October 17, 2024 • 34 mins
Shiloh Bates talks to Stephen Anderberg, the sector lead for U.S. CLOs at Standard and Poor's Global Ratings, about how CLOs are rated, trends in upgrades and downgrades and defaults. They also discuss how the September 18, 2024, interest rate cut may move the market. And they talk about how CLO ratings have performed relative to the more well-known corporate rating system. For a full transcript, or to learn more about Flat Rock Global, please visit flatrockglobal.com.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:06):
Hi, I'm Shiloh Bates and welcometo the CLO Investor podcast.
CLO stands for collateralizedloan obligations,
which are securities backed by poolsof leverage loans. In this podcast,
we discuss current newsin the CLO industry,
and I interview key market players.
Today I'm speaking with Steven Anderberg,
(00:28):
the sector lead for US CLOs atstandard and POS global ratings.
We discuss how CLOs arerated and trends in upgrades,
downgrades and defaults.
The federal reserve cut interestrates by 50 bips on September 18th,
so we discuss how that isexpected to move the CLO market.
We also discuss how CLO ratings haveperformed relative to the more well-known
(00:52):
corporate rating system.
Many of the securities we discuss arerated speculative grade or spec grade by
s and p, which means not investmentgrade or more specifically.
That's when a borrower has theability to repay but faces significant
uncertainties such as adverse businessor financial circumstances that could
(01:14):
affect credit risk. That'saccording to the s and p website.
If you enjoy the podcast, pleaseremember to share like and follow.
And now my conversationwith Steve Anderberg.
Steve, thanks so much forcoming on the podcast.
Shilah. Thank you very much.
(01:34):
Really appreciate the invite and I'mlooking forward to the conversation.
Likewise,
why don't we start off by just goingthrough your background and how you ended
up as a CLO ratings analyst at s and p?
So my current role ats and p is sector lead.
I've been with s and P now for 25 yearsand before that worked with the city of
New York in the office and managementand budget with an s and p.
(01:55):
I've always been withunstructured finance.
I started in a BS surveillance back inAugust of 99 looking after the ratings of
collateralized bond obligations.
And from there pretty quicklystarted working on CLOs,
which at that point I think literallyhad six transactions outstanding,
something like that.
From that I've seen the CLO market evolveand grow through the 1.0 era before
(02:16):
the financial crisis andthe 2.0 era that followed.
And obviously today it's a trillion dollarasset class and the core part of most
investor portfolios, a structuredfinance investor portfolios.
You were rating CLOsin 1999, is that right?
I was in the surveillance group,
so doing the monitoring and rating changeson collateralized bond obligations,
(02:36):
which actually did see a lot ofrating changes. And then the CLOs,
which really didn't.
I was just curious because I started onthe buy side working for CLO managers in
2000. So like you,
I've been around since the very beginningand seen this and experienced this
huge growth.
It's been amazing to see it evolve froma bank loan market and then to grow into
a trillion dollar market that wehave today. It's really astonishing.
(02:58):
Agreed. So Steve, how are CLOs rated?
The CLO ratings at s and p are built ona foundation of the corporate ratings,
and by corporate ratings,
I mean both the fundamental creditrating assigned to the company a single B
minus double B minus, whatever it is,
and then also a recovery rating that'sassigned to each loan in the collateral
pool. And in B-S-L-C-L-O,
something like 98% of the collateralcarries a public rating and the company
(03:22):
rating is used to infer a likelihoodof default in the modeling.
And the recovery rating obviously isused to infer the recovery assumptions in
the cashflow modeling. So when welook at ACL O, when we rate ACL O,
we put the portfolio into ourcredit model CDO evaluator,
which simulates default rates forportfolios that are commensurate with our
different rating levels.
And then it provides something calleda scenario default rate at each CLO
(03:44):
tranche rating level. And you can thinkof these as the scenario default rates
is our hurdle rate at eachrating level. So for example,
within a typical B-S-L-C-L-O collateralpool, if you run that through the model,
you might get a scenario default rateof say 68% at the AAA rating level,
which just means that for a tranchefrom a CLO to be rated AAA that's
(04:05):
collateralized by that pool of assets,
the tranche is going to have to beable to withstand 68% at the cloud or
defaulting over the life ofthe CLO without the tranche missing any interest or
principle.
So each alone has its ownprobability of default and also
the loans have somecorrelation with each other.
So they're in differentindustries or similar industries.
(04:27):
So you're running simulations throughthe loan pool and just trying to figure
out what the AAA and a AA survive,
and then that ties back to theactual ratings that you give.
Is that how to think about it?
That's exactly how to think about it.
With the one caveat that we'renot looking at the debt rating,
we're looking at the company rating.
And the reason for that is that the debtratings get notched up and down based
(04:49):
on the recovery prospects.
So a senior secured loan is typicallygoing to be one or two notches higher than
the company rating, and we alreadycapture recovery elsewhere.
So we look at the companyrating across our CLO analysis.
So when A CLO is coming to life andit's not maybe getting the ratings
that is targeting,
the solution to that is either tohave more equity or to have a more
(05:11):
diversified pool of loans or a higherrated pool of loans are those key
options.
You could either rejigger the collateralpool and maybe put some higher rated
assets in,
or you could change the capital structureon the CLO to make it work under the
methodology.
One thing that's nice is an assetclass is that the models are available
externally, including to thearrangers. So for the most part,
(05:32):
when they're submittinga transaction to us,
they have an idea of what's going towork and what's not going to work,
so we don't have to goback and forth as much.
And then one of the things I've noticedthat maybe wasn't intuitive to me right
away was just when something's ratedAAA and sold as such through banks,
a lot of times it seems like the ratingagencies would actually be willing to
provide more leverage at the AAAor AA in that it's actually the cap
(05:57):
on how deep the attachment point isactually comes from end investors,
not the rating agency.
That is true up and down the capitalstack that sometimes you end up with a
tranche that could pass at a higher ratinglevel that's going out at a different
rating level. And then it couldbe based on investor preference.
There's also you need a cushion in place.
You're not going to want a AAA thatgoes out and has a very tight rating
(06:20):
cushion,
both for just downgrade prospects ofthe tranche itself or tranche rating,
but also because the manager does notwant to get cut off from trading if they
fail their CDO monitor test,which affects reinvestment.
So they want some cushion in place there.
So nobody wants to have ratings thatare close to where they would be
potentially downgraded if the CLO LOhas some missteps. And then at Flatrock,
(06:41):
we're an investor both in broadlysyndicated CLOs and middle market CLOs.
How does the rating approachdiffer between the two?
It's a really good question.
Middle market CLOs are an area that'sseen tremendous growth over the past
several years. In the aftermathof the GFC, it was typically,
if you look at the overall issuance,
it was in the maybe high single digitsand then after 2016 it bumped up a little
(07:03):
bit. Now it's more like20% of total CLO issuance.
We use the same criteria for bothBSL CLOs and middle market CLOs.
There are a lot of structuraland collateral differences between the two that
get captured in the quantitative modeling.
There are two key differencesthough from a rating approach,
even under the same criteria,both having to do with the assets.
So if you look at BSL CLOs,
(07:24):
probably 98% of the assets carry apublic rating and middle market CLOs,
that's not the case.
The majority of the assets are notrated and yet to run CDO evaluator,
you can need a rating on for each companyin the portfolio. And so for those,
we do credit estimates, which arebasically for the CLO manager's purposes,
just an estimate of what a rating wouldbe if the company had a rating and they
(07:44):
could use that in theirmanagement of the CLO,
and we could use it in our analysis.So credit estimates instead of ratings.
And then the second differenceis recovery assumptions.
The spec grade public loanshave a recovery rating tied to them that guides the
recovery assumption, the CLO modeling,
but the credit estimatesdon't come with that.
So there we have a standardtable for recovery assumptions.
So you have one set of assumptions,senior secured, non cover light loans,
(08:08):
and other set of assumptions for seniorsecured cover light loans and so on down
the stack.
