Agencies Assess CLO Security Risk?

The growth of CLOs over the past two decades has been driven by their ability to use

overcollateralization and tranching to create investment grade financial claims using a below investment

grade loan portfolio. A key determinant of investor acceptance of CLOs relies on investors’ trust in the

credit ratings assigned to CLO debt tranches. Some recent studies have questioned whether the methods

used by the rating agencies properly account for default risk. Specifically, three questions have been

raised about the efficacy of the rating agencies to properly capture the risk of CLO investing: (1) Are CLO

28

Electronic copy available at: https://ssrn.com/abstract=4207996

collateral risk and tranche ratings disconnected? (2) What are appropriate default correlations for

structured products? and (3) Do CLO debt tranches face extremely high levels of systematic risk?

Possible Disconnect between CLO Collateral and Tranche Ratings

A 2020 study by John Griffin and Jordan Nickerson reported discrepancies between the credit

ratings of CLO tranches and collateral during the Covid-19 crisis. They point out that between March and

August 2020, S&P and Moody’s downgraded approximately 25% of the collateral loans fed into CLOs but

only 2% of tranche values, with the downgrades concentrated in junior tranches. From this they conclude

that CLOs are riskier than their current ratings would suggest. But Cordell et al., while agreeing that the

collateral pools became riskier over this period, also documented that CLO debt tranches were secured

by significantly more collateral—and were thus less leveraged—than before the 2007-2008 financial crisis

in response to post-crisis tightening of rating agencies’ criteria for structured products.

Improper Estimation of Default Correlations

In an earlier study, Griffin and Nickerson (2016) proposed several frameworks to estimate the

appropriate default correlations for structured products, each of which jointly considers the role of co-

movements in modeled risk characteristics and unmodeled systematic risk, or what they call “fragility.”

The agencies’ credit rating default correlation assumption for Collateralized Loan Obligations (CLOs) was

only 0.01 before the global financial crisis and increased to 0.03 after the crisis. In contrast, Griffin and

Nickerson’s joint consideration of observable risk factors and frailty leads to a substantially higher

estimate of 0.12. The authors argue that this higher estimate for default correlation translates into a

roughly 26% underestimation of CLO credit risk by the rating agencies.

In a separate study, Foley-Fisher et al (2020a) argue that privately produced safe debt, like the

AAA debt tranches of CLOs, can be expected to work so long as there is no incentive for investors to collect

private information about its fundamentals and all other parties involved in the securitization process

29

Electronic copy available at: https://ssrn.com/abstract=4207996

understand this. The idea is that these securities are senior, complicated, and opaque, making the

“information costs” of understanding them high. The high information costs associated with AAA CLO

tranches in turn makes their valuations insensitive to information, which has the benefit of providing

protection against adverse selection and so has the paradoxical effect of reinforcing their AAA rating.

This could likely change, however, in the event news casting doubt on the fundamental value of

the collateral came to light. In that case, these securities once believed to be safe would, as happened to

MBS during the global financial crisis, become information sensitive, making the entire class of assets

riskier. This is the source of fragility to the market for CLO securities posited by Griffin and Nickerson

(2016).

Ratings Reflect Default Risk Not Systematic Risk

Whereas the risk priced in financial markets tends to be systematic, or non-diversifiable, the rating

agencies focus on the risk of default. An early study of securitization (Coval et al. (2009)) argued that an

overlooked feature of the securitization process is its substitution of risks that are largely diversifiable

(variance in total cash flow) for risks that are highly systematic. Where the volume of leveraged loans in

1998 was approximately $240 billion, it had risen by the time of the GFC to more than $1.2 trillion, and

roughly 60% of these loans were held by CLOs. The default risk that matters to the CLO debt tranches is

not the risk of default to a single security, which is largely idiosyncratic, but the risk of adverse states of

the economy in which the entire portfolio of loans in the pool of assets defaults. This risk to the CLO

collateral pool is thus systematic.

