CLO Investing: How Collateralized Loan Obligations Work and Where the Risk Sits

Collateralized loan obligations went from an institutional niche to a retail product in a few years, mostly through ETFs. That makes it worth understanding what sits inside them: loans to leveraged, often private-equity-owned companies, packaged so the senior slices are very safe and the junior slices carry a lot of risk. The same structure can be a conservative cash alternative or a leveraged credit bet depending on which tranche you own.

How a CLO is built

A CLO is a special purpose company, usually incorporated offshore, that buys roughly 150 to 400 senior secured loans. It funds the purchase by issuing debt tranches rated AAA down to BB or B, plus unrated equity. A collateral manager selects the loans and can trade them during a reinvestment period, typically up to five years for new deals, after which principal repayments pay down the debt, starting with AAA.

An illustration of a simplified $500 million CLO:

S&P has noted that AAA tranches in many recent deals have 35% or more par subordination, compared with around 26% for a typical 2006 deal.

The tests that protect senior tranches

• Overcollateralization (OC) tests compare the par value of the loans with each debt class. If losses push a ratio below its trigger, cash that would have gone to junior tranches and equity is redirected to pay down senior debt.
• Interest coverage (IC) tests do the same for interest income.
• CCC limits. Loans rated CCC above a set share of the portfolio, commonly 7.5%, are counted at market value rather than par in the OC tests, which trips the tests sooner when credit weakens.

These features mean the senior tranches get paid down faster exactly when conditions deteriorate, at the expense of equity distributions.

How much loss it takes to hurt each tranche

A rough illustration using the structure above and ignoring excess interest and OC diversions, which add protection: AAA holders lose money only if portfolio losses exceed 38% of par. With 50% recovery on defaulted loans, that needs about 76% of the loans to default. With 30% recovery, it takes about 54%. Recovery assumptions matter a lot, and they have been moving the wrong way (see below). BlackRock has published a similar estimate of about 75% defaults at 50% recovery.

The equity is the other extreme. It is roughly 11 times levered in this example ($500 million of loans against $45 million of equity), so each 1% of portfolio par lost costs the equity about 11% of its value.

Equity returns: an illustration

Suppose the loans pay SOFR plus 3.4%, the debt costs SOFR plus 1.6% on average, and management and other fees take 0.4% of the portfolio. All numbers are illustrative, not current market levels.

• Loan spread income: 3.4% of $500m = $17m
• Debt spread cost: 1.6% of $455m = about $7.3m
• Fees: 0.4% of $500m = $2m
• Left for equity: about $7.7m, or roughly 17% on $45m, plus SOFR earned on the equity's share of the assets

Then subtract credit losses. If 2% of the portfolio defaults in a year and recovers 60%, the loss is 0.8% of $500m, or $4m, which takes almost nine points off the equity return. Loan prepayments and repricings also matter: when borrowers refinance at lower spreads, the income falls while the debt cost stays fixed until the CLO itself is refinanced or reset.

The track record

According to S&P Global Ratings and managers citing its data, no AAA-rated CLO tranche has defaulted since the market began. Of 4,322 ratings on pre-2010 "CLO 1.0" deals, 40 defaulted, 15 of which started as investment grade. S&P's 2023 study found no investment-grade default among its rated CLOs issued since 2010, which it attributes to more senior loan portfolios, shorter reinvestment periods, and more subordination.

That record comes with caveats. It covers a period with only a few severe recessions. A 2026 academic paper points out that a zero-default history does not prove zero risk, and that AAA coverage ratios fell sharply in early 2009. And CLO 2.0 tranches rated below investment grade have defaulted.

What changed in the loan market

The underlying loans have changed:

• Covenant-lite loans now make up most of the market, so lenders get fewer early warnings and less bargaining power.
• Liability management exercises, where a struggling borrower moves assets or reorders creditor priority out of court, have become common. PitchBook LCD reported a payment default rate of 0.87% for the Morningstar LSTA loan index in August 2026 but a 2.88% rate once distressed exchanges were included. Fitch's measure, which includes distressed exchanges, was 4.8% for the 12 months to August 2026.
• Recoveries are lower. Bloomberg reported in August 2026, citing S&P data, that first-lien lenders recovered less than 30 cents on the dollar in 23% of bankruptcies over the past three years, compared with 7% of cases from 2008 to 2022.

Lower recoveries matter more for junior tranches and equity than for AAA, but they erode every layer's cushion.

CLO ETFs in practice

CLO ETFs let individuals own diversified tranches with daily liquidity. Janus Henderson's AAA CLO ETF (JAAA), the largest, passed $30 billion in August 2026.

They were tested in April 2025, when tariff announcements rattled markets. On April 7, JAAA had its largest one-day outflow since launch, close to $600 million, and CLO ETFs as a group lost well over $1 billion that month. The underlying market absorbed the selling, partly because large holders redeemed in kind, and AAA CLO prices dipped only briefly. Lower-rated CLO ETFs fell more. The episode showed that ETF liquidity depends on the liquidity of the tranches underneath, which is good for AAA and weaker further down.

Who should own which tranche

• AAA: A floating-rate alternative to short-term bonds or cash, with a modest spread over SOFR. Price falls in a panic are usually brief. Fits conservative income portfolios.
• AA to BBB: More spread and more sensitivity to the credit cycle. Mark-to-market losses in a recession can be meaningful even if the tranche eventually pays in full.
• BB and B: Equity-like drawdowns are possible. Treat these as high-yield credit, sized accordingly. Compare with high-yield bonds.
• Equity: A levered bet on loan defaults, recoveries, and the manager. Closed-end funds that hold CLO equity often pay high distributions that include return of capital, and their net asset values can fall over time. Read the fund's distribution sources and NAV history before buying.

Checking a manager or fund

• Manager history: default and loss rates across past deals, performance through 2008-09 and 2020 if available, and how much of its own equity it holds. US risk retention rules no longer apply to most open-market CLO managers after a 2018 court decision, but EU and UK rules still require 5% retention for deals sold there.
• Portfolio: CCC share, concentration in sectors such as software and healthcare, average loan price, and exposure to borrowers that have done liability management deals.
• Structure: remaining reinvestment period, OC test cushions, and when the deal can be refinanced or reset.
• For ETFs: tranche ratings held, expense ratio, spread over SOFR, and how the fund behaved in April 2025.

Taxes

Coupons and ETF distributions are ordinary income, which favors holding CLO funds in IRAs or other tax-deferred accounts. CLO equity held directly is usually a passive foreign investment company interest for US taxpayers, which requires elections and extra filings.

For related areas, see our guides to structured credit, private credit, distressed debt, and alternative investments.

This guide is for informational purposes only and does not constitute investment, tax, or legal advice. CLOs and CLO funds can lose value, including principal, and lower tranches can be illiquid. Consult a qualified advisor before investing.