So for middle market loans and for broadlysyndicated loans that are going into
CLOs, either way they get rated,
just the question is whether or not it'sa public rating available to all market
participants or it'sa less formal process.
The credit estimatethat s and p also does.
(08:28):
In order to do the analysis, youneed a rating or estimated rating,
implied rating on every company in theportfolio. It doesn't work without that.
So how has your ratingsframework changed over time?
The criteria does not change that often.
It changes when our view ofthe fundamentals changes.
So we did a really bigupdate in September of 2009,
which affected almost all theassumptions within the criteria,
(08:52):
the assumed default rates,recovery rates, correlation,
and a host of other things.If you look at the 1.0 CLOs,
you could get to a AAA rating withmaybe a 26% par subordination.
But then coming out of the GFC in partbecause of rating agency changes in part
because of market changes, itwas more like mid thirties now.
So the criteria changes.
(09:13):
September of 2009 was part of that justrequired more subordination to get to a
aaa. And then after that there wasa smaller update in June of 2019,
which is to take into account thetotality of data from the last of the 1.0
transactions, which werepaying down at that point.
So it sounds like since the GFCthat to get the ratings that CLO
investors want,
(09:34):
there has to be more subordination orreally equity really at each level.
So the AAA has more equity supportingit and the AA same all the way down the
stack, so that's beneficial. Obviouslywe're an investor in double Bs,
so the double Bs get more equity,
that's great for them it meanslower probability of default.
But given how CLO securitiesperformed through the GFC and beyond,
(09:57):
why was it necessary to make your modelingassumptions or your framework more
conservative from the perspectiveof the debt investors.
Given what we're going through at thatpoint and the data we're looking at
during the GFC,
we were rethinking how correlated poolsof credit might perform under different
stresses. So revisited thecriteria based on that,
and there was a lot that came out of it.
(10:18):
The modeling assumptionsbecame more punitive,
especially at the higher rating levels,
especially for AAA shouldpoint out by the way,
CLO AAA has performed really wellin the history of the market.
There's never been a AAA default,
but the criteria change did requiremore subordination for those.
And there were other changes as well interms of what we expect to see in the
CLO transaction documents and otherthings all just making sure the ratings
(10:39):
produced and the criteriawould be commensurate with the economic environments
we thought they should be able to survive.
The Fed has been in a hiking cycle.How has that affected CLO ratings.
Given the 50 basis pointrate cut on September 18th?
This is obviously a timely topic.
The key issues looking forward are goingto be the path of interest rate cuts,
but also what happens with earningsgrowth in the spec rate corporate space,
(11:01):
the rate cuts this week provided atailwind for leveraged loan issuers since
they'll be devoting less cash interestpayments and able to use it for other
things including investingin future growth.
And if we were to end up in a situationwhere EBITDA growth turned negative,
the rate cuts will providesome cushion there.
So definitely at least a modestcredit positive for CLO collateral.
In terms of CLO issuance outlook,
(11:24):
I think the impact of the ratecuts is a little bit more opaque.
I'd be interested in yourperspective as well here,
but resets and refis aren't just goingto continue to dominate issuance this
year and maybe next year just given theoverhang of transactions that are out
there that are outside the no-callperiod with spreads that are wider than
current market spreads for C new issuance,
we think that insurance demandfor floating rate product is going to remain very
(11:47):
strong and that will have a benefitfor CLOs in general. And anecdotally,
we've heard that insurance companieshave raised a trillion dollars in
annuities,
and that's a lot of dry powder to deployinto the CLO market and Japanese banks,
the ones we've spoken to, remain wellpositioned to continue to buy CLOs.
So between those two things we thinkthat would be supportive for CLO issuance
(12:08):
next year.
One of the things that has surprisedme about interest rates and the market
reaction is just that if you would'veasked me in 2021 and broadly these
syndicated CLOs were printing at,call it, there's a LO back then,
but 115 basis points overwas where some aaas were
getting done. And asthe Fed started hiking,
(12:29):
actually the spreads on aaas went wider.
And I would've thought people would beclamoring for aaas because the higher
base rate the return offered was higher.
But actually the marketreaction I think was twofold.
It was banks sensing potentially thebeginning of a recessionary period just
for putting the brakeson buying new aaas one,
(12:50):
but also because the base rate was higher.
A lot of investors just didn't want totake the extra risk and earn the AAA
spread over IOR or SOFR for example.They're happy just earning the base rate.
So now I think what'll happenis as the fed lowers rates,
I think what's going to happen is peopleare going to be very interested more in
the spread than they were in the pastbecause the base rate isn't going to be as
(13:12):
high.
And I think that'll continueto put downward pressure on spreads up and down the
stack from AAA to double B. That'show I think this is going to play out.
I think you make aninteresting point, Charlo.
So I fully agree thatto some extent at least,
investors look at the all in yield on atrach so that as the base rate becomes
less a driver of that, they're goingto focus more on the credit spread.
(13:33):
So that could push for wider creditspreads going forward or at least put a
damper on further credit tightening.Spread tightening. It is interesting,
there are other factors that drive allthis as well. And you look back at 2021,
which was so far the alltime record for new issuance,
185 billion of CLO newissuance. And in that case,
the banks were just flush with depositsfrom covid stimulus checks and needed a
(13:54):
place to deploy. The cashrates were very tight.
So CLOs just offered an attractivespread above the base rate,
and that's just where alot of the money went,
and that's just droveissuance for that year.
But I think I see it maybe slightlydifferent though I think as the fed cuts
that actually spreads aregoing to continue to decline.
And the reason is I think it's maybeopposite of the theory. You'd be like,
(14:15):
okay, well if the fed's cutting,I care about all in yield,
so the spreads needs to be higher.
But actually that's definitelynot what happened on the way up.
On the way up the Fed was hiking andinvestors were acquiring at least for a
while more spread.
So here I think what's going to happenis AAA is going to be all the more
interesting now because they'regoing to need to earn it.
They're not going to be ableto just take the base rate.
(14:37):
So that'll be more competitionfor aaas and down the stack.
I think that's true ifyou look at 2022 and 2023,
which showed very wide spreads onaaas and other charges and CLOs,
economic growth or lack thereof andfear of a recession just drives a lot of
this. So maybe that would be the dominantfactor in driving credit spreads going
forward.
Prospects for recession definitelywould be a big factor in there as well.
(15:01):
So changing topics.
The CLOs do have bondbuckets usually is 5%.
The typical cap on bonds that can be putinto A ZLO, I think that's the number.
And a lot of times these arefloating rate secured bonds,
so they look a lot like loans anyways.
But what are you seeing in terms ofmanager usage of bond buckets today?
Interesting question. So wethought with the rate cut,
(15:23):
it was a good time to take alook and to see where we were.
So obviously the addition ofbonds to CLO portfolios is a
phenomenon we've seen over the past coupleof years as higher interest rates to
hold.
And there was really a way for managersto pick up part the cost of giving up
some spread.
And there was also a pretty substantialweighted average rating factor or wharf
benefit since a lot of the bonds camefrom slightly higher rated issuers,
(15:46):
in some cases a lot higher rated issuersthan the typical leveraged loan issuer.
So I think we're at like2% of total collateral.
So still a modest in terms of actual usageof the bond buckets, but if you look,
the median whaf or SP wharf of thebonds is 2165 against a median whaf
of the CLO collateral overall of 2,700.
(16:06):
So that's a pretty significant difference.
So the wharf is the weightedaverage rating factor,
and that ties into probability ofdefault for each underlying loan.
That's exactly right for eachof the underlying issuers.
So a lower wharf indicates, atleast from a rating perspective,
a higher quality collateral pool.And if you look at the bond holdings,
I think we just did aquick snapshot yesterday,
(16:27):
82% of them are from spec gradeissuers, so it's as expected,
but 14% were from triple B issuers andthen 4% actually came from single A minus
or higher,
which you don't usually see those higherrated obligors within C collateral
pools. It was just aphenomenon for the bonds.
