Upgrades and downgrades on leveraged loan issuers have been relatively balanced in recent months, but a handful of U.S. broadly syndicated loan (BSL) collateralized loan obligation (CLO) obligors--several of them widely held--have seen their ratings lowered into the 'CCC' category (see table 2 below). Many BSL CLOs have experienced an uptick in 'CCC' buckets, and our CLO index has experienced an increase to 6.2% on average, a level not seen since May 2024 (see table 1 below). Currently, about 22% of CLOs in our BSL CLO index have breached the 7.5% 'CCC' threshold. Additionally, several issuers have also recently experienced a downgrade to 'CC', 'SD', or 'D', causing CLO default exposures to rise to a recent high as well.

Given that weighted average loan prices have ticked downwards in recent months to levels not seen since November 2023, some CLOs may see pressure on their overcollateralization (O/C) test cushions in the near future. Currently, none of the CLOs in our index of seasoned reinvesting transactions are failing one or more O/C tests, but 1.53% have less than 1% cushion. Across our rated amortizing CLOs, 13% are currently failing one or more O/C tests, while 39% have less than 1% cushion.

Despite all of the talk around tariffs and other disruption, the average proportion of CLO loans from obligors with a negative rating outlook remains moderate at about 13%, while exposure to 'B-' rated obligors with a negative outlook has declined to 2.39%, potentially a good omen for 'CCC' basket sizes in the future.

Table 1

Table 2

U.S. CLO 2.0 Defaults And Recoveries

CLO tranche defaults remain rare, with 25 CLO 2.0 tranches out of about 17,900 rated defaulting, a record that almost every other class would be envious of (or should be if they're not). All of these were junior CLO notes rated in the 'BB' or 'B' categories at close, and all but one of them (24 out of the 25) are from the 2014 and prior vintage CLO transactions. These defaulting CLO tranches came from early vintage CLO 2.0 transactions that had high exposure to energy-, commodities-, and retail-related issuers before ending their reinvestment periods amidst the pandemic-related economic downturn in 2020, limiting the managers' ability to course correct. One of the 2014 vintage CLOs was reset in 2018, with a two-year reinvestment period ending right before the peak of COVID-19.

Nearly all defaulting CLO 2.0 tranches have a few things in common, something we highlighted several years back when there were 11 defaulting CLO 2.0 tranches (see "The Dirty (Almost) Dozen: What Separates Defaulting U.S. CLO 2.0 Tranches From The Rest," published July 7, 2022). First, all of the CLO 2.0 defaults occurred after the peak of the pandemic, with the first occurring in mid-2021. All were junior tranches that had experienced multiple downgrades as rising 'CCC' baskets and declining O/C test cushions signaled stress well ahead of the default. Given junior CLO notes are deferrable, we generally don't lower our rating to 'D' until the default has occurred after the clean-up call (i.e., liquidation of the portfolio assets and final distributions have been made resulting in a principal shortfall to one or more rated notes). In some cases where the CLO notes were significantly (and in our view, permanently) undercollateralized, we may lower our tranche rating to 'D' ahead of the clean-up call after we get confirmation from the manager that the liquidation proceeds are unlikely to pay off the junior CLO notes in full.

More recently, we did an exercise to estimate the recovery proceeds for these defaulting CLO 2.0 tranches. To calculate our estimated recovery value, we used the outstanding tranche balance of the notes reported on the latest trustee report available to us (as a percentage of the original issuance amount). We find across the full list of 25 defaults, the average recovery (100% minus outstanding balance as of latest trustee reported balance) was about 38%; average of 45% across the 13 CLO notes originally rated within the 'BB' category and an average of 31% across the 12 CLO notes originally rated within the 'B' category.

Table 3

In cases where there is still residual collateral reported in the latest trustee report available to us, we assume the reported market value of those assets will be used to pay down the outstanding balance of the defaulted CLO note to estimate a post-liquidation recovery (with zero credit given to equity positions held). With this assumption, the average recovery across the 25 defaults was 43%; across the 13 CLO notes originally rated within the 'BB' category, the average was 54%; and across the 12 CLO notes originally rated within the 'B' category, the average was 31%. There is wide variation in the recovery estimates, ranging from as high as 98.5% (amendment made prior to final distribution date to allow principal cash to pay fees after paying the note down to 98.5% on the final payment date) to less than 0% (outstanding balance as of latest trustee report was greater than the issuance amount due to interest deferrals on the junior CLO note as a result of coverage test cures). A majority of the tranches that defaulted with an outstanding balance of over 100% were junior CLO tranches originally rated within the 'B' category.