1. Fed forward guidance, or not?

2. Revisiting Apollo’s epic short of First Brands

3. Significance of Thoma Bravo’s Proofpoint refi

4. PIMCO’s take on BDC default trends

5. Touching on Mark Walter’s insurance mess

6. BDC tender offer follow-up discussion

7. Notable deals

8. Bulletin board

Disclaimer: there are 4,600+ of us here discussing private credit and leveraged loans, which is not a big world. At some point, I will probably throw punches at your company. I am blunt and sarcastic and, as friends like to remind me, too mouthy to live to an old age, so my take likely won’t please you. But don’t worry, next week it will be someone else in the spotlight and you will quietly nod along in satisfaction.

Also, none of this is investment advice.

1. Tinfoil Fed

Alright, so Fed Chairman Warsh scared the heck out of everyone when he took over the throne, suggesting we shouldn’t expect much in the way of forward guidance anymore. Then, in his Jackson Hole speech on Friday, he still didn’t spell out what to expect per se… but he did say the Fed has “work to do” on inflation, reiterated that 2% remains a “firm, fixed target,” and called interest rates the “predominant tool.” I mean… I’m terrible at reading clues, but even I can’t help but put these pieces together, you know?

But that’s what the more or less serious part of my brain thinks. Then there is the less serious part, which for some reason went full conspiratorial/tinfoil-hat mode and spotted an odd choice of words in Warsh’s speech, where he brings up anecdotes about Kohn and Bernanke:

Listen, I know it sounds crazy, but either he is a troll who wanted to mess with the algorithms, or we are going to get two hikes this year, in which case I’ll become famous for this boneheaded take!

Both scenarios sound equally entertaining, though…

TIME OUT

If your company is in the SF Bay Area and/or NYC and has a suitable space that can accommodate up to 100 people for an event with PEs, LPs, lenders, lawyers, etc. (basically, our ecosystem)… reach out. And if your company doesn’t have one but is feeling particularly generous and wants to rent a space for the benefit of such a group… also reach out.

Aznaur.Midov@KierLior.com

2. Apollo: Shorting First Brands

Alright, this piece started as a reaction to a Financial Times (Alphaville) article about Apollo’s short of First Brands, with a bunch of thoughts I wanted to share. But while researching it, I dug into the court filings and ended up spending the next few hours reading through them with my jaw on the floor. I will need to write another piece on that, otherwise this letter would be far too long. For now, here is a quick tidbit from the court docs.

None of it is particularly surprising, but the fact that it started in 2018 is, at least to me. I mean, it is pretty impressive that management was able to mislead a couple of dozen lenders for almost a decade. There are always some experts who will shake their heads and say something like “I’m not surprised, lenders are dumb.” I know because I’ve seen plenty of this kind of commentary, and I used to think the same way when I was younger. I had read a couple of books on investing and listened to Munger and Buffett make similar arguments, so I thought I knew better.

Then I spent years working in the industry and meeting all kinds of people, and one category I have not really come across is “dumb.” Sure, there were some questionable characters, but they usually disappeared from the institution before we met again. And you can certainly bet that no institution has a management team full of dumb people. We’ll get back to this point.

Anyway, the Financial Times got hold of a few slides from an Apollo internal presentation in which they make the case for shorting First Brands’ debt. As a reminder, Apollo and Diameter Capital (minority-owned by Apollo) bet against the debt via CDS and obviously got it right. So, the slides seemingly show the logic behind the bet.

It’s probably hard to read, but there are a few key points covered in this particular slide:

1. The cash flow of the business has never matched the Company’s claims about its margins nor its ability to drive cost savings in the business;” (b) “It’s not uncommon for management teams/sponsors to exaggerate run-rate impacts, synergies, etc., but in the case of First Brands, it is not just pro forma numbers that don’t make sense”; (c) “The Company’s income statement paints a picture of a company with 20% operating margins, which is difficult to believe for this business, especially when First Brands has been acquiring sub 5 and 10% EBITDA margin businesses.

2. Company is 100% controlled by CEO Patrick James ... The CEO has controlled dozens of entities over the years, many of which have ended up in litigation or liquidation (or both). Simple Google searches reveal almost nothing about him.

The Financial Times provides some commentary on the subject.

So, was it some ninja corporate finance skills that allowed Apollo to profit from the First Brands debacle? According to these slides, Apollo’s edge came from a) knowing how to do a Google search; b) understanding how the balance sheet, cash flow statement and P&L are linked; and c) simply observing that the company’s group margins didn’t resemble those of the acquisitions that they were making.

FT’s Alphaville is being sarcastic, and it oversimplifies what actually went into the decision. The issues discussed above and in the slide may have been enough for Apollo to walk away from the deal, but going short required something much more concrete. First Brands is not Oracle, where CDS trade over the counter and can be bought by any qualified investor. Apollo had to find a counterparty willing to structure and sell them CDS in the first place. It’s similar to the scene in The Big Short where asks Goldman to create CDS on mortgage bonds. And because these were bespoke contracts, finding a counterparty was only one problem; the significant premium these CDS would have carried was another.

In fact, FT itself covers a couple of other points that carry more weight.

For one, Apollo wasn’t simply another lender that received a package from the lead arranger, Jefferies, glanced at the P&L and thought, “OMG, 20% EBITDA margin, short it now!!!” Apollo had already spent time digging into First Brands’ financials as a loan participant through one of its subsidiaries, Gulf Stream, which happened to have exposure to First Brands, although it exited that position well before things went south.

There was another angle as well, which comes up in the court filings. Katsumi, one of First Brands’ trade finance providers, tried to offload some of its exposure to Apollo, but Apollo declined after reviewing the opportunity. There is also a fascinating detail about Katsumi that appears in the filings and was highlighted in the FT piece:

On September 22, 2023, Katsumi asked First Brands to provide copies of seven invoices Katsumi had already funded, as part of an audit. On September 25, 2023, a First Brands accounts receivable analyst based in Romania sent Katsumi the requested invoices without funneling the response through Andy Brumbergs or Edward James [ed: top First Brands executives who have since been charged with fraud].

Katsumi compared these underlying invoices to the amounts it had actually funded, and the discrepancy was unmistakable: not one of the seven sampled invoices matched. For example, an invoice actually worth $240.35 had been funded by Katsumi as if it were worth $434,997.58, an inflation of more than 180,000%.

I mean, I can’t add much to this, but if Apollo reviewed collateral that remotely resembled what’s discussed above, it’s no wonder they passed on the deal and probably started putting together that infamous thesis.

There is also another point that isn’t discussed in the article or the filings, but may have played a role here. As you can see on the slide, Apollo was on a DQ (Disqualified Lender) list. That means the credit agreement specifically prohibited existing lenders from selling their loans to Apollo and a number of other firms on the list. There is nothing unusual about DQ lists, and lenders end up on them for all sorts of reasons. Some because they are difficult to deal with, others because there is some kind of conflict of interest. We don’t know why Apollo was on the list (although we can probably guess), but we do know that Apollo owns Tenneco and ABC Technologies, some of First Brands’ largest competitors. You can bet that’s how management justified not wanting Apollo as a lender.

The point I’m getting at is that if you are a lender in a syndication, you are largely limited to the information that every other lender receives. But if you compete with a company selling niche products, you have a much better sense of what the numbers should look like, and you are likely to hear things that lenders don’t.

And if you are Apollo, which owns Tenneco and ABC, and is already a bit suspicious of First Brands, you can call your portfolio companies and say: “Hey, am I crazy, or do these numbers not make sense?” You might hear back: “They definitely don’t make sense, and also my buddy who used to work for First Brands says there’s something fishy going on.” At which point you gotta go: “OMG, short it now!!!” Well, that’s my speculation, but if I thought of it, make no mistake, Apollo thought of it too.

At this point, you might be wondering: “OK, so lenders clearly didn’t have that extra layer of information that Apollo did, but why didn’t they simply pass on the loan when they saw those fat, irregular margins, the shady CEO with a history of litigation, etc.? Because they are dumb, right?”

No. They are in the business of originating loans. When Jefferies brings you a rated loan with attractive spreads (and First Brands’ loans had those) and the deal is oversubscribed, you are fighting to get into the syndicate. That’s how the business works. I know this looks dumb with hindsight, but in real life, frauds, especially frauds on this scale, are so incredibly rare that even considering the possibility is a little nutty. Read my pieces on Credit Culture and Liquid vs. Private Credit, which provide some additional color on how this business works.

“OK, they got into the loan, but couldn’t they notice that something was off and start asking management some tough questions?”

Well, speaking of which, here is an excerpt from the court filings:

Jefferies and its affiliates, including LAM [Leaukadia Asset Management / Pint Bonita], are another example of a party that conducted diligence and uncovered irregularities and concrete examples of the fabricated invoices. Specifically, on January 13, 2023, LAM sent First Brands a series of due diligence questions related to LAM’s existing accounts receivable facility, including the aggregate dollar amount of all inventory financing, accounts payable, and accounts receivable securitization facilities in place at First Brands. LAM also requested that First Brands schedule an onsite visit for LAM’s annual review, which it noted was “past due,” and asked for a call with First Brands’ CFO. Rather than simply provide the requested diligence information, Edward James pushed back on the scope of LAM’s requests. In a response on January 14, 2023, he wrote that he was “surprised by the length of the questions” LAM had posed, and reminded LAM that Jefferies had been “rewarded handsomely” for its role in FBG’s prior transactions. He added that he would “need to keep any future recommendations in mind” if LAM expressed concerns about providing additional factoring capacity.

I used to operate in the lower middle market, where deals were small club deals with two or three lenders, and I’ve experienced the agent calling to kindly let me know that the CFO was complaining that my team was asking too many questions. It’s annoying, but you know, you limit the number of questions next time.

Here, we’re talking about a completely different scale, with 20 lenders on the deal, and you can bet their follow-up questions weren’t even answered.

Again, most of the strange things you might find in financials are easily explained away by management, and there is nothing illegal or shady about it. Most of the time, everything turns out just fine. Well, 0.0001% of the time, things blow up, and I get to feature you in my letter, you know?

“OK, what about the arranger, Jefferies? It was closest to the deal, so how could they miss it?”

As the arranger, Jefferies was getting paid 1-2% on these deals, and I’m sure it was at the higher end here because First Brands needed an arranger that knew not to rock the boat. At filing, First Brands had about $5B of term loans (not including off-balance-sheet debt), which we can fairly assume were arranged by Jefferies. So, say $100m in fees, right?

Well, First Brands used that debt to fund acquisitions, so every time it bought a company, it needed to hire an investment bank to represent it, generating another round of fees. If I had to speculate, I think I know which IB was getting that business, so we can probably safely double our estimate.

In other words, Jefferies had roughly 200 million reasons to keep its long-term client happy. The numbers are purely speculative, but the underlying motivation is just common sense.

One last thing. This piece is ultimately about Apollo’s short. Apollo and Diameter had been carrying that expensive CDS for about a year, and it was written against the existing First Brands debt. That debt was supposed to be refinanced in late summer 2025, which would have caused the CDS to expire. But lenders began asking questions about First Brands’ off-balance-sheet financing, and the refinancing stalled. The debt was still trading at 95 cents, suggesting that the market viewed the delay as temporary and expected the refinancing to go ahead once the questions about the financials were resolved. Then, in September, the Financial Times published its story about Apollo’s short position, and the loans fell to 80 cents. By the end of the month, they were trading in the 40s.

Can you imagine how lucky Apollo was that the FT published that story at that time?

3. Thoma’s 40 Little Gifts

Bloomberg’s Thoma Bravo Conceded 40 Deal Sweeteners as Debt Talks Heat Up is not a headline I’d pass on. The piece focuses on Proofpoint, Thoma Bravo’s cybersecurity portfolio company, which recently extended $4.3B of syndicated debt by two years, from 2028 to 2030 (Bloomberg says $5B, but that includes $700m that was not extended). The significance is obvious: a software loan, the 2028 maturity wall discussed in the press, and a well-known sponsor in Thoma Bravo.

Before we discuss the piece, it would be useful to discuss Proofpoint. To start, cybersecurity is one of the software verticals generally considered relatively resistant to AI disruption, at least for now. Proofpoint serves 80% of Fortune 500 companies, providing email security, which may not be as mission-critical as ERP (Oracle, SAP), but is up there, meaning revenue retention (a key metric for investors and lenders) is very strong. Its ARR in 2025 was $2.5B and grew “double digits” (likely low teens), based on its “press release” right after SaaSpocalypse. Unlike many SaaS companies, especially in the lower middle market, Proofpoint has a sizable EBITDA of ~$800m-$900m, based on my back-of-the-envelope calculation using Fitch’s rating analysis. We’re talking about a real-deal company here, folks.

Now, Fitch rated Proofpoint “B,” which very comfortably places it in junk territory without much hope of escaping. The obvious reason is leverage. It has $4.3B of first-lien term loan, which was just extended, another $700m first-lien term loan (not extended yet), and a $1.2B second-lien term loan. There is also a $300m unfunded revolver. Fitch projected 2026 leverage at 7.1x (hence the junk rating), which is how I reverse-engineered the EBITDA ($6.2B funded debt / 7.1x = $873m).

Okay, so we’ve got all we need to check the Bloomberg piece.

Thoma Bravo executives went into talks with lenders knowing they’d have to give ground — the only question was how much. On the other side were some of Wall Street’s biggest money managers — names like BlackRock and Invesco — and they were well aware the balance of power had shifted in their favor.

The final tally: some 40 creditor-friendly changes to the deal documents, including tighter limits on future borrowing and regular check-in calls, as well as an extra $60 million in annual interest to extend most of the $5 billion loan by two years.

The piece goes into some detail on the creditor-friendly changes, most of which are baskets within the credit agreement that are typically used as negotiation points anyway. That said, having Thoma agree to tighten the credit agreement on a high-quality company, even if junk-rated, is a strong signal.

Bloomberg noted:

Lenders said that two of the biggest concessions won were a so-called omni blocker and mandatory quarterly calls.

Still, Thoma held the line on one important point: the debt remained covenant-lite. That means Proofpoint doesn’t have to meet regular financial tests tied to measures such as leverage or earnings in order to remain in compliance with the loan. Thoma also didn’t have to put any new equity into the company.

A couple of things stand out here.

First, Thoma essentially gave up its ability to LME the lenders through the omni blocker and agreed to quarterly calls with management. While that may be a major concession in the syndicated loan market, it is fairly standard in direct lending. In a sense, Thoma went from a piece of paper saying “Proofpoint owes money to these lenders” to a credit agreement with some actual protections for lenders.

Second, the absence of financial covenants shows there was still significant room to tighten the credit agreement, meaning Thoma retained quite a bit of leverage in the negotiations. The fact that Thoma was not required to inject equity further supports this. An equity contribution could have been used to pay down debt if liquidity was sufficient.

Although Thoma does not have a reputation for LME-ing lenders, it almost feels like lenders wanted to address the LME provisions in the credit agreement now, and were willing to accept that trade-off in exchange for a 2-year extension. This looks like a win-win for both sides.

One less obvious but very interesting part of the amendment is the extra $60m of interest expense. The Financial Times, in a piece about the transaction, noted:

The new debt would yield about 9.3 per cent, or an extra 1.5 percentage points above the secured overnight financing rate compared with its existing debt

A 1.5% rate increase is massive in the BSL market and is essentially equivalent to the premium for illiquid direct loans. Using current rates, the new debt appears to be priced at roughly SOFR + 5.50%, which is private-credit-type pricing.

Bloomberg adds:

Market watchers say the transaction could serve as a blueprint for other private equity firms, which collectively have more than $55 billion of broadly-syndicated software debt coming due over the next couple years.

I sort of agree, although it is worth noting that Thoma’s reputation as a premium software-focused sponsor and Proofpoint’s strong position in the sector clearly helped, so these terms will not be available to most deals. But they don’t have to be, because what we needed was for deals to start moving, and the media coverage helps bring awareness to that. More hairy deals will come with tighter structures, including financial covenants and equity infusions, so I think we are seeing movement in the right direction.

4. Defaulting the Definition

I like reading PIMCO’s The Credit Market Lens letter because it offers a useful perspective on private and liquid credit that differs from what you typically read in alts. Last week’s newsletter, Narrowing the Visibility Gap in Defaults, focused on a familiar topic: defaults across private and public credit markets.

Specifically, PIMCO estimated the BDC default rate and compared it with default rates in BSL and HY. This was obviously a tricky exercise because, unlike BSL and HY, BDC defaults are not tracked by rating agencies but are instead negotiated privately between lenders and borrowers (something I covered in my previous letter).

PIMCO then tried to make an apples-to-apples comparison across the three leveraged credit markets (BDCs, BSL, and HY) by using a broader, common definition of default that includes certain default-type events. They admit the comparison is imperfect, but believe it is still useful enough to share the results.

The chart basically shows that BDCs’ estimated default rate is about 16%, although it was around 18% fairly recently. The finding is quite shocking, despite all the noise coming from BDCs.

If your associate brings you this chart, the first thing you should ask, before even looking into how “default” is defined, is:

“Wait, how the entire industry was able to operate at a 10%+ default rate for a decade and not only didn’t blow up, but generated equity-like returns.”

The answer you will receive is some variation of the following:

“Hmm… yeah… let me double-check the data. Maybe there is a formula break.”

Well, the chart was published, so clearly the data was accurate. Which raises the question of how defaults were defined, because the result makes no sense. I mean, “default” is one of the scariest words in credit, but if you are telling me that on a good day, 1-in-10 loans are in default, we need to adjust the definition of “scary.”

The report suggests that PIMCO was focused on five types of event defaults (no double-counting) and provides the breakdown of their contribution to the total default amount (see chart).

1. Default payment

2. Non-accrual (i.e., collection of contractual payments is deemed doubtful)

3. Post-origination cash-to-PIK conversion (PIK optionality at origination is excluded)

4. Material maturity extension of the original loan

5. Debt-to-equity swaps

Payment defaults and non-accruals are no-brainers. “Bad PIK” has some exceptions, but it is fair to include it. The debt-to-equity swap is also non-controversial, but the chart showing more swaps than payment defaults and non-accruals combined is suspicious, because a debt-to-equity swap is logically a last-resort option. I have a strong feeling the chart may have the colors mixed up, but without granular data, we just have to take it as presented.

The maturity extension is where I have an issue, because it sounds logical on the surface but overlooks an important nuance of BDC loans.

Assuming that by “material maturity extension” PIMCO means something like two years (we don’t know), the logic would be that a borrower reaching the end of its loan tenor without repaying or refinancing the debt is under stress and has to kick the can down the road. But that logic misses an important dynamic of BDC loans.

Before diving in, I think it is fair to assume that the maturity extension data was collected something like this: a typical lower middle market loan has a 5-year tenor, and if the loan was put in place 7 years ago, it would be counted as a 2-year maturity extension.

Most of these loans are provided to sponsor-backed companies, and one thing to know about them is that they are very acquisitive (buy-and-build is the main strategy), often using debt to finance acquisitions. During the first 1-2 years, add-on acquisitions typically do not change the initial maturity. So, if the company was acquired in 2020 using a $50m loan with a 2025 maturity, and the next year there is a $10m upsize for an add-on, the maturity would typically remain at 2025 for the combined $60m facility.

However, starting around 2023, every upsize will come with a, say, two-year maturity extension request. Not because the sponsor is worried about the company’s performance, but because they have learned that capital markets are cyclical, and they don’t want to ask for extensions when lenders are in panic mode (boy, were they right to do so).

So, if the determination of the “maturity extension” was as I described, based on the difference between the current maturity and the original maturity, it was not distress that led to it. It was simply the result of sponsors extending maturities as they upsized their loans for acquisitions.

Yet, even if we had a situation where a bunch of 5-year loans were reaching maturity and borrowers were asking for a 2-year extension rather than repaying the loan or refinancing it for another 5 years, there are logical reasons for that as well. The reason not to repay is straightforward: none of these loans were expected to be amortized to zero (they typically have 1-5% annual amortization), and they were not expected to be repaid from operating cash flow. The loans are typically repaid from the proceeds from selling the company or refinance with another lender.

What happens when maturity is approaching but the sponsor wants to wait another 1-2 years, either to complete the integration of a recently acquired add-on or simply to wait out market turbulence from tariffs? The company clearly doesn’t have enough cash to repay the loan outright, but that was never the plan (which is important). Well, they ask for, say, 2-year extension. They are not going to put in place a new 5-year loan, because it would come with, say, a 2% closing fee, another 1-2% early termination fee if they were to sell after a 1-year holding period, plus the need to negotiate another credit agreement if the lender changes. The point is, for most of the last decade, a 2-year maturity extension was not the result of distress, but rather convenience. And it was also in the lender’s interest. You would typically rather retain a yield-producing borrower you know and understand than replace it with a new one.

Of course, it would be disingenuous to say that extensions are never a sign of distress. It is fair to assume that many extensions completed in, say, 2026 were driven by distress or broader lender concerns about the future of the business, paired with broader market sentiment (Thoma’s Proofpoint, anyone?).

I understand that PIMCO had to come up with some sort of criteria for defaults, and there is nothing wrong with classifying maturity extensions as defaults for the purposes of its analysis. It is, however, misleading to treat them the same way as payment defaults, because doing so inflates the number of problem loans (and there are surely plenty of genuine problem loans without adding all these extensions to the count).

5. Insurance, Under Pressure

“Hey Aznaur, why aren’t you covering what’s going on with life insurers and this whole Mark Walter mess?” you may ask. Well, it’s a great question. The answer is that things are moving so fast that by the time I digest what is going on, there are already new pieces discussing new developments in the story, with new parties getting involved. Like, last week we found out that a “CLO” can also stand for Collateralized LeBron Obligations, so I’d rather wait for the full scope of the issue to be revealed instead of revisiting it every week.

I do find the situation very interesting and important to follow because PE-owned life insurers have been considered for some time an overlooked but potentially destabilizing factor in the financial markets. There were a couple of angles that people had on it:

1. PE-owned life insurers loaded up on private credit, which is full of overvalued assets, and that will make it all blow up (not my position).

2. PE-owned life insurers have operating entities that are levered to the eyeballs, often 50x+ (assets/equity), which can blow up whether or not they are invested in private credit (my position).

Both of these angles also share the view that captive reinsurers, which are a signature feature of these PE-owned life insurers, are not transparent enough and add an additional layer of risk.

Apparently, there is a third angle: PE-owned life insurers may not fully disclose the extent of their investments in affiliated entities, which is a big deal because these transactions receive regulatory scrutiny due to the potential conflict of interest.

I’m not going to get too granular on the situation, but here is a very, very high-level overview. As you know, in one of Walter’s insurance companies, those affiliated transactions were reportedly restated from roughly 3% of assets to more than 40%, which is, you know, pretty awkward, to say the least. Walter is now working to reduce those affiliated-party exposures, while also selling assets (LA Lakers among them). And of course, DOJ and SEC are all over the case, which is the only reason why anyone would agree to sell a basketball team. There are reports that more entities may have been exposed than what has been covered in the media, but that’s another rabbit hole we are not going down today

In my view, there are really several things that can go wrong here and that I am watching closely.

1. Life insurance companies are generally believed to have relatively sticky funding sources, particularly annuities. Policyholders may face surrender charges for withdrawing money early, which is why they are generally thought to be less likely to withdraw than bank depositors. That said, if policyholders lose confidence in an insurer’s ability to pay, they could start pulling their money out, leading to a bank-run-like situation. So, I’m watching to see if people actually start pulling money from Walter’s insurers or, more broadly, from other life insurers.

2. A bank-run situation would be obvious, and we are clearly not there yet. What I’m watching instead is whether the negative headlines start keeping new money away. Existing policyholders may stay put, but if fewer new customers buy annuities, we should see it in a sharp slowdown in annuity growth on the insurers’ balance sheets. I think that kind of slowdown could leave them one bad piece of news away from policyholders rushing to get their money out.

3. Separately, I’m watching whether the negativity around Walter’s insurers starts to contaminate sentiment toward publicly traded insurers more broadly.

Regarding #1, one of my subscribers shared this public chat of annuity policyholders, which is some sort of Reddit-type massive discussion board. You can see people discussing withdrawing money from insurers, but mostly those whose annuities have reached maturity and who are trying to avoid the fees. There is one comment where a person is willing to lose 2% or so, but that’s what I saw. It is a small sample, but better than relying on my gut. Based on that, the situation doesn’t look ideal, but is generally stable.

Regarding #2, it is too early to see the impact on insurers’ balance sheets, but Bloomberg just ran a piece suggesting that Truist and Fifth Third have paused distributions from Mark Walter-linked insurers. We don’t know how much this will affect annuity sales, but this is bad. If blue-chip banks are pulling back, smaller distributors will likely follow suit.

Which leads to #3. I think the best way to see how much this issue could spread to the broader industry is to look at how negative and positive events affect public companies in the same industry. There is no specific life insurance ETF, so I created a mock portfolio with a few U.S.-based publicly traded life insurance companies: MetLife, Prudential, F&G Annuities, CNO Financial, and Globe Life. Below is its YTD chart. The first arrow points to July 20, marking the date when Bloomberg broke the news that the DOJ was examining Mark Walter’s insurers.