So it sounds like some CLO managershave bought high quality bonds,
probably low coupon, but bought'em at a discount to par.
(16:51):
And that helps with the CLOsover collateralization tests.
And just generally offset par lossesthat might've come from elsewhere.
And then one of the trends we've seenin the market really over the last two
years is the rise of theliability management exercise.
So I did a podcast recently withDrew Sweeney of TCW on this,
but the LME is the out of courtrestructuring where the private
(17:15):
equity investor that owns the firmand the first lien debt agree to some
restructuring, whichprobably entails a haircut,
the first lien lenders takinga haircut on their debt.
How has the rise of the LME affectedyour business and your ratings?
So you're right in pointing out thatmost of the LMEs have come from private
equity owned or sponsored companies.
(17:36):
There have been a couple withpublicly owned companies as well,
but one point I would make is thatCLO portfolio are very diverse.
A median portfolio has somethinglike 300 obligor in it.
So the impact of any one obligor in acollateral pool going through a management
exercise is fairly muted. Andwe have noted, by the way,
a difference in approach betweenthe bigger scale managers,
(17:56):
the higher A UM managers and the smallermanagers where the larger managers are
more likely to leverage their scale andengage in the LME process and try to
come through with a more positive outcome,
something that's morecreative to the recoveries,
the smaller managers would be much morelikely to sell and just get out of the
situation because they may not have toscale to ensure they get our seat at the
table.
(18:16):
And then what's the trends in loanupgrades and downgrades, for example,
today?
So the higher interest rates and slowinggrowth we've seen over the past couple
of years have definitely put pressureon lower rated corporate borrowers,
and some of them at least have beenstraining under the burden of higher debt
service costs. And by late 2023,
the downgrade rate for specgrade companies had risen considerably and peaked at
(18:38):
three and a half downgradesfor each corporate rating upgrade as have October of
last year. And then it'smoderated a bit since,
although it's beenchoppy month over month.
I think at this point most corporate loanissuers have adjusted to higher rates
by cutting costs, differing CapExexpenditures and doing other things.
And then another benefit of the currentmarket is a spec grade companies have
(18:58):
been able to refinance or repricetheir outstanding loans in some cases,
cutting their interest expensesby 50 basis points or more.
So they've got the benefit of tworate cuts just by refinancing their
outstanding debt.
So we're definitely in bettershape than we were say a year ago.
The fed beginning to cut.
Should we infer from that that maybemore upgrades than downgrades are on the
(19:19):
come here?
So you are absolutely right that thedecrease in rates is going to provide a
tailwind for highly leveraged borrowersand it'll free them up at least to some
extent from the burden of interestpayments and allow them to redeploy that
money into other things, including CapEx,
which could benefit growth goingforward. In terms of the ratings,
it's already largely baked in. And whenthe analysts look at a company rating,
(19:40):
they're already taking into account theforward curve of rates and taking that
into account when assigning the rating.
So at the very fact of justseeing the rate decrease,
unless something has changedthe path going forward,
it makes us change our assumptions.
I don't think you would see a largenumber of upgrades based solely on that.
So that was upgrades and downgradesfor the underlying loans and CLOs,
but what about upgrades anddowngrades for CLO securities?
(20:03):
So we haven't downgraded a AAA CLOtrache rating probably since 2012.
It's been a long time.
It really takes a considerable amountof economic stress before that happens.
And I took a quick lookat the past four years,
going back to the beginning of 2020 atwhat the CLO trache ratings were doing,
and obviously 2020 giventhe pandemic, the shutdowns,
(20:24):
the impact on spec rated corporateissuers was pretty significant.
And there were a lot of downgrades in theCLO collateral pools and the liability
ratings followed.
So something like 13% of B-S-L-C-L-Oratings from s and p got lowered
in 2020 and a lot fewer middle marketers,
1% middle market ratingsgot lowered that year,
but every year since from 2021 onward,
there've actually been more CLO trancheupgrades rather than downgrades.
(20:47):
So I think it's a pretty good record,
especially given the level ofcorporate downgrades in 2022 and 2023.
The CLO managers were able to repositionthe portfolios in most cases and go out
against the downside and just thestructural mechanics protected the
transactions and they've done prettywell in the course of a stressed economic
period.
So transitioning from CLOupgrades and downgrades,
(21:08):
the more important thing tome would just be defaults.
So have we seen an uptick in CLOsecurity defaults since the covid period?
The short answer is not really.
We're just about to update a pieceand show that we've had 61 CLO tranche
defaults through both the 1.0 and the2.0 era that the asset class is large
enough now that every year you're goingto see a small handful of double B
(21:30):
tranches that might default.
Usually these are right now what we'reseeing is transactions that originated
before the pandemic and then sufferedagain in 2022 and 2023 and just weren't
able to cover, at leaston a projected basis,
the debt outstandingat the double B level.
But 61 traunch defaults out of 18,000or so rated tranches over the past
30 years is a pretty good record.
(21:52):
Agreed. So the default statisticsthat you just mentioned,
how did those compare to the loan ratings?
Does a double BCLO,
is it comparable to a double B corporateor how should we think about the
ratings between the two?
The short answer is CLO ratingshave that done very well,
and in every case perform with fewerdefaults and comparably rated corporates.
(22:15):
You have to be a little bit careful inhow you do it, because fundamentally,
if you're looking at CLO ratings,
they're originated and thenseven or eight years later,
or maybe sooner if it gets reset orwhatever, they're not there anymore.
With corporates, you have ratingsoutstanding for a long time,
but Meredith Coffee from the LSDApublished a piece back in July of
2022,
actually using some s and p data andcomparing the default rates for CLO
(22:37):
tranches against the default rates forrated corporates. And if you take a look,
for example, at double Bfor CLO 1.0 transactions,
the double B default rate for CLO tranchesrated double B was something like 4%
for CLO 2.0. So far the double B traunchdefault rate has been well under 1%.
And then if you look at corporateratings over the same period,
it's more like 9%. So there's a prettysignificant difference between those two.
(23:01):
But you're quoting there thecumulative default rates.
How I would think about it is like anannual default rate. So if the cumulative,
you said at 4% for the one point ohs,
so maybe if they were outstanding forseven or eight years, call it eight years,
then it's a 50 BIP default rate per year.
And then for say the two pointohs, you said 1% cumulative.
(23:21):
Well,
how many years do you think those doubleB's were outstanding for maybe five,
something like that? So it's 20 bips peryear would be the default rate there.
Yeah, for an individual transaction, ofcourse it's going to be extraordinary,
unlikely you'd see a default in year one.
But if you want to look acrossthe universe of tranches,
you could look at it thatway. Any way you stack it up.
CLO tranche ratings come in prettywell below the comparable corporate.
(23:45):
So one of the questions I'mfrequently asked, I tell them, Hey,
the double B defaultrate has been quite low,
and their responses well are aroundthe precipice of a massive default wave
is the future can be verydifferent from the past.
And maybe embedded in that is thequestion about higher base rates.
But I think your answer on upgradesand downgrades answers that question.
(24:08):
If the CLO securities are, asyou said, for the most part,
upgrades is the trend, not downgrades.
That's true. I mean, even duringa stressed economic period,
if you want to get a sense of how theCLO ratings might respond under different
levels of economic stress, weactually publish a series annually,
a series of stress tests,
and the DSL iteration of that shouldbe out this month or maybe early next
(24:30):
month.
We take all of the CLO tranche ratingsand run them through basically four sets
of stresses with successively,
more collateral defaults andsuccessively bigger triple C baskets.
And then we look at the impact forour rated universe of these stresses.
So if you want to see what wouldhappen to the bbs, for example,
under a scenario where 20%of loans default and the triple C baskets blow up to
(24:51):
40%, as unlikely as either of thosethings would be, you could do that.
So Charlotte, if I could ask you acouple of questions and turn the tables,
how do you use ratings into yourinvestment process or processes?
At Flatrock?
We're not really ratingsconstrained. So we own CLO equity.
Obviously CLO equity is non-rated,
(25:12):
so we'd like the CLO debt securitiesin the CLO to be rated well,
but we don't have a regulator ora capital charge regime. Well,
our regulator is FINRA and the SEC,
but we're not an insurance company or abank where we need to maintain a overall
rating or portfolio quality.So for equity, you're not rating those securities,
(25:32):
so we make our own judgments.And then for double B,
your opinion on the credit qualityof the node is certainly important,
but we're doing our ownscenarios and work ourselves.
So you said your for stress cases, we'vegot a bunch of stress cases as well,
but at the end of the day,sometimes we would buy double Bs.
That would be a weaker end ofmaybe where you guys would rate,
(25:54):
and we would do that because they're cheapin the market and we think they offer
good returns. Maybe they're later intheir life and they don't need all the
equity subordination thatthey might've had originally.
If a double B is upgraded, which I don'tthink we've seen too many of those,
I don't think it really matters that muchin terms of trading level because it's
funny.
One difference in CLOs for other marketsis that when a CLO O'S security is
(26:17):
issued as double B, I think the industrynomenclature just keeps it as double B.
That's where it was initially rated.
That's the part of thecapital stack that it is.
And if you guys upgrade it to triple B,
I think people would continueto call it the double B.
And if it was downgraded a single B,
maybe somebody wouldstill call it a double B,
but acknowledge that there's beensome losses on the loans ratings.
(26:41):
It's very important part of the market,but it's really affecting, I think,
especially the guys who are playing inthe investment grade parts of the stack.
So the AAA investor needsthat rating for the bank,
wants to make sure that rating ismaintained through the life of the CLO.
But some of the debts downgraded in oneof our deals definitely means there's
been some par losses onthe loans. But for us,
(27:01):
that could just mean we boughtthe equity cheap in the past.
It wouldn't necessarily correlateto how we're doing down the stack.
Interesting.
And how constructive are you on OS justas an asset class say going into next
year? I mean,
how do you see the combined impacts ofrate cuts and either economic growth or
lack of economic growthimpact in the asset class?
(27:22):
So I think one of the things that willchange now with the Fed beginning,
what we think it's going tobe this cutting cycle here is just that the relative
attractiveness of double Bs has reallybeen pronounced over the last two years.
So the Fed hiked a lot, andyou could get at least in,
let's talk about middle market CLOs forexample. The yields on those securities,
(27:42):
CLO middle market bbshave been in the 13% area.
So a lot of people looked at thatand said, Hey, that's a great return.
Why would I want to take firstloss risk and be in the equity?
So some of 'em made thatrotation or that new allocation,
and then now as the Fed cuts, the yields,
the double Bs are going to be goingdown one and then two for CLO equity.
(28:04):
The projected returns there are not goingto look that different even though the
Fed is cutting. And the reason is thatI think most people who sit in my seat,
they're running an IRR calculationthrough the cash flows that the CLO equity
is supposed to produce oris expected to produce.
So there's already baked intothat a declining SOFR curve.
So there's already a budget there.
(28:26):
If you're buying equity today and you'retargeting mid to high teen returns,
you're budgeting for that decline inSOFR over time. Whereas for the double B,
those payments are made quarterly andthey reset quarterly and they're going to
be resetting lower, and that's going toresult in less cashflow to the double B.
So my point in that is just I think you'regoing to see a rotation out of double
(28:46):
BCLO equity is going to becomerelatively more interesting.
And I think the two things that maybenefit CLO equity here are that one,
the lower interest expenseburden for a lot of companies,
it's not going to matter.The loan's money good,
and maybe you'd rather havethe higher base rate there,
but some weaker borrowers in the marketwill actually benefit and have more
(29:07):
cash, more liquidity, better interestcoverage. And for CLO equity,
that could mean a marginal difference interms of default rate more favorable to
us. So I think that's a potential benefit.And then the other part of that is,
again, if you're targeting mid tohigh teen returns for CLO equity,
as the fed cuts,
other competing asset classes are goingto offer lower comparative returns and
(29:31):
CLO equity the rates, wedon't see them coming down.
So on a relative value basis,
I think it's going to make equity lookmore appealing than maybe it has over the
last year and change.
Interesting.
Well, Steve, thanks so muchfor coming on the podcast.
Really enjoyed the conversation.
The content here is for informationalpurposes only and should not be taken as
(29:54):
legal business tax or investmentadvice or be used to evaluate any
investment or security.
This podcast is not directed at anyinvestment or potential investors in any
Flat Rock Global Fund definition.
Section A UM refers toassets under management,
LMT or liability managementtransactions are an out of court
(30:16):
modification of a company's debt Layeringrefers to placing additional debt with
a priority above the first lien term loan.
The secured overnight financing rate,
SOFR is a broad measure of thecost of borrowing cash overnight,
collateralized by treasury securities.The global financial crisis,
GFC was a period of extreme stress inglobal financial markets and banking
(30:40):
systems between mid 2007 and early 2009.
Credit ratings are opinions aboutcredit risk for long-term issues or
instruments,
the ratings lie on a spectrum rangingfrom the highest credit quality on one end
to default or junk on the other.A AAA is the highest credit quality A
C or D,
depending on the agency issuing therating is the lowest or junk quality.
(31:04):
Leveraged loans are corporate loans tocompanies that are not rated investment
grade broadly. Syndicated loansare underwritten by banks,
rated by nationally recognized statisticalratings organizations and often
traded by market participants.
Middle market loans are usuallyunderwritten by several lenders with the
intention of holding the investmentthrough its maturity spread is the
(31:27):
percentage difference in current yieldsof various classes of fixed income
securities versus treasury bondsor another benchmark bond measure.
Our set is a refinancing andextension of A CLO investment.
EBITDA is earnings beforeinterest, taxes, depreciation,
and amortization.
An add back would attempt toadjust EBIT DA for non-recurring
(31:51):
exchange traded funds. libor,
the London Interbank offer ratewas replaced by software on
June 30th, 2024.
Deliver means reducing the amount ofdebt financing. High yield bonds are
corporate borrowings ratedbelow investment grade that are usually fixed rate
and unsecured default refers tomissing a contractual interest or
(32:14):
principle payment.
Debt has contractual interestprinciple and interest payments,
whereas equity representsownership in a company.
Senior secured corporateloans are borrowings from a company that are backed by
collateral.
Junior debt ranks behind seniorsecure debt in its payment priority.
Collateral pool refers to the sum ofcollateral pledge to a lender to support
(32:39):
its repayment.
A on-call refers to the time inwhich a debt instrument cannot be
optionally repaid.
A floating rate investment has an interestrate that varies with an underlying
floating rate index.General disclaimer section,
references to interest rate movesare based on Bloomberg data.
(33:00):
Any mentions of specific companies arefor reference purposes only and are not
meant to describe the investment meritsof or potential or actual portfolio
changes related to securities of thosecompanies unless otherwise noted.
All discussions are based on USmarkets and US monetary and fiscal
policies.
Market forecasts and projections arebased on current market conditions
(33:24):
and are subject to change without notice,
projections should not beconsidered a guarantee.
The views and opinions expressed by theFlat Rock global speaker are those of
the speaker as of the dateof the broadcast and do not necessarily represent the
views of the firm as a whole.
Any such views are subject to changeat any time based upon market or other
(33:45):
conditions,
and Flat Rock Global Disclaims anyresponsibility to update such views.
This material is not intended to berelied upon as a forecast, research,
or investment advice.
It is not a recommendation offeror solicitation to buy or sell any
securities or to adoptany investment strategy.
Neither Flat Rock Global nor the FlatRock Global Speaker can be responsible for
(34:10):
any direct or incidental loss incurredby applying any of the information
offered.
None of the information provided shouldbe regarded as a suggestion to engage in
or refrain from any investment relatedcourse of action as neither Flat Rock
Global nor its affiliates are undertaking.
To provide impartial investment advice,
act as an impartial advisor or giveadvice in a fiduciary capacity.
(34:34):
Additional information about this podcastalong with an edited transcript may be
obtained by visiting flat global.com.
Hi, I'm Shiloh Bates and welcometo the CLO Investor podcast.
CLO stands for collateralizedloan obligations,
which are securities backed by poolsof leverage loans. In this podcast,
we discuss current newsin the CLO industry,
and I interview key market players.
Today I'm speaking with Steven Anderberg,
(00:28):
the sector lead for US CLOs atstandard and POS global ratings.
We discuss how CLOs arerated and trends in upgrades,
downgrades and defaults.
The federal reserve cut interestrates by 50 bips on September 18th,
so we discuss how that isexpected to move the CLO market.
We also discuss how CLO ratings haveperformed relative to the more well-known
(00:52):
corporate rating system.
Many of the securities we discuss arerated speculative grade or spec grade by
s and p, which means not investmentgrade or more specifically.
That's when a borrower has theability to repay but faces significant
uncertainties such as adverse businessor financial circumstances that could
(01:14):
affect credit risk. That'saccording to the s and p website.
If you enjoy the podcast, pleaseremember to share like and follow.
And now my conversationwith Steve Anderberg.
Steve, thanks so much forcoming on the podcast.
Shilah. Thank you very much.
(01:34):
Really appreciate the invite and I'mlooking forward to the conversation.
Likewise,
why don't we start off by just goingthrough your background and how you ended
up as a CLO ratings analyst at s and p?
So my current role ats and p is sector lead.
I've been with s and P now for 25 yearsand before that worked with the city of
New York in the office and managementand budget with an s and p.
(01:55):
I've always been withunstructured finance.
I started in a BS surveillance back inAugust of 99 looking after the ratings of
collateralized bond obligations.
And from there pretty quicklystarted working on CLOs,
which at that point I think literallyhad six transactions outstanding,
something like that.
From that I've seen the CLO market evolveand grow through the 1.0 era before
(02:16):
the financial crisis andthe 2.0 era that followed.
And obviously today it's a trillion dollarasset class and the core part of most
investor portfolios, a structuredfinance investor portfolios.
You were rating CLOsin 1999, is that right?
I was in the surveillance group,
so doing the monitoring and rating changeson collateralized bond obligations,
(02:36):
which actually did see a lot ofrating changes. And then the CLOs,
which really didn't.
I was just curious because I started onthe buy side working for CLO managers in
2000. So like you,
I've been around since the very beginningand seen this and experienced this
huge growth.
It's been amazing to see it evolve froma bank loan market and then to grow into
a trillion dollar market that wehave today. It's really astonishing.
(02:58):
Agreed. So Steve, how are CLOs rated?
The CLO ratings at s and p are built ona foundation of the corporate ratings,
and by corporate ratings,
I mean both the fundamental creditrating assigned to the company a single B
minus double B minus, whatever it is,
and then also a recovery rating that'sassigned to each loan in the collateral
pool. And in B-S-L-C-L-O,
something like 98% of the collateralcarries a public rating and the company
(03:22):
rating is used to infer a likelihoodof default in the modeling.
And the recovery rating obviously isused to infer the recovery assumptions in
the cashflow modeling. So when welook at ACL O, when we rate ACL O,
we put the portfolio into ourcredit model CDO evaluator,
which simulates default rates forportfolios that are commensurate with our
different rating levels.
And then it provides something calleda scenario default rate at each CLO
(03:44):
tranche rating level. And you can thinkof these as the scenario default rates
is our hurdle rate at eachrating level. So for example,
within a typical B-S-L-C-L-O collateralpool, if you run that through the model,
you might get a scenario default rateof say 68% at the AAA rating level,
which just means that for a tranchefrom a CLO to be rated AAA that's
(04:05):
collateralized by that pool of assets,
the tranche is going to have to beable to withstand 68% at the cloud or
defaulting over the life ofthe CLO without the tranche missing any interest or
principle.
So each alone has its ownprobability of default and also
the loans have somecorrelation with each other.
So they're in differentindustries or similar industries.
(04:27):
So you're running simulations throughthe loan pool and just trying to figure
out what the AAA and a AA survive,
and then that ties back to theactual ratings that you give.
Is that how to think about it?
That's exactly how to think about it.
With the one caveat that we'renot looking at the debt rating,
we're looking at the company rating.
And the reason for that is that the debtratings get notched up and down based
(04:49):
on the recovery prospects.
So a senior secured loan is typicallygoing to be one or two notches higher than
the company rating, and we alreadycapture recovery elsewhere.
So we look at the companyrating across our CLO analysis.
So when A CLO is coming to life andit's not maybe getting the ratings
that is targeting,
the solution to that is either tohave more equity or to have a more
(05:11):
diversified pool of loans or a higherrated pool of loans are those key
options.
You could either rejigger the collateralpool and maybe put some higher rated
assets in,
or you could change the capital structureon the CLO to make it work under the
methodology.
One thing that's nice is an assetclass is that the models are available
externally, including to thearrangers. So for the most part,
(05:32):
when they're submittinga transaction to us,
they have an idea of what's going towork and what's not going to work,
so we don't have to goback and forth as much.
And then one of the things I've noticedthat maybe wasn't intuitive to me right
away was just when something's ratedAAA and sold as such through banks,
a lot of times it seems like the ratingagencies would actually be willing to
provide more leverage at the AAAor AA in that it's actually the cap
(05:57):
on how deep the attachment point isactually comes from end investors,
not the rating agency.
That is true up and down the capitalstack that sometimes you end up with a
tranche that could pass at a higher ratinglevel that's going out at a different
rating level. And then it couldbe based on investor preference.
There's also you need a cushion in place.
You're not going to want a AAA thatgoes out and has a very tight rating
(06:20):
cushion,
both for just downgrade prospects ofthe tranche itself or tranche rating,
but also because the manager does notwant to get cut off from trading if they
fail their CDO monitor test,which affects reinvestment.
So they want some cushion in place there.
So nobody wants to have ratings thatare close to where they would be
potentially downgraded if the CLO LOhas some missteps. And then at Flatrock,
(06:41):
we're an investor both in broadlysyndicated CLOs and middle market CLOs.
How does the rating approachdiffer between the two?
It's a really good question.
Middle market CLOs are an area that'sseen tremendous growth over the past
several years. In the aftermathof the GFC, it was typically,
if you look at the overall issuance,
it was in the maybe high single digitsand then after 2016 it bumped up a little
(07:03):
bit. Now it's more like20% of total CLO issuance.
We use the same criteria for bothBSL CLOs and middle market CLOs.
There are a lot of structuraland collateral differences between the two that
get captured in the quantitative modeling.
There are two key differencesthough from a rating approach,
even under the same criteria,both having to do with the assets.
So if you look at BSL CLOs,
(07:24):
probably 98% of the assets carry apublic rating and middle market CLOs,
that's not the case.
The majority of the assets are notrated and yet to run CDO evaluator,
you can need a rating on for each companyin the portfolio. And so for those,
we do credit estimates, which arebasically for the CLO manager's purposes,
just an estimate of what a rating wouldbe if the company had a rating and they
(07:44):
could use that in theirmanagement of the CLO,
and we could use it in our analysis.So credit estimates instead of ratings.
And then the second differenceis recovery assumptions.
The spec grade public loanshave a recovery rating tied to them that guides the
recovery assumption, the CLO modeling,
but the credit estimatesdon't come with that.
So there we have a standardtable for recovery assumptions.
So you have one set of assumptions,senior secured, non cover light loans,
(08:08):
and other set of assumptions for seniorsecured cover light loans and so on down
the stack.
So for middle market loans and for broadlysyndicated loans that are going into
CLOs, either way they get rated,
just the question is whether or not it'sa public rating available to all market
participants or it'sa less formal process.
The credit estimatethat s and p also does.
(08:28):
In order to do the analysis, youneed a rating or estimated rating,
implied rating on every company in theportfolio. It doesn't work without that.
So how has your ratingsframework changed over time?
The criteria does not change that often.
It changes when our view ofthe fundamentals changes.
So we did a really bigupdate in September of 2009,
which affected almost all theassumptions within the criteria,
(08:52):
the assumed default rates,recovery rates, correlation,
and a host of other things.If you look at the 1.0 CLOs,
you could get to a AAA rating withmaybe a 26% par subordination.
But then coming out of the GFC in partbecause of rating agency changes in part
because of market changes, itwas more like mid thirties now.
So the criteria changes.
(09:13):
September of 2009 was part of that justrequired more subordination to get to a
aaa. And then after that there wasa smaller update in June of 2019,
which is to take into account thetotality of data from the last of the 1.0
transactions, which werepaying down at that point.
So it sounds like since the GFCthat to get the ratings that CLO
investors want,
(09:34):
there has to be more subordination orreally equity really at each level.
So the AAA has more equity supportingit and the AA same all the way down the
stack, so that's beneficial. Obviouslywe're an investor in double Bs,
so the double Bs get more equity,
that's great for them it meanslower probability of default.
But given how CLO securitiesperformed through the GFC and beyond,
(09:57):
why was it necessary to make your modelingassumptions or your framework more
conservative from the perspectiveof the debt investors.
Given what we're going through at thatpoint and the data we're looking at
during the GFC,
we were rethinking how correlated poolsof credit might perform under different
stresses. So revisited thecriteria based on that,
and there was a lot that came out of it.
(10:18):
The modeling assumptionsbecame more punitive,
especially at the higher rating levels,
especially for AAA shouldpoint out by the way,
CLO AAA has performed really wellin the history of the market.
There's never been a AAA default,
but the criteria change did requiremore subordination for those.
And there were other changes as well interms of what we expect to see in the
CLO transaction documents and otherthings all just making sure the ratings
(10:39):
produced and the criteriawould be commensurate with the economic environments
we thought they should be able to survive.
The Fed has been in a hiking cycle.How has that affected CLO ratings.
Given the 50 basis pointrate cut on September 18th?
This is obviously a timely topic.
The key issues looking forward are goingto be the path of interest rate cuts,
but also what happens with earningsgrowth in the spec rate corporate space,
(11:01):
the rate cuts this week provided atailwind for leveraged loan issuers since
they'll be devoting less cash interestpayments and able to use it for other
things including investingin future growth.
And if we were to end up in a situationwhere EBITDA growth turned negative,
the rate cuts will providesome cushion there.
So definitely at least a modestcredit positive for CLO collateral.
In terms of CLO issuance outlook,
(11:24):
I think the impact of the ratecuts is a little bit more opaque.
I'd be interested in yourperspective as well here,
but resets and refis aren't just goingto continue to dominate issuance this
year and maybe next year just given theoverhang of transactions that are out
there that are outside the no-callperiod with spreads that are wider than
current market spreads for C new issuance,
we think that insurance demandfor floating rate product is going to remain very
(11:47):
strong and that will have a benefitfor CLOs in general. And anecdotally,
we've heard that insurance companieshave raised a trillion dollars in
annuities,
and that's a lot of dry powder to deployinto the CLO market and Japanese banks,
the ones we've spoken to, remain wellpositioned to continue to buy CLOs.
So between those two things we thinkthat would be supportive for CLO issuance
(12:08):
next year.
One of the things that has surprisedme about interest rates and the market
reaction is just that if you would'veasked me in 2021 and broadly these
syndicated CLOs were printing at,call it, there's a LO back then,
but 115 basis points overwas where some aaas were
getting done. And asthe Fed started hiking,
(12:29):
actually the spreads on aaas went wider.
And I would've thought people would beclamoring for aaas because the higher
base rate the return offered was higher.
But actually the marketreaction I think was twofold.
It was banks sensing potentially thebeginning of a recessionary period just
for putting the brakeson buying new aaas one,
(12:50):
but also because the base rate was higher.
A lot of investors just didn't want totake the extra risk and earn the AAA
spread over IOR or SOFR for example.They're happy just earning the base rate.
So now I think what'll happenis as the fed lowers rates,
I think what's going to happen is peopleare going to be very interested more in
the spread than they were in the pastbecause the base rate isn't going to be as
(13:12):
high.
And I think that'll continueto put downward pressure on spreads up and down the
stack from AAA to double B. That'show I think this is going to play out.
I think you make aninteresting point, Charlo.
So I fully agree thatto some extent at least,
investors look at the all in yield on atrach so that as the base rate becomes
less a driver of that, they're goingto focus more on the credit spread.
(13:33):
So that could push for wider creditspreads going forward or at least put a
damper on further credit tightening.Spread tightening. It is interesting,
there are other factors that drive allthis as well. And you look back at 2021,
which was so far the alltime record for new issuance,
185 billion of CLO newissuance. And in that case,
the banks were just flush with depositsfrom covid stimulus checks and needed a
(13:54):
place to deploy. The cashrates were very tight.
So CLOs just offered an attractivespread above the base rate,
and that's just where alot of the money went,
and that's just droveissuance for that year.
But I think I see it maybe slightlydifferent though I think as the fed cuts
that actually spreads aregoing to continue to decline.
And the reason is I think it's maybeopposite of the theory. You'd be like,
(14:15):
okay, well if the fed's cutting,I care about all in yield,
so the spreads needs to be higher.
But actually that's definitelynot what happened on the way up.
On the way up the Fed was hiking andinvestors were acquiring at least for a
while more spread.
So here I think what's going to happenis AAA is going to be all the more
interesting now because they'regoing to need to earn it.
They're not going to be ableto just take the base rate.
(14:37):
So that'll be more competitionfor aaas and down the stack.
I think that's true ifyou look at 2022 and 2023,
which showed very wide spreads onaaas and other charges and CLOs,
economic growth or lack thereof andfear of a recession just drives a lot of
this. So maybe that would be the dominantfactor in driving credit spreads going
forward.
Prospects for recession definitelywould be a big factor in there as well.
(15:01):
So changing topics.
The CLOs do have bondbuckets usually is 5%.
The typical cap on bonds that can be putinto A ZLO, I think that's the number.
And a lot of times these arefloating rate secured bonds,
so they look a lot like loans anyways.
But what are you seeing in terms ofmanager usage of bond buckets today?
Interesting question. So wethought with the rate cut,
(15:23):
it was a good time to take alook and to see where we were.
So obviously the addition ofbonds to CLO portfolios is a
phenomenon we've seen over the past coupleof years as higher interest rates to
hold.
And there was really a way for managersto pick up part the cost of giving up
some spread.
And there was also a pretty substantialweighted average rating factor or wharf
benefit since a lot of the bonds camefrom slightly higher rated issuers,
(15:46):
in some cases a lot higher rated issuersthan the typical leveraged loan issuer.
So I think we're at like2% of total collateral.
So still a modest in terms of actual usageof the bond buckets, but if you look,
the median whaf or SP wharf of thebonds is 2165 against a median whaf
of the CLO collateral overall of 2,700.
(16:06):
So that's a pretty significant difference.
So the wharf is the weightedaverage rating factor,
and that ties into probability ofdefault for each underlying loan.
That's exactly right for eachof the underlying issuers.
So a lower wharf indicates, atleast from a rating perspective,
a higher quality collateral pool.And if you look at the bond holdings,
I think we just did aquick snapshot yesterday,
(16:27):
82% of them are from spec gradeissuers, so it's as expected,
but 14% were from triple B issuers andthen 4% actually came from single A minus
or higher,
which you don't usually see those higherrated obligors within C collateral
pools. It was just aphenomenon for the bonds.
So it sounds like some CLO managershave bought high quality bonds,
probably low coupon, but bought'em at a discount to par.
(16:51):
And that helps with the CLOsover collateralization tests.
And just generally offset par lossesthat might've come from elsewhere.
And then one of the trends we've seenin the market really over the last two
years is the rise of theliability management exercise.
So I did a podcast recently withDrew Sweeney of TCW on this,
but the LME is the out of courtrestructuring where the private
(17:15):
equity investor that owns the firmand the first lien debt agree to some
restructuring, whichprobably entails a haircut,
the first lien lenders takinga haircut on their debt.
How has the rise of the LME affectedyour business and your ratings?
So you're right in pointing out thatmost of the LMEs have come from private
equity owned or sponsored companies.
(17:36):
There have been a couple withpublicly owned companies as well,
but one point I would make is thatCLO portfolio are very diverse.
A median portfolio has somethinglike 300 obligor in it.
So the impact of any one obligor in acollateral pool going through a management
exercise is fairly muted. Andwe have noted, by the way,
a difference in approach betweenthe bigger scale managers,
(17:56):
the higher A UM managers and the smallermanagers where the larger managers are
more likely to leverage their scale andengage in the LME process and try to
come through with a more positive outcome,
something that's morecreative to the recoveries,
the smaller managers would be much morelikely to sell and just get out of the
situation because they may not have toscale to ensure they get our seat at the
table.
(18:16):
And then what's the trends in loanupgrades and downgrades, for example,
today?
So the higher interest rates and slowinggrowth we've seen over the past couple
of years have definitely put pressureon lower rated corporate borrowers,
and some of them at least have beenstraining under the burden of higher debt
service costs. And by late 2023,
the downgrade rate for specgrade companies had risen considerably and peaked at
(18:38):
three and a half downgradesfor each corporate rating upgrade as have October of
last year. And then it'smoderated a bit since,
although it's beenchoppy month over month.
I think at this point most corporate loanissuers have adjusted to higher rates
by cutting costs, differing CapExexpenditures and doing other things.
And then another benefit of the currentmarket is a spec grade companies have
(18:58):
been able to refinance or repricetheir outstanding loans in some cases,
cutting their interest expensesby 50 basis points or more.
So they've got the benefit of tworate cuts just by refinancing their
outstanding debt.
So we're definitely in bettershape than we were say a year ago.
The fed beginning to cut.
Should we infer from that that maybemore upgrades than downgrades are on the
(19:19):
come here?
So you are absolutely right that thedecrease in rates is going to provide a
tailwind for highly leveraged borrowersand it'll free them up at least to some
extent from the burden of interestpayments and allow them to redeploy that
money into other things, including CapEx,
which could benefit growth goingforward. In terms of the ratings,
it's already largely baked in. And whenthe analysts look at a company rating,
(19:40):
they're already taking into account theforward curve of rates and taking that
into account when assigning the rating.
So at the very fact of justseeing the rate decrease,
unless something has changedthe path going forward,
it makes us change our assumptions.
I don't think you would see a largenumber of upgrades based solely on that.
So that was upgrades and downgradesfor the underlying loans and CLOs,
but what about upgrades anddowngrades for CLO securities?
(20:03):
So we haven't downgraded a AAA CLOtrache rating probably since 2012.
It's been a long time.
It really takes a considerable amountof economic stress before that happens.
And I took a quick lookat the past four years,
going back to the beginning of 2020 atwhat the CLO trache ratings were doing,
and obviously 2020 giventhe pandemic, the shutdowns,
(20:24):
the impact on spec rated corporateissuers was pretty significant.
And there were a lot of downgrades in theCLO collateral pools and the liability
ratings followed.
So something like 13% of B-S-L-C-L-Oratings from s and p got lowered
in 2020 and a lot fewer middle marketers,
1% middle market ratingsgot lowered that year,
but every year since from 2021 onward,
there've actually been more CLO trancheupgrades rather than downgrades.
(20:47):
So I think it's a pretty good record,
especially given the level ofcorporate downgrades in 2022 and 2023.
The CLO managers were able to repositionthe portfolios in most cases and go out
against the downside and just thestructural mechanics protected the
transactions and they've done prettywell in the course of a stressed economic
period.
So transitioning from CLOupgrades and downgrades,
(21:08):
the more important thing tome would just be defaults.
So have we seen an uptick in CLOsecurity defaults since the covid period?
The short answer is not really.
We're just about to update a pieceand show that we've had 61 CLO tranche
defaults through both the 1.0 and the2.0 era that the asset class is large
enough now that every year you're goingto see a small handful of double B
(21:30):
tranches that might default.
Usually these are right now what we'reseeing is transactions that originated
before the pandemic and then sufferedagain in 2022 and 2023 and just weren't
able to cover, at leaston a projected basis,
the debt outstandingat the double B level.
But 61 traunch defaults out of 18,000or so rated tranches over the past
30 years is a pretty good record.
(21:52):
Agreed. So the default statisticsthat you just mentioned,
how did those compare to the loan ratings?
Does a double BCLO,
is it comparable to a double B corporateor how should we think about the
ratings between the two?
The short answer is CLO ratingshave that done very well,
and in every case perform with fewerdefaults and comparably rated corporates.
(22:15):
You have to be a little bit careful inhow you do it, because fundamentally,
if you're looking at CLO ratings,
they're originated and thenseven or eight years later,
or maybe sooner if it gets reset orwhatever, they're not there anymore.
With corporates, you have ratingsoutstanding for a long time,
but Meredith Coffee from the LSDApublished a piece back in July of
2022,
actually using some s and p data andcomparing the default rates for CLO
(22:37):
tranches against the default rates forrated corporates. And if you take a look,
for example, at double Bfor CLO 1.0 transactions,
the double B default rate for CLO tranchesrated double B was something like 4%
for CLO 2.0. So far the double B traunchdefault rate has been well under 1%.
And then if you look at corporateratings over the same period,
it's more like 9%. So there's a prettysignificant difference between those two.
(23:01):
But you're quoting there thecumulative default rates.
How I would think about it is like anannual default rate. So if the cumulative,
you said at 4% for the one point ohs,
so maybe if they were outstanding forseven or eight years, call it eight years,
then it's a 50 BIP default rate per year.
And then for say the two pointohs, you said 1% cumulative.
(23:21):
Well,
how many years do you think those doubleB's were outstanding for maybe five,
something like that? So it's 20 bips peryear would be the default rate there.
Yeah, for an individual transaction, ofcourse it's going to be extraordinary,
unlikely you'd see a default in year one.
But if you want to look acrossthe universe of tranches,
you could look at it thatway. Any way you stack it up.
CLO tranche ratings come in prettywell below the comparable corporate.
(23:45):
So one of the questions I'mfrequently asked, I tell them, Hey,
the double B defaultrate has been quite low,
and their responses well are aroundthe precipice of a massive default wave
is the future can be verydifferent from the past.
And maybe embedded in that is thequestion about higher base rates.
But I think your answer on upgradesand downgrades answers that question.
(24:08):
If the CLO securities are, asyou said, for the most part,
upgrades is the trend, not downgrades.
That's true. I mean, even duringa stressed economic period,
if you want to get a sense of how theCLO ratings might respond under different
levels of economic stress, weactually publish a series annually,
a series of stress tests,
and the DSL iteration of that shouldbe out this month or maybe early next
(24:30):
month.
We take all of the CLO tranche ratingsand run them through basically four sets
of stresses with successively,
more collateral defaults andsuccessively bigger triple C baskets.
And then we look at the impact forour rated universe of these stresses.
So if you want to see what wouldhappen to the bbs, for example,
under a scenario where 20%of loans default and the triple C baskets blow up to
(24:51):
40%, as unlikely as either of thosethings would be, you could do that.
So Charlotte, if I could ask you acouple of questions and turn the tables,
how do you use ratings into yourinvestment process or processes?
At Flatrock?
We're not really ratingsconstrained. So we own CLO equity.
Obviously CLO equity is non-rated,
(25:12):
so we'd like the CLO debt securitiesin the CLO to be rated well,
but we don't have a regulator ora capital charge regime. Well,
our regulator is FINRA and the SEC,
but we're not an insurance company or abank where we need to maintain a overall
rating or portfolio quality.So for equity, you're not rating those securities,
(25:32):
so we make our own judgments.And then for double B,
your opinion on the credit qualityof the node is certainly important,
but we're doing our ownscenarios and work ourselves.
So you said your for stress cases, we'vegot a bunch of stress cases as well,
but at the end of the day,sometimes we would buy double Bs.
That would be a weaker end ofmaybe where you guys would rate,
(25:54):
and we would do that because they're cheapin the market and we think they offer
good returns. Maybe they're later intheir life and they don't need all the
equity subordination thatthey might've had originally.
If a double B is upgraded, which I don'tthink we've seen too many of those,
I don't think it really matters that muchin terms of trading level because it's
funny.
One difference in CLOs for other marketsis that when a CLO O'S security is
(26:17):
issued as double B, I think the industrynomenclature just keeps it as double B.
That's where it was initially rated.
That's the part of thecapital stack that it is.
And if you guys upgrade it to triple B,
I think people would continueto call it the double B.
And if it was downgraded a single B,
maybe somebody wouldstill call it a double B,
but acknowledge that there's beensome losses on the loans ratings.
(26:41):
It's very important part of the market,but it's really affecting, I think,
especially the guys who are playing inthe investment grade parts of the stack.
So the AAA investor needsthat rating for the bank,
wants to make sure that rating ismaintained through the life of the CLO.
But some of the debts downgraded in oneof our deals definitely means there's
been some par losses onthe loans. But for us,
(27:01):
that could just mean we boughtthe equity cheap in the past.
It wouldn't necessarily correlateto how we're doing down the stack.
Interesting.
And how constructive are you on OS justas an asset class say going into next
year? I mean,
how do you see the combined impacts ofrate cuts and either economic growth or
lack of economic growthimpact in the asset class?
(27:22):
So I think one of the things that willchange now with the Fed beginning,
what we think it's going tobe this cutting cycle here is just that the relative
attractiveness of double Bs has reallybeen pronounced over the last two years.
So the Fed hiked a lot, andyou could get at least in,
let's talk about middle market CLOs forexample. The yields on those securities,
(27:42):
CLO middle market bbshave been in the 13% area.
So a lot of people looked at thatand said, Hey, that's a great return.
Why would I want to take firstloss risk and be in the equity?
So some of 'em made thatrotation or that new allocation,
and then now as the Fed cuts, the yields,
the double Bs are going to be goingdown one and then two for CLO equity.
(28:04):
The projected returns there are not goingto look that different even though the
Fed is cutting. And the reason is thatI think most people who sit in my seat,
they're running an IRR calculationthrough the cash flows that the CLO equity
is supposed to produce oris expected to produce.
So there's already baked intothat a declining SOFR curve.
So there's already a budget there.
(28:26):
If you're buying equity today and you'retargeting mid to high teen returns,
you're budgeting for that decline inSOFR over time. Whereas for the double B,
those payments are made quarterly andthey reset quarterly and they're going to
be resetting lower, and that's going toresult in less cashflow to the double B.
So my point in that is just I think you'regoing to see a rotation out of double
(28:46):
BCLO equity is going to becomerelatively more interesting.
And I think the two things that maybenefit CLO equity here are that one,
the lower interest expenseburden for a lot of companies,
it's not going to matter.The loan's money good,
and maybe you'd rather havethe higher base rate there,
but some weaker borrowers in the marketwill actually benefit and have more
(29:07):
cash, more liquidity, better interestcoverage. And for CLO equity,
that could mean a marginal difference interms of default rate more favorable to
us. So I think that's a potential benefit.And then the other part of that is,
again, if you're targeting mid tohigh teen returns for CLO equity,
as the fed cuts,
other competing asset classes are goingto offer lower comparative returns and
(29:31):
CLO equity the rates, wedon't see them coming down.
So on a relative value basis,
I think it's going to make equity lookmore appealing than maybe it has over the
last year and change.
Interesting.
Well, Steve, thanks so muchfor coming on the podcast.
Really enjoyed the conversation.
The content here is for informationalpurposes only and should not be taken as
(29:54):
legal business tax or investmentadvice or be used to evaluate any
investment or security.
This podcast is not directed at anyinvestment or potential investors in any
Flat Rock Global Fund definition.
Section A UM refers toassets under management,
LMT or liability managementtransactions are an out of court
(30:16):
modification of a company's debt Layeringrefers to placing additional debt with
a priority above the first lien term loan.
The secured overnight financing rate,
SOFR is a broad measure of thecost of borrowing cash overnight,
collateralized by treasury securities.The global financial crisis,
GFC was a period of extreme stress inglobal financial markets and banking
(30:40):
systems between mid 2007 and early 2009.
Credit ratings are opinions aboutcredit risk for long-term issues or
instruments,
the ratings lie on a spectrum rangingfrom the highest credit quality on one end
to default or junk on the other.A AAA is the highest credit quality A
C or D,
depending on the agency issuing therating is the lowest or junk quality.
(31:04):
Leveraged loans are corporate loans tocompanies that are not rated investment
grade broadly. Syndicated loansare underwritten by banks,
rated by nationally recognized statisticalratings organizations and often
traded by market participants.
Middle market loans are usuallyunderwritten by several lenders with the
intention of holding the investmentthrough its maturity spread is the
(31:27):
percentage difference in current yieldsof various classes of fixed income
securities versus treasury bondsor another benchmark bond measure.
Our set is a refinancing andextension of A CLO investment.
EBITDA is earnings beforeinterest, taxes, depreciation,
and amortization.
An add back would attempt toadjust EBIT DA for non-recurring
(31:51):
exchange traded funds. libor,
the London Interbank offer ratewas replaced by software on
June 30th, 2024.
Deliver means reducing the amount ofdebt financing. High yield bonds are
corporate borrowings ratedbelow investment grade that are usually fixed rate
and unsecured default refers tomissing a contractual interest or
(32:14):
principle payment.
Debt has contractual interestprinciple and interest payments,
whereas equity representsownership in a company.
Senior secured corporateloans are borrowings from a company that are backed by
collateral.
Junior debt ranks behind seniorsecure debt in its payment priority.
Collateral pool refers to the sum ofcollateral pledge to a lender to support
(32:39):
its repayment.
A on-call refers to the time inwhich a debt instrument cannot be
optionally repaid.
A floating rate investment has an interestrate that varies with an underlying
floating rate index.General disclaimer section,
references to interest rate movesare based on Bloomberg data.
(33:00):
Any mentions of specific companies arefor reference purposes only and are not
meant to describe the investment meritsof or potential or actual portfolio
changes related to securities of thosecompanies unless otherwise noted.
All discussions are based on USmarkets and US monetary and fiscal
policies.
Market forecasts and projections arebased on current market conditions
(33:24):
and are subject to change without notice,
projections should not beconsidered a guarantee.
The views and opinions expressed by theFlat Rock global speaker are those of
the speaker as of the dateof the broadcast and do not necessarily represent the
views of the firm as a whole.
Any such views are subject to changeat any time based upon market or other
(33:45):
conditions,
and Flat Rock Global Disclaims anyresponsibility to update such views.
This material is not intended to berelied upon as a forecast, research,
or investment advice.
It is not a recommendation offeror solicitation to buy or sell any
securities or to adoptany investment strategy.
Neither Flat Rock Global nor the FlatRock Global Speaker can be responsible for
(34:10):
any direct or incidental loss incurredby applying any of the information
offered.
None of the information provided shouldbe regarded as a suggestion to engage in
or refrain from any investment relatedcourse of action as neither Flat Rock
Global nor its affiliates are undertaking.
To provide impartial investment advice,
act as an impartial advisor or giveadvice in a fiduciary capacity.
(34:34):
Additional information about this podcastalong with an edited transcript may be
obtained by visiting flat global.com.
Popular Podcasts
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com