Corporate rating downgrades continue to be felt within U.S. broadly syndicated loan (BSL) collateralized loan obligations (CLOs), but metrics were generally stable over the past month. The average 'CCC' basket (including 'B-' obligors with ratings on CreditWatch negative) nudged upward ever so slightly to 6.47% from 6.39% a month ago, and up more meaningfully from a post-pandemic low of 3.62% a year ago. CLO assets with ratings with a negative outlook, which we view as an important forward-looking indicator of potential corporate rating changes, was at 17.04%, up from 16.26% a month ago and less than 11.0% a year ago. Interestingly, exposure to loans from 'B-' rated companies, which has increased markedly over the past six years (see slide 24 in our most recent published quarterly BSL CLO slide deck, "SLIDES: U.S. BSL CLO And Leveraged Finance Update: Steep Rate Climb Erodes Credit Metrics Of Lower-Rated Firms," published May 15, 2023), has actually ticked downward for the second month in a row, to 29.45% from 31.09% in late April. The recent decline in 'B-' exposure across the index is mostly driven by downgrades into the 'CCC' category as well as a recent slowdown in downgrades to 'B-' from 'B' or above.

Given the increase in exposure to 'CCC' and nonperforming assets and the average loan price, unsurprisingly, the average BSL CLO junior overcollateralization (O/C) test cushion has declined to 4.11% from 4.63% a year ago, which, by historical standards, is still a high level of cushion. As usual, however, when speaking about average metrics, it's important to note that metrics for individual CLOs can vary significantly from the mean. On this note, four out of the 583 reinvesting BSL CLO transactions in the index this month are failing their most recent junior O/C tests, while another six transactions have less than 50 basis points (bps) of cushion (all pre-pandemic transactions). Meanwhile, just seven post-pandemic transactions from the index this month have less than 3% junior O/C cushion (minimum 2.3%).

Table 1

Table 2

Pre-Pandemic Transactions Closer To Breaching The 7.5% 'CCC' Threshold, Though The Difference With Post-Pandemic Transactions Is Decreasing

Over the past year, the average 'CCC' bucket amongst pre-pandemic BSL CLOs have increased to 6.9% (compared to 6.47% for the reinvesting BSL CLO universe as a whole) from 4.7%, while the average 'CCC' bucket amongst the post-pandemic transactions have increased to 6.1% from 2.7%. The difference in 'CCC' exposures between the two cohorts has decreased, from 2.0% a year ago to under 0.9% today. We think this is partly because the pre-pandemic transactions are more likely to experience more manager intervention to limit growth of their 'CCC' buckets, which, consequently, will impact the junior O/C test cushion. We note that the current average exposure to 'B-' with a negative outlook (which, under our methodology, we count as part of the BSL CLO 'CCC' bucket) were nearly identical across the two cohorts: 5.39% for pre-pandemic and 5.37% for post-pandemic.

Chart 1

Different Future O/C Cushions For Pre- And Post-Pandemic Transactions

As a result of the above, we believe different O/C cushions for pre- and post-pandemic transactions will diverge. Pre-pandemic had lower average junior O/C cushions a year ago; since then, they have declined by a larger margin (by 0.81%) when compared to the average decline across the post-pandemic transactions (just 0.25%).

Chart 2

Pre-Pandemic Transactions Experience Slight Par Loss, While Post-Pandemic Transactions Experience Slight Par Gain

Over the past year, the average par balance (not including any haircuts) of pre-pandemic transactions have declined slightly to an average of 99.24% of target par from 99.41% of target par, while post-pandemic transactions have increased slightly to an average of 100.36% from 100.27%. As pre-pandemic transactions had lower credit quality, they were more likely to experience par loss due to higher default exposure (equity received from exchange do not receive par credit in CLOs) and, we believe, higher likelihood of de-risking activity from the manager, both resulting in a higher likelihood of par loss.

Chart 3

Par Gain From Below-Par Purchases?

In table 3, we focus on the transactions that experienced the largest changes in par balance over the past year (top and bottom 10th percentile by change in par across index). The transactions that built the most par over the past year were more likely to experience an increase in junior O/C cushion (as expected, the top 10th percentile of par gainers experienced a 20 bps average increase in junior O/C cushion while the overall average change across the index saw a decline of 52 bps). However, these transactions saw more credit deterioration (larger increases in S&P Global Ratings' weighted average rating factor [SPWARF]), while the change in the weighted average price of the portfolios saw below-average increases. Meanwhile, the transactions that lost the most par saw large drops in junior O/C cushion; although, they experienced less credit deterioration in their portfolio (smaller increase in SPWARF) and a larger improvement in the liquidity of the portfolio (larger increase in weighted average loan price), highlighting the different approaches CLO managers have in preparing for the future.

Table 3