As you can see, my index initially reacted to the news, but the move quickly reversed. I wanted to believe that was the end of the episode, and the subsequent decline in the portfolio could have been driven by other macro events happening in the world. But that second sharp drop bothered me, so I checked the date, August 17–18, and looked at what events were developing in the industry. My LLMs were pointing out that the major insurance-related events during those two days included reports about the DOJ investigation into Walter’s companies and a selloff in Sammons bonds, a Walter-affiliated entity. The S&P 500 did not show the same drop, so it does look like these events were affecting the broader life insurance industry.

So, what do I make of all this? We are not seeing a bank-run-type event yet, but there are signs that people are pulling money from Walter’s life insurers. I have little doubt that the statutory filings for Walter’s insurers will show at least a slowdown in annuity growth, if not an outright decline. But I am more interested in whether this has affected other players in the industry. We will have to wait for Q3 results from publicly traded life insurers to see how annuities performed. Frankly, I don’t expect a significant change, but it is worth checking.

And yes, if one of Walter’s insurance companies were to fail, it would not be some minor blip. Even relatively mild news, such as the investigations we have seen so far, appears to have moved the stock prices of its public peers. A failure would be a much bigger event.

6. Cox Gets Specific

About a month ago, I wrote a post about Cox Capital’s tender offers for blue-chip BDCs (Apollo, HPS, and Ares), where it offered to buy shares from existing investors at a 15% to 30% discount to NAV. I speculated about the thesis behind the trades and provided some high-level math to illustrate the bets.

A couple of weeks later, I was actually able to interview Cox Capital founder John Cox (must listen!) about the trade. John took me a couple of levels deeper into the strategy, touching on subjects like diligence, catalysts, and more. It was a particularly interesting episode given how unique the strategy is.

After the episode was published, one listener commented with a few follow-up questions addressed to me. And because I’m quite skilled at diverting attacks (a skill I picked up in my prior corporate life), I tagged John on the post and asked him to answer the questions himself. He kindly did.

I thought his answers were very useful, so I’m sharing them below, both for subscribers who have listened to the episode and those who haven’t yet.

Q: Given that many of the underlying BDC portfolio companies are private and have limited publicly available information, how deep can you realistically go when underwriting the underlying credits?

A: The honest answer is that we generally cannot underwrite every private borrower to the same depth as the originating lender. Public filings provide useful information (investment type, coupon, maturity, cost, fair value, non-accrual status, and often borrower identity) but generally not complete financials, covenant packages, or lender reporting.

We do bottom-up work on the larger positions where information permits, but most of the analysis is necessarily conducted at the portfolio and fund levels. We look at NAV movement and mark credibility; concentration and vintage exposure; non-accruals, PIK income, amendments and realized losses; leverage; fees and expenses; distribution coverage; and how the portfolio’s marks and performance compare with public BDCs and other managers holding similar credits.

Limited visibility becomes unacceptable when we cannot get comfortable that reported NAV is directionally reliable, or when the investment return depends on a favorable credit outcome we cannot independently assess. A discount to NAV is a cushion, but it is not a cure for stale or incorrect marks. Greater uncertainty requires a greater discount; at some point, we simply do not buy.

This is also inseparable from the exit analysis. If a listing is a possible outcome, we need to anticipate how public-market investors will value the portfolio, manager, fee structure, leverage and asset mix…not merely whether the underlying loans ultimately repay at par.

Q: This was also touched on, but how do you think about the ultimate exit when buying positions in perpetual-life vehicles?

A: I think “there is no guarantee it will trade at NAV” understates the issue. Many externally managed BDCs trade below NAV consistently. Discounts are more the rule than the exception. Moreover, if sustained liquidity pressure is what ultimately causes a perpetual non-traded BDC to list, that same pressure may create substantial selling overhang at listing.

We therefore do not assume that NAV will be our terminal value. We model multiple paths: issuer repurchases, a third-party secondary sale, a listing or merger, and a longer hold. For each scenario, we consider the cash distributions received, a realistic terminal discount to NAV, and the time required to get there. The purchase price needs to produce an acceptable return across a reasonable range of outcomes…not only if the redemption queue clears quickly or the fund lists at NAV.

The longer and less certain the duration, the greater the entry discount we require.

Q: How do you think about the cyclicality of this opportunity set? The current dislocation seems particularly attractive because redemption requests are exceeding the liquidity that some evergreen vehicles are willing or able to provide. If redemption pressure eventually subsides and repurchase programs can once again satisfy most investor requests, presumably the discount required by sellers would compress significantly.

A: I think the answer is both cyclical and structural.

The magnitude of the current dislocation is cyclical. If net flows improve and repurchase requests decline and proration should ease. Like any secondaries strategy, there will be periods of heavier deployment, periods of harvesting, and periods when the right decision is simply not to invest.

The underlying liquidity mismatch, however, is structural. The vehicle may be perpetual (for now); the shareholder is not. These funds own illiquid assets while providing limited and generally discretionary liquidity. Retail capital is also raised in waves, and secondary supply tends to follow those fundraising waves with a lag. Eventually, some portion of each investor cohort needs liquidity because of age, estate planning, adviser changes, portfolio reallocation, or ordinary cash needs.

They need their capital back because they need their capital back…not because they have developed a nuanced view of an opaque private-credit portfolio.

Accordingly, I expect retail secondary opportunities to persist even in more normalized markets. What will change throughout the cycle is the volume, the pricing, and which liquidity constrained vehicles (feeders, BDCs, REITs, CEF/tender funds, interval funds) offer the best risk-adjusted returns.

Anyway, these are just follow-up questions, so I recommend listening to the full episode.

7. Deals on the Block

Download below: the full Excel file covering deals completed in 3Q26 (plus the prior three quarters).

CLICK to Download 3Q26 Details

KIÉR LIÓR – Providing outsourced private and leveraged loan underwriting and portfolio management to institutional investors, family offices, and direct lenders.

www.KierLior.com

8. Bulletin Board

• Empirasign, a provider of secondary market trading data for structured products, covering ABS, MBS, CMBS, CLOs with BWIC history dating back to 2010. They also carry real-time market data for IG/HY bonds as well as secondary loans. DEBT SERIOUS readers get a 1-month completely FREE trial. Click the LINK and select DEBT SERIOUS during registration

• Bloomberg is offering 60% off its annual subscription ($180 vs $399). If you are in Business, Bloomberg is a MUST, with no real close alternative LINK

• WSJ Online is offering 65% off its annual subscription ($4 per week vs $11) LINK

• The Financial Times is offering a 4-week trial for $1 (total, not per week), but make sure to cancel before it renews to avoid paying $75/month LINK

• Shortcut.AI, an AI tool for Excel that lets you build models by simply typing requests (similar to ChatGPT). It is free to use up to a certain number of tasks per day (LINK). If you hit your limit before finishing your project, simply upgrade to the Pro plan ($20/month) and enter promo code DEBTSERIOUS (all capital letters) to get 50% off

• LoanEdge, a BDC research tool that lets you look into BDC loan compositions, individual loans, which BDCs own a specific loan, and more, is kindly providing DEBT SERIOUS readers with a free 3‑month trial (the regular price is ~$3k per year, and it’s higher for institutions). Here is a LINK to the tool. Send an email to Sadaf Khan at sadaf@theloanedge.com and let her know you are from the DEBT SERIOUS community

As a reminder, I don’t personally benefit from either of these offers. I am just saving you money.

Lastly, our community is over 4,600 strong and includes a broad cross-section of mid- to senior-level professionals across private credit, private equity, LPs, and debt advisory. Please use the Bulletin Board at the end of each weekly letter to share events, discounts, or deals that benefit the community. I will continue to feature these opportunities and personally connect interested parties at no cost.

That’s the bell — round over. See you in the next.

Aznaur

Aznaur.Midov@KierLior.com

News

Articles I found interesting. Doesn’t mean I agree with their takes.

1. Warsh said the Fed will return inflation to its 2% goal, which he called a “firm and fixed target,” and that interest rates are its main tool. Investors raised the odds of a September hike from 36% to 50%+ (Bloomberg)

2. Bessent’s plan to double buybacks of longer-dated bonds was debated, but it is already having an impact, with the 30-year spread narrowing to its smallest since February (Bloomberg)

3. “The worsening fiscal outlook leads to higher risk premia and financing costs. But the way to deal with this is not manipulating markets at the margin — it’s taking steps to curb the nation’s budget deficits” – Op-ed by Bill Dudley, fmr. president of the NY Fed, on Bessent’s interventions (Bloomberg)

4. BlackRock is seeking buyers for TCPC’s remaining $671m loan portfolio and has approached Ares and other firms. TCPC previously sold $523m to a Pantheon-backed vehicle (Bloomberg)

5. Thoma Bravo gave Proofpoint lenders 40 creditor-friendly changes to extend most of its $5B loan by two years. The deal could set a template for other PE firms as lenders gain leverage over software debt (Bloomberg)

6. Apollo confirmed a data breach in which hackers accessed SSNs, dates of birth, and contacts, though it did not specify whose. Google says PE firms have been targeted, and hackers netting up to $750k in ransom (TechCrunch)

7. Evergreen buyout, infrastructure and credit funds are attracting institutional investors, with lower returns than drawdown funds but also lower fees and regular withdrawals. Blackstone and KKR are expanding in the space (Financial Times)

8. PE firms are using structured equity to return cash to LPs while keeping portcos they cannot sell. The deals blend debt and equity, with investors earning mid-teens returns (Bloomberg)

9. European direct lending hit a record €63B in H1, despite slowing in Q2 to €28B from €35B in Q1, putting the market on course for a record-breaking full year (ACI)

10. Mark Walter’s TWG maintains that there has been no “fire sale” or “fraud” at its companies, as the SEC and DOJ continue to probe undisclosed affiliated investments (Financial Times)

11. LeBron James borrowed $300m from two Mark Walter-linked life insurers in 2018 through bonds backed by his future non-basketball earnings. The bonds paid 4.8% and mature in 2049 (Bloomberg)

12. Truist and Fifth Third paused annuity sales from Mark Walter-related Delaware Life amid a US probe. S&P and Fitch lowered the A- insurer’s outlook, signaling a potential downgrade (Bloomberg)

13. Insurers nearly doubled SRT risk cover for banks to €4.7B in 2025 from €2.7B, helping banks free capital for lending; Fitch warns banks could face higher capital needs if insurers stop renewing the protection (Financial Times)

14. Broadcom’s bond yields rose 14 bps and five-year CDS costs rose 28 bps as it backs large AI financing deals. Its guarantees for AI chip financing are raising concerns about hidden balance-sheet risks (Bloomberg)

15. MBS investors could see higher-rate income cut off sooner as AI helps homeowners refinance at lower rates by identifying and reaching eligible borrowers, potentially raising refis to 60% from about 30% (Bloomberg)

16. The First Brands case was converted from Chapter 11 to Chapter 7 after a judge rejected a plan to repay creditors with funds recovered from lawsuits against insiders. The $1.1B DIP loan trades at 16 cents (Bloomberg)

17. Aston Martin £1.3B bondholders filed for discovery over a £550m HPS deal that transferred some of Aston’s most valuable assets, including branding and naming rights, to a subsidiary securing HPS’ loan (Financial Times)

18. Eurozone business lending fell to €22B in July from €34B in May, suggesting the June rate hike from 2% to 2.25% may be affecting economic activity and investment. Investors expect another hike to 2.5% in September (WSJ)

19. Mexico is at risk of a junk downgrade as $130B in government bailouts of Pemex have nearly doubled the budget deficit. Mexican dollar bonds now yield more than debt from lower-rated Guatemala and Panama (Bloomberg)

20. Zambia’s borrowing costs fell in the first bond auction since President Hichilema won reelection, with seven-year yields dropping 101 bps to 14.79%. Local-currency bonds have returned 37% this year (Bloomberg)

Transactions

1. Ares will lead a $2.2B direct loan to MedImpact to finance its planned acquisition of Medical Card System. The deal would be one of the year’s biggest private-credit loans (LA Business Journal)

2. Guggenheim’s affiliate bought its asset management unit’s debt, pushing the $1.2B loan to 84 cents from below 70 cents last week. The $1.2B loan fell after Mark Walter’s Guggenheim held a call with lenders (Financial Times)

3. SoftBank plans to issue ¥1T ($6.3B) of seven-year yen retail bonds, with a 4.3%-4.9% indicative coupon. The bonds will be priced Sept. 4 (WSJ)

4. JPMorgan has begun talks with potential lenders for a $5B debt package to fund Volta Infrastructure’s AI data-center buildout (Bloomberg)

5. Dublin-based AIB Group is working with Santander and Howden on an SRT tied to €2.5B ($2.9B) of project-finance loans, as it expands the use of SRTs to hedge its loan book (Bloomberg)

6. Uplifters Foundation plans to issue as much as $250m in municipal bonds to build new $3m houses in the Pacific Palisades disaster area (Bloomberg)

7. Mexico returned to the Samurai bond market after two years, raising ¥282.8B ($1.77B) before Japanese rates may rise further. The four tranches mature in 3.5 to 20 years, priced at 115-210 bps over yen swaps (Bloomberg)

8. JPMorgan and Santander are leading a $14B-$15B financing for Argentina LNG, a project to liquefy and export the country’s shale gas. The project is estimated to cost $24B (Bloomberg)

Chart of the Week

“The wages of workers in the construction industries and the goods producing industries (mining, manufacturing, construction) has suddenly taken off since late 2025, and their growth exceeds the average service-sector wage growth by a lot. In view of a weakened housing construction activity, this might only be explained by the surfeit of demand for non-housing construction workers” – Macquarie.

• Bloomberg is offering 60% off its annual subscription ($180 vs $399) LINK

  • If you’re not a subscriber and the article is behind a paywall, search for the article title online. It’s often reprinted on Yahoo Finance for free.

• WSJ Online is offering 65% off its annual subscription ($4 per week vs $11) LINK

• The Financial Times is offering a 4-week trial for $1 (total, not per week), but make sure to cancel before it renews to avoid paying $75/month LINK

These are NOT sponsored links, just money-saving tips.

News

Articles I found interesting. Doesn’t mean I agree with their takes.

1. Treasury may raise bond buybacks above $4B per operation after 30-year yields hit 5.3%, near a 20-year high. The move briefly pushed yields down, but half the drop was later reversed (Reuters)

2. Bond traders are hedging the risk of Fed rate cuts in 2027 as U.S. economic data weakens. September hike odds have fallen sharply, while traders buy options that benefit from future cuts (Bloomberg)

3. The AI debt binge is crowding out Treasuries as hyperscalers flood the market with bonds. Treasuries may need to pay higher yields, with big tech borrowing ~$200B, or 25% of Treasury net issuance (Bloomberg)

4. Apollo’s famous short of First Brands’ debt was based on realizing that its cash flow and P&L didn’t add up. FB also overstated receivables borrowing in one instance by 180,000% of the amount it was entitled to (Financial Times)

5. Private-credit BDC fundraising fell 82% to $2B in Q2, while investors sought a record $23B in redemptions. Direct lending is shrinking as managers diversify into IG data centers and other private-credit strategies (Bloomberg)

6. Carlyle is shifting back to Washington under CEO Harvey Schwartz, focusing on government-linked investments. Its DC roots are an edge as defense, infrastructure and supply-chain spending grow (Semafor)

7. Mark Walter built Guggenheim by moving insurers into private assets, but now faces a federal probe over $20B of undisclosed affiliate investments. Walter is selling sports assets and raising billions to shore up the insurers (Financial Times)

8. UK insurers are targeting $1.35T of defined-benefit pension liabilities from pension plans, as higher rates pushed plans into surplus. 40% of retirement assets are in private markets, with one-third in private credit (WSJ)

9. HPS and Oaktree seized MBS Group after its debt default. Lower TV spending and shrinking streaming seasons (from 22 to 10 episodes) hit the film-equipment supplier; creditors converted $100Ms of debt to equity and injected $40M (Financial Times)

10. CD&R’s aggressive Multi-Color restructuring has damaged its reputation as a creditor-friendly sponsor. Lenders to its other portcos are now organizing into cooperation groups and pushing for tighter protections (WSJ)

11. Erebor Bank, founded by Palmer Luckey and backed by Peter Thiel, aims to replace SVB but risks repeating its concentration on tech clients. Its tech, defense, crypto and AI focus could amplify losses in a downturn (Bloomberg)

12. Europe’s banks are pushing the EU to ease Basel rules as U.S. deregulation what puts them at a competitive disadvantage. The changes could unlock hundreds of billions of euros in capital and liquidity (Bloomberg)

13. Bond investors are wary of $70B of off-balance-sheet AI liabilities as Nvidia, Broadcom and Meta backstop AI debt. Rating agencies warn these guarantees could become costly in a downturn (Bloomberg)

14. SRT issuance reached $18B in H1 2026 and is on track for a sixth straight annual record. Manulife CQS is raising about $1B for its fourth SRT fund, targeting a ~13% IRR after the prior fund returned over 11% (Bloomberg)

15. European direct-lending volume rose to €9.8B from €8.5B in the prior three months. Software loans widened to E+525-550 from E+475 six months ago. €4.8B of direct loans were refied into BSLs, the 2nd-highest level (PitchBook)

16. Canadian institutions hold C$500B ($360B) of private credit, mostly in the US; pensions and insurers hold most of it. Private credit is 15% of Canadian corporate credit, vs. 75%+ from banks and debt markets (Bloomberg)

17. India’s private credit market is expanding into mid-market financing, with $10m-$60m deals reaching 61% of deal value as borrowers seek alternatives to banks. Domestic funds account for 74% of deal value (Economic Times)

18. Emerging Markets inflation-linked bonds are up 11.1% in 2026, vs. 1.6% for local bonds. The $886B market is benefiting from sticky inflation and stronger currencies, with Brazil and Mexico over half (Bloomberg)

Transactions

1. Anthropic is increasing its revolver from $2.5B to $10B as it prepares for an IPO. Bookrunners are expected to provide $1.25B each, followed by $1B and $750M from other participating banks (Bloomberg)

2. Alphabet raised A$5.5B ($3.89B) in Australian debt, becoming the first AI hyperscaler to tap the market. The bonds run 3, 5, 10 and 20 years, with a 6.9% coupon on the 20-year bond and A$18B+ in bids (Reuters)

3. KKR-owned Internet Brands plans to issue bonds to refinance syndicated debt, with RBC already sounding out major lenders. Its loans have recovered to near par as business improves after an AI-driven selloff (Bloomberg)

4. Ripple Prime raised $275m via a private placement of senior unsecured notes at an 8.25% coupon, rated BBB by KBRA. Proceeds will fund growth and working capital (FM)

5. Blackstone’s BCRED raised $750m of 5-year notes vs. a $500m target, at a 2.05% spread, 25bps tighter than initial talk. Blue Owl Tech Finance raised $400m vs. a $200m target, at a 2.50% spread vs. 2.65% on its existing bonds (Bloomberg)

6. Franklin Templeton closed its first Collateralized Fund Obligation (CFO), raising $1.5B for exposure to private equity secondaries, continuation vehicles and US middle-market direct lending (Yahoo Finance)

7. Thoma Bravo-backed Sophos is seeking to refi $2B+ of loans with leveraged-loan investors after private credit passed, likely offering higher coupons, amort. and tighter covenants. Thoma rebuffed requests for fresh equity (Bloomberg)

Chart(s) of the Week

1. The secondary market for private fund stakes has seen wider discounts in recent years, with the U.S.-Iran war and concerns over AI’s impact on software and other companies contributing to the increase.

2. Of the 13,509 PE-backed companies in US sponsor portfolios as of June 30, some 3,332 have been held for five or more years and have not completed a deal of any kind since the end of 2021.

   

• Bloomberg is offering 60% off its annual subscription ($180 vs $399) LINK

  • If you’re not a subscriber and the article is behind a paywall, search for the article title online. It’s often reprinted on Yahoo Finance for free.

• WSJ Online is offering 65% off its annual subscription ($4 per week vs $11) LINK

• The Financial Times is offering a 4-week trial for $1 (total, not per week), but make sure to cancel before it renews to avoid paying $75/month LINK

These are NOT sponsored links, just money-saving tips.

John founded Cox Capital Partners in 2020. Before that, he served as Head of Strategy at Alternative Investment Exchange (AIX) and was a Director at Future Standard. He began his career in private banking at Credit Suisse.

In this interview, we discuss the investment thesis behind John’s tender offers, what makes these opportunities attractive, and how he evaluates the risks and returns.

Subscribe to DEBT SERIOUS at https://debtserious.substack.com

In Round #57:

1. NVIDIA and Wall Street’s $500B deal

2. DEBT SERIOUS and Nick Nemeth on ARR loans

3. Software loans update

4. Jane Street’s $15B loss and private credit deal

5. Unexpected CDS spread drivers

6. “Questionable” support campaign for private assets in 401(k)s

7. Notable deals

8. Bulletin Board

Disclaimer: there are 4,500+ of us here discussing private credit and leveraged loans, which is not a big world. At some point, I will probably throw punches at your company. I am blunt and sarcastic and, as friends like to remind me, too mouthy to live to an old age, so my take likely won’t please you. But don’t worry, next week it will be someone else in the spotlight and you will quietly nod along in satisfaction.

Also, none of this is investment advice.

1. Jensen’s $500 Billion

So, I guess not enough money is flowing into AI, so Nvidia and several alts decided to team up on a new $500B financing initiative for compute.

Bloomberg provides some color:

In doing so, he [Jensen Huang] was seeking to assure Nvidia’s investors that there are plenty of deep-pocketed firms ready to finance his clients, particularly AI startups such as Anthropic PBC and OpenAI that are key to Nvidia’s future demand. While he’s bullish on AI spending overall, his company has been seeking to broaden its customer base beyond hyperscalers including Microsoft and Amazon, many of which are trying to create their own components.

So, Nvidia found itself in an interesting situation where it is at the center of this AI boom, but many of its customers are growing too fast to keep financing GPUs and the infrastructure around them at the current pace. That is obviously bad news for Nvidia, whose revenue growth would decelerate if that were to happen.

So what did they do? They reached a deal with Apollo, KKR, Blackstone, BlackRock, Brookfield, and Goldman Sachs, with these firms providing financing for AI infrastructure, including Nvidia GPUs. And on top of the lucrative paper and fees, Nvidia will backstop up to 25% of the financing, effectively helping absorb some of the risk.

Last week, all the participants joined Becky Quick on CNBC to discuss this union for 35 minutes. I’ll be honest, it was somewhat uncomfortable to watch, and then I figured out why. It’s like a CEO of a firm you’ve never met (Becky) suddenly gathers the MD (Jensen) and VPs (alts) to walk through a deal you’re all working on, but nobody had time to prepare. One of them even dialed in (Larry Fink), and, as you know, there is always someone who couldn’t be in the room but somehow needed to be present (even though the contribution is questionable).

The MD opens the presentation, the VPs reluctantly answer follow-up questions, constantly glancing at him for approval like they’re waiting for a nod. Then they loosen up and start offering uninvited commentary, throwing around stats about how big the market is, how amazing Nvidia is, how many cool projects they’ve done over the years. They are basically trying to impress the MD while making sure they don’t screw up in front of the CEO. I’m pretty sure Jensen told them “Good job” after they left the studio, and they probably walked away feeling quite pleased with that wonderfully vague, all-purpose compliment.

You can watch it yourself, but I wish one of them had just said: “Look, we don’t want to miss the AI frenzy, and we’ve got a good chunk of capital we’d love to put to work for those fat fees and spreads. But we’re also getting nervous about all these cash-burning AI firms that need financing. Then Jensen comes to us with a 25% first-loss proposal, and we’re thinking, “Holy cow, of course we’re in.” And all this stuff we just threw at you? That was basically us making Jensen look good on CNBC so he’ll prioritize us for future financing deals. Anyway, let’s wrap this up. Jensen, let’s reconnect later today.”

I mean, that’s what I would do… which is probably why I’m not running Blackstone.

2. Geeking Out on ARR

Readers may have noticed that my takes on private credit differ from those of the media. I am not the only person who writes about the subject, but most of the time, I also don’t see eye to eye with other authors, especially when it comes to certain nuances. Some of these authors have strong points I actually share, but we may view the ultimate outcomes differently, so I respectfully disagree with them (there are others, whom I disagree with quite disrespectfully… so follow me on Twitter for that).

So, Nick Nemeth, one of the bestselling Substack authors, has written about the interconnection between life insurance and private credit long before it became fashionable. He and I had one of those respectful disagreements on Twitter, and instead of trying to fit the full argument into 280 characters, Nick invited me onto his podcast so we could debate it in a longer form. I am a better writer than a talker, but you know, when someone calls you out publicly, it’s kind of hard to back off, so I reluctantly agreed.

We planned to debate insurance, private credit, and a bunch of related things that were brewing in the media. I prepared accordingly, getting plenty of statistics at my fingertips because I was expecting a full-blown Piers Morgan Uncensored-type showdown. But it was anything but that. In fact, as soon as I mentioned in the introduction that I had spent most of my career working on those now-infamous ARR software loans, Nick zeroed in on the topic, and we quickly abandoned everything we had planned to discuss. We ended up spending 1.5 hours discussing the evolution of ARR loans, the challenges facing the market today, and where it’s headed next, including the impact of AI.

Check out the episode on YouTube. Also, in case you are wondering, I’m the better-looking one in the thumbnail above…

3. Software, Still Standing

OK, speaking of the AI’s effects on software, but less from my perspective and more from what others said/wrote last week, and there’s been plenty. I’ll start with a quick anecdote from my conversation with a very successful small growth fund focused on software investments.

Generally speaking, growth funds are something between VC and PE. They typically invest in companies that no longer burn as much cash as VC-backed startups, but also aren’t growing as fast (say, 40% is plenty for them). But they also don’t do a lot of the financial engineering that PEs do.

Here is how our conversation went, and I’m paraphrasing:

DEBT SERIOUS: So, how much of an impact has this AI thing had on your portfolio?

Growth Fund: Zero.

DS: Right. Seriously?

GF: I’m serious. Zero. One of our companies had an issue with a big client that was pushing for a large discount and threatened to cancel and go with an “AI-first” solution. We had an emergency board meeting to discuss it, but ultimately decided to hold the line, fully expecting them to churn. But they renewed for three years, which was very surprising. Keep in mind, though, that our portfolio companies are not these easily replaceable applications. They are more ingrained in their customers’ operations.

DS: Hm. Are you saying there is no problem and everything is overblown?

GF: I’m definitely not saying that. I’m only suggesting there hasn’t been an impact on our portfolio so far. Frankly, in my almost 20-year software investment career, this is the most challenging period because of how uncertain things are.

Yes, existing portfolio companies are performing well, but multiples have reset lower, and we don’t know what multiples we’ll be exiting at when the time comes.

The bigger problem is with new investments. We have a fund to deploy, and we’re meeting interesting AI-first companies growing at astronomical rates, but the valuations in these deals just don’t make sense. Besides that, the problem is that they are just a couple of years old, and we have no idea whether this growth is driven by a strong solution they are offering or simply by customers experimenting with different AI products.

This Growth Fund’s portcos don’t translate directly into the leveraged loan or private credit market because its companies are smaller and don’t even carry debt. But the comments about the existing portfolio, multiple compression, and AI risk are remarkably consistent with what I’ve been reading in the financial press and research papers.

For example, last week Bloomberg ran a fairly detailed piece on software loans, touching on essentially the same points.

“If you strip away the capital structure and the AI disruption risk, the operating profile of most of these companies looks really healthy,” says Alen Lin, a senior director at Fitch Ratings covering North America technology credit. “It is that high financial leverage, the high interest expense, that’s dragging down the overall financial metrics.”

I mean, on one hand you might say, “well, if you strip away leverage from Lehman Brothers, the bank is still around.” On the other hand, though, we are talking about an industry that was expected to be disrupted by AI, and so far it is holding up quite well. In fact, it is holding up so well that, based on Fitch’s report published last week as well, software has experienced the lowest default rate in its sample:

Healthcare providers remained the sector with the highest number of unique defaulters in the July TTM period. Fitch recorded 17 unique defaulters in the sector, resulting in a 9.5% default rate, up from 9.4% in June 2026 and 6.9% in July 2025. The technology software sector remained the sector with the lowest default rate among the largest PCDR sectors at 1.2% in July, unchanged from the prior month. The most recent default in the software sector within the PMR portfolio was in July 2024.

I think there are two forces in play here, though. First, when we talk about software companies, we mean SaaS companies, which have recurring revenues based on contracts prepaid a year in advance. Those prepayments are then recognized gradually every month, so you are not going to see the top line (and subsequently the bottom line) suddenly going away. Second, it’s too early. Even those customers that are considering switching to AI-first solutions are still testing them alongside the software they already use.

I was historically focused on the first point, and thought that it would take a year from the SaaSpocalypse (until the full cohort goes through a renewal cycle) to understand to what extent the business was disrupted. Frankly, that was based on a naive assumption that customers immediately started testing alternatives right after SaaSpocalypse. So, it would take longer than a year, especially for larger customers where the complexity of the business just wouldn’t allow them to move that fast.

Yet, whereas the fundamentals of these businesses seem to be intact, the perception and, subsequently, valuations have obviously changed. Apollo has put together a very nice report that includes the historical multiples for software businesses.

As you can see, valuations have fallen not just back to pre-Covid levels, but all the way back to the “pre-SaaS-is-cool” days of a decade ago. The good news is that even at the 2021 peak, lenders generally didn’t go beyond 3.0x ARR leverage (occasionally stretching to 4.0x, as with Medallia). So, as long as the top line holds, most of these loans should still have enough LTV cushion. Loan marks take a hit, but that doesn’t necessarily translate into actual losses. The bad news is that PE sponsors are now underwater, giving them even more reason to kick the can down the road and avoid defaults at almost any cost rather than transact at valuations that would crystallize those losses.

This leads us to the maturity wall, of course, as software loans will need to be refinanced over the next few years. I’ve covered the subject a few times already, and basically said that maturity walls tend to get pushed out without any issues, but software loans are a harder case due to the uncertainty around the underlying businesses. But when you look at the chart below, where Bloomberg uses BDC software loans as a proxy for private credit more broadly, the picture is somewhat unexpected.

With U.S. direct lending (part of private credit) and U.S. leveraged loans being approximately the same size, at $1.2T-$1.4T each depending on who you ask, there are two observations here:

1. Software loans seem to be a fairly insignificant part of total loans, especially when you compare the amount of maturing loans to the market itself. In other words, the $19B and $41B of loans maturing in 2028 represent just 1.5% and 3%, respectively, of the direct lending and leveraged loan markets. Even writing them all down to zero doesn’t sound catastrophic, ignoring, of course, the second-order effects.

2. Since the markets are approximately the same size, it looks like the upcoming maturity wall is more of a leveraged loan market problem than a direct lending problem, which is quite surprising if you are focused on the headlines. Knowing how sensitive the leveraged loan market is to external disruptions (see my favorite chart below), that should be more concerning.

Again, on an absolute basis, it would just be an inconvenience even if all these loans went to zero, but the second-order effects here would be more consequential than if the same amount of private loans went to zero. These loans have observable market values and make up the majority of the debt held in CLOs, while CLO securities are also held on the balance sheets of insurance companies and pension funds.

Just some food for thought.

4. Jane Street’s Private Pivot

It seems like things and fortunes are moving very fast. Just a few months ago, we were all blown away by how the notoriously secretive Jane Street, which seems to be some hybrid of a market maker (buying and selling securities to provide liquidity to the market) and a proprietary trading firm (meaning it trades its own capital rather than managing outside investors’ money like a hedge fund), reported $40B in trading revenue for 2025, more than JPMorgan’s $36B and Goldman Sachs’ $31B.

Although this is all I planned to say as an introduction before moving to the main point, I found Jane Street quite interesting, so let’s double-click on them first.

It was launched in 2000 by three traders from Susquehanna International Group and one IBM developer. Over time, only one of those traders remained at the company: Rob Granieri, who is known for his love of Burning Man and for apparently, by accident, financing the purchase of AK-47s, Stinger missiles, and grenades to topple the South Sudanese government, after being duped by some activists he knew. Granieri’s picture is below… you know, just a typical Wall Street trader…

Then again, Jane Street is not your typical Wall Street investment shop. Just visit its website, which gives you a glimpse into a firm resembling a Silicon Valley startup with offices full of amenities and young T-shirt-wearing, blue-haired members of the team (btw, years ago, Sam Bankman-Fried and Caroline Ellison of FTX were among them). In their public engagements, you won’t see conferences or CNBC Squawk Box appearances, but rather a podcast with Dwarkesh (the tech Joe Rogan, I guess), chess championships, and some odd posts about machine learning and some stuff that only PhDs in quantum physics seem to appreciate.

There is also a page where those PhDs probably hang out: Puzzles… literally. Not the kind of puzzles you are thinking of, but the kind that make me want to stop thinking altogether… like the one below.

“…truncated tetrahedron”… I mean, surely I’m not their audience. If you solve the puzzle, you can submit your answer to Jane Street, and they’ll publish the names of those who solved it underneath. I counted 821 people, so it’s fair to say that at least that many people tried to solve it, right? Clearly, they are popular.

Alright, let’s be Serious and talk Debt… On Friday, August 7th, Bloomberg ran an article about Jane Street working to refinance $11B of bonds in the private market (emphasis mine):

Jane Street is negotiating with a group of investors in a private-credit deal that would rework the firm’s $11 billion debt load and limit its financial disclosures to market participants.

Terms of the deal — which could total as much as $15 billion and be in the form of a private placement — are expected to be finalized in the coming days.

Four days later, Tuesday, August 11th, S&P assigned a BB rating to $14.6B of Senior Secured Notes, which is likely when the transaction closed. I’m noting the dates to emphasize how quickly the transaction was executed after the announcement. Also, as of this writing, all its financial information seems to be scrubbed from the internet, except that Google still shows the now-defunct Investor Relations page.

I did find a somewhat recent graph of its capitalization, which breaks down its debt tranches, which conveniently add up to about $11B. It doesn’t have much value beyond the $29.9B of “member equity,” which tells you how profitable the company has been since inception.

If you are wondering why I care about Jane Street, and its financials in particular, it’s because you might have missed the news right before the weekend, reported in a WSJ article: Jane Street Suffers Loss of About $15 Billion Following Troubles at Situational Awareness.

WSJ notes that the loss was the worst in Jane Street’s history and elaborates on the causes:

Jane Street originally invested $2.5 billion in Situational Awareness, a figure that soared to nearly $10 billion earlier this year, according to people familiar with the matter. But the investment dropped to between $3 billion and $3.5 billion following Situational Awareness’s July losses.

In addition to that loss of nearly $7 billion, Jane Street suffered from the selloff in AI-related stocks last month and the firm’s move to cut its risk in those stocks.

Ouch! That Leopold guy again, and the AI trade… I mean, can you imagine what would happen to the markets if the AI investments suddenly stop for one reason or another? I can’t.

Back to Jane Street: their rush to refinance makes perfect sense now. They didn’t want to provide details around the loss and likely didn’t want to deal with any volatility in their bonds, which would have definitely happened after the news. That’s probably why they chose a private credit route, which is actually more costly for the company, and the FT estimated that they will be paying 1.5% more on the new debt, or about $220m in additional interest expense, plus another one-time closing fee of ~$200m.

I’ve seen some anons online wondering if Jane Street even disclosed the losses to the new lenders. Of course they did. As a borrower, you sign a bunch of depressing “representations” or whatever it’s called around material information, and not disclosing a $15B loss certainly falls under that category. Plus, they did the same with the rating agency, so that BB rating, which puts Jane Street in the junk category, would have been based on the information available to S&P, including the loss.

Obviously, the fact that the transaction was closed implies the lenders are comfortable with Jane Street’s financial health. Remember I mentioned that the company had $40B in revenue last year? FT in another piece suggested that Jane’s adjusted EBITDA is $31.2B, or a 77% EBITDA margin, which sounds insane and likely has a bunch of dubious add-backs, perhaps related to those Situational Awareness gains or something (which is why I wanted to review the financials).

You’d think the $15B loss contributed to S&P assigning the BB rating to Jane Street, but if you look at how the company’s recent senior secured notes were rated, they were also BB, as you can see in the table below (source: Empirasign… check the Bulletin for a free trial).

The rating was done in July 2025, so let’s use the 2024 revenue figure, which Bloomberg places at $20B. At a 77% margin, you get $15B of adjusted EBITDA. I mean, it’s rated at BB with what looks like 0.73x leverage ($11B/$15B)? But frankly, at any margin, even 25%, it sounds too odd. And of course, it’s because I walked into the trap of applying a metric I saw (EBITDA) to a financial company. Financials are a different animal altogether, and for a company like Jane Street, it’s all about liquidity, loss-absorbing capital, trading risk, and other metrics that you and I never think about.

That said, I still like that 77% EBITDA margin!

5. CDS, Decoded

This is the second time in recent memory that I am reading a piece on credit and don’t believe what I am seeing. The title of the Reuters article Soaring AI hyperscaler default hedges aren’t what they seem introduces the intrigue, and then the author does the work for me summarizing what’s up in one paragraph:

The ​explosion in many of AI giants' credit default swap (CDS) spreads makes for striking charts and juicy headlines. They have borrowed too much, free cash flows are evaporating, and ‌a financial reckoning of some sort is looming.

…CDS markets are incredibly thin. These derivatives provide protection against the risk that a bond issuer fails to meet its debt ​obligations, effectively acting as insurance that pays out if it misses interest or principal payments. But trading volumes are tiny and markets illiquid.

What? I’ll be frank with you, even I was looking at the CDS as a strong signal that hyperscalers’ debt was getting risky because of leverage and uncertainty around the AI economics. But Reuters makes an interesting point: it’s not necessarily the risk of default driving the premium, but rather the fact that so many lenders want to hedge their exposure to the sector as they continue writing new loans to hyperscalers, pushing up the premium in a market with thin supply.

You might ask, how do you know that’s the case? Well, if lenders were genuinely getting worried about losses, you’d expect them to pull back from lending to hyperscalers first and then load up on CDS. But that’s not what we’re seeing.

This doesn’t mean the risk hasn’t increased, though, and there’s a good reason Oracle’s CDS is trading so much wider than the rest of the crowd. I’m sure disclosures like the ones in Oracle’s recently published 10-K have something to do with it:

As of May 31, 2026, we had $260 billion of additional lease commitments, substantially all related to data center arrangements, that are generally expected to commence between the first quarter of fiscal 2027 and fiscal 2029 and for terms of fifteen to nineteen years that were not reflected on our consolidated balance sheet as of May 31, 2026 or in the maturities table above.

But Reuters makes an interesting point nonetheless.

6. 401(k) Support, Now Available

I’d love to pretend I’m neutral when it comes to private credit, but I doubt the readers would buy it. The media took such an extreme and factually wrong view of the asset class that I spent the better part of a year correcting them, which made me sound like a cheerleader (well, fine).

But there is a line I draw, and it’s the inclusion of private credit in 401(k) plans. We’ve just witnessed what a wave of BDC redemption requests can do to the entire space, and it weathered the episode just fine, thanks to its structural protections (well, gates) and, I’d argue, a fairly sophisticated investor base made up of wealth advisors managing money primarily for accredited investors. The fact that the takeaway from this is that we should include private assets in 401(k)s for regular Joes and Janes, who can rebalance their portfolios whenever they want, is a disaster waiting to happen. It would obviously do well for many years while new money keeps pouring in, but eventually you’ll create a problem that you can’t deal with.

Then there is a new dimension to the issue: the methods being used to push these assets into pensions, and it is not coming from Joes and Janes demanding it, but rather from the asset managers themselves. And apparently, they don’t play it clean.

Bloomberg’s article Trump’s 401(k) Proposal Shows Evidence of Phony Public Support goes into some details:

A review of almost 12,000 comments in favor of the administration’s proposed rule found submissions in the names of dead people.

So, what happened here is that after the proposal to include private assets in 401(k)s was released, the Department of Labor started receiving comments from the general public in support of or opposition to the initiative. There’s nothing unusual about that process. It’s standard practice for many government initiatives. What was unusual in this case, besides the fact that some supportive comments came from people who were deceased or who said they never submitted them, is that a significant portion of the comments supporting the initiative appeared to have been fabricated.

Roughly 12,000 form-letter comments backing the proposed rule were almost evenly split across five templates. Each comment that followed a template bore a different person’s name but otherwise was remarkably consistent — down to the number of characters, punctuation and line breaks.

The comments following these five templates were submitted in almost identical amounts to one another each day between April 29 and May 5, then ceased. The other submissions for the campaign opposing the rule came in at more random intervals and stretched out over more of the two-month comment period.

The article shows the chart (below) with more than 30,000 opposing comments and roughly 12,000 suspicious supporting comments, and even to a non-statistician, it looks off. The supporting comments are way too concentrated within a short period of time, there are just five templates, and, unlike almost all of the comments opposing the initiative, they don’t have email addresses attached to them. After reporters were able to track people down based on their unique names (like, not many Aznaurs in the U.S.) and contact them, the result was that most didn’t respond, but those who did denied they had submitted the comments or said the comments couldn’t have been submitted by their deceased relatives.

Although I’m a supporter of individual choice, I’ve said in the past that there are some things that should be kept away from people, even if they ask for them… like drugs from kids, or private assets from Joes’ and Janes’ 401(k)s. In this case, trying to push these assets down the throats of a majority of people who didn’t ask for them, and in fact mostly opposed them, while the apparent support seems to have been manufactured… it’s a whole new level of absurdity.

7. Deals on the Block

Download below: the full Excel file covering deals completed in 3Q26 (plus the prior three quarters).

CLICK to Download 3Q26 Details

KIÉR LIÓR – Providing outsourced private and leveraged loan underwriting and portfolio management to institutional investors, family offices, and direct lenders.

www.KierLior.com

8. Bulletin Board

• Four debt/equity requests: $35m, $12m, $9m, and $7m (LINK)

• Empirasign, a provider of secondary market trading data for structured products, covering ABS, MBS, CMBS, CLOs with BWIC history dating back to 2010. They also carry real time market data for IG/HY bonds as well as secondary loans. DEBT SERIOUS readers get a 1-month completely FREE trial. Click the LINK and select DEBT SERIOUS during registration

• Bloomberg is offering 60% off its annual subscription ($180 vs $399). If you are in Business, Bloomberg is a MUST, with no real close alternative LINK

• WSJ Online is offering 65% off its annual subscription ($4 per week vs $11) LINK

• The Financial Times is offering a 4-week trial for $1 (total, not per week), but make sure to cancel before it renews to avoid paying $75/month LINK

• Shortcut.AI, an AI tool for Excel that lets you build models by simply typing requests (similar to ChatGPT). It is free to use up to a certain number of tasks per day (LINK). If you hit your limit before finishing your project, simply upgrade to the Pro plan ($20/month) and enter promo code DEBTSERIOUS (all capital letters) to get 50% off

• LoanEdge, a BDC research tool that lets you look into BDC loan compositions, individual loans, which BDCs own a specific loan, and more, is kindly providing DEBT SERIOUS readers with a free 3‑month trial (the regular price is ~$3k per year, and it’s higher for institutions). Here is a LINK to the tool. Send an email to Sadaf Khan at sadaf@theloanedge.com and let her know you are from the DEBT SERIOUS community

As a reminder, I don’t personally benefit from either of these offers. I am just saving you money.

Lastly, our community is over 4,500 strong and includes a broad cross-section of mid- to senior-level professionals across private credit, private equity, LPs, and debt advisory. Please use the Bulletin Board at the end of each weekly letter to share events, discounts, or deals that benefit the community. I will continue to feature these opportunities and personally connect interested parties at no cost.

That’s the bell — round over. See you in the next.

Aznaur

Aznaur.Midov@KierLior.com

News

Articles I found interesting. Doesn’t mean I agree with their takes.

1. NVIDIA teamed with private credit firms to mobilize $500B for hyperscalers and frontier labs to build data centers and buy NVIDIA GPUs. AI compute is emerging as a new asset class, akin to 1970s CMBS (CNBC)

2. Blackstone is considering abandoning its plan to raise $3B through a collateralized fund obligation (CFO) to return cash to investors after struggling to find a buyer for the CFO’s equity tranche (Bloomberg)

3. SEC exempts data center ABS from the disclosure, investor protection, and risk retention requirements that apply to traditional ABS, making it easier for data center operators to issue debt (Bloomberg)

4. Lenders began diligencing community and political opposition before committing to data center projects to avoid wasting time and resources. In 1H26, at least 75 projects worth $130B faced local opposition (Reuters)

5. Review of 12k comments supporting private assets in 401(k) found signs of a fake campaign, including comments submitted on behalf of deceased people. The 30k opposing comments appear legitimate (Bloomberg)

6. Signs of stress are emerging in private credit despite the industry’s upbeat tone, with loan defaults at their highest since 2021 and returns declining (WSJ)

7. High leverage and high interest expense weigh on Software’s otherwise very healthy operations. This deep dive examines AI’s impact on software businesses and software ARR loans (Bloomberg)

8. Private credit has been tightening loan structures, reducing the use of PIK among other things. PIK provisions appeared in 13.5% of new deals originated in Q2’26, down from 25% in Q4’25 (WSJ)

9. Private credit finances ABS (“Anything But Software”) deals, showing a preference for industrials with tangible assets and remaining highly selective when it comes to software (Bloomberg)

10. Mark Walter’s Delaware Life and Clear Spring are working to unwind $20B+ of private credit investments reclassified as affiliated assets amid a U.S. prosecutors’ probe (Financial Times)

11. Banks charge BDCs a premium on loans during rate hikes, despite their more favorable risk profile than non-BDC borrowers, according to a study by Fed economists (Financial Times)

12. BofA commits $250B to U.S. infrastructure through direct lending, underwriting and investment in areas such as AI and energy, as part of Trump initiatives and U.S.’s 250th anniversary (Financial Times)

13. Regional banks see stronger C&I loan demand, with 16% reporting higher demand from small and mid companies, up from 5% last quarter, as AI-driven capex spending trickles down (Yahoo Finance)

14. The U.S. government sold $25B of 30Y bonds at the highest interest rate since 2001, 5.216%, as investors demand greater compensation for inflation and fiscal risks (Bloomberg)

15. Deutsche Bank is the first European bank appointed as a yuan clearing house, as China seeks to challenge USD dominance. Low rates have fueled a surge in yuan borrowing by foreign companies (Financial Times)

16. China is shifting AI funding from state-sponsored to cap. market-driven, as U.S. competition intensifies. Tech firms raised $38B in 2026, the most since 2016, at an average yield of 1.9%, 300bps below the U.S. (Bloomberg)

17. African local-currency debt has outperformed emerging-market debt this year, returning 5.5% vs. 3.2%. Returns vary widely, with Zambia up ~35%. Yields reach 16% in Uganda, 18% in Zambia, and 21% in Nigeria (Bloomberg)

18. Alternative investment firm 777 Partners filed Chapter 11 with $2.7B of debt, after being under restructuring advisors since 2024. Prosecutors accuse its founder of using fake invoices to inflate the firm’s finances (Bloomberg)

19. DOJ and CFTC are probing transactions involving Radiant World, one of the world’s largest iron ore traders, over concerns that the company provided banks with falsified trade documents (Bloomberg)

Transactions

1. Apollo reached a $2.6B financing deal with the New York Yankees. Proceeds will be used to refinance existing debt and fund franchise growth (WSJ)

2. JPMorgan led a $441m debt financing for Global AI, a two-year-old AI infrastructure company with $6.2B of contracted revenue, including $1B from infrastructure already built and delivered to clients (Bloomberg)

3. UBS provided a 12-month extension on software company ION’s $460m revolver, pushing its maturity to July 2027. ION is the most indebted software company in Europe (Bloomberg)

4. Blue Owl raised $750m of BBB+ 10-year notes to repay its revolver, pricing at +2.20% over Treasurys, tighter than the initial +2.50% talk. Strong demand allowed the deal to be upsized from $500m (Bloomberg)

5. NVIDIA-backed neocloud Lambda raised a $926m, 4.4-year loan, led by Morgan Stanley. The loan was priced at 99.5 cents on the dollar with a SOFR + 3.00% spread, with proceeds funding GPU acquisitions (Yahoo Finance)

6. ZenithArc, a data center developer, sold $2.25B in “green bonds” to fund construction of a facility leased to Jane Street. The five-year notes are priced at 99.5 cents on the dollar and yield ~9% (Bloomberg)

7. Kenya to raise $5.4B over the next 9 months, including $800m USD-denominated, $500m Sukuk (Islamic), $300m Panda (Chinese yuan), $900m Samurai (Japanese yen), and the rest from World Bank, AfDB and others (Bloomberg)

Chart of the Week

Few things matter more to markets right now than the AI buildout. Its sudden scale has become a major US economic stimulus, shifting macro data. Columbia Business School’s Stijn Van Nieuwerburgh estimates AI infrastructure investment at 2.8% of GDP, larger than the railroad boom and still rising. He argues that without it, the US would be in recession.

• Bloomberg is offering 60% off its annual subscription ($180 vs $399) LINK

  • If you’re not a subscriber and the article is behind a paywall, search for the article title online. It’s often reprinted on Yahoo Finance for free.

• WSJ Online is offering 65% off its annual subscription ($4 per week vs $11) LINK

• The Financial Times is offering a 4-week trial for $1 (total, not per week), but make sure to cancel before it renews to avoid paying $75/month LINK

These are NOT sponsored links, just money-saving tips.

1. VC’s DPI history

2. Radiant World’s creditors pull facilities

3. BlackRock TCP’s CV sale to Pantheon

4. BDC’s defy expectations

5. European maturity wall is moving

6. 9fin’s covenant report

7. Notable dals

8. Bulletin board (New discounts and Four investment opportunties)

Disclaimer: there are 4,400+ of us here discussing private credit and leveraged loans, which is not a big world. At some point, I will probably throw punches at your company. I am blunt and sarcastic and, as friends like to remind me, too mouthy to live to an old age, so my take likely won’t please you. But don’t worry, next week it will be someone else in the spotlight and you will quietly nod along in satisfaction.

Also, none of this is investment advice.

1. DPI Gets Historical

Does anybody know Kyle Stanford from PitchBook? I don’t, but I came across his report on U.S. VC returns over the past decade and I’d like to meet the guy. I typically don’t care much about VC, but the intro he posted about the report (below) intrigued me enough to read the whole thing for fun:

“…the lowest five-year DPI multiple this CENTURY.” What a choice of words!!! I mean, with intros this violent, he might need to borrow my disclaimer above, you know? Although, he should have gone full savage and called it “the lowest this millennium,” which would have been just as accurate.

I don’t care if you’re not interested in VC returns. That intro alone makes you want to check out the report!

Download "the CENTURY" report

2. Radiant, Meet Reality

Well, there seems to be a new “First Brands” brewing in the media: Radiant World. The two have a few things in common: both grew too quickly, both are being accused of fabricating invoices, their founders have virtually no online presence, and Jefferies’ Point Bonita hedge fund was among their lenders.

Most people haven't heard of Radiant World, but if you happen to run a steel mill, you definitely know the company as a major iron ore trader. Not a hedge fund-type trader, but the type of trader that buys iron ore from Brazilian or Australian mines and resells it to Chinese steel mills, while also taking care of the logistics to ensure it is delivered on time.

Radiant World was founded in India in 2003 by 23-year-old Pinkesh Nahar. India was already one of the world’s major iron ore producers and exporters, so it was a natural place to start an iron ore trading business. The company gradually expanded before entering China in 2008, and that’s when things really took off, as it hit the jackpot riding China’s rapid industrialization and the resulting surge in demand for steel and iron ore. By then, it had also established relationships with mining giants like Vale, Rio Tinto, and BHP Group.

Most metals trading is dominated by Glencore and Trafigura, but iron ore was a relatively niche business that wasn’t worth building out as a separate operation for these powerhouses. That gave Radiant World a natural niche. By 2012, Glencore and Trafigura were already using Radiant to round out their own offerings without having to build a dedicated iron ore trading business.

Nahar, who has kept a very low profile and is barely mentioned online, has been anything but conservative when it comes to growing the firm. By 2014, Radiant was trading about 7 million tons of iron ore a year. A decade later, that had grown 10x to 70 million tons. The company’s latest reported figures show annual revenue of roughly $12B. One interesting thing about Radiant is that, unlike most trading houses, which primarily act as intermediaries connecting buyers and sellers and typically hedge the commodities they trade, Radiant would occasionally make outright bullish bets by buying iron ore and holding it unhedged for a period of time, betting on price appreciation. And historically, those bets seem to have worked pretty well.

The trading business of Radiant World is sizable and, unsurprisingly, very well-banked, with significant credit exposures across a number of financial institutions. Commodity traders like Radiant typically rely on banks not only for traditional corporate loans, but also for trade finance facilities tied directly to their underlying transactions. In Radiant’s case, trade invoices and shipping receipts were used as collateral for its financing lines. If you are wondering just how well banked Radiant was, below is the list of financial institutions that worked with the company as of 2025.

That’s especially important given the latest developments at the company. A couple of weeks ago, Bloomberg reported that Cargill, Glencore, Rio Tinto and Vitol had stopped trading with Radiant. Glencore’s exposure to Radiant may be as high as $500m-$800m, making its decision to stop trading with the company particularly significant. Bloomberg reported that the companies had discovered Radiant was providing banks with falsified invoices and other documents related to iron ore trades. Since then, it has been revealed that Arab Bank, one of the company’s key financiers, has stopped issuing letters of credit. ICBC has suspended repo financing, while Société Générale has reduced its exposure. Deutsche Bank and KBC have also frozen some of Radiant’s assets in Singapore. Meanwhile, Intesa Sanpaolo and Jefferies’ Point Bonita fund are reviewing their exposures, which are estimated at $200m-$300m each.

The thing that bothers me most about these credit frauds (or alleged frauds, as of now) is not that they occur, because no one can completely prevent fraud from happening. It’s that significant red flags are typically missed:

Two of the world’s biggest commodity trade financiers, Cooperatieve Rabobank UA and ING Groep NV, stopped financing Radiant World several years ago amid concerns about certain documents involved in some of its trades.

Look, you’ve done your due diligence, and it turned out that major banks had already exited the relationship because they were concerned about potential fraud. Yet you still proceeded with the deal. This was clearly surfaced during diligence, but some smart nose on the origination team convinced everyone that there had been a misunderstanding and that the documents weren’t actually fabricated. Well, you could have fired the person on the spot for pushing a dumb deal, except you also signed off on it under pressure from other members of the committee. What can I say? Hopefully, it’s a misunderstanding this time as well. Otherwise, the whole team should be kicked out, you know?

Even more interesting, though, is the following point Bloomberg made about Radiant in a report last year:

The trading house’s [Radiant’s] scale means that any change in its fortunes could be felt throughout the iron ore market — a commodity whose role as the key ingredient of steel makes it crucial to the global economy and the key driver of profits for the world’s largest mining companies

3. BlackRock Origami

In last week’s letter, I wrote the following:

What if the private credit crowd borrowed a page from the Tbilisi airport’s currency exchanges, except instead of rotating the winners, they rotated the losers? They could start separate funds whose sole purpose was to refinance the really bad deals. Let’s call those funds, idk, BlackRockTCPs (…sorry… just couldn’t help myself), and rotate which one refinances the bad loans each quarter? I mean, who cares about performance, collect your fees, repackage the loans into some sort of CLO or AMAPS or whatever, and voilà, the maturity wall is gone!

The following day, the Wall Street Journal published a piece discussing how BlackRock TCP, which has been in the news for all the wrong reasons, is transferring nearly half of its loans, including the collateral underlying its CLO, into a separate continuation vehicle and selling a majority stake in that Frankenstein of a vehicle to Pantheon.

I mean, I’m not saying I predicted the whole thing, but boy, isn’t this hilarious? Now, if you’ve read the WSJ piece, you are probably wondering what’s actually going on here, and I don’t blame you. The article throws a bunch of numbers at you as if everything is simple and perfectly logical. It absolutely isn’t.

To try to understand it, you have to go back to the source: the SEC filings and the company’s earnings call presentation. And even then, it feels a bit like reconciling adjusted EBITDA on a compliance certificate with the numbers in the audit (if you know, you know).

May 2026

BlackRock TCP bundled 70 of its loans with a total value of $535m, into a CLO. It initially sold the A1, A2, and B tranches to outside investors, followed later by the C tranche, raising approximately $406m in total (remember this figure). It retained the D and LLC Int. tranches, representing the subordinated and equity interests, on its balance sheet for a total of approximately $130m. The $406m of proceeds were used, among other things, to repay outstanding bank credit lines with Citi and ING.

None of this information is discussed in the article or in BlackRock TCP’s recent SEC filings. It is simply useful background for understanding which CLO the article is referring to.

August 2026

Now, onto the actual transaction that was announced. As reported by the WSJ:

Under the Aug. 4 deal, BlackRock TCP is transferring roughly two-thirds of 78 middle-market loan investments into a new continuation vehicle and Pantheon will purchase a 95% interest in it. The Santa Monica, Calif.-based BDC expects to receive gross proceeds of about $152 million from the deal and will retain the remaining 5% interest in the continuation vehicle.

Frankly, this is all you need to know, because getting deeper into the details opens up a rabbit hole that is impossible to make sense of. Basically, BlackRock created a new CV, moved 78 loans into it, including the 70 loans serving as collateral for the CLO plus several more, and transferred the $406m of CLO obligations along with them (among other liabilities), reducing and cleaning up the balance sheet. It then sold 95% of the CV to Pantheon for $152m.

When all is said and done, including the upcoming major loan paydown from one of its portfolio companies (Domo), the leverage ratio will drop to 0.3x from 1.38x before the transaction.

If you really want to see how the transaction looks on the financials, see below. But don’t try to reconcile it, and certainly don’t contact management for further details either, as they just got hit with another shareholder lawsuit, essentially accusing the company of mismanagement that led to a 2/3 decline in its stock price.

4. Credit Gets Boring

You know how some things are too good to be true to the point where you just want to pass on them, because surely there’s a catch? Bloomberg’s article titled “Private Credit Funds Avert Worst Fears and Rebound From Lows” gave me exactly that feeling, cuz, you know, they are not supposed to be saying this.

The piece is a collection of bullet points on a bunch of public BDCs’ Q2 earnings calls, and begins with the following:

In the eye of the private credit storm, outlooks called for the $1.8 trillion market to either bring about the next financial crisis or for the pressure to quickly peter out. Nearly half a year later, it seems the truth lies somewhere in the middle.

Publicly traded BDCs have been reporting their second-quarter results in recent days, providing the clearest state-of-play yet for the industry. That’s because unlike their non-traded peers, which are mired in a multibillion-dollar backlog of redemption requests, these vehicles have set pools of money, putting the focus squarely on their loan valuations and portfolio credit quality.

The article elaborates that, based on Q2 calls, these BDCs are focused on trimming underperforming loans, working out non-accruals, and reducing leverage. They are also avoiding dividend cuts, which the market tends to view negatively. In other words, lenders are doing exactly what they are supposed to do: dealing with problem loans and de-risking, and apparently that caught the media off guard. Well, too bad.

The market reacted swiftly to what it perceived as positive developments, and BDCs have rallied over the past week or two, although they remain below their pre-SaaSpocalypse levels (Feb. 2026) and, in some cases, even below pre-First Brands levels (Sept. 2025). I’m a bit worried, though, that if Q3 results are similarly less negative than expected, some people’s worldviews might start melting down, which would be concerning.

And frankly, there were some negative developments at certain BDCs: Ares’ non-accruals grew by 15% to $708m (2.4% of total loans); Blackstone marked down its NAV to the lowest level since 2020; and senior executives at BlackRock and Goldman stepped down, which is rarely a good sign. But the negatives were clearly less severe than the market had feared, particularly at the beaten-down BlackRock and Blue Owl BDCs, which helps explain the rally.

We’ll see what Q3 brings…

5. Refi, Extend, Repeat

A few months ago, in Round #39, I wrote that concerns around the 2028 maturity wall of leveraged and private loans were overblown. As with previous maturity walls, I expected it to get smoothed out as we approach 2028, with the only wildcard being software loans. And even there, I suggested that the issue was not so much the fundamentals, but the uncertainty around them, which would take some time to figure out.

Well, it looks like Europe very much agrees, as discussed in PitchBook’s article, “Europe tackles 2028 loan maturity wall faster than expected, but software lags.” And, of course, I’m going to discuss it!

The following paragraph and chart are basically summarize the piece:

Despite several big-ticket buyout deals, refinancing and extensions led supply in European loans this year, making up 67% of total volumes of €110.6 billion excluding repricings. The pace accelerated in the second half, which has whittled away at the European loan maturity wall. This debt was identified as a potential threat to the market at the start of the year, with a chunk of borrowers’ financing needs dating from the height of the QE-fuelled LBO boom in 2021.

I think the chart tells the story quite well, as the value of loans maturing in 2028 has shrunk considerably since December 2025. Obviously, we can’t simply extrapolate this refinancing trend into the future, because it is only logical that the cleanest, strongest loans were the ones that got extended first. The piece acknowledges as much:

The next block of work to extend the 2028 need may also prove tougher than the first part of the year. “What is left? There is usually a reason if something is not already addressed,” said one manager. “Probably leverage is a bit higher than it should be, or the sector could be stuck in a long downturn cycle that has yet to recover,” they added.

That said, the pressure to refinance or face loan acceleration (the demand to repay the loan) will force sponsors to either put in more equity or bring in junior lenders to bridge the gap between the amount of equity they are willing to contribute and the amount of debt senior lenders are willing to extend. Everyone in the capital structure is incentivized to refinance, or at least extend the maturity, so you are not betting only on the companies themselves, but also on the incentives of each player in the capital structure. And I wouldn’t bet against the latter.

Now, the piece covers leveraged loans, specifically European ones. However, it suggests that in certain situations, private credit refinanced some of these loans, meaning the competition between BSL and private credit is another variable that could help facilitate the refinancing process.

Lastly, the piece is about European loans, while it was US loans that have made most of the noise so far. That said, I view the European situation as a guide to what we should expect in the US, especially given how much deeper and more competitive the US capital markets are.

So, if you were to ask me, we are getting closer to something like the following:

6. Covenant Season

9fin recently published its US Covenant Trends Report 1H26, covering leveraged loans and high-yield bonds. The report is worth checking out, but below are a few tidbits on leveraged loans (skipping HY altogether).

First, the Margin Ratchet, which is a fancy way of saying “leverage-based pricing grid” (which in itself is a fancy way of saying that if leverage is 5x, the price is S+5.00%; if it is 4.5x, the price is S+4.75%, etc.):

Margin ratchets appear in 40% of sponsored deals and 21% of non-sponsored deals. This stands in contrast to the European market, where margin ratchets appear in almost every deal.

I’m a bit surprised that ratchets are on almost every European deal, as I’ve been under the impression that European deals are more lender-friendly and, generally speaking, ratchets favor the borrower. From a Finance 101 perspective, the riskier the asset, the more premium the investor should demand, and vice versa. So, when the loan deleverages over time and there is less time to maturity, surely the loan has de-risked, and technically, lower pricing is justified. But as a lender, I don’t care about Finance 101. I want the highest pricing I can get, and I want it to stay through maturity. As a borrower, especially a sponsor-backed one, you don’t really care about my sentiments toward the pricing I want because they conflict with your deep desire to have the lowest price possible.

That’s how we find a compromise: we start closer to where I want pricing to be (still lower than I wished) and gradually get to where you want it to be (still higher than you wished), with the reduction contingent on deleveraging.

Anyway, seems like in Europe they skip the negotiation and start with: “OK, it will be a grid.” Meanwhile, US deals seem to follow my logic, although it’s a bit surprising to see that some non-sponsored deals got three step-downs. Perhaps they opened at a high enough spread to allow for it, or these were large, competitive deals where lenders had less negotiating leverage.

Second, EBITDA adjustment caps, which limit how much those “one-time expenses,” “pro forma adjustments,” and other fancy EBITDA add-backs can increase reported EBITDA. That’s important not only for leverage and fixed-charge coverage ratios, but also because many “baskets” are sized based on EBITDA.

The proportion of deals with uncapped EBITDA adjustments has dropped across the board:

• A third of sponsored deals have uncapped EBITDA adjustments, down from 53% in H2 25

• 15% of non-sponsored deals have uncapped EBITDA adjustments, down from 27% in H2 25

It’s nice to see uncapped adjustments down to 32%, although anyone who has seen the kinds of add-backs that typically make it into the calculation and what adjusted EBITDA does to your ability to analyze a company’s true cash flow would tell you that 32% is still 32% too many. And forget uncapped adjustments. Even a 30% adjustment, which seems to be somewhere in the middle of the chart, can work some serious magic on covenants.

Think of a fixed charge coverage ratio (FCCR), whose purpose is to ensure that the business generates enough cash to cover its fixed expenses. Obviously, the worst you can allow is an FCCR of 1.0x, which means the borrower is just breaking even. But being the prudent lender you are, you set the covenant above 1.0x, say at 1.25x.

Now, if the borrower’s fixed charges are $8 and adjusted EBITDA is $10, it’s in compliance right at that 1.25x level (10 / 8 = 1.25x). But if that EBITDA includes 30% of adjustments ($3), suddenly your true cash generation is only $7, which doesn’t cover the $8 of fixed charges. That borrower better have a solid cash pile on the balance sheet to cover the shortfall, and those adjustments better truly be “one-time,” with all pro forma synergies fully realized now (riiiiiight), so the expenses are covered in the next reporting period.

Lastly, the report includes a number of other changes to credit agreements, including LME blockers, reporting requirements, and a bunch of baskets that are too technical and confusing to get into here. Baskets are weird animals in general and can be surprisingly annoying to negotiate. It’s hard to conceptualize why an obscurely worded basket sized at 40% of EBITDA could be materially more dangerous than a 30% one, so they often end up being driven more by precedent deals with a particular sponsor than by any strong conviction one way or another. They often serve as negotiating currency: you give the sponsor the basket size they are asking for in exchange for getting what you need somewhere else. Of course, when the deal blows up, you suddenly understand exactly why that basket mattered, and why 30% instead of 40% could have made all the difference in the world.

So, perhaps read this 9fin report…

KIÉR LIÓR – Providing outsourced private and leveraged loan underwriting and portfolio management to institutional investors, family offices, and direct lenders.

www.KierLior.com

7. Deals on the Block

Download below: the full Excel file covering deals completed in 3Q26 (plus the prior three quarters).

CLICK to Download 3Q26 Details

8. Bulletin Board

• Four debt/equity requests: $35m, $12m, $9m, and $7m (LINK)

• Empirasign, a provider of secondary market trading data for structured products, covering ABS, MBS, CMBS, CLOs with BWIC history dating back to 2010. They also carry real time market data for IG/HY bonds as well as secondary loans. DEBT SERIOUS readers get a 1-month completely FREE trial. Click the LINK and select DEBT SERIOUS during registration

• Bloomberg is offering 60% off its annual subscription ($180 vs $399). If you are in Business, Bloomberg is a MUST, with no real close alternative LINK

• WSJ Online is offering 65% off its annual subscription ($4 per week vs $11) LINK

• The Financial Times is offering a 4-week trial for $1 (total, not per week), but make sure to cancel before it renews to avoid paying $75/month LINK

• Shortcut.AI, an AI tool for Excel that lets you build models by simply typing requests (similar to ChatGPT). It is free to use up to a certain number of tasks per day (LINK). If you hit your limit before finishing your project, simply upgrade to the Pro plan ($20/month) and enter promo code DEBTSERIOUS (all capital letters) to get 50% off

• LoanEdge, a BDC research tool that lets you look into BDC loan compositions, individual loans, which BDCs own a specific loan, and more, is kindly providing DEBT SERIOUS readers with a free 3‑month trial (the regular price is ~$3k per year, and it’s higher for institutions). Here is a LINK to the tool. Send an email to Sadaf Khan at sadaf@theloanedge.com and let her know you are from the DEBT SERIOUS community

As a reminder, I don’t personally benefit from either of these offers. I am just saving you money.

Lastly, our community is over 4,400 strong and includes a broad cross-section of mid- to senior-level professionals across private credit, private equity, LPs, and debt advisory. Please use the Bulletin Board at the end of each weekly letter to share events, discounts, or deals that benefit the community. I will continue to feature these opportunities and personally connect interested parties at no cost.

That’s the bell — round over. See you in the next.

Aznaur

Aznaur.Midov@KierLior.com

News

Articles I found interesting. Doesn’t mean I agree with their takes.

1. Public BDCs rebounded from lows as Q2 results showed stabilization and improvement. The piece summarizes calls from ARCC, BXSL, TCPC, OBDC, GSBD, FSK, OCSL, MFIC, MSDL, TSLX, and CGBD (Bloomberg)

2. Ares scaled back its €1B private credit CV to €400m after investors demanded deeper discounts on the loans than Ares was willing to accept (Financial Times)

3. BlackRock TCP sold a 95% interest in a CV to Pantheon for $152m, which, after debt repayment, will reduce its leverage from 1.38x to 0.3x. The CV includes 80 middle-market loans totaling $523m (WSJ)

4. Two Blue Owl BDCs reported improved results, with OBDC posting higher NII and its lowest leverage in two years, though non-accruals rose from 2% to 2.8% QoQ. The BDCs also repurchased $90m of shares (Bloomberg)

5. PIK loans grew from 6% to 10% of BDC portfolios over the past four years, suggesting growing pressure on borrowers’ cash flow (Semaphor)

6. Google’s $200B AI infrastructure machine: Google guarantees data centers, Broadcom supplies and helps finance chips, while Apollo and Blackstone fund hardware leased to Anthropic (Financial Times)

7. AI data center CMBS spreads widened over the past 12 months, with AAA up 15bps to 168bps and BBB- up 11bps to 321bps, as heavy AI debt supply drives investor fatigue (Bloomberg)

8. Deutsche Bank froze Radiant World’s bank accounts, while Arab Bank and ICBC suspended financing amid concerns that the company had provided banks with falsified documents related to iron ore trades (Bloomberg)

9. Jefferies discovered fake invoices among those backing its financing to Sapphire Minmetals. Sapphire was owned by Radiant World until 2015 (Bloomberg)

10. U.S. insurers hold $40B of debt privately graded by Egan-Jones, a rating agency sued by former employees who allege they were pressured to inflate ratings (WSJ)

11. EchoStar’s Hughes Satellite filed for Chapter 11 after lacking the cash to repay $1.5B of bonds due August 1, as competition from SpaceX’s Starlink and Amazon Leo drove a 22% YoY drop in broadband subscribers (WSJ)

12. Recovery rates for first-lien lenders in bankruptcy have declined dramatically, driven by shrinking junior tranches, collateral increasingly dominated by IP, LMEs, and other factors. See Chart of the Week (Bloomberg)

13. “Sell America” trade re-emerges as investors fear the Fed is losing its grip on debt markets without an inflation strategy. 30Y yields topped 5%, the highest since 2007, while the USD weakened against G10 currencies (Bloomberg)

14. $100B of bonds in the USD and EUR IG indexes are trading at spreads close to junk-bond levels, including Oracle and Stellantis. Potential “fallen angels” (IG downgraded to junk) also include BDCs exposed to SaaS (Bloomberg)

15. European green bond sales hit a record $242B in 1H26, on track to surpass the $385B full-year record set in 2024, as the Middle East conflict revived interest in energy security and infrastructure investment (Financial Times)

16. European loan refinancing accelerated in the second half of the year, defying expectations, as borrowers moved to get ahead of the 2028 maturity wall (Yahoo Finance)

17. Citi-led African sovereign bond sales are up 70% YoY despite the Middle East conflict, with Citi leading every deal but one (Angola’s $1.5B raise), totaling $6.2B vs. $7.3B for all of 2025 (Bloomberg)

18. A record $8.8B of PE, VC, and Private Credit capital has flowed into Emerging Markets YTD to fund AI data center buildouts (Bloomberg)

19. France and the UK will help Zimbabwe restructure its $21B debt to the IMF, World Bank, AfDB and Paris Club, aiming to restore capital-markets access after being locked out since 1999 (Bloomberg)

Transactions

1. Jane Street is seeking $15B of private debt to refinance $11B of bonds, giving it more investment flexibility while avoiding quarterly financial disclosures required for public bonds (Bloomberg)

2. Ares is leading a $2.2B private debt package to finance MedImpact’s acquisition of Medical Card Systems, one of the largest direct loans this year. The second-lien loan is expected to price at SOFR+8.0% (Bloomberg)

3. HPS minority-owned Authentic Brands participated in Aston Martin’s recent £550m loan. Of that, £100m is a DDTL available if Aston transfers 50.1% of its non-automotive IP to Authentic (Bloomberg)

4. Alphabet sold $25B of AI-related IG bonds across 10 tranches, with maturities from 2 to 40 years. The yield premium was 1.3% over Treasuries, tighter than the initial 1.55% talk (Bloomberg)

5. Blackstone is pitching investors on a second $36B debt package tied to Anthropic’s use of Google’s TPU chips. The first deal was a $35B package arranged by Blackstone and Apollo (Yahoo Finance)

6. Kazakhstan plans to raise ¥3.4B ($500m) in Panda bonds, taking advantage of lower rates in Chinese capital markets. It will be the second ¥3.4B tranche, following a 3-year, 1.90% Panda issued earlier this year (Bloomberg)

7. Laos state-owned power producer EDL-Generation issued $300m of five-year, 11.125% dollar-denominated junk bonds (CCC) to refinance existing notes, a rare deal in Asia’s dollar bond market (Bloomberg)

Chart(s) of the Week

1. Recovery rates are plummeting, with more first-lien lenders recovering next to nothing in bankruptcy.

   

2. The carry trade — the strategy of borrowing in yen and parking in a higher-yielding currency like the Mexican peso — has been an ultra-reliable source of profits. Over five years, it has strongly beaten even the S&P 500’s total return.

• Bloomberg is offering 60% off its annual subscription ($180 vs $399) LINK

  • If you’re not a subscriber and the article is behind a paywall, search for the article title online. It’s often reprinted on Yahoo Finance for free.

• WSJ Online is offering 65% off its annual subscription ($4 per week vs $11) LINK

• The Financial Times is offering a 4-week trial for $1 (total, not per week), but make sure to cancel before it renews to avoid paying $75/month LINK

These are NOT sponsored links, just money-saving tips.

In Round #55:

1. From Georgia with love

2. How private credit became the dominant story in the media

3. Looking into how PE socializes risk through insurers

4. Q2 update on private equity and direct lending

5. Notable deals

6. Bulletin Board: includes new FREE access to EMPIRASIGN (a structured credit dataset) and FOUR live deals

Disclaimer: there are 4,400+ of us here discussing private credit and leveraged loans, which is not a big world. At some point, I will probably throw punches at your company. I am blunt and sarcastic and, as friends like to remind me, too mouthy to live to an old age, so my take likely won’t please you. But don’t worry, next week it will be someone else in the spotlight and you will quietly nod along in satisfaction.

Also, none of this is investment advice.

1. Lari-Go-Round

Ok, I told you in the last letter that I’m in Tbilisi (country of Georgia), but I didn’t share the story of currency exchange, because there was already enough about Georgia in that letter.

So, as you arrive at the airport, you are greeted by a dozen currency exchange windows where you can get the local currency, the lari. I was looking at the exchange rates, and they all seemed identical to the penny, so I figured it didn’t really matter which window I used. But then I noticed a couple of windows offering significantly better rates (see below).

That made no sense, because they offer the most commoditized service possible, exchanging one type of currency for another. Plus, they are located right next to one another, so you can compare all their rates without even moving from where you are standing. As I was approaching the window with the better rate, I was genuinely puzzled about why anyone would be such a sucker as to exchange their money at the other windows.

Before exchanging any money, I confirmed with the teller that they were, in fact, exchanging currency and not offering some sort of transfer or another service that could explain the rate difference, because at that point I was truly bamboozled. She confirmed that it was just a regular currency exchange, so of course I had to follow up by asking why anyone would exchange money at the window to the right. Before she could answer, the guy standing behind me jumped in: “Because they rotate. Today it is this window with favorable exchange rates, tomorrow it is the one on the right.” The teller laughed and denied it, but it was the kind of laugh that pretty much confirmed the guy was right!

It was actually funny to me, because once he said it, it seemed like the most logical explanation possible, yet somehow it was the one thing I hadn’t considered! I mean, I thought of differences in transaction volume, potential marketing tricks the other windows might be using, and basically everything I’ve learned over the years about competitive tactics, except for the most obvious explanation: a dozen players had simply fixed their prices for the benefit of two others, and were rotating throughout the day/week!

Where is Lina Khan when you need her?!

Anyway, that got me thinking. You know how people keep bringing up the maturity wall coming in 2028 and wondering who is going to refinance all those loans? What if the private credit crowd borrowed a page from the Tbilisi airport’s currency exchanges, except instead of rotating the winners, they rotated the losers? They could start separate funds whose sole purpose was to refinance the really bad deals. Let’s call those funds, idk, BlackRockTCPs (…sorry… just couldn’t help myself), and rotate which one refinances the bad loans each quarter? I mean, who cares about performance, collect your fees, repackage the loans into some sort of CLO or AMAPS or whatever, and voilà, the maturity wall is gone!

Ok, since the maturity wall issue was just fixed, moving on to the next problem…

2. Private Credit: Building a Boogeyman

And the next problem is trying to make sense of how we got here with private credit and an observation I haven’t seen discussed much.

When I started DEBT SERIOUS a year ago, before all the hysteria about the space, the media was still at a crossroads over whether to call the space “Private Credit” or “Private Debt” (even though there are technically differences that we don’t care about). The chart below shows search results for “Private Credit” (blue line) and “Private Debt” (red line) over the past five years. As you can see, sometime around the middle of last year, there was a sudden move to “Private Credit,” and the term has been a click magnet ever since. Interestingly, both terms appear to have peaked in March of this year and have been declining ever since, so maybe we will finally stop hearing about them.

We all know the story that private credit exploded after the 2008 financial crisis, when new regulations made certain types of lending more expensive for banks, particularly to leveraged middle-market companies, and funds stepped in to fill the void. Something like that… no one really explains the exact regulatory details or what requirements actually changed.

Obviously, if we are talking about funds stepping in and doing the lending that banks used to do, we are really talking about the Direct Lending part of private credit, even though we kept calling the whole thing private credit. The reason I am specifying that is because private credit has turned into a pretty large umbrella of strategies, as we’ve noted before, and you can see them in the chart below (ignore the percentages, which are from a survey about LPs’ allocation focus).

Bear with me, but keep this chart in mind. So, if you were in banking or direct lending at the end of 2024, you’d hear a fair amount of frustration about intensifying competition, loosening covenants, and compressing pricing. Newspapers, meanwhile, were primarily discussing M&A in the space, covering some basic facts about private credit, and publishing the occasional skeptical take, as the industry was still quite mysterious.

Then, around April/May 2025, you started seeing an increasing number of bearish takes on the space (see the FT headlines below), which somewhat coincided with tariffs, I guess. This is not a knock on the Financial Times, which just happens to have all of these articles in one place. That said, I do believe the FT was a major contributor to the hysteria that was about to kick in and has continued ever since.

The triggering events were the back-to-back blowups of Tricolor and First Brands. That’s when private credit became the center of attention, and when I believe the media realized how many clicks it could generate. If you recall, those deals were actually frauds, and even though we’ve had a few more similar frauds since then (across hundreds of thousands of loans), Tricolor and First Brands became the “I told you so” examples of what private credit would bring, cited by a bunch of pundits from the credit world (often biased because of competitive dynamics), including Dimon with his “cockroaches” comment, which certainly didn’t help.

With headlines getting more dramatic over the following months, non-traded BDCs started experiencing redemption requests above their typical 5% quarterly repurchase limits. The financial media began covering this heavily because they had to keep using the magic words “Private Credit,” which generated clicks. Remember when Blue Owl received redemption requests equal to 17% of shares in its BDC at the beginning of this year and met them in full? The media covered it extensively. They also started tracking how redemption requests increased each quarter for certain other BDCs, even though they either remained below 5% or were fully met by the funds. This was apparently important enough to track, and given the number of funds out there, you could generate an endless stream of articles and clicks.

Then came SaaSpocalypse in February, and we learned how exposed private credit was to those loans. A new wave of articles about the exposure of each fund, the hidden exposure of each fund, and what people thought about the exposure and hidden exposure of each fund kept popping up.

Then came more redemptions, now above 5%, that were gated. The media started covering each gated fund, turning gating into something inherently inappropriate. They conveniently forgot that just a couple of months earlier, the same funds received no credit for going beyond their requirements and meeting redemptions in full.

Recently, we’ve started seeing another angle of the private credit story: insurance. And this one is becoming increasingly relevant. So, there seems to be no end in sight to this private credit debacle.

And now, let me explain why I brought back all these memories. The private credit hysteria started with Tricolor and First Brands, right? But the irony is that Tricolor had absolutely nothing to do with private credit. First Brands had very little to do with private credit: it had about $10B-$12B of debt, of which roughly $1.2B was provided by hedge funds through supply-chain finance, with the balance largely made up of BSLs and bank loans. Remember that chart above? Supply-chain finance is more of a Niche strategy, although some may consider it a subcategory of Asset-Based Finance (ABF).

Why does it matter? Because by calling Tricolor’s bank debt and First Brands’ niche finance exposure “private credit,” the media helped cause a run on BDCs, which represent a Direct Lending strategy that had nothing to do with either Tricolor or First Brands.

SaaSpocalypse was definitely a solid reason to worry about direct lending, which is why BDCs continue to struggle. That’s fair.

But now, the discussion has shifted to private credit in insurance companies. The issue is that, when it comes to private credit, life insurers primarily invest in asset-backed finance (like consumer or auto loans, etc.), strategies like NAV or SRT, infrastructure debt, and direct lending, which is typically investment grade (whereas BDCs do direct lending for non-investment-grade borrowers). There are obviously BSL CLOs and private CLOs, and the latter contain a non-investment-grade direct lending component, but insurers typically invest in the rated senior tranches of those instruments.

The point is that all the noise about problems in private credit started with reporters ignoring crucial nuances and lumping loosely related strategies into one category, “Private Credit,” even though those strategies aren’t even correlated. That eventually turned the situation into something of a self-fulfilling prophecy, stigmatizing the entire private credit umbrella in the process.

Let me use a simple analogy to bring the absurdity of the situation home. Imagine that a few Volkswagen Golfs experienced brake problems that resulted in injuries. Newspapers started running stories about German cars causing injuries. Then they interviewed salespeople from Toyota and Ford, who explained that German cars have strange braking systems and said they weren’t surprised by the accidents, as they’d been warning about those systems for years. The stories gained traction, and suddenly people started selling their Mercedes’ and BMWs because reporters glossed over the rather important nuance that the problems involved Volkswagens, specifically Golfs, and actually just a handful from, say, the 2021 vintage.

Now, that doesn’t mean the Volkswagen brake problems should be dismissed. Maybe there is a broader issue with a particular model or model year that warrants a recall. But surely that doesn’t mean every German car has a brake problem and is dangerous to drive.

3. Private Credit, Public Risk

Speaking of insurance! A couple of weeks ago, a paper called Private Credit’s State Backstop: How Private Equity Socializes Risk Through Insurers, started making rounds, with certain excerpts circulating online.

I went into the paper fully prepared to lose my temper within the first few paragraphs, expecting the usual story about how private credit is very risky, how it experiences stress (everything I discussed in the prior piece), and how insurance companies should never invest in it or the world will come to an end. Turns out I was wrong. It was anything but that, and the paper was actually quite interesting.

There are several key themes in the paper, which I’m going to oversimplify, of course: (i) PE firms have been acquiring life insurance companies in part as a tool to maximize their own fees, not necessarily to improve insurance operations; (ii) these life insurance companies tend to shift allocations away from traditionally conservative, liquid investments and toward illiquid, hard-to-value private credit assets, potentially adding risk; (iii) the unique way insurance companies are liquidated could ultimately leave taxpayers on the hook for insolvent insurers; and (iv) the authors propose regulatory reforms aimed at reducing the risk of insurer failures and avoiding taxpayer bailouts.

Let’s go through each theme briefly.

The first theme, the acquisition of life insurers by PE firms, is pretty self-explanatory. It is no secret that alts like Apollo or KKR have acquired insurers and gradually converted them into machines that generate quasi-perpetual capital. Unlike traditional drawdown funds that last 8-10 years, where LPs may choose not to reinvest the proceeds into a new fund upon maturity, especially if the fund underperforms, insurers’ capital is sticky, which also means recurring fee generation.

Besides the fees, though, the paper discusses other benefits a PE firm can get from owning an insurer. For example, it can sell assets from one of its funds to the insurer’s balance sheet on terms favorable to the PE firm. It uses the following statistics:

Recent empirical evidence suggests that PE-owned insurers buy identical assets at higher prices when the seller is an affiliate [PE]. By comparing purchases of the same structured security on the same day, [a study] shows that PE- owned insurers pay 7bps more when buying from affiliates, and 40bps more when the transaction involves privately-placed securities.

Essentially, assets that are sold to third parties for $99.00 are sold by affiliates to insurers for $99.07 (structured securities) and $99.40 (privately-placed securities). Amortize the difference over a 3-5 year tenor, and the premium gets fairly small. I’m sure if you were to confront management, they’d say that $99.40 is a fair price, that they were comfortable paying it because they know how diligent the affiliate is, etc. The issue is different, though. If these types of self-dealing transactions are part of the ordinary course of business, how do we know distressed assets aren’t also being sold to insurers at inflated prices? The paper uses a recent Blue Owl transaction as an example, where it sold assets near par to its affiliated insurer, Kuvare. That raised some eyebrows because Blue Owl’s public BDC was trading at a significant discount to NAV at the time, suggesting the market didn’t trust the marks.

A separate point is that PE-owned insurers can cede liabilities to captive reinsurers (discussed in Round #43), which can provide tax benefits and result in lower capital requirements for the insurer. In other words, the insurer can operate with more leverage.

The second theme is closely related to the first, and in fact is part of the initial point in the paper itself. I am separating it for convenience. PE-owned insurers have been shifting their allocations away from traditional liquid investments and toward more opaque private credit assets. Private credit provides higher yields, but it is also illiquid and harder to value. That means less certainty around the price at which you can exit a position when you actually need to, which adds risk.

The paper uses some statistics showing that PE-owned insurers allocate significantly more to private credit than independent insurers, and argues that further increases in those allocations reduce the margin for error and, consequently, increase risk.

In 2024, 49.5% of new PE-owned insurers’ investment was in privately-placed instruments, compared to just 13.9% for independent insurers…

…The exact extent to which the rise of private credit on insurer balance sheets raises insolvency risk is contestable, but trendlines point towards greater hazard. The most optimistic study, using a relatively narrow definition of private credit, finds that greater private credit exposure has not been historically associated with higher estimated insolvency risk. We note that, as private credit continues to sharply rise each year to new historical highs of insurer ownership, projections based on historical data will be contending with economic environments that insurers have never actually experienced.

Despite my rant about the media’s misunderstanding of private credit, I can’t disagree with this take. More importantly (for me!), the paper doesn’t obsess over the potential collapse of private credit (it actually does a good job of distinguishing between direct lending and ABS), so it is more of a “look, we are fine for now, but this is concerning, and if the trend continues, it could blow up in our faces” take.

The third theme is the most important one, and is something I didn’t know about. So, what happens when one of these insurance companies blows up? Much like banks, insurers don’t go through a traditional bankruptcy process. Instead, they end up in receivership. With banks, the priority is protecting depositors; with insurers, it’s protecting policyholders. Unlike banks, where the FDIC is a federal agency that takes over a failed bank and works out its obligations, insurance doesn’t have a federal FDIC-equivalent receiver; it is primarily regulated and resolved at the state level, as the industry is regulated state by state.

Another big difference is that every bank pays premiums to the FDIC (into the Deposit Insurance Fund), and when one of them blows up, those funds are used to cover the infamous $250k per depositor. Insurance doesn’t have an equivalent pre-funded pool. Instead, when an insurance company blows up and ends up in receivership run by its state regulator, other insurance companies (its competitors) that operate in the same line of business (auto, life, etc.) are assessed a predetermined amount to make sure policyholders get covered up to $300k. Crazy, right? You put your competitor out of business and then have to bail out its policyholders!

But that’s not the main part. The main part is that the vast majority of states then provide tax credits to the insurers that bailed out their failed competitor, and over the next 5-10 years, they can recoup most or all of their contributions. Well, if those contributions are ultimately offset by tax credits, the bailout is effectively being funded by taxpayers.

This is the main point of the paper, and where it gets its title, “How PE Socializes Risk.” The argument is that PE firms buy life insurance companies, get access to permanent capital on which they charge fees, allocate that capital to higher-yielding and potentially riskier investments, and if the insurer blows up, the PE firm isn’t the one required to fund the bailout. Instead, taxpayers can ultimately end up on the hook.

The fourth theme is about the authors’ proposed reforms. They don’t recommend banning PE acquisitions of insurers or reducing their allocations to private credit, which is where I thought the paper was going. Instead, they suggest fairly logical (but surely unpopular among PE-owned insurers) reforms: (i) a surcharge for balance sheet opacity; (ii) recourse to PE owners and holding companies; (iii) charging insurers premiums to create a Deposit Insurance Fund-like pool that would be used to pay policyholders when an insurer fails; and (iv) ending the tax credits discussed above. They propose several other reforms and go into much more detail on why each is important, but this is the high-level summary.

P.S. I agree with the authors on many points, and I also have concerns about PE-owned life insurance companies. In fact, just a few days before they published the paper, I posted the following note on Substack:

Yes, in my humble opinion, Apollo and KKR are Systemically Important Financial Institutions (SIFIs), because if their insurance arms blow up, it would feel like a Lehman moment (I said “feel,” not that it would be on the same scale). But I don’t think they are necessarily where the problems will start. As I’ve noted many times in the past, the media and pundits are overly focused on the strongest players, who, ironically, have the most experience and resources and are in the best position to pick the healthiest deals. I think the bigger risk is the second-tier players trying to emulate the Apollo/KKR model without the same experience, resources, or access to the best deals, and instead financing the stuff the big guys passed on.

I’m also less concerned about insurers’ exposure to private credit and more concerned about their leverage. Insurers are levered around 20:1 on a consolidated basis (including holdcos, reinsurers, etc.), but their operating entities are often levered 50:1, if not higher, as we discussed in this interview with Tom Gober. If you are 50x levered and we hit a prolonged recession, you better make sure the holdco has enough liquidity on hand to bail you out.

Weirdly enough, if we get some major short-term market dislocation where liquid assets suddenly show significant unrealized losses, those private holdings, with their slower repricing, might actually buy these insurers some time, at least on paper. Of course, the values will be fake, but as long as the market stabilizes before they need to start liquidating assets, the lag in repricing might actually provide a temporary buffer. That’s just my hypothesis, though.

4. Deal Flow, Slow

PitchBook published its mid-year report on PE, which conveniently includes a couple of slides on direct lending (cuz, you know, someone needs to provide the debt). There are a number of interesting slides, but I will cover a few.

The first one covers Q2 PE activity based on deal volume ($$$) and deal count.

This doesn’t look particularly good, with both deal value and deal count down significantly over the past few quarters. One small silver lining is that average deal size appears to be coming down (the line is getting closer to the bars), which, in my book, is more important than sheer value.

I’m not sure why value is the metric that gets quoted in the press so often. $200B of transaction value that includes five $20B acquisitions isn’t necessarily healthier than 400 deals at $500m each. What we really need is for the lower middle market to start transacting and work through the backlog, not just a handful of healthy upper-market deals boosting the headline numbers.

Direct lending typically correlates with PE, so you could have predicted how it looked. Both value (or volume as they chose to say here) and deal count are down.

I don't know why they thought it was a good idea to switch the presentation (unlike the prior chart, deal count is now bars, not the line), but luckily there's no need to read too much into the chart: basically, we are back to 2023 value and transaction count.

The last chart shows the number of exits through secondaries. Given all the discussion about the lack of liquidity and exits, along with the numerous articles about the growth in secondaries, I fully expected 2026 to be a record year for exits through this route.

So far, though, it looks like exits will come in below 2025 and roughly in line with 2024, which I find very strange. Well, it is what it is, I guess.

Overall, a nice little report. Not too many words, plenty of charts, and perfect for when you need a break from actual work but still want something work-related on your screen while daydreaming about the weekend.

5. Deals on the Block

Download below: the full Excel file covering deals completed in 3Q26 (plus the prior three quarters).

CLICK to Download 3Q26 Details

6. Bulletin Board

• Four debt/equity requests: $35m, $12m, $9m, and $7m (LINK)

• NEW: Empirasign – the true leading provider of secondary market trading data for structured products, covering ABS, MBS, CMBS, CLOs with BWIC history dating back to 2010. They also carry realtime market data for IG/HY bonds as well as secondary loans. Used across desks at every major dealer, hedgefund and asset managers, Empirasign offers the most complete data and correct data set for credit practioners, and DEBT SERIOUS readers get a 1-month completely FREE trial. Click the LINK and select DEBT SERIOUS during registration

• WSJ Online is offering 65% off its annual subscription ($4 per week vs $11) LINK.

• Bloomberg is offering 60% off its annual subscription ($180 vs $399). If you are in Business, Bloomberg is a MUST, with no real close alternative (LINK).

• Shortcut.AI, an AI tool for Excel that lets you build models by simply typing requests (similar to ChatGPT). It is free to use up to a certain number of tasks per day (LINK). If you hit your limit before finishing your project, simply upgrade to the Pro plan ($20/month) and enter promo code DEBTSERIOUS (all capital letters) to get 50% off.

• LoanEdge, a BDC research tool that lets you look into BDC loan compositions, individual loans, which BDCs own a specific loan, and more, is kindly providing DEBT SERIOUS readers with a free 3‑month trial (the regular price is ~$3k per year, and it’s higher for institutions). Here is a LINK to the tool. Send an email to Sadaf Khan at sadaf@theloanedge.com and let her know you are from the DEBT SERIOUS community.

As a reminder, I don’t personally benefit from either of these offers. I am just saving you money.

Lastly, our community is over 4,400 strong and includes a broad cross-section of mid- to senior-level professionals across private credit, private equity, LPs, and debt advisory. Please use the Bulletin Board at the end of each weekly letter to share events, discounts, or deals that benefit the community. I will continue to feature these opportunities and personally connect interested parties at no cost.

That’s the bell — round over. See you in the next.

Aznaur

Aznaur.Midov@KierLior.com

News

Articles I found interesting. Doesn’t mean I agree with their takes.

1. The FOMC voted 9-3 to keep the Fed rate unchanged at 3.50%-3.75%, with three dissenting voices advocating for a hike to address inflation (CNBC)

2. The Bank of England kept the UK rate at 3.75%, as anticipated after inflation fell to a 15-month low of 2.6%. Still, three of nine committee members voted for a 25bps hike (CNBC)

3. BlackRock’s acquisition of HPS, even at a high valuation, is paying dividends, with the combined platform winning deals (i.e., Meta) neither could secure alone. The deal increased BLK’s private credit AUM 4x (Bloomberg)

4. Blue Owl said redemptions are easing, but fundraising also slowed. The company raised $7.6B in Q2 vs. $12.1B a year ago, while credit AUM fell from $159B to $158B (Benzinga)

5. Cracks are appearing in private credit, although there is no explosion of stress. Q&A with PitchBook’s head of private credit and private equity research on conflicting trends in the market (Morningstar)

6. Revolut will offer EU customers access to private equity and private credit investments for as little as €1 through a Revolut-managed feeder. Apollo, Ares, and Hamilton Lane funds are among those selected (Bloomberg)

7. Fitch reported private credit defaults hit a record 6%, up from 5.7% QoQ, driven by maturity extensions of stressed loans, not PIK or interest deferrals. Defaults were highest in Industrials at 10.4% and lowest in Software at 1.2% (Bloomberg)

8. PIMCO is anchoring the debt for Oracle’s $16B data center build-out, but on its own terms. Instead of shorter-term, lower-yielding debt, it pushed for higher-yield, long-term bonds matching the long-term lease (Bloomberg)

9. U.S. insurance regulators are concerned about the complexity and circular ownership of multi-asset securitized products, such as Apollo’s AMAPS (Round #44), held by PE-backed life insurers like Apollo and KKR (Financial Times)

10. The SEC is probing Dodgers owner’s insurer, Delaware Life, for allegedly failing to disclose $16B (42% of assets) in affiliated private credit investments, reporting just $1B (3%) instead (WSJ)

11. European loans are priced ~4bps tighter than U.S. loans, after historically being 22bps wider, making Europe more attractive to borrowers. U.S. private credit woes / BDC redemptions have pushed U.S. pricing higher (Bloomberg)

12. Big UK banks challenged the BoE’s comparison of UK and U.S. capital rules, arguing their 12.6% CET1 requirement exceeds the adjusted U.S. level of 10.7%, tying up £23B of capital and costing £750B of lending capacity (Financial Times)

13. European banks’ AT1 bond spreads (a type of bank debt) hit a record low of 211bps, but investor demand persists. Some view investors as far too complacent (Bloomberg)

14. The UAE sold a record $30B of USD- and EUR-denominated bonds YTD, $3.7B above the 2020 record, despite the Middle East conflict. Spreads widened to 77bps over U.S. Treasurys from 65bps before the war (Bloomberg)

15. Asian private credit fundraising hit a 10-year low. Just five Asia-focused funds raised $1.2B in 1H26, vs 29 ($10B) last year and 23 ($20B) in 2022. Also, LPs continue to favor U.S. managers over Asian firms (Financial Times)

16. Hong Kong launches Chinese government bond futures after two failed attempts (2017 and 2023). Significantly higher foreign ownership and demand for hedging make conditions more favorable this time (Bloomberg)

17. Bain-backed U.S. Renal Care plans to return assets stripped from lenders in a 3-year-old LME as it seeks to refinance a $1.2B loan due in 2028 (Bloomberg)

18. User Acquisition financing, provided by PvX Partners, offers capital to mobile apps. Instead of EBITDA, lending is based on custom metrics such as Return on Ad Spend and User Retention Rate (Bloomberg)

Transactions

1. Blackstone will acquire $25B of Australian home and personal loans from HSBC, subject to regulatory approval. The portfolio will be held across its credit, insurance, real estate, and tactical opportunities strategies (Reuters)

2. Fortress signed a $1.5B forward flow loan deal with Irish fintech Wayflyer, which provides unsecured loans to e-commerce businesses (Irish Times)

3. CoreWeave raised a $2.6B loan tied to its Anthropic computing deal after conceding 1%-1.25% higher pricing to lenders. The deal is priced at SOFR + 5.50% and discounted to 96-97 (Financial Times)

4. ING completed $10B of SRTs, including $6B tied to various U.S. and EU project financings, such as data centers, and $4B tied to small and medium-sized Dutch companies (Bloomberg)

5. Philippines’ BDO Unibank raised ₱132B ($2.1B) in the country’s largest-ever corporate debt issuance, 26x the initial offer due to strong institutional demand. The bond yields 6.26% (Business Inquirer)

6. NVIDIA will guarantee $250B of OpenAI debt to cover data center leases and facility build-out, but not the NVIDIA chips that would go inside (WSJ)

7. Germany’s soccer/football league Bundesliga is raising a €1B, 20-year loan from U.S. investors, including Apollo (Goal)

Chart of the Week

In the absence of forward guidance, the market is trying to figure out the Fed’s next move.

Per Apollo: “One interpretation of [the rate movement], is that markets understand the Fed’s commitment to 2% inflation, but with no forward guidance, the market does not understand how the Fed will get to 2% inflation. Is the way to 2% inflation through higher rates, a smaller balance sheet or tighter financial conditions? The answer to this has significant implications for the yield curve and how and when we will achieve 2% inflation. The lack of clarity about how to get there is what’s pushing yields higher because there is now a risk that it may take longer or involve a policy mistake.”

• Bloomberg is offering 60% off its annual subscription ($180 vs $399) LINK

  • If you’re not a subscriber and the article is behind a paywall, search for the article title online. It’s often reprinted on Yahoo Finance for free.

• WSJ Online is offering 65% off its annual subscription ($4 per week vs $11) LINK

These are NOT sponsored links, just money-saving tips.

1. Why Georgia’s GDP has doubled

2. Kevin Warsh and the leveraged basis trade

3. Wells Fargo’s quarterly direct lending report

4. Eagle Point on software loans and CLO equity investing

5. Notable deals

6. Bulletin board

Disclaimer: there are 4,400+ of us here discussing private credit and leveraged loans, which is not a big world. At some point, I will probably throw punches at your company. I am blunt and sarcastic and, as friends like to remind me, too mouthy to live to an old age, so my take likely won’t please you. But don’t worry, next week it will be someone else in the spotlight and you will quietly nod along in satisfaction.

Also, none of this is investment advice.

P.S. There’s a new energy deal posted this week (see the Bulletin Board).

1. Doubled GDPs

So, where was I? Ok, yesterday, the National Bank of Georgia (Georgia the country, not Georgia the state) kept its interest rate unchanged at 8.25%, despite inflation running significantly above its 3% target at 5.8%. The NBG is expected to keep rates unchanged for now, with inflation beginning to decline in the second half of 2026 and returning to its 3% target over the medium term. Who cares, right? Well, I do, because I’m currently visiting Tbilisi and have already gone through the “Wait a minute, what’s going on?” phase a couple of times.

The first time was when I boarded my Amsterdam-Tbilisi flight, fully expecting 95% of the passengers to be Georgians. Instead, it seemed like 95% were Westerners traveling to Tbilisi. This wasn’t a connecting flight, either, so Tbilisi was clearly their final destination.

The second time was when I decided to look into the country’s economy and came across the 10-year GDP chart below. I’m not an economist, but seeing GDP more than double in 5 years, after a fairly ordinary growth rate in the previous 5 years, blew my mind.

Well, I then looked at where the country is located, remembered the major conflict involving its much larger neighbors, and figured the two were probably connected. It was somewhat counterintuitive, though, because I thought security concerns would hurt the region, not help it. I was wrong. It turns out Georgia’s post-2021 economic boom was fueled in part by a recovery in tourism, but mostly by a surge in immigration, financial inflows, and transit trade following the Russia-Ukraine conflict.

The tourism part is easy to explain: mountains, great food and wine, and low prices. That said, the tourism recovery was driven by the end of COVID, not the conflict, obviously. The immigration part is pretty straightforward too. A significant influx of Russian, Ukrainian, and Belarusian immigrants escaping the conflict increased demand for goods and services, but with the population growing by only about 200,000 people over five years to 3.9 million, it clearly wasn’t enough to explain GDP more than doubling.

Financial inflows were a much bigger contributor. Because of sanctions and broader security concerns, entire businesses relocated from Russia, Ukraine, and Belarus to Georgia. They brought capital, created jobs, increased demand for the Georgian currency (the lari), boosted exports of goods and services, increased demand for loans, generated higher tax revenues, and everything else you learned in Economics 101.

Then there was transit trade. As the Russia-Ukraine conflict made some of the traditional trade routes between Europe and Asia much harder to use, more goods started flowing through Georgia instead, turning it into one of the region’s main transit hubs.

What’s also quite interesting is that Georgia wasn’t the only country to benefit from the conflict. Several neighboring former Soviet republics did as well, especially the smaller economies, where it takes much less to move the GDP needle. Below are the main beneficiaries (I also included cumulative inflation, since it contributes to nominal GDP growth).

So, is there a private credit opportunity? Not yet. Some of the largest U.S. private credit funds have more AUM than the entire GDP of these countries, although slapping “Armenia Fund I” on a pitch deck is certainly more memorable than launching yet another lower-middle-market direct lending fund. So why should you care? Well, IDK, these countries have mountains, great food and wine, and low prices, so the next time you’re planning a European vacation, skip Paris and come to Tbilisi…

This post was definitely not sponsored by the Ministry of Tourism of Georgia… although it probably should have been.

2. Warsh: Forward Guidance vs. Basis Trades

Back on home turf, since we touched on interest rates earlier. The Fed kept rates unchanged at 3.50%-3.75%. The Financial Times had an interesting piece last week on Kevin Warsh’s shift from transparent forward guidance to taking a much more tight-lipped approach. As the Financial Times puts it, Warsh said:

“Financial market prices are probably the most important source of information to guide central bankers,” he [Warsh] said at his inaugural press conference last month. “But when all the financial markets are doing is reflecting back what we’ve said, then we’re taking the most important source of information and we’re being blind to it.”

I like how Ajay Rajadhyaksha, global chair of research at Barclays, laid out his take on the shift

Forward guidance has turned markets into a mirror. The Fed watches markets; markets watch the Fed. And no one’s actually watching the economy.

It’s hard to argue with that logic, as it sounds bulletproof. The problem is that a significant number of market participants (including myself) haven’t operated under Greenspan and his ambiguity policy. So, there is a pretty good chance this perfectly sound libertarian-ish view turns into another example of “be careful what you wish for.” Because we recall what happened to many crypto bros in 2022 who wanted independence from the financial system but as soon as their altcoins blew up, asked for a bailout and regulation.

And there are other things to worry about besides just the Fed’s new communication approach. The article points out that the ownership of the Treasury market has changed quite a bit. Historically, U.S. Treasurys were primarily held by patient, long-term investors, such as insurance companies and pension funds. Hedge funds have always been active in the market, but their Treasury holdings have grown dramatically in recent years. They now own about $2.5T of Treasurys (for reference, China holds around $660B). Even more interesting, their Treasury exposure has doubled over the past two years, growing much faster than the Treasury market itself.

The concern is that many of these funds are running leveraged basis trades (discussed below). If those trades start going against them, hedge funds could be forced to unwind their positions, creating a liquidity crunch in what’s supposed to be the most liquid market in the world.

I’m gonna be honest, I have no idea what the repercussions of that would be. That’s more of a question for economists and the liquid credit crowd. But, at least optically, the chart above doesn’t look good. The buildup in these positions has been incredibly steep (at least to my liking).

Also, if you have no idea what a basis trade is, don’t worry. Until a few days ago, neither did I. I got tired of reading about it without actually understanding how it worked, but never cared enough to dig into the mechanics. Well, until I started writing this letter. I can now proudly say I’m about 5% less clueless than I was before.

Basis Trade

This might have been the most confusing concept I tried to wrap my head around, mostly because different sources explain it differently. At its core, the trade is actually pretty simple: you buy, say, a 10Y Treasury (go long) and short 10Y Treasury futures. Then you come across futures quotes that look like this, and suddenly you start wondering whether you understood any of it correctly:

So, with the 10Y Treasury priced at, say, $98.50, the futures quote of $108.125 sounds like way too rich of a spread. Well, two things:

1. Those 108s are not the prices of 10Y Treasurys, but rather the quoted prices for the Treasury futures contract. To convert that quote into the price of the Treasury that can actually be delivered into the contract, I need to multiply it by something called the Conversion Factor, then account for accrued interest and a few other adjustments. For simplicity, let’s ignore those other adjustments and just multiply 108 by the first conversion factor of 0.9127 (which may or may not be the cheapest-to-deliver bond…long story). That gives us $98.74, which makes the comparison to the Treasury’s $98.50 price a lot more intuitive.

   

2. That said, relative value is what ultimately matters. We are not betting on a specific price target for the 10Y Treasury, but rather focusing on the spread between our long and short positions. Since we initiated the trade with a spread of -$0.24 ($98.50 minus $98.74), the position becomes profitable as that spread narrows.

Leveraged Basis Trade

Now, that $0.24 spread is before accrued interest and a few other adjustments, so the actual profit opportunity is even smaller (say, it is $0.05). How do hedge funds make money on such tiny spreads? Leverage.

Specifically, repo. And you remember how repo works, right? Suppose a hedge fund wants to buy $100m of 10Y Treasuries from a dealer, say Goldman Sachs. It doesn’t have $100m sitting in its bank account. Instead, before the trade settles, it arranges repo financing with JPMorgan. JPMorgan agrees to lend the hedge fund $99m, with the Treasuries serving as collateral for the loan. When the trade settles, Goldman Sachs delivers the Treasuries, JPMorgan wires $99m to Goldman Sachs on the hedge fund’s behalf, and the hedge fund contributes the remaining $1m. The hedge fund now owns $100m of Treasuries, despite only putting up $1m of its own capital.

On the other side, the hedge fund enters into $100m of Treasury futures. Unlike buying Treasurys, it doesn’t need an additional $100m to do so. Instead, the futures exchange requires the hedge fund to post only a small amount of collateral, typically only a small fraction of the contract’s value. By posting just a few million dollars of margin, the hedge fund can establish a $100m short futures position.

So, with only a few million dollars of its own capital, the hedge fund has effectively created a $100m long position in Treasuries and a matching $100m short position in Treasury futures.

The Concern

With all that background, let’s go back to the Financial Times piece and see why Warsh chose to frame his comments around hedge funds’ Treasury exposure. The article continued:

An interest-rate surprise from the Warsh Fed might not only unsettle Treasury yields but set off a vicious circle of margin calls — when a sudden price movement in assets bought with borrowed money requires an injection of capital — and fire sales by hedge funds that could create a dash for cash.

Torsten Sløk, chief economist at Apollo, the private equity giant, says that there is more chance of such a scenario — which he labels a “leverage unwind” — because of the Fed’s change of stance. He adds: “It is indeed the case that when you have less forward guidance, things can happen all of a sudden.

I don’t think I can add much more color to that paragraph.

Time Out

So, a few months ago, you spoke to your boss about your compensation. You thought you were underpaid, but when the dust settled, your pay barely moved. Well, that’s because you probably didn’t realize just how underpaid you were and ended up asking for too little. Like, if you’re a Director at a private credit fund, your base salary alone can range anywhere from $215k to $300k… and don’t even look at the bonus. See? You’d know that if you used BuysideHub, where I shamelessly stole the data below.

Anyway, below is the word from BuysideHub, which happen to be the sponsor of this Round.

Buyside Hub is a compensation analytics platform heavily used by Private Credit and Leveraged Finance professionals.

It is the one-stop shop for a detailed compensation breakdown for Buyside Credit professionals.

And it’s 100% free for those who join and contribute to the platform. You can join 15,000+ other Buyside Hub community members here:

www.buysidehub.com/welcome

3. Direct Lending Check-Up

Wells Fargo CIB’s Q2 report on direct lending is out (download below), laying out the key themes and trends. Historically, volatile markets tend to slow down loan origination but lead to higher-quality loans. These dynamics reinforce each other, as lenders become more selective and, when they do extend credit, demand stronger terms. The Q2 results were largely in line with what you’d expect, although it’s worth putting some numbers behind the trend.

Total origination fell QoQ to its lowest level since Q3’23, which isn’t exactly surprising. We covered this in Round #52, where Reuters noted that the slowdown in deal activity came just as direct lending funds were raising record amounts of capital. Those two trends rarely coexist for long. Eventually, managers have to deploy that capital, and that usually means competition for deals increases and underwriting standards begin to loosen.

Oddly enough, that might not be such a bad thing. Private equity dealmaking has already been slow, and if lenders stay this picky, it’s hard to see transaction volumes picking up anytime soon. A little more competition among lenders could be exactly what the market needs to get deals moving again, even if it comes with a bit more flexibility on terms.

Pricing improved across three of the five market segments (below), while Large Cap and Opportunistic Credit stayed flat QoQ. Interestingly, last quarter it was the exact opposite, with Large Cap and Opportunistic spreads tightening by 25 bps and 12 bps, respectively. So, over the past two quarters, we’ve basically had a perfect catch-up.

On stronger terms, Wells’ report shows that both average senior net leverage and interest coverage on new loans improved by 0.2x, to 4.4x and 2.6x, respectively. Media-beloved “good PIK” also fell to its lowest level in the past two years.

Lastly, let’s talk about the maturity wall. On its own, the chart below pretty much matches what the media has been saying: the 2028-2029 cohorts are shaping up to be the fun ones to deal with. But as I’ve written before, I’m less pessimistic about refinancings than most. Historically, these maturity walls have a way of smoothing themselves out as borrowers refinance ahead of schedule.

Software loans are definitely the wildcard because of how much uncertainty AI has created around the sector. Even so, if you compare the Q1 and Q2 charts, you’ll notice that the 2026 and 2027 cohorts have already started to shrink, suggesting at least some borrowers are refinancing early (see the chart).

Ironically, I actually think the 2026 and 2027 cohorts are the tougher ones. Those loans will need to be refinanced before we have a much clearer view of what AI will actually do to the software industry, which makes the early refinancing activity even more encouraging. Of course, it’s probably the stronger credits extending first. The real question is whether that trend eventually trickles down to the rest of the market.

One other thing that caught my eye is the increase in 2033 maturities. That’s probably driven mostly by new issuance, with some contribution from refinancings of the 2026-2027 vintages, although I’d guess those are still the minority.

To be honest, I’m surprised lenders are still comfortable writing 7-year loans to software companies. If AI has taught us anything, it’s how quickly technology can change. Personally, I’d argue that even the traditional 5-year term is too long for software businesses. ChatGPT has been around for just 3.5 years, yet it has already forced the entire software industry to rethink its future. If you asked me, 3-year loans should probably become the new standard for the sector (and I’m in the camp that AI’s impact on software has been overblown).

The problem is that no lender can do that on their own. Push for a 3-year maturity, and the sponsor will simply find someone else willing to write a 5-year deal, and you probably won’t get another call from them. At the same time, lenders obviously can’t coordinate on shorter maturities without raising antitrust concerns, or at the very least triggering backlash from the private equity crowd. So the only realistic path is to slowly shift the narrative by talking about it on podcasts, writing about it on social media, and bringing it up in interviews until everyone accepts that 3-year loans should become the new standard for software companies.

I guess I just fired the first shot.

Download Wells Fargo Q2 Report

4. CLO Equity, Software, and AI

The Credit Edge podcast is one I listen to quite often, and every now and then they put out a good episode. I enjoyed their recent conversation with Tom Majewski of Eagle Point Capital, a firm best known for investing in CLO equity. A few interesting tidbits below:

• On AI’s impact on the software industry

  AI disruption is real, but the market may be getting ahead of itself. Like Amazon transformed retail without destroying it, AI will reshape software over time, with mission-critical applications likely proving far more resilient than lower-value tools. There will be an impact, likely more severe than many expect, but it will come later because the implementation of innovation takes time.

• On software loans

  Most loans are refinanced well before their contractual maturity, so borrowers often have another opportunity to refinance before facing a true maturity event. Companies can also move between the syndicated and private credit markets depending on their circumstances, giving them multiple refinancing options. Companies with revenue and EBITDA may continue accessing the capital markets despite deteriorating performance. There will eventually be a day of reckoning, but time is a lender’s friend, as borrowers can refinance from SOFR +300bps to +600bps or even 800bps, because paying higher spreads is preferable to losing all their equity.

• On rising CLO default fears

  There are always outliers. Even in the CLO 1.0 era, 96% of CLOs delivered positive equity returns, with median IRRs well above marketed base cases. Historically, the biggest risk to CLO equity hasn’t been defaults but spread compression. In 2025, Nomura estimated CLO equity returned -15%, driven less by credit losses than by loans repricing tighter (e.g., 375 bps to 325 bps), reducing asset income. Managers can offset some of this by refinancing and resetting CLO liabilities, but liability spreads tightened far less than asset spreads, causing net interest margins to compress despite proactive management. Ironically, bull markets and loan repricing have historically hurt CLO equity more than defaults. Even a hypothetical 10% default rate over the next 12 months may be less damaging than severe spread compression, depending on where loans trade.

• On CLO equity returns this year

  Banks reported non-trivially negative 1H CLO equity returns, driven more by credit volatility than spread compression. Most of the spread compression appears to have played out, issuance has become more disciplined, and third-party equity participation is expected to increase in 2H after a 2025 market dominated by manager-sponsored deals. While loan price declines weighed on CLO equity valuations, realized credit losses have remained limited despite restructurings and modifications. Secondary CLO equity and existing portfolios currently offer more attractive opportunities than new issue CLOs.

• On private credit loan marks

  Private credit valuation opacity remains frustrating, particularly when loans are held at 100 for multiple quarters and then suddenly marked at 30. While differences such as 98 versus 95 across BDCs are reasonable for Level 2/3 assets, marks of 60 versus 100 on the same loan should not happen. At the same time, revenue across many portfolio companies continues to grow, EBITDA is generally flat to growing, and 10%-30% revenue declines remain isolated rather than widespread. Even so, most BDCs continue to trade at discounts to book value, reflecting market expectations of future credit losses that are viewed as overpriced.

Overall, it’s worth a listen.

5. Deals on the Block

Download below: the complete Excel file covering deals completed in 3Q26 (plus the prior three quarters).

CLICK to Download 3Q26 Details

KIÉR LIÓR – Providing outsourced private and leveraged loan underwriting and portfolio management to institutional investors, family offices, and direct lenders.

www.KierLior.com

6. Bulletin Board

• Four debt/equity requests: $35m, $12m, $9m, and $7m (LINK)

• WSJ Online is offering 65% off its annual subscription ($4 per week vs $11) LINK.

• Bloomberg is offering 60% off its annual subscription ($180 vs $399). If you are in Business, Bloomberg is a MUST, with no real close alternative (LINK).

• Shortcut.AI, an AI tool for Excel that lets you build models by simply typing requests (similar to ChatGPT). It is free to use up to a certain number of tasks per day (LINK). If you hit your limit before finishing your project, simply upgrade to the Pro plan ($20/month) and enter promo code DEBTSERIOUS (all capital letters) to get 50% off.

• LoanEdge, a BDC research tool that lets you look into BDC loan compositions, individual loans, which BDCs own a specific loan, and more, is kindly providing DEBT SERIOUS readers with a free 3‑month trial (the regular price is ~$3k per year, and it’s higher for institutions). Here is a LINK to the tool. Send an email to Sadaf Khan at sadaf@theloanedge.com and let her know you are from the DEBT SERIOUS community.

As a reminder, I don’t personally benefit from either of these offers. I am just saving you money.

Lastly, our community is over 4,400 strong and includes a broad cross-section of mid- to senior-level professionals across private credit, private equity, LPs, and debt advisory. Please use the Bulletin Board at the end of each weekly letter to share events, discounts, or deals that benefit the community. I will continue to feature these opportunities and personally connect interested parties at no cost.

That’s the bell — round over. See you in the next.

Aznaur

Aznaur.Midov@KierLior.com

News

Articles I found interesting. Doesn’t mean I agree with their takes.

1. The Fed’s shift away from forward guidance is debated. But the Treasury market’s structure has changed: hedge funds hold 2x more Treasuries than two years ago, increasing the risk of a margin call spiral (Financial Times)

2. Goldman launched a product consisting of a basket of 18 equal-weighted AI-related HY issuers (CoreWeave, Cipher, etc.), tradable in $50m-$250m sizes. Average bond yield is 7.45%. JPM launched a similar product with 11 equal-weighted IG issuers (Microsoft, Meta, Amazon) (Bloomberg)

3. Wellington views most AI data-center debt as unattractive due to spreads at least 2% tighter and LTVs of 90% vs. 60% four to five years ago (Bloomberg)

4. SoftBank’s $40B bridge loan for its OpenAI investment will generate $100m+ in fees for JPM and Goldman, with other banks earning tens of millions. Banks are staying close to SoftBank to secure future IPO and deal mandates (Bloomberg)

5. In parallel with 2008, financial engineering in debt markets is on the rise. One example is collateralized fund obligations that allow life insurers to reduce regulatory capital requirements from 30% to 19% (Bloomberg)

6. UBS, Cantor, and Partners Group are pitching investors senior tranches of collateralized fund obligations, along with other insurance-enhanced structured products that carry A2 credit ratings (Bloomberg)

7. Insurance regulators closed a CLO investment loophole for life insurers after working on it for four years. By then, many had already shifted to other structured credit products with similar risk characteristics (WSJ)

8. S&P warns of rising risks for UK pension insurers, whose portfolios are composed of 10%+ private assets. Some hold 10%+ in private credit alone, in addition to other private assets (Financial Times)

9. There is a record $350B of 10+ year-old assets stuck in U.S. PE funds, 3.5x 2015 levels and 100x 2005. Many were acquired in 2020-21 by funds raised in the mid-to-late 2010s (WSJ)

10. Some wealth managers are rebranding “semi-liquid” funds as “conditional liquidity” or “periodic liquidity” funds after investors were gated in non-traded BDCs (Bloomberg)

11. Blackstone’s Jon Gray sees a material deterioration in redemption activity, though he admits it’s early in the quarter. Q2 inflows also dropped 70% YoY to $1B (Financial Times)

12. KKR is considering strategic alternatives for its REIT (KREF), whose shares have fallen 65% over the past five years (Bloomberg)

13. Big banks are refocusing on CRE to grow loan books, targeting multifamily housing and industrial properties tied to data center projects. BofA and U.S. Bancorp grew CRE loans 8% YoY, PNC 15%, and Truist 25% (WSJ)

14. BlackRock’s bond division, which holds Aston Martin’s debt and is party to a lender co-op, could face an internal clash with HPS. Aston is seeking to move assets away from existing lenders and raise new debt from HPS (Bloomberg) [As of this writing, Aston has raised $740m from HPS]

15. Distressed U.S. companies are opting for UK restructurings over Chapter 11, attracted by a more surgical, less disruptive, and more confidential process (Bloomberg)

16. CashCall filed for Chapter 11 after two judgments totaling $400m. The subprime consumer lender once charged APRs as high as 135% (Bondoro)

17. Credit secondaries surpassed $20B in 1H26, more than doubling YoY and already exceeding full-year 2025 volume. GP-led transactions accounted for 83% of transaction volume (ACI)

18. Japan will ease bank lending caps to finance larger acquisitions and AI data centers. The move backs the government’s push into 17 priority industries, including AI, semiconductors, and shipbuilding (Bloomberg)

19. Developed economy government debt to hit a record $76T by 2026, up $4T this year on budget deficits, geopolitical tensions, and higher spending. Debt-to-GDP has climbed to 104% from 68% two decades ago (Reuters)

Transactions

1. BlackRock’s GIP is to sell bonds backed by stakes in its funds. It will bundle $1B+ into a collateralized fund obligation (CFO), with the senior tranche at SOFR + 250 bps and the mezzanine at SOFR + 475 bps (Bloomberg)

2. Blackstone-owned AirTrunk is raising $3B (A$4.3B) from banks to finance the construction of a data center in Australia. Lenders include Morgan Stanley, DBS, ING, HSBC, Crédit Agricole, and others (Bloomberg)

3. JPMorgan and other U.S. banks to help fund Japan’s $550B U.S. investment pledge, made to Donald Trump. Japanese banks passed as funding long-term projects with yen is too costly (Reuters)

4. In one of India’s largest private credit deals, Shapoorji Pallonji Group raised 151B rupees ($1.6B) through a private bond placement. Investors included local firms, as well as Farallon, Cerberus, and others (Bloomberg)

5. Despite Iran’s strikes, Kuwait raised $6B in USD-denominated bonds. The deal included 3-, 5-, and 10-year tranches, priced at 70-85 bps over Treasurys (Bloomberg)

Chart of the Week

Global bonds were hit this week as energy prices surged again. UK gilt yields this week posted their longest run of daily closes above 5% in almost two decades, while Germany’s 10-year yield hit its highest level since 2011. The U.S. 30-year yield finished just shy of its highest level since 2007.

• Bloomberg is offering 60% off its annual subscription ($180 vs $399) LINK

  • If you're not a subscriber and the article is behind a paywall, search for the article title online. It's often reprinted on Yahoo Finance for free.

• WSJ Online is offering 65% off its annual subscription ($4 per week vs $11) LINK

These are NOT sponsored links, just money-saving tips.

1. DEBT SERIOUS: One Year Later

2. Goldman Sachs pitches LPs to buy back their own capital call lines

3. Cox Capital’s tender offer for Apollo, Ares, and HPS BDCs

4. Notable Deals

5. Bulletin Board (including THREE new deals)

Disclaimer: There are 4,300 + of us here discussing private credit and leveraged loans, which is not a big world. At some point, I will probably throw punches at your company. I am blunt and sarcastic and, as friends like to remind me, too mouthy to live to an old age, so my take likely won’t please you. But don’t worry, next week it will be someone else in the spotlight and you will quietly nod along in satisfaction.

Also, none of this is investment advice.

P.S. If you lend / invest in Latin America, don’t miss the Bulletin Board. There are three Energy deals posted this week.

1. Debtiversary

This week marks the first anniversary of DEBT SERIOUS. That’s right, I’ve been ranting about private credit and leveraged loans for a year now. And the community of people who voluntarily put themselves in the position of listening to those rants has grown to 4.3k subscribers and ~500 followers (followers are basically just shy subscribers). There are almost 100 of you debtageeks who even decided it was worth paying for my free newsletter, which is really flattering (and I, of course, appreciate it, and am working on something for you, as promised).

Perhaps the most fascinating thing is how diverse this community is, from students to senior executives in alts... including people from firms I've thrown punches at, so kudos for being good sports and not sending me angry emails or unsubscribing.

Jokes aside, I just wanted to thank you all for letting these letters into your inbox, helping make the newsletter one of the largest in the private credit/LevFin space, and pushing the podcast into the top 2-3% of business podcasts globally.

Now that my LinkedIn-style humblebrag is over, I’d like to ask you to make this a two-way conversation: send me an email with a few words about yourself (aznaur.midov@kierlior.com), add me on LinkedIn, forward the newsletter to your junior staff so they can nerd out, or to your bosses... so they can do what the big guns I mentioned two paragraphs ago are already doing.

Also, start using the Bulletin Board. It’s a free resource to downsell exposure (if you’re a lender), raise debt (if you’re a debt advisor or in PE, idk), announce a tender offer for a BDC (hey, you never know), or post anything else you think the community could benefit from.

Alright, one more thing: this letter is shorter than usual because I’m traveling, but I didn’t want to break the weekly ritual.

2. Goldman Sachs: Capital Call Remix

If you are not intrigued by the article titled Goldman Pitches Investors on Loans to Their Own Private Funds, I am not sure you are a human. Goldman’s new pitch fascinates me and, to be honest, mildly confuses me as well. Here is what Bloomberg says:

Goldman Sachs’ fund-finance team is pitching a trade that would allow large investors to buy back so-called capital call lines to their own funds. The bank would free up balance sheet capacity and keep some of the extra spread as part of the transaction, while limited partners would receive interest on capital that would otherwise be sitting idle.

I still remember the first time I learned about capital call lines (aka subscription lines) and then proceeded to ask my senior about 50 “Wait, but why?“ questions. For those who don’t know, most mid- to large-cap PE funds have one (well, GPs for that matter, but let’s stick to PE). It’s a revolving credit facility that the fund typically draws on when it makes an investment. In other words, the “equity” piece is initially funded by debt.

The primary reason is simple: convenience. Buying a company is a chaotic process with dozens of moving pieces. The last thing anyone wants is to add another one by coordinating capital calls with 50 different LPs while trying to get a deal across the finish line.

Without a subscription line, the instructions would sound something like this: "Reach out to all 50 LPs. Ask them to be ready to wire their pro rata share for an investment we're about to make, even though the purchase price is still tentative and likely won't be finalized until 3-7 days before closing. So, tell them to earmark roughly X, but expect it to change. If they send too much, we'll wire the difference back. Oh, and make sure all 50 wires hit our account no later than the day before closing. Go."

Alternatively, they can just tell Goldman, with whom they already have a subscription line in place, that the acquisition is coming a few days before the transaction closes and receive one wire for the entire amount needed. After the transaction closes and the chaos subsides, they can reach out to the LPs and request the funds. The difference is that LPs know exactly how much they need to wire, and the PE fund can collect those funds gradually, as not all LPs send their money at the same time. Once all the capital has been received, the line is repaid.

Two things before going back to the article:

• These subscription lines are underwritten based on the credit quality of the LPs because, ultimately, the LPs are the ones on the hook. If the PE fund doesn’t repay the line, the LPs are still required to wire their share of the committed capital. That’s why these are considered very safe loans and typically yield around SOFR + 150-200 bps.

• These lines are not repaid right after closing. They are typically outstanding for up to 6 months or even longer. PE funds are not in a rush to repay them because doing so helps boost their IRR by delaying when LP capital is called. Since IRR starts accruing only once LPs have wired their capital, delaying those capital calls artificially inflates reported IRRs.

With that background, here is what seems to be happening. Goldman basically told LPs: “Look, you guys fully guarantee these lines, but you do not even get paid for it. Why not earn some spread?” And this sounds like a no-brainer pitch.

If you think about it, as it stands today, LPs are indirectly lending money to PE funds at 0% (since they guarantee the facility), which makes it seem like they are doing it simply to boost PE funds’ IRRs, which is pretty strange. Under Goldman’s proposal, they can at least get paid for doing it, so people like me do not write pieces like this one.

Here is how I think it would work. Say a PE fund (”Bhoma Travo”) needs $50m. Bhoma still calls Goldman, with whom it has the facility in place, and Goldman funds $50m for the acquisition. Goldman then calls Bhoma’s LP (PamPERS), who agreed to participate in this new arrangement discussed in Bloomberg. The deal is that PamPERS will wire Goldman, say, 90% (just guessing) of the $50m ($45m), with Goldman keeping the remaining $5m, which I assume it would need to retain as the agent. PamPERS is now earning 5.50% (SOFR of 3.70% + 1.80%).

When Bhoma finally decides to repay the loan, it would reach out to PamPERS, asking it to fund its $50m commitment, because on Bhoma’s books, that transaction between Goldman and PamPERS didn’t happen. Then PamPERS, Bhoma, and Goldman hop on a call, and Goldman says: “Hey, Bhoma, don’t worry about it, we will work with PamPERS directly. PamPERS, you already sent us $45m, so just send us the remaining $5m.” All agree. PamPERS sends Goldman $5m, makes a few book entries in its accounting system, and that’s it. Everyone is happy. BTW, Bhoma was most likely aware of the transaction, so they didn’t go, “Wait, what are you two talking about?”

This is just my take, and I’m sure the transaction will be structured with more nuances, but directionally it should be correct.

There is one clarification in the piece:

This trade is meant for large investors who have committed capital through separately managed accounts where they are the sole investor in the fund, the people said. Groups of limited partners, working in coordination with fund managers, could also participate, they added.

The point is that Goldman wants to deal with only one PamPERS, or a few, because dealing with 50 would be a nightmare.

3. Cox Capital

Reuters ran a piece titled Private credit roundup - Discounts show the cost of getting out, which worked as intended, and I clicked. The article is basically a collection of loose facts tied together by their proximity to BDCs, but it was the second paragraph that I was most curious about:

Pressure in private markets is showing up in an ​uncomfortable place: the gap between what funds say their assets are worth and what investors may have to accept to ‌get out now.

Cox Capital Partners this week launched tender offers for shares in non-traded business development companies run by Apollo, Ares, and HPS, offering to buy them at discounts of 15% to 30% to their May-end net asset values.

Those of you following the BDC developments will recognize Cox Capital’s name because, a few months ago, they teamed up with Saba Capital and launched a tender offer for three of Blue Owl’s BDCs at 20-35% discounts to NAV. To my surprise, despite Blue Owl being featured in virtually every negative private credit headline, less than 1% of shares were tendered.

So, Cox is back, but this time it’s targeting BDCs managed by the largest alts, which also happen to have received the largest redemption requests this quarter. Cox’s press release provides the clearest summary of the tender offer and the rationale behind it:

“Comparable traded BDCs trade at meaningful discounts to net asset value. We priced each offer on three things — portfolio quality, the liquidity characteristics, and those observable traded-market discounts — not on a negative view of these funds,” said John Cox, CEO and CIO of Cox Capital. “Repurchase demand ran well ahead of issuer program capacity again last quarter, and we believe the industry needs a standing utility for the unfilled remainder. For a holder seeking full liquidity, the practical arithmetic is a blended exit: shares accepted by an issuer program are repurchased at or near net asset value, remaining shares can be sold to us.”

There are a few things to note in the above:

First, Cox makes a very clever case to investors, basically saying, “be realistic: if you need money quickly, get some prorated at par and tender some to us.” On a blended basis, that’s not a bad outcome. Look at Ares, for example, where they are offering to buy shares at a 15% discount to NAV, which basically means that if, for every share you redeem at par, you tender one share at a 15% discount, you receive 92.5% of your money on a blended basis. Not perfect, but for those who need funds quickly, that might be attractive.

Second, they don’t bash the funds directly. In fact, Cox clarifies that this is not a negative view on the funds themselves. That’s likely true, although the fact that they offer different discounts to different funds may offend some managers (I don’t think Cox cares, though). They do note that they based the offer on portfolio quality, liquidity, and comps. The last two are straightforward, but the first is probably harder to figure out because there is no public information about the borrowers’ financial health.

Third, the assigned discounts are somewhat surprising. Apollo, which is often regarded as one of the leaders in the private credit space, is being offered the highest discount, which kind of stinks for them if they happen to see the table (well, they obviously don’t care, but it’s still optically annoying).

What’s also important, and not immediately intuitive, is what those discounts actually mean. I’ve seen people online throwing around their estimated values for the loans in these BDCs, saying they’re worth 60-70 cents on the dollar, without realizing how scary those numbers would be in reality. That would basically imply wiped-out equity across virtually all deals, which would affect not only the funds, but even your 401(k) because of the shock such a situation would send through the markets, reminding us that, in a panic, all assets become correlated.

The reason, in my view, is that people are just used to hearing that BDCs trade at, say, 0.7x NAV, and they intuitively assign the same discount to the underlying loans. They forget that BDCs are typically levered roughly 1:1 ($1 of debt for every $1 of equity, where equity is BDC NAV). So a 0.7x NAV valuation doesn't mean a 30% discount to the assets; it's 15%, because leverage amplifies the effects of asset changes on NAV.

I’m going to over-explain this because, in my opinion, these dynamics are at the heart of Cox’s investment thesis. Let’s use made-up numbers, but tie them to Cox’s tender offer for HPS shares at a 25% discount to NAV.

In the chart below, HPS used $100 of equity (NAV), levered it 1:1 by borrowing $100, and originated $200 of loans, which it currently carries at par. Well, Cox is likely skeptical that those loans are worth par, but they don’t know their exact values, so they use the redemption panic to offer to buy the shares at a 25% discount to NAV, or $75. Because the debt amount is unambiguous, Cox is implicitly valuing the loan portfolio at $175 ($100 of debt + $75 of equity). So, divide $175 (Cox’s valuation) by $200 (HPS’ valuation), and you get 87.5%, which implies only a 12.5% discount (not 25%). Make sense, right?

Let’s say 100 thousand shares were tendered at $75 each, for a total of $7.5m. The best part of this strategy, at least the way I see it, is what comes next. Cox is not going to be an activist, trying to remove management and hire KIÉR LIÓR (my shameless plug) to manage those loans. When the redemption window opens, they will request to redeem all 100k shares they acquired. Well, they will be gated, and only 30% of their redemption request will be granted, so they will redeem 30k shares at $100 each, earning a 33% return ($25 gain on a $75 purchase price). Multiply the 33% quarterly return by 4 to annualize it, and you get an ~132% IRR (don’t nitpick). The following quarter they will do the same thing, be gated again, but redeem another pro rata portion of their shares. The IRR will be lower, however, since you only multiply the return by 2 to annualize it, as two quarters have passed. They’ll keep repeating the process until all of the shares are redeemed.

But that’s not all. HPS BDC also pays a 9% dividend, or $9 for every $100 of NAV. Cox will receive the same $9, but its cost basis is $75, so its dividend yield is 12%. That’s right, Cox will be gradually redeeming HPS shares at NAV, earning a juicy IRR while also collecting a 12% yield on the shares that haven’t been redeemed yet. By the way, that 132% IRR above is understated because it doesn’t include the dividend. So yeah, Cox will do just fine.

4. Deals on the Block

Download below: the complete Excel file covering deals completed in 3Q26 (plus the prior three quarters).

CLICK to Download 3Q26 Details

KIÉR LIÓR – Providing outsourced private and leveraged loan underwriting and portfolio management to institutional investors, family offices, and direct lenders.

www.KierLior.com

5. Bulletin Board

• Three LATAM debt/equity requests: $35m, $12m, and $9m (LINK)

• WSJ Online is offering 65% off its subscribers ($4 per week vs $11) LINK.

• Bloomberg is offering 60% off its annual subscription ($180 vs $399). If you are in Business, Bloomberg is a MUST, with no real close alternative (LINK).

• Shortcut.AI, an AI tool for Excel that lets you build models by simply typing requests (similar to ChatGPT). It is free to use up to a certain number of tasks per day (LINK). If you hit your limit before finishing your project, simply upgrade to the Pro plan ($20/month) and enter promo code DEBTSERIOUS (all capital letters) to get 50% off.

• LoanEdge, a BDC research tool that lets you look into BDC loan compositions, individual loans, which BDCs own a specific loan, and more, is kindly providing DEBT SERIOUS readers with a free 3‑month trial (the regular price is ~$3k per year, and it’s higher for institutions). Here is a LINK to the tool. Send an email to Sadaf Khan at sadaf@theloanedge.com and let her know you are from the DEBT SERIOUS community.

As a reminder, I don’t personally benefit from either of these offers. I am just saving you money.

Lastly, our community is over 4,300 strong and includes a broad cross-section of mid- to senior-level professionals across private credit, private equity, investment banking, LPs, law, and other related fields. Please use the Bulletin Board at the end of each weekly letter to share events, discounts, or deals that benefit the community. I will continue to feature these opportunities and personally connect interested parties at no cost.

That’s the bell — round over. See you in the next.

Aznaur

Aznaur.Midov@KierLior.com

News

1. Profile of Apollo, covering its evolution from a private equity firm to an investment giant rivaling the biggest Wall Street banks. The article includes notable deals, such as Intel, Anthropic, and Vonovia (Bloomberg)

2. Cox Capital launched a tender offer for shares of non-traded BDCs managed by Apollo, Ares, and HPS at 15% to 30% discounts (Reuters)

3. Per survey, 57% of global insurers plan to increase private credit allocations over the next 12-24 months, up from 37% in 2024. Interest is concentrated in IG assets, including Direct Lending, ABS, and NAV (Financial Times)

4. Citi’s turnaround under Jane Fraser, marked by asset sales, headcount cuts, and outside hires, finally restored the bank’s pre-GFC market cap. Analysts are bullish, but former/current employees have concerns (Financial Times)

5. Private credit underwrote 2022-23 deals assuming rates would quickly decline. Now the market is pricing fresh hikes, which no one underwrote for, signaling more pain ahead (CNBC)

6. Private credit is expected to outperform PE over the medium term as the higher-for-longer interest rate environment persists (Planadviser)

7. Goldman is pitching capital call lines to institutional investors, allowing them to buy back facilities extended to their own funds (Bloomberg)

8. Private credit is providing preferred equity to PE firms as slower exits drive liquidity needs. Issuance rose from $6B in 2024 to $9B+ in 2025, funding LP distributions and new investments (9fin)

9. Apollo to invest ~$20B in Mexico to finance projects, including infrastructure and other debt. The Mexican government has been courting private capital to finance the country's infrastructure (Bloomberg)

10. Hyperscalers’ bonds make up 4-5% of the IG index, but 8.3% of its Duration Times Spread (risk measure), meaning their impact on risk exceeds what their market-value footprint alone would suggest (Bloomberg)

11. Oracle's credit rating was cut to BBB-, just one notch above junk, as the company burns cash faster than it can generate revenue while funding a $250B data center buildout (Bloomberg)

12. Green bonds are on the rise amid backlash against data centers. Proceeds can finance data centers that meet energy-efficiency standards or include the development of renewable power such as solar and wind (Bloomberg)

13. BofA and the City of Chicago scrapped plans to sell the city's $1B+ of past-due receivables, including parking tickets. Lack of precedent, privacy concerns, legal risk, and reputational concerns killed the deal (Bloomberg)

14. Despite the 10Y+ IG corporate bond index yielding around 6%, its spread over Treasurys is only about 50 bps wider than that of 1-3Y IG bonds, making the long end less of a no-brainer (WSJ)

15. Middle East bond spreads hit a 4-year high of 402bps over Treasuries amid the U.S.-Iran conflict, making it the worst-performing EM dollar bond region, with spreads 163bps wider than peers (Bloomberg)

16. Indian private credit and ABS are rising, driven by favorable market conditions and 14-17% IRR potential (FinanceAsia)

17. Wall Street banks say consumers remain healthy as spending fuels consumer loan growth, with delinquencies below expectations (Reuters)

18. New York’s one- and two-year rent freezes will affect 1 million rent-stabilized apartments, capping revenue while operating costs continue to rise. Near-term CMBS risk is modest, but likely to increase over time (Fitch)

Transactions

1. The New York Yankees and Apollo are working on a $3B financing package, consisting mostly of debt and some equity, for the baseball team. Proceeds will refinance existing debt and fund growth investments (Bloomberg)

2. Blackstone-owned QTS Realty (data centers), which had been in the process of raising a $1B term loan and $1B bond issuance, instead opted for a $3.25B loan sale led by JPMorgan (Bloomberg)

3. Robinhood to sell $400m ABS backed by consumer credit card receivables, with the highest-rated tranche yielding SOFR+0.8%. It’s the company’s first offering of this kind (Yahoo Finance)

4. Connecticut public utility RWA to issue $2.4B of Munis to finance its acquisition of Aquarion Water Company from publicly traded Eversource Energy (Bloomberg)

5. Mitsubishi UFJ lined up $1.5B of debt to finance Aditya Birla’s acquisition of Shell’s renewable assets in India (Bloomberg)

6. Taiwan-based GMI Cloud is raising NT$20.45B ($635m) in a GPU-backed loan, primarily from Asian banks active in similar U.S. transactions (Bloomberg)

7. The African Dev. Bank approved a $400m loan to South Africa, which will be guaranteed by the UK, to fund water and electricity projects in coal mining and power plant regions under a climate finance program (Financial Post)

Chart of the Week

• Bloomberg is offering 60% off its annual subscription ($180 vs $399) LINK

  • If you're not a subscriber and the article is behind a paywall, search for the article title online. It's often reprinted on Yahoo Finance for free.

• WSJ Online is offering 65% off its subscription ($4 per week vs $11) LINK

These are NOT sponsored links, just money-saving tips.

In Round #52:

1. PitchBook's forecast for the size and composition of private credit in 2030

2. Is liquid credit's criticism of private credit justified?

3. HSBC cuts exposure to risky private credit funds

4. Interpreting KBRA's statistics on recurring revenue loans and ARR ABS

5. Direct lending raises more but deploys less

6. Notable deals

7. Bulletin board

Disclaimer: there are 4,300+ of us here discussing private credit and leveraged loans, which is not a big world. At some point, I will probably throw punches at your company. I am blunt and sarcastic and, as friends like to remind me, too mouthy to live to an old age, so my take likely won’t please you. But don’t worry, next week it will be someone else in the spotlight and you will quietly nod along in satisfaction.

Also, none of this is investment advice.

1. Vision 2030

PitchBook just published a forecast for private market growth through 2030, covering various assets from PE to RE, and of course, private credit. There isn't a lot to discuss, but I like the charts they provided, as they remind us where private credit stands and where it is going.

We are at $3.8T today and projected to hit $5.6T by 2030, which means I'll have something to write about for the foreseeable future. Now, that $3.8T is a rather elusive number because you'll see different outlets cite different figures, with Bloomberg, for example, insisting it is $1.8T. Part of the discrepancy is that many outlets keep using outdated numbers, and part of it is that the definition of private credit has been evolving.

PitchBook has also provided a breakdown of private credit by strategy, which is very helpful.

Assuming the current breakdown of private credit is directionally similar to what it will look like in 2030, most of the noise around private credit comes from the 8.3% allocated to wealth-focused evergreen funds (redemptions). That 8.3%, together with the 20% allocated to direct lending and 2.5% to mezzanine, are also where the other major headache associated with private credit sits: software loans.

So, when we hear that 20% of BDC loans are software loans, I guess that's 20% of 31%, or about 6% of total private credit. Let's round that up to 10% for conservatism, since other strategies have software exposure as well. Hey, maybe this PC thing isn't a systemic risk after all!

2. Pile-On: Liquid Credit on Private Credit

Readers know that I’m often critical of the media’s coverage of private credit, which is largely driven by fearmongering and clickbait. The media doesn’t make up facts, but it often presents them without context, making situations appear far more dramatic than they actually are. If you read the financial press and base your views solely on what you read, you will almost certainly come to the conclusion that private credit is in trouble.

But there is another group that has been vocal about private credit troubles, and they carry far more credibility than the media: public credit investors, or liquid credit investors, if you prefer that term. Whether they speak on a conference panel, appear on a podcast, or post on social media, their words carry more weight because they are experts in a closely related field. After all, credit is credit, so they know what they are talking about, right? And they do, but they, too, miss important nuances.

It is worth addressing the elephant in the room. For years, particularly during the low interest rate era, institutional investors increased their allocations to private credit in pursuit of higher yields, often at the expense of traditional public fixed income. Several years later, a different dynamic is emerging. I've read about it a few times, but since I'm not in the allocation game, I'll let BlackRock do the talking:

We also believe private credit is increasingly being considered by investors in the context of their broader fixed income allocations. As a result, we see scope for the fluidity of investor allocations between public fixed income and private credit to increase over time.

Show me the incentive and I’ll show you the outcome, right? Well, show me the asset class that's taking capital that was historically allocated to mine, and I'll show you the asset class I despise.

The criticism is the same: misleading marks, misleading default numbers, and misleading performance figures that are largely the result of the prior two issues. Basically, in their view, private credit generates all those fees by making up its numbers, which I'm sure is infuriating to the liquid credit crowd. Again, they have a point, but there are also nuances that explain many of these dynamics. Let's review.

Misleading Marks: Even people within private credit would acknowledge there is some truth to this criticism. They would tell you that marks are often too sticky and are not adjusted as quickly or as much as they should be. But there is a big leap between saying, "Yeah, I think, based on recent performance, this loan should probably be marked a few points lower," and claiming that private credit loans are "marked to myth," as critics often do.

The liquid credit crowd is used to marking portfolios to market continuously, with valuations fluctuating based on macro factors and market sentiment, not just changes in the underlying borrower. To them, keeping a loan at the same value for several consecutive quarters is a foreign concept.

Well, to private credit, constant marking has been a foreign concept. Until recently, when some funds started moving to daily NAVs, few actually thought it was a problem. When a private credit origination team presents a deal to the investment committee, everyone assumes the loan will stay on the books until maturity. As a result, the discussion centers on the fundamentals: whether the borrower is strong enough to support the debt, whether the loan structure is sound and includes the right covenants to bring everyone back to the table if performance deteriorates, and whether the pricing adequately compensates for the risk. Those value fluctuations are somewhat of an afterthought.

Well, I guess they are important if we're talking about fees, specifically fees that BDCs charge based on NAV. But the fee discussion is less relevant for traditional drawdown funds, where the manager is compensated based on committed or deployed capital. Since BDCs represent only a small portion of the private credit market, I'm setting aside the discussion of fees tied to fluctuating NAVs and focusing instead on the ultimate value of the loans at maturity, and why fluctuations along the way are not top of mind. Hopefully, we're clear here.

Misleading Default Numbers: This point has two separate arguments: (i) private credit lenders often prevent formal defaults by amending loans ahead of projected stress; and (ii) PIK structures can mask the true extent of underlying defaults or distress. Frankly, both arguments are technically true, and they do prevent the occurrence of formal defaults. But the motivation behind them matters. In lending, these are often routine loan management tools rather than evidence of anything shady.

Let’s start with the amendment point. The borrower knows it is going to miss a leverage covenant, so it notifies the sponsor. The sponsor reaches out to the lender saying, “Hey, can we hop on a call?” which is usually a cue for, “I’m about to ask for something.” These situations rarely come out of the blue. Based on the borrower’s performance, everyone was expecting the call. The conversation typically goes something like this:

PE: (after some small talk) Look, don’t want to take up too much of your time, but as you know, things are taking longer than expected. Some synergies haven’t materialized, but it’s a timing issue. [Fill in the blank] took longer than expected. We’ve worked closely with management and are very confident we’ll get there, so we need some covenant relief. Here’s our updated model and the revised covenants we’re proposing.

PC: Cool, let us discuss internally and revert.

Obviously, this is a cordial conversation, not a one-sided monologue.

The lender then discusses the deal internally, tightens the cushions on the proposed covenants, and at a minimum charges amendment fees. Depending on the severity of the situation, the lender may also require an equity injection, a partial loan paydown, or outright refuse the request, instead waiting for an actual default, which gives lenders significantly more leverage to demand more substantial changes to the deal structure.

You need to put yourself in the shoes of a lender. More often than not, these covenant breaches are not severe, and even after the reset, you’ll still catch any further deterioration if the borrower’s performance continues to weaken. You are also compensated for providing that relief. And as I said, depending on the severity of the situation, the lender may decide it needs more than just amendment fees, or respond, “We don’t think we can do it, so take us out” (well, you wouldn't actually say that, but rather overstructure the deal and demand so much that the sponsor would simply refinance you). In fact, anyone who has ever sat in the lender’s seat has gone through the exercise of recalculating the covenants, carefully comparing the definitions in the credit agreement against the borrower’s compliance certificate to confirm that the borrower remains within the required limits. If not, the lender comes back to the negotiating table with a completely different counterproposal.

The tone of the conversation will change dramatically from that point on, of course, but at no point are lenders trying to avoid a formal default simply to keep it out of S&P's or Fitch's statistics. As noted, a formal default actually gives them significant leverage in the negotiation.

And this is nothing new. It’s exactly how banks have managed borrower relationships since the beginning: have the conversation before the default and determine whether, and on what terms, the borrower should be supported. That's why lending has long been called a relationship business.

Are there exceptions? Of course. But that's not what we're discussing here.

PIK Masking the Default: Sometimes, the borrower is burning significantly more cash than projected and needs additional liquidity, in which case the sponsor would propose to upsize the loan to provide liquidity to the business, or alternatively, PIK some or all of the interest, with an upsize typically being the preferred option (because the cash is fungible and can be deployed outside of debt service). Again, for lenders, this is simply another tool to stabilize the borrower, not some conspiratorial way of avoiding defaults, which, as we just established, few people outside of Fitch, S&P, and perhaps distressed debt funds actually care about.

You have the same scenario of reviewing the model and determining, based on the assumptions, whether the proposed PIK period would bridge the company to positive cash flow and a recovery from the situation it found itself in.

Generally speaking, this is not how the public credit crowd operates, including high-yield managers (let alone investment-grade investors). They have no relationship with the borrower or the sponsor, so everything is very transactional. As such, the idea of working something out or coming up with different structures, such as PIK, is foreign to them. If the price of the bond is moving in the wrong direction, they check the fundamentals, and if the company appears to be heading toward a default, they are out. Heck, they are often out long before there is even a realistic possibility of default, because as soon as default becomes probable, the bond price will reflect it. There are exceptions, of course, where a small group of bondholders works things out with the issuer, but those cases are relatively rare. Which is why public credit is allergic to all this "working things out with the sponsor" and view it as some sort of conspiracy, or simply another way to kick the can down the road.

That's not a crazy idea, though. But again, you have to think like a lender. The sponsor reached out ahead of a default. They have a plan. You agreed to a PIK, but always in exchange for something else, such as fees, an equity infusion, or tighter documentation. If you can get the sponsor to put in more cash and start executing on a plan, which may ultimately involve selling the company, that's almost always the preferred route. Every other alternative, such as an LME, bankruptcy, or having the keys thrown back to you, tends to suck. So, that so-called "bad PIK" is often the better alternative to whatever would have come next.

Diagnosis of the Problem

I think the disconnect lies in how difficult it is to consistently outperform in public credit, where your performance is measured against the market every minute, and the moment you stop outperforming, investors start fleeing. Meanwhile, they look at the private credit crowd, which seems to raise seemingly unlimited capital, mark its own NAVs, and generate all these fees. At least that's my read of the situation. But I think the topic deserves a bit more discussion because the other side is rarely as easy as it looks.

Start with portfolio building. If you are in public credit and just closed, say, a $100m fund, you can wake up tomorrow morning, open your Bloomberg Terminal, and deploy the entire fund within a couple of days. You can choose from hundreds of bond issuers and screen them by rating, industry, yield, you name it. You can also exit any position at any time, with or without a reason. Industry outlook worsened? Out. Competition is catching up? Out. The CFO got a stupid haircut? Out.

Now think about private credit. You raised a $100m fund (too small, but whatever), and now you have to deploy it deal by deal. You start calling on sponsors, but they are well-banked, with multiple go-to lenders they already work with. So you’ve been wining and dining them for months, making the case for why they should give you a shot, all to no avail. One day, you finally get the call you’ve been waiting for. But they are not bringing you into a brand-new deal. Those go to lenders with whom they already have established relationships, because the sponsor needs certainty that the acquisition will close on time and without surprises

Instead, they ask you to refinance an existing deal. Remember those deals lenders overstructure to the point that the sponsor wants to refinance them altogether? Congratulations, that's what you are about to get: a hairy, poorly performing deal that their lending partners don't really want anymore.

But you still need to find a way to make it work. You need to structure the loan in a way that protects your downside while also showing the sponsor you are someone they can work with (PIK upfront?). That’s how you earn your way into becoming one of their go-to lenders, so the next time they have a clean, competitive financing, your phone rings too.

You finally closed the deal, but now you are married to it for 5-7 years. You can only get out at maturity or at a refinancing if the company gets acquired (assuming a change of control triggers repayment). If the company starts underperforming the new, sandbagged projections you underwrote to, you will get to know the other side of the sponsor: how supportive they are of the borrower during distress, and whether there is another new lender, that has been wining and dining them for months, and is willing to refinance you out of the deal. You never know how that will turn out, and you certainly would love the option to just go to your Bloomberg Terminal and get out of that deal, even at a loss.

P.S. Yes, private credit does things differently, and from the outside, some of those practices might look shady, but they aren’t. They’re simply standard tools used to achieve financing objectives. It's like seeing a group of cars speeding and bumping into each other. You turn to your friend and say, "Gee, these guys are completely reckless. They're speeding, cutting each other off. We should call the police." Your friend replies, "Well, we're at the racetrack watching NASCAR. That's exactly what these guys are supposed to be doing."

3. Risk à la HSBC

This one from the Financial Times is interesting: HSBC pulls back from riskier private credit lending:

Europe’s biggest lender told clients in recent weeks that it would not renew their facilities after deciding to stop lending to private credit funds that did not provide sufficient returns to justify the risk, according to three people familiar with the matter. Instead, HSBC will focus on lending to less risky private credit funds, the people added.

On one hand, this supports a point I've made before: if private credit cracks, it won't be the Apollos and Blackstones of the world you need to worry about, but the second-tier firms lending to less stable borrowers.

Well, but then FT continues:

Britain’s two largest banks reconsidered their risk appetite in the wake of the blow-up of the bridging lender Market Financial Solutions, which stung both banks. MFS collapsed in February amid fraud allegations owing more than £2bn to some of the world’s biggest banks and private credit firms. Barclays provisioned £228mn to cover losses from its loans to MFS, while HSBC took a $400mn charge because of its lending to Apollo’s asset-backed lending unit Atlas SP, which had lent to MFS.

I covered the MFS fiasco in Round #38, and based on the discussion in the UK, some strong actions by local banks were expected. To be clear, MFS was not a private credit fund. It was a specialty lender that collapsed due to fraud, and a number of its creditors lost money.

Barclays had provided MFS with a warehouse facility and ended up losing money. HSBC wasn’t involved in the MFS situation, but apparently still managed to lose money through its back leverage to Apollo’s Atlas.

The irony is that when HSBC says it’s cutting exposure to risky private credit, I highly doubt it’s referring to Atlas, the very platform through which it actually lost money. It’s far more likely talking about some smaller, lesser-known funds you’ve never heard of. That’s just my speculation, of course, but I can’t imagine HSBC picking up the phone and telling Atlas/Apollo it's cutting their credit line. If it does, I'd do just about anything to be a fly on the wall for that conversation.

4. KBRAcadabra

KBRA just published its Q1’26 Recurring Revenue Loan Metrics Dashboard, covering those notorious ARR software loans. The dashboard is based on weighted-average statistics from 102 ARR loans and 23 KBRA-rated recurring revenue ABS (I'll come back to those). The release feels a bit late for Q1, and the publication only includes a bullet-point summary (need to be a client to access the full report). Still, it's worth a look because it provides one of the first snapshots of the market roughly two months after the SaaSpocalypse. That's probably too early for the full impact to show up in the data, but it should give us an early read on where things stand.

Let's go through a couple of points:

On an aggregate portfolio basis, annual recurring revenue (ARR) for the dashboard population is $177.6 million, down 0.9% quarter-over-quarter (QoQ) and 0.7% year-over-year (YoY), but remains above the historical average of $175.1 million.

Right out of the gate, the decline in ARR growth is bad news. These companies have historically been valued at a multiple of ARR, with higher growth leading to higher multiples. When ARR growth slows, the multiple compresses. When it turns negative, the multiple can collapse. Before the SaaSpocalypse, companies growing ARR at 20%-30% with strong retention could trade at 7-10x ARR. If growth slows, the multiple might fall to, say, 4-6x. Once ARR starts contracting, valuations can fall much further.

That said, the average ARR in this portfolio is $175m, which provides a strong floor. Out of tens of thousands of SaaS companies globally, only about 200-300 reach that scale. Companies that large have historically been very desirable, and many of them went public around those ARR levels. Even with flat ARR, they should hold their value much better than smaller lower-middle market software companies, which can trade at, or even below, 1x ARR once revenue starts contracting.

Without going too far down the rabbit hole, there is also a good chance these companies already generate solid EBITDA because they are no longer growing as fast. Maybe not the 30%-40% margins they could theoretically achieve, but perhaps 20% ($35m EBITDA), if not more. If that’s the case, investors will gradually start valuing them based on EBITDA rather than ARR, which should help support valuations. That said, I don’t think this is simply a shift from top-line growth to EBITDA generation. Last year’s report showed 31% YoY ARR growth, and a drop this big in just one year is simply too much.

The debt-to-ARR ratio increased 2.5% QoQ to 1.7x but is down 5.1% YoY. The average loan-to-value ratio decreased 11 basis points (bps) QoQ but increased 260 bps YoY to 29.7%.

First, on leverage. At 1.7x, leverage in private credit is actually very low. I’d have thought the leverage would be closer to 2x-2.25x on average, so that’s good news. Perhaps the reason is the composition of the portfolio, which might include not just pure SaaS companies, but also so-called tech-enabled businesses that still generate recurring revenue and provide software, but where people, rather than the software itself, provide most of the value.

Just to make the point clear, think of a cybersecurity software company where you simply install the software. The gross margins could be 80%-85% because, outside of hosting costs and something called customer success (let’s not get into it), there isn’t much else. Now, think of a cybersecurity company that also has software, but where the main value is a team of people dedicated to managing your cybersecurity full time. Because of labor costs, their gross margins will be 40%-50%, and they usually get lower valuations and, therefore, lenders provide lower leverage.

That's just my speculation, though, since 1.7x ARR is more in line with historical bank lending than private credit (banks typically lent 1-1.5x ARR, occasionally up to 1.75x).

Second, on LTV of 29.7%. Let’s do math, ok? So, assume a company has $100m of ARR and carries $170m of debt (1.7x ARR). A 29.7% LTV implies an EV of about $570m, or 5.7x ARR... for a company with a declining top line and what many see as an existential threat from AI (whether that's real or not)... come on, folks, what are we doing here?

To be fair, if you do the same math using the numbers from the last report, which showed 31% ARR growth, 1.8x ARR leverage, and a 27.1% LTV, the implied value is $660m, or 6.6x ARR, so there has been solid compression. But I am still skeptical about the current valuation multiple, and I am willing to bet almost anything that you won’t exit these companies with negative growth at that multiple. Now, I don’t think they are underwater, but perhaps an LTV of 50%-60% sounds a bit more realistic today? Maybe the Q2 numbers will tell a different story. We'll see.

23 KBRA-rated recurring revenue ABS

Ok, I told you we will get back to this. Look, I’ve spent the better part of my career in the ARR lending world, and I’ve never seen these things. Apparently, there is such an instrument as recurring revenue loan ABS (RRL ABS), which is simply a bunch of ARR loans chopped into tranches, rated, and sold to investors. Well, there is a word for such a thing, a CLO, right? Although there is limited information about the structure of this instrument, I found out that there are a couple of differences between CLOs and RRL ABS.

1. While a CLO collateralizes either a portfolio of BSL loans or private credit loans (for private credit CLOs), RRL ABS is concentrated in ARR and tech loans, which tend to have higher yields than traditional direct lending loans, but also may include some BSL and/or middle market loans.

2. Unlike a CLO, where interest and principal payments are based on a waterfall, going first to senior tranches and then the remainder to junior tranches, in RRL ABS everything is pro rata. However, there is typically a ~5% default threshold, which upon breaching turns the payment structure into a traditional CLO-like structure: senior tranches get paid first, then junior.

If you want to see a real-life example, below is the structure of Vista-sponsored RRL ABS. You can dig into the docs HERE, if you are really bored.

And if you've worked on one of these yourself, reach out. I'd love to hear more about how they're structured.

5. Glass Half Levered

Here is a piece from Reuters: US direct-lending activity falls even as private credit firms raise more cash, which goes against how the media should be operating. Instead of breaking the article into two, one for the majority pessimistic crowd (US direct-lending activity falls) and one for the 20-30 optimists (Private credit firms raise more cash), they decided to combine both into one article and upset everyone. No wonder their stock is down more than 50% since the private credit turmoil began last year, while every other news outlet is printing money by selling "the next financial crisis" stories about Tricolor, BDC redemptions, and SaaS lending. Gotta change it!

Anyway, back to the article. It opens with some pretty interesting statistics:

North America-focused closed-end direct-lending funds raised $16.25 billion in the quarter, ‌up from $1.3 billion in the first quarter, according to Preqin data, the highest in two years.

…and it continues with an equally interesting statistic pointing in the less encouraging direction, and then even shows us the chart with the combined data:

U.S. direct-lending volume, a measure of loans made directly by private-credit funds to companies, fell about 55% quarter-on-quarter ​to $33.59 billion in the second quarter from $74.67 billion in the first, the lowest level since ​the second quarter of 2023, according to PitchBook/LCD data. The deal count declined to ⁠154 from 217.

The immediate reaction is that such a drastic QoQ decline in direct lending is bad news. It coincides with a 56% QoQ decline in PE activity, with the two showing an almost perfect correlation. What's probably even more concerning, though, is that excess capital and a scarcity of opportunities to deploy it rarely lead to good outcomes. We won't know whether that's the case this time for another 3-4 years, once today's capital has been deployed and we can evaluate the performance of this loan cohort.

On the other hand, I do see a silver lining. Excess capital in direct lending at least guarantees that, when and if PE activity accelerates, there will be enough capital to finance those transactions and build on that momentum. In addition, if you recall the chart from the first piece in the letter, Mezzanine and Special Situations funds represent 11.5% of the market combined, and I believe they will play a very important role over the next 2-3 years by bridging the gap between what PE firms are willing to pay for a company and what direct lenders are willing to provide. Mezz and Special Sits, with their junior debt and more creative capital solutions, might be the missing layer of capital needed to get the PE flywheel spinning again.

If you are raising a Mezz or Special Sits fund and plan to use the paragraph above in your LP deck, I expect a fee. Thx.

6. Deals on the Block

Download below: the complete Excel file covering deals completed in 3Q26 (plus the prior three quarters).

CLICK to Download 3Q26 Details

KIÉR LIÓR – Providing outsourced private and leveraged loan underwriting and portfolio management to institutional investors, family offices, and direct lenders.

www.KierLior.com

7. Bulletin Board

• WSJ Online is offering 65% off its subscribers ($4 per week vs $11) LINK.

• Bloomberg is offering 60% off its annual subscription ($180 vs $399). If you are in Business, Bloomberg is a MUST, with no real close alternative (LINK).

• Shortcut.AI, an AI tool for Excel that lets you build models by simply typing requests (similar to ChatGPT). It is free to use up to a certain number of tasks per day (LINK). If you hit your limit before finishing your project, simply upgrade to the Pro plan ($20/month) and enter promo code DEBTSERIOUS (all capital letters) to get 50% off.

• LoanEdge, a BDC research tool that lets you look into BDC loan compositions, individual loans, which BDCs own a specific loan, and more, is kindly providing DEBT SERIOUS readers with a free 3‑month trial (the regular price is ~$3k per year, and it’s higher for institutions). Here is a LINK to the tool. Send an email to Sadaf Khan at sadaf@theloanedge.com and let her know you are from the DEBT SERIOUS community.

As a reminder, I don’t personally benefit from either of these offers. I am just saving you money.

Lastly, our community is over 4,300 strong and includes a broad cross-section of mid- to senior-level professionals across private credit, private equity, LPs, and debt advisory. Please use the Bulletin Board at the end of each weekly letter to share events, discounts, or deals that benefit the community. I will continue to feature these opportunities and personally connect interested parties at no cost.

That’s the bell — round over. See you in the next.

Aznaur

Aznaur.Midov@KierLior.com

News

1. UBS helped launch Blue Owl's tech BDC, with 60% of capital from its wealth management clients. It then advised clients to cut private credit exposure, triggering Blue Owl redemptions (Financial Times)

2. US direct lending volume fell to $34B in Q2’26, the lowest since Q2’23. Compared with Q1’26, volume declined 55%, while deal count fell from 217 to 154 (Reuters)

3. US direct lending funds raised $16B+ in Q2'26, marking the second-strongest quarter in four years, driven by strong demand from large institutional investors as retail capital pulled back (Financial Times)

4. BDC redemption requests rose QoQ across most non-traded BDCs, averaging 10.3% in Q2 vs 9.7% in Q1. The increase was driven by follow-on requests prorated in the prior quarter. Full table included (Fitch)

5. TPG and TA-backed software company Planview failed to refinance its $1.5B BSL term loan with private credit despite offering a higher interest rate. The first-lien loan is quoted at 86 cents, but up from 74 cents in mid-April (SCW)

6. Part of the largest private credit deal, Anthropic’s $35B debt package arranged by Apollo and Blackstone, will be offered to other money managers in the coming months. By 2027, roughly $15B should be tradable (Bloomberg)

7. Apollo outbid Castlelake for EasyJet with a $7.6B offer versus $7.3B. Apollo’s surprise bid came after Castlelake raised its offer five times. A bidding war is likely (Bloomberg)

8. Private placements in CEEMEA totaled $33B, 10x higher than in 2025, with 2/3 coming from the Middle East following the conflict with Iran. Average yield was 5.12% vs. 3.18% on emerging market bonds (Bloomberg)

9. Big Tech has priced 7 bond deals exceeding $25B YTD, more than the previous six years combined. Demand may be cooling, with Amazon's latest issue attracting 1.6x orders, below this year's average (Bloomberg)

10. Australian bond market attracted a record A$52B ($36B) of foreign bond issuance in YTD 2026, up 30% YoY, as strong investor demand capped borrowing costs amid intensifying competition for deals (Bloomberg)

11. HSBC cut back leverage to riskier private credit clients but will continue extending it to stronger firms (Financial Times)

12. EU banks' SRT volumes climbed 36% YoY in Q1'26, reaching ~€500B of protected loans, or 2-3% of the European lending market (Financial Times)

13. EU regulators sought private credit data from the US Treasury but were rebuffed over confidentiality. Some EU officials warn that limited transparency could lead to stricter bank capital requirements (Reuters)

14. United Securities-backed United PF, the largest Planet Fitness operator, is negotiating a debt-for-equity swap with lenders. Meanwhile, its $529m first-lien term loan due in December trades at 89 cents (Bloomberg)

15. As LMEs evolve, lender co-ops are admitting second- and third-tier creditors while carving out fees to compensate steering committees for the heavy lifting (ACI)

16. Serta Simmons Bedding’s majority lenders that participated in the infamous 2020 debt restructuring were ordered by a bankruptcy judge to pay $400m to minority lenders. The ruling could have implications for other LMEs (WSJ)

17. M&A is expected to surpass 2021, hitting $6.4T in 2026. Morgan Stanley says Q2 announced deals jumped 64% YoY, led by software, utilities, energy, and healthcare (Reuters)

18. EU regulators sought private credit data from the US Treasury but were rebuffed over confidentiality. Some EU officials warn that limited transparency could lead to stricter bank capital requirements (Reuters)

Transactions

1. Blackstone-owned QTS Realty (data-center) is raising $2B to refinance debt and fund a dividend recap (Bloomberg)

2. Bank of America extended $520m to OpenAI despite previously avoiding cash-burning companies. IPO preparations helped make BofA comfortable with the loan (Bloomberg)

3. Banks lined up a $49B debt package backing Paramount’s takeover of Warner Bros., consisting of $30B of IG bonds, $12B of second-lien bonds, and $7.5B of IG BSLs (Bloomberg)

4. Online lender Lendable (Goldman-backed) is raising $670m through an ABS backed by personal loans. Demand was at least 2x oversubscribed, with some tranches more than 8x oversubscribed (Bloomberg)

5. Autodoc, a German online car parts retailer, is raising a ~$600m Term Loan B to buy back Apollo's 15% minority stake (Bloomberg)

6. Apollo Debt Solutions BDC issued $750m of 7-year senior unsecured notes with a 6.35% coupon (AltsWire)

7. Italy raised $6B in its first dollar bond sale since 2021, drawing $20B of demand. The Italy-Germany 10-year spread has tightened to 77bps from 250bps when Giorgia Meloni took office in 2022 (Bloomberg)

Chart(s) of the Week

1. Amazon has raised over $100B in debt since 2025, more than any other company driving the AI boom.

   

2. “It’s not levered ETFs that keep you awake at night,” says Alex Altmann (global head of equities at Barclays). “What keeps you awake at night is that you [might] have a structural impairment to equities that effectively no economist on the planet has a cell in their econometric model that says 20% impairment to the S&P, that basically destroys, let’s just call it, around about $16 trillion of wealth. So let’s just say that’s half of US GDP, right? And that is an instant impairment to US consumption. It’s your recession straight out the gate.”

In Round #51:

1. When loans lose 50 points

2. Private credit’s financing of BNPLs

3. Allianz on the interconnection between IG, HY, BSLs, and Private Credit

4. Revisiting Medallia and philosophizing about PIK

5. AI and chip-backed debt

6. Notable deals

7. Bulletin Board

Disclaimer: there are 4,200 + of us here discussing private credit and leveraged loans, which is not a big world. At some point, I will probably throw punches at your company. I am blunt and sarcastic and, as friends like to remind me, too mouthy to live to an old age, so my take likely won’t please you. But don’t worry, next week it will be someone else in the spotlight and you will quietly nod along in satisfaction.

Also, none of this is investment advice.

1. French Connection

Last Thursday, July 2, as most of you were preparing for a long holiday weekend, or had already checked out, some of your colleagues holding the Term Loan B of Santé Cie, an Ardian-owned home care provider, received news that French authorities had seized €46m from the company's bank accounts as part of a fraud investigation.

Bloomberg elaborated on the situation:

The preliminary investigation, which started in June 2024, concerns suspected offenses of “active and passive corruption, money laundering of the proceeds of corruption, money laundering of the proceeds of aggravated tax fraud, presumed money laundering and breaches of France’s anti-gift law,” the Parquet National Financier (French financial prosecutor) said in its statement.

Well, that sounds quite serious. So, as soon as lenders learned about the account seizures, they immediately checked the loan price and saw the chart below.

We look at this chart and see the price dropping from par to 60 cents (it actually fell into the 50s). They looked at the chart and saw a ruined weekend, calls with the sponsor, calls with lenders, calls with legal teams, internal calls, and subconsciously calculated the WWH formula (which I just invented) that goes like this:

WWH stands for Weekend Work Hours and represents the number of hours you will spend dealing with a specific issue. So, (100 – 50) / 2 = 25. The good news is that this was a 3-day weekend, so that’s just over 8 hours a day. Manageable!

Now, after spending the entire weekend working, they received the following update (per Bloomberg):

On Monday, the company, owned by private equity firm Ardian SAS, said it requested to start a mandat-ad-hoc with the commercial court of Lyon. This process would allow the firm to confidentially negotiate with its creditors outside formal insolvency proceedings and under the supervision of an expert.

As a result, the price recovered into the 80s. You know, when you spend all weekend working on something urgent, you almost wish it would subsequently go to zero for all the trouble it caused you. Oh, well.

Also, I like how Bloomberg used this random photo taken in Denver for the article (below) because Sante Cie is a home care provider. But I think it ended up being quite symbolic...

…But it is not a nurse holding a patient's hand... it is a VP holding the hand of an associate whose weekend was sacrificed to a fire drill, only for things to somewhat stabilize by Monday, making all that work seem pointless.

2. BNPL: Private Credit Goes Shopping

As a parent, I have some rules for my kids: don't do drugs and don't use Buy Now Pay Later (BNPL). Oh, and don't commit crimes. So, yeah, I'm not exactly a fan of BNPL. Which is why I was very curious to read Bloomberg's Private Credit Is Making Bets on Consumer Debt at a Precarious Time, about private credit firms financing BNPL providers through forward flow agreements, and dig into how these deals actually work.

Before we get to private credit’s role, let’s start at the beginning.

As you know, BNPL is essