- Between 2017 and the start of 2020, the number of outstanding credit estimates was on the rise.
- A significant portion of the assigned credit estimates between 2017 and 2020 are from the information technology and healthcare-related sectors.
- Due to the impact of COVID-19 pandemic, we reviewed issuers operating in consumer-facing sectors, as well as issuers that experienced some form of credit event and issued "specified amendments." Overall, more than half of the credit estimate issuers held within middle-market collateralized loan obligations were reviewed between March and September of 2020.
- The pandemic-related reviews resulted in a significant uptick in non-performing ('cc', 'sd', and 'd') and 'ccc' category ('ccc+', 'ccc', and 'ccc-') credit estimates assigned in 2020.
Before the onset of the COVID-19 pandemic, favorable credit conditions, along with prolonged periods of low interest rates in the U.S., steered investors towards a range of alternative asset classes in pursuit of higher yields and returns. One asset class that attracted greater investor interest during this time were collateralized loan obligations backed by middle-market loans (MM CLOs), as reflected by the continued increase in credit estimates issued for middle-market entities since 2017 (see chart 1). Unsurprisingly, there was a decline in the number of newly issued credit estimates during the second quarter of this year as the leveraged loan and CLO market took a pause due to COVID-19 before issuance picked up again. As of today, there is a robust pipeline of new credit estimates in the works as a number of MM CLOs aim to price and/or close by year end.
Before the pandemic, middle-market loan issuers, like their broadly syndicated loan (BSL) issuer counterparts, experienced low levels of default (see "Middle Market Loan Performance: A Decade In Review," published Sept. 11, 2017). Unlike their broadly syndicated counterparts, middle-market issuers, often smaller in scale and scope, tend to be unrated. They are also privately owned but the loans they issue offer richer spreads. Recent surveys have also shown middle-market loans, especially those at the lower end of the middle market, tend to have stronger covenants and documentation than their BSL loan counterparts.
Growth Of MM CLO Credit Estimates
In the years following the Great Financial Crisis, the demand for middle-market loans and MM CLOs remained stable (between 2012 and 2016, the average count of middle-market entities for whom S&P Global Ratings had an outstanding credit estimate was around 650), with the traditional MM CLO issuers printing a few deals per year. Starting in 2017, however, the increase in both the CLO investor base along with the thirst for yield spawned an increase in demand amongst investors for MM CLOs. Based on S&P's Leveraged Commentary & Data (LCD), S&P Global Ratings rated six U.S. MM CLOs in 2016; this number more than quadrupled in 2019 when we rated 27 U.S. deals. Consequently, the number of loans for which S&P Global Ratings provided credit estimate scores increased steadily from 2017 through early 2020.
The pace of first-time credit estimates (i.e., entities whose loans appear within MM CLOs for the first time) also increased after 2017 (see chart 2). This was due to a combination of the low interest rate environment that saw many middle-market companies tap CLO financing for the first time, and an uptick in the number of MM CLOs that were issued among existing and new MM CLO issuers.
Sector breakdown of credit estimates
Our CLO rating analysis (for both MM CLOs and BSL CLOs) uses the sectors outlined in the Global Industry Classification System (GICS) to assess sector diversity within collateral pools. Chart 3 below, which tracks the changes in composition of MM CLOs pools since 2012 using somewhat broader industry categories for purposes of readability, shows the following trends:
- Like their counterparts in the broadly syndicated universe, software/tech/IT- and healthcare-related sectors have seen an increased representation among middle market CLOs. At the start of 2012, IT-related issuers only made up about 6.3% of the credit estimates outstanding then; by the end of third-quarter 2020, this has increased nearly threefold to 18.7%.
- As of end of third-quarter 2020, industrials-related issuers represent the largest sector amongst the outstanding credit estimates, followed by IT, healthcare, and then consumer discretionary.
- Consumer discretionary continues to make up a significant proportion of the credit estimates, though it has declined to about 18.3% from 26.0% as of the start of 2012.
