What Is a CLO? Following One Dollar of Loss Through Structured Credit

Almost everything confusing about collateralized loan obligations resolves once you stop asking what a CLO owns and start asking where the first dollar of loss goes. The ownership question has a short answer: corporate loans. The loss question has a long answer, and that long answer is the entire product.

This matters now because the structure has arrived in an ETF. The first CLO ETF launched in 2020. What was a private, institutional, minimum-of-millions market is now something an adviser can put in a client account in a single trade, which means a lot of people are meeting the structure for the first time through a ticker rather than through a prospectus.

Start with the L

The letters are worth taking in reverse. Obligation is a bond. Collateralized means something backs it. The interesting letter is the one in the middle.

Danielle Gilbert of Panagram, who spent eighteen years on the sell side in fixed income before joining the firm, walked through it on Behind the Ticker in the plainest terms anyone has used on the show. Start with the L, the loan. The loans in question are senior secured loans made to companies. Senior and secured are both doing work. Senior means these lenders get paid before other lenders. Secured means the loan is backed by the borrower’s assets, so if the company fails, this claim is the one attached to something you can sell.

The borrowers are mostly companies rated below investment grade. That sounds alarming until you look at what happens when one of them stops paying. Because the loan sits at the top of the borrower’s capital structure and is tied to real assets, recovery on a default is comparatively high, though recoveries have fallen below their long run average as capital structures have become more loan heavy. Default frequency is low as well. As Gilbert put it, "the default rates are actually quite low for senior secured loans."

There is a second feature of these loans that changes how the finished product behaves, and we will come back to it: they float. The interest rate on a leveraged loan resets periodically against a short-term benchmark rather than sitting fixed for ten years.

Now the structure

A CLO manager buys a few hundred of these loans. Call it a pool of a billion dollars. Then, instead of selling investors an equal slice of that pool, the manager sells slices of different seniority against it.

That is the whole trick, and Gilbert described it as taking a pool of contractual cash flows and creating different risk profiles from it. One pool, many products.

The slices are called tranches, and they are rated. At the top sits the AAA tranche, which is the largest, usually somewhere around two thirds of the deal. Below it, in descending order: AA, A, BBB, BB. At the bottom sits the equity tranche, which carries no rating at all.

Money flows through this stack in two directions, and getting the directions straight is the point of this entire article.

Interest and principal flow down from the top. Every quarter, the loans in the pool pay interest. That cash goes first to the AAA holders, in full. Whatever remains goes to AA, then A, then BBB, then BB. Whatever is left after every rated tranche has been paid belongs to the equity.

Losses travel up from the bottom. When a borrower in the pool defaults and the recovery falls short, that shortfall is absorbed by the equity tranche first. Only when the equity has been entirely consumed does the next loss reach BB. Only when BB is gone does it touch BBB.

So the equity holders receive whatever is left over, and they absorb whatever goes wrong. They are paid last and hit first. The AAA holders are paid first and hit last, and for a loss to reach them, every tranche beneath has to be wiped out completely.

How thick is the floor

This is where the abstract structure becomes a number you can actually reason about.

Gilbert quantified the cushion sitting beneath the rated tranches: there are "eight to 10 points of cushion" in the equity layer. That means eight to ten percent of the pool can be lost outright before any rated tranche takes a dollar of damage. And that is only the first layer of defence. A BB holder sits above the equity. A BBB holder sits above the equity and the BB. A AAA holder sits above everything.

The historical record follows the structure. Gilbert cited a hard figure for one of the riskier rated slices: "The cumulative default rate for CLO BBs is 47 basis points." Forty seven basis points is under one half of one percent, and that is for the lowest rated tranche in the structure, four steps below the top.

None of this makes a CLO tranche riskless, and the loss ordering does not eliminate loss. It relocates it, deliberately and by contract, to the people who signed up to receive it.

The 2008 question

Every conversation about this product eventually reaches the same place, so it is worth addressing directly rather than waiting for it.

CLO and CDO differ by one letter, and in 2008 the collateralized debt obligation was at the centre of the wreckage. The structures rhyme. Both pool debt and tranche it. The difference is what went into the pool.

Gilbert drew the line cleanly. "That was subprime collateral. It was consumer risk, just consumers that weren’t going to pay." The pool, in other words, was the problem, and the tranching simply distributed a loss that was always going to arrive.

The 2008 vehicles were built on residential mortgages made to borrowers who could not service them, and worse, on second-order instruments built out of other such vehicles. Consumer risk of that kind was correlated in a way the models did not capture. When house prices fell nationally, everything in the pool deteriorated together, and diversification that existed on paper did not exist in fact.

A CLO holds senior secured loans to several hundred operating companies across many industries. Those companies fail for their own separate reasons in ordinary times. The correlation is real but it is not the correlation of a single national housing market. That is a genuine structural difference, and it is the reason the two products performed very differently through 2008 and again through 2020.

Why it does not behave like a bond fund

Remember that the underlying loans float. This gives a CLO a personality that surprises people who file it mentally under fixed income.

When short-term rates rise, the coupons on the loans in the pool reset upward, and the cash flowing through the structure increases. A conventional bond fund does the opposite: its holdings pay a fixed coupon, so when rates rise, the market value of those bonds falls. Duration, the sensitivity to interest rates that dominates most bond portfolios, is close to absent here.

What replaces it is credit risk. You have traded exposure to the rate cycle for exposure to the corporate default cycle. In a rising rate environment that has been a comfortable trade. In a genuine recession, where defaults cluster and recoveries compress, it is the trade that gets tested.

Gilbert framed the appeal as a way to get floating rate exposure and capture the front end of the curve alongside a credit spread. That is an accurate description of what you are buying and, read carefully, an accurate description of what you are exposed to.

Reading a CLO ETF

The wrapper does not change any of the above. It changes access, liquidity and price transparency, and it leaves the structure underneath entirely intact. Four things are worth checking.

Which tranche. This is the single largest determinant of what you own. A fund that buys only AAA tranches and a fund that reaches down to BB are not variations on a theme, they are different asset classes wearing the same three letters in their category name. The yield difference between them is the compensation for sitting closer to the losses.

Whether the manager is buying or building. Some funds buy tranches issued by third party CLO managers, while others are affiliated with a manager and may hold their own deals, and each arrangement carries its own set of questions about selection and incentives.

What the underlying loans are. The pool is corporate credit, so industry concentration and average borrower quality matter. A pool weighted toward one sector inherits that sector’s cycle.

How it trades. CLO tranches are over the counter instruments. In calm markets that is invisible to you. In stressed markets the market for the underlying can thin out, and an ETF’s spread and its premium or discount to net asset value will tell you when that is happening. Watch them in the weeks you would rather not be watching anything.

The short version

A CLO is a pool of senior secured corporate loans, sliced into layers that are paid in order from the top and damaged in order from the bottom. Everything about the product follows from that ordering. The rating on a tranche is a statement about how much has to go wrong beneath you before anything reaches you, and the yield is the price of your position in that queue.

Ask which tranche, and you have asked the only question that determines what you own.

This is educational content and not investment advice. It is not a recommendation regarding any security. Investing involves risk, including possible loss of principal. Past performance does not guarantee future results.