Another way to look at this is that the ratings of rating agencies assess the default risk of a CLO

tranche like that of an individual security when in fact the default risk of the tranche relates to the default

risk of a pool of assets whose default risk is highly correlated with catastrophic event in the entire

economy. As a consequence, the securities produced by structured finance activities have far less chance

30

Electronic copy available at: https://ssrn.com/abstract=4207996

of surviving a severe economic downturn than traditional corporate securities of equal rating. Moreover,

because the default risk of senior tranches is concentrated in systematically adverse economic outcomes,

investors should demand far larger risk premia for holding structured claims than for holding comparably

rated corporate bonds. If correct, their argument suggests that tranches of CLO are overpriced relative

to the pricing of the individual securities contained in them.

But do CLOs in Fact Add Risk to the Economy?

CLOs are credited with helping support the rapid growth of a category of below investment grade

business loans that would otherwise not be available through the banking system. In fact, CLOs are the

principal institution in what is commonly referred to as “the shadow banking system.”35 This raises two

related questions: has the increase in funding to this market had a material effect on the quality of the

loans being funded; and to the extent the answer is yes, has this borrowing increased the risk of financial

crisis in the economy?

The growth in demand for leveraged loans in response to CLO formations since their origination

has been accompanied by reductions in their reliance on loan covenants. In 2018 it was estimated that

80% of leveraged loans were “covenant-lite” transactions, as compared to only 30% a decade earlier. Such

a weakening of lender protections combined with the lower credit quality of the issuers at least suggests

that the growth in CLO funding of bank loans could be increasing risk in financial markets. At the very

least, it has led regulators to express concern over the prospects of excessive leveraging in the economy

by less creditworthy companies.36

35 Memorandum to the members of the House Committee on Financial Services, “Emerging Threats to
Stability: Considering the Systemic Risk of Leveraged Lending”, June 4, 2019.
36 Governor Lael Brainard, “Assessing Financial Stability over the Cycle,” Peterson Institute for

International Economics, Washington, D.C. (December 7, 2018),

https://www.federalreserve.gov/newsevents/speech/brainard20181207a.htm.

31

Electronic copy available at: https://ssrn.com/abstract=4207996

 Two possible explanations have been offered for why securitization of leveraged loans might

result in decreased loan quality, and both are related to incentives that securitization provides to bankers

who originate new loans: (i) knowing that they are going to sell loans into securitization pools could induce

banks to extend credit to higher risk borrowers, and (ii) banks have a reduced incentive to monitor

securitized loans they no longer own. In a study addressing both these possibilities, Kara et al. (2015)

found that, at the time of issuance, banks did not select and securitize loans of lower credit quality, but

following securitization, the credit quality of securitized loans deteriorated more than loans in the control

group. The latter finding was construed as tentative evidence suggesting that poorer performance by

securitized loans could be linked to reduced bank monitoring incentives.

At a macro level, some economists express concern that the securitization of business loans by

the shadow banking system37 could lead to excessive credit being granted to less creditworthy borrowers,

in turn which would have a destabilizing effect on the economy. 38 In addition, Foley-Fisher et al. (2020b)

show that the largest U.S. life insurers have entered private debt markets focusing on commercial banking

to capture the illiquidity premium associated with CLO securities. As a consequence, the authors argue

that life insurers have become more vulnerable to an aggregate shock to the corporate sector.

But as both Nobel laureate Merton Miller (1998) and his former University of Chicago colleague

Chris Culp (2013) have argued, the shadow banking system effectively complements and generally works

to strengthen the stability of the banking system. Miller and Culp suggest that the impact of banking

crises on economic activity is actually thus limited rather than amplified through greater reliance by

market participants on non-banking financial products and structures. The functioning and success of the

37 Shadow banking system is a term for the collection of non-bank financial intermediaries that provide
services similar to traditional commercial banks but outside normal banking regulations. CLOs being a prime
example of such an intermediary.
38https://www.bloomberg.com/news/articles/2017-10-27/shadow-banking-gets-bad-rap-so-treasury-

wants-to-erase-the-term.

32

Electronic copy available at: https://ssrn.com/abstract=4207996

leveraged loan market both throughout and since the GFC seem consistent with this thesis. Although this

market initially experienced a boom followed by a bust, CLOs, unlike private subprime mortgages and

related asset backed securitizations, have experienced a spectacular recovery since the global financial

crisis.

Lessons Learned

Interest in CLOs has grown in recent years as CLOs have become the largest nonbank lender in

the U.S. Technically, CLOs connect the bank’s lending mechanism with financial markets and, in so doing,

enable banks to originate additional loans. The fact that this added source of financing to the banking

system lies outside the purview of the bank regulatory system has given rise to the notion, with typically

unfavorable connotations, of a shadow banking system. Some have expressed the concern that such

growth might contribute to a financial crisis like the GFC of 2007-2008 that was triggered by the

securitization of sub-prime mortgages. The most important contributor to this concern may well be the

general lack of information about and understanding of the securitization process. Thus, our objective in

writing this primer is to provide a first look for anyone wanting to better understand CLOs and their

distinctive features as a securitization vehicle that allowed them not only to weather the GFC but thrive

in the post-GFC world.

We offer several observations about loan securitization using CLOs that we think will make them

much more accessible to interested readers:

4. CLOs fulfil a demand, or they would not exist. Specifically, they offer a variety of

securities with different risk-return profiles that would otherwise impose

substantial due-diligence requirements on investors that most are either ill-

equipped or unwilling to undertake.

33

Electronic copy available at: https://ssrn.com/abstract=4207996

5. CLOs have a “secret sauce” that is not all that mysterious. Like most

securitizations, CLOs create new securities with a range of different risks using

pooling, tranching, and overcollateralization. Below investment grade floating

rate loans are converted into an array of financial claims spanning the default

rating spectrum, including AAA-rated floating rate securities down through an

unrated equity tranche. The highest rating securities are marketed to institutional

investors that require the AAA rating for their investments, and the riskier

tranches are marketed to investors, notably private equity and hedge funds, with

greater sophistication and risk tolerance.

6. CLOs are quite different from the CDOs that are based on portfolios of

nonconforming mortgage-backed securities that triggered the sub-prime

mortgage crisis. Specifically, there are three fundamental differences that

suggest that CLOs do not pose the same risks to investors and the economy.

a. First, the asset being securitized is different: CLOs securitize portfolios of

short-term, floating rate corporate loans made to borrowers from a

diverse set of industries. The CDOs that securitized sub-prime mortgages

held portfolios of long-term, sub-prime mortgages (often carrying fixed

rates of interest) based on a single industry, residential real-estate. Also,

unlike residential mortgages, leveraged loans have the scale that justifies

the scrutiny of institutional analysts. Residential mortgages, on the other

hand, get one set of eyes from the mortgage originator.

b. Second, CLOs are more transparent. CLOs provide monthly reports to

their investors detailing their loan portfolios, including purchases and

sales for the period as well as current loan prices and ratings. The quality

34

Electronic copy available at: https://ssrn.com/abstract=4207996

 of the collateral pool is monitored regularly by rating agencies and

investors using both overcollateralization and interest coverage tests.

Failures to correct breaches of indenture covenants can trigger purchase

of additional collateral using interest proceeds. Mortgage backed CDO

reporting was generally not linked to the underlying portfolio, which can

lead to potential problems especially considering the longer duration of

mortgages.

c. Finally, CLO asset managers actively manage the composition of the

collateral portfolio. The CLO asset manager is actively engaged in

managing the contents of the collateral assets (loans) not only during the

building of the portfolio, but afterwards through the end of the

reinvestment period. Although some mortgage backed CDOs had

managed collateral portfolios, most were static. Active management of

collateral has the potential to mitigate emerging risks as the managers

can react to changing market conditions and take evasive action on

behalf of CLO investors.

7. There is much that we do not know about CLOs. Two fundamental questions have

been raised about the risks posed by CLOs to investors: One is related to the

reliability of credit ratings provided by the major rating agencies. Because

investors in CLO tranches rely on the credit ratings, it is crucial to the health of

the market for CLOs that these ratings be dependable. Second, CLOs are

securitizations of “subprime” bank loans and, as such, they could pose a source

of risk to the economy. Although the collateral loan portfolio is diversified across

35

Electronic copy available at: https://ssrn.com/abstract=4207996

 industries, this form of diversification does not reduce economy-wide or systemic

risk.

36

Electronic copy available at: https://ssrn.com/abstract=4207996

References

Bates, Shiloh, 2020, An Introduction to CLO Equity, Flat Rock Global, (February),

https://flatrockglobal.com/wp-content/uploads/2020/04/An-Intro-to-CLO-Equity-February-2020-

2.pdf.

Cordell, Larry, Michael R. Roberts, and Michael Schwert, 2020, CLO Performance, Working Paper, Wharton

School, University of Pennsylvania.

Coval, Joshua, Jakub Jurek, and Erik Stafford, 2009, The Economics of Structured Finance, Journal of

Economic Perspectives, 23, 1, 3-25.

Culp, Christopher L., 2013, Syndicated Leveraged Loans During and After the Crisis and the Role of the

Shadow Banking System, Journal of Applied Corporate Finance, 25, 2, 63-85.

Dang, Tri Vi, Gary Gorton, and Bengt Holmström, 2019, “The Information View of Financial Crises,” Annual

Review of Financial Economics, 11, 1, 39-65.

Foley-Fisher, Nathan, Gary Gorton, and Stephane Verani, 2020a, “Adverse Selection Dynamics in Privately-

Produced Safe Debt Markets,” Working Paper, Revised October 2020.

Foley-Fisher, Nathan, Gary Gorton, and Stephane Verani, 2020b, “Capturing the Illiquidity Premium,

Woring Paper, February 2020.

Griffin, John M. and Jordan Nickerson (2017) Debt Correlations in the Wake of the Financial Crisis: What

are Appropriate Default Correlations for Structured Products? Journal of Financial Economics, 125,

451-474.

Harris, Robert S., Tim Jenkinson, and Steven N. Kaplan, 2014, Private Equity Performance: What do we

Know? Journal of Finance, 69 (5), 1851-1882.

37

Electronic copy available at: https://ssrn.com/abstract=4207996

Kaplan, Steven N. and Antoinette Schoar, 2005, Private Equity Performance: Returns, Persistence, and

Capital Flows, Journal of Finance, LX (4), 1791-1823.

Kara, Alper, David Marques-Ibanez, and Steven Ongena, 2015, Securitization and Credit Quality, Board of

Governors of the Federal Reserve System, International Finance Discussion Papers, Number 1148.

Liu, Emily and Tim Schmidt-Eisenlohr, 2019, "Who Owns U.S. CLO Securities?" FEDS Notes. Washington:

Board of Governors of the Federal Reserve System, July 19, 2019, https://doi.org/10.17016/2380-

7172.2423.

Moody’s Investor Service, 2019, Structured Finance: Impairment and loss rates of global CLOs: 1993-

2018. https://www.moodys.com/research/Structured-Finance-Impairment-and-loss-rates-of-global-

CLOs-1993--PBS_1164579

Standard & Poor’s, 2020, LCD’s Quarterly Leveraged Lending Review, 2Q,

https://www.spglobal.com/marketintelligence/en/news-insights/research/lcd-quarterly-review-q1-

2020-leveraged-loans-suffer-historic-distress-losses-comparing-the-coronavirus-crisis-to-2008.

Trivedi, Himani and Loren Sageser, 2021, CLO Equity: Where does it Fit in Investor Allocations?

https://documents.nuveen.com/Documents/Nuveen/Default.aspx?uniqueId=d2448128-f349-4fa9-

892d-64e405fbd808.

38

Electronic copy available at: https://ssrn.com/abstract=4207996