- Other sectors that saw a decline in outstanding credit estimates over the years include telecom, which, as of the end of third-quarter 2020, represented 1% of the outstanding credit estimates, down from 7.2% at the start of 2012.
MM CLO Credit Estimate Credit Quality Over The Years
Many of the companies we assign credit estimates to are financial sponsor owned and highly levered. On account of their weaker business and financial risk profiles, these companies tend to have credit estimate scores at the lower end of the credit spectrum. Between 2017 and 2019, over 75% of the credit estimates outstanding (as defined in Chart 1) have a score of 'b-' (see chart 4). This contrasts with BSL CLO collateral pools, for which loans from 'B-' rated obligors comprised roughly 20% of total assets prior to COVID-19, and about 24.6% currently.
The credit profile of MM CLO collateral pools with greater exposure to lower rated (or credit estimated) obligors tends to produce higher levels of subordination than that of typical BSL CLOs. The MM CLO 'AAA' tranche, for example, might have par subordination of 40% to 45%, compared to about 35% par subordination for a typical BSL CLO 'AAA' tranche.
Review of credit estimates for MM CLOs
Credit estimates are a point-in-time score and are not monitored regularly in the conventional sense the way a corporate rating would be. In order to maintain an outstanding credit estimate, CLO documents generally require the CLO manager to furnish financial information on the loan-issuing company to S&P Global Ratings at least once every 12 months so we can update the credit estimate to reflect current conditions and operating performance. Sometimes, loans from the same issuer are held across CLOs issued by multiple MM CLO managers. As of third-quarter 2020, of the roughly 1,100-plus credit estimated companies held within S&P Global Ratings-rated MM CLOs, about one-third have loans held within more than one MM CLO manager, and some issuers have loans held across CLOs from as many five or more managers. Since CLOs from different managers are issued on different schedules, various managers will request credit estimates at different intervals in order to maintain their own credit estimate scores, and S&P Global Ratings often ends up reviewing these entities more than once a year.
Further, we may also review credit estimates at other times, as we deem appropriate. MM CLO transaction documents typically contain provisions that require CLO managers to notify S&P Global Ratings of "specified amendments" such as non-payment of interest or principal, changes in payment terms (e.g., addition of payment-in-kind terms (PIK), changes in maturity dates, and changes in coupon rates); breach of covenants; restructuring of debt; pushing back of maturities, among others. Even absent this, we may also review credit estimates if we think a given entity might have seen an adverse change in its credit profile.
Credit estimates take a hit from COVID-19
Along with BSL loan issuers, earnings and revenue of MM loan issuers were also impacted by the COVID-19 downturn. Within the second and third quarters of 2020, we received and reviewed multiple specified amendment notifications relating to credit estimated issuers from different CLO managers. Additionally, we reviewed credit estimates from issuers operating within consumer facing sectors whose performance we viewed as particularly vulnerable to pandemic containment measures. For these entities, we considered their liquidity, leverage, and creditworthiness, as well as their upcoming financial obligations over the next 12 months. Generally, we took the following actions on the affected issuers:
- In the absence of specific refinance plans within the next six or 12 months, we lowered a number of credit estimate scores to 'ccc' or 'ccc-'.
- Where we received specified amendments that caused the investors to receive less than the promise of original security (conversion of cash to cash and PIK, push back of maturity, rescheduling amortization payment, etc.), we lowered the credit estimate score to an 'sd' and subsequently refreshed our score based on the revised loan terms and company financials.
- Where we received notifications of a bankruptcy filing or payment default with no forbearance agreement, we lowered the credit estimate to 'd'.
|Credit Estimate Transition During COVID-19 (March 1, 2020 Vs. Sept. 30, 2020)|
|Credit estimate as of Sept. 30|
|Credit estimate as of March 1, 2020||Proportion of outstanding credit estimates as of March 1, 2020||bb+||bb-||b+||b||b-||ccc+||ccc||ccc-||cc/sd/d||NR|
Of the roughly 1,100 credit estimates held across MM CLOs at the start of COVID-19, over 50% were reviewed between March and September. At the end of the first quarter, a total of 12.5% of the entities had credit estimates in the 'ccc' range, but on account of all the COVID-associated actions, the number went up to 19.2%. At the end of the third quarter, a total of 92% of the entities had a credit estimate score of 'b-' or below. The corresponding number for BSLs, which have a rating of 'b-' or below at the end of the third quarter, is about 40%.
Liquidity and selective defaults during COVID-19:
Liquidity was a major issue for several smaller companies in the MM pool on account of the shut downs as a response to the pandemic. While there were instances of sponsor infusion to address liquidity needs, MM managers also took other measures to preserve liquidity during the advent of the pandemic. Some examples of liquidity preservation included reduction of interest payments, swapping cash interest payments for PIK payments, providing for interest deferrals, reduction of scheduled amortizations, and maturity date extensions. In S&P Global Ratings' view, some of these measures resulted in investors getting less than the promise of original securities without adequate and offsetting compensation and hence constituted a selective default. Accordingly, we lowered the credit estimate score to 'sd' and subsequently refreshed our score based on the revised terms of payment.
In all, about 7.3% of the credit estimates outstanding before COVID-19 (March 2020) were lowered to a non-performing credit estimate score ('cc', 'sd', or 'd') between March and September; of which, 5.5%, were lowered to 'sd' for reasons cited above, and subsequently, had the credit estimate score refreshed based on the revised terms (see chart 6).
One-year lagging non-performing rate increased sharply if we include the selective defaults:
After gradually declining for the past several years to a relatively low 1% at the start of 2020, the one-year lagging non-performing rate for credit estimates increased sharply to about 8% by September 2020 largely due to selective defaults recorded from the notices of specified amendments (see chart 7). If we exclude selective defaults and focused on the conventional defaults, the one-year lagging default rate amongst our sample of outstanding credit estimates increased to about 2.5% in July.
As we continue to review credit estimates, we see the pace of credit deterioration has slowed in the second half of 2020, mirroring what we see within the BSL market. We will, however, continue to review estimates when we get notification of material events via specified amendments. Credit estimates will also be reviewed when they come up for updates or annual review to see if they reflect current credit conditions.
While Credit Estimates Took A Turn For The Worse During The Pandemic, MM CLO Portfolios Continue To Show Some Resilience
MM entities showed resilience after being hit hard by the pandemic. Relative to BSL CLOs, MM CLOs' liability structures typically close with higher tranche subordination, and the cushions on their junior overcollateralization (OC) ratio tests were no exception. As of March 2020, across a sample of 71 S&P Global Ratings-rated reinvesting MM CLOs, the average junior OC cushion was 5.2% (as opposed to 3.8% for the CLO Insights 2020 Index of 410 U.S. BSL CLOs). Partially due to the changes in credit estimates discussed above, by September, the average junior OC tests for these MM CLOs have declined to 3.7%, with only a handful of transactions failing their junior tests (as opposed to about 25% of the CLO Insights 2020 Index failing at some point between the second and third quarter). Additionally, the par balance of MM CLO portfolios reinvesting through COVID have remained relatively stable during this time.
In short, following our review of credit estimate scores outlined above, we find that:
- The pandemic-related economic shutdown and downturn had a significant impact on the credit estimates we have assigned to the loan issuers within MM CLOs.
- It also led to a significant number of payment defaults and, in particular, selective defaults. However, relative to obligors within BSL CLO transactions, MM CLO obligors have held up relatively well.
- Additionally, the ratings assigned to our MM CLO transactions have been fairly resilient so far; out of the 464 downgrades taken on U.S. CLO ratings in the third quarter due to COVID, only seven tranches were downgraded from MM CLOs (the rest were downgrades to ratings from US BSL CLO ratings; see "S&P Global Ratings COVID-19 Related Actions On U.S. CLO Ratings", published Sept. 17, 2020).
- As with BSL CLOs, future MM CLO rating transitions will be dependent upon the performance of the companies that issue the loans in the collateral pools, including the potential credit impact of a second wave of COVID-19 cases.
Appendix I: Annual Credit Estimate Transition (2008-2019)
Within our sample of outstanding credit estimates, from 2008 through 2019 (pre-COVID), on average, we noted that about 70% of credit estimates outstanding at the start of the year are refreshed by the end of the calendar year. Of these, half are assigned the same credit estimate while more than two-thirds are assigned a new credit estimate within the same category. When cross-category changes do occur, they are more likely to be assigned a lower credit estimate. Moreover, the loans with an estimate at the lower end of the credit spectrum had a higher penchant to default (the lower the credit estimate, the higher the percentage of nonperforming transitions), as expected (see Appendix I).
Several entities have their estimates set to not refreshed ('nr') within our sample because no update was received within the year. This could be partly because the loan was removed from the CLO portfolio, thus no longer needed a credit estimate, or because, in some instances, it was paid down. It could also be partly due the fact that there is less incentive for a manager to get an estimate refreshed for an issuer that has experienced significant credit deterioration.
|One-Year Credit Estimate Transitions (2008-2019)|
|Count at start of year (no.)||IG (%)||bb (%)||b (%)||ccc (%)||cc/sd/d (%)||nr (%)|
Transitions: ratings and credit estimates
S&P Global Ratings calculates transition rates in our default studies based on groupings called static pools, where we include all issuers with an active rating as of a given date (typically January 1 of a given year) within a static pool. Transitions are determined based on a comparison of the issuer's rating at the beginning of the static pool period, with the rating at the end of the period. Entities that have had ratings withdrawn--that is, revised to not rated ('NR')--are surveilled with the aim of capturing a potential default. If an entity has its rating withdrawn after the start date of a particular static pool and subsequently defaults, we will include it in that static pool as a default and categorize it in the rating category it was a member of at that time.
In the above tables, we follow a similar approach for our sample of outstanding credit estimates. For each annual transition summary, we start with a static pool of outstanding credit estimates at the start of the year and compare with the credit estimates outstanding at the end of the year. If the credit estimate has been refreshed at some point within the year, the latest estimate will be captured for the credit estimate value at the end of the year. If the credit estimate has not been refreshed within the year, the transition will be to 'nr' at the end of the year. In addition to the financials received as part of a request for a refreshed credit estimate, we rely on notification of "specified amendments" as required by the CLO indentures. We also rely on trustee reports to determine if a default has occurred.
Appendix II: Assumptions To Calculate Default And Transition Data
After a credit estimate is assigned, we assume it remains valid for one year.
Given that, within our sample between 2008 and 2019, on average, 70% of the credit estimates outstanding at the start of the year, were "refreshed" at least once during the year. If we treat these new estimates as an "affirmation" or "action" versus the prior credit estimate, we can create transition and default statistics.
The credit estimates within our sample that do not get "refreshed" within the year, we treated as a "transition" to 'nr', or a "withdrawal" action at the end of the year (we did not treat these expired credit estimates as a transition to 'CCC-', even though that is what our CLO criteria assume). We note that this credit estimate withdrawal assumption is very different from the transition of a surveilled S&P Global Ratings credit rating to 'NR', which is generally due to a paydown or at the issuer's request.
Even if we are not requested to renew a credit estimate, our practice is to keep track of (selective) defaults when we are alerted of them.
This report does not constitute a rating action.
|Primary Credit Analysts:||Daniel Hu, FRM, New York + 1 (212) 438 2206;|
|Ramki Muthukrishnan, New York + 1 (212) 438 1384;|
|CLO Sector Lead:||Stephen A Anderberg, New York + (212) 438-8991;|
|Ratings Performance Analytics:||Evan M Gunter, New York + 1 (212) 438 6412;|
No content (including ratings, credit-related analyses and data, valuations, model, software or other application or output therefrom) or any part thereof (Content) may be modified, reverse engineered, reproduced or distributed in any form by any means, or stored in a database or retrieval system, without the prior written permission of Standard & Poor’s Financial Services LLC or its affiliates (collectively, S&P). The Content shall not be used for any unlawful or unauthorized purposes. S&P and any third-party providers, as well as their directors, officers, shareholders, employees or agents (collectively S&P Parties) do not guarantee the accuracy, completeness, timeliness or availability of the Content. S&P Parties are not responsible for any errors or omissions (negligent or otherwise), regardless of the cause, for the results obtained from the use of the Content, or for the security or maintenance of any data input by the user. The Content is provided on an “as is” basis. S&P PARTIES DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, ANY WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE OR USE, FREEDOM FROM BUGS, SOFTWARE ERRORS OR DEFECTS, THAT THE CONTENT’S FUNCTIONING WILL BE UNINTERRUPTED OR THAT THE CONTENT WILL OPERATE WITH ANY SOFTWARE OR HARDWARE CONFIGURATION. In no event shall S&P Parties be liable to any party for any direct, indirect, incidental, exemplary, compensatory, punitive, special or consequential damages, costs, expenses, legal fees, or losses (including, without limitation, lost income or lost profits and opportunity costs or losses caused by negligence) in connection with any use of the Content even if advised of the possibility of such damages.
Credit-related and other analyses, including ratings, and statements in the Content are statements of opinion as of the date they are expressed and not statements of fact. S&P’s opinions, analyses and rating acknowledgment decisions (described below) are not recommendations to purchase, hold, or sell any securities or to make any investment decisions, and do not address the suitability of any security. S&P assumes no obligation to update the Content following publication in any form or format. The Content should not be relied on and is not a substitute for the skill, judgment and experience of the user, its management, employees, advisors and/or clients when making investment and other business decisions. S&P does not act as a fiduciary or an investment advisor except where registered as such. While S&P has obtained information from sources it believes to be reliable, S&P does not perform an audit and undertakes no duty of due diligence or independent verification of any information it receives. Rating-related publications may be published for a variety of reasons that are not necessarily dependent on action by rating committees, including, but not limited to, the publication of a periodic update on a credit rating and related analyses.
To the extent that regulatory authorities allow a rating agency to acknowledge in one jurisdiction a rating issued in another jurisdiction for certain regulatory purposes, S&P reserves the right to assign, withdraw or suspend such acknowledgment at any time and in its sole discretion. S&P Parties disclaim any duty whatsoever arising out of the assignment, withdrawal or suspension of an acknowledgment as well as any liability for any damage alleged to have been suffered on account thereof.
S&P keeps certain activities of its business units separate from each other in order to preserve the independence and objectivity of their respective activities. As a result, certain business units of S&P may have information that is not available to other S&P business units. S&P has established policies and procedures to maintain the confidentiality of certain non-public information received in connection with each analytical process.
S&P may receive compensation for its ratings and certain analyses, normally from issuers or underwriters of securities or from obligors. S&P reserves the right to disseminate its opinions and analyses. S&P's public ratings and analyses are made available on its Web sites, www.standardandpoors.com (free of charge), and www.ratingsdirect.com and www.globalcreditportal.com (subscription), and may be distributed through other means, including via S&P publications and third-party redistributors. Additional information about our ratings fees is available at www.standardandpoors.com/usratingsfees.
Any Passwords/user IDs issued by S&P to users are single user-dedicated and may ONLY be used by the individual to whom they have been assigned. No sharing of passwords/user IDs and no simultaneous access via the same password/user ID is permitted. To reprint, translate, or use the data or information other than as provided herein, contact S&P Global Ratings, Client Services, 55 Water Street, New York, NY 10041; (1) 212-438-7280 or by e-mail to: email@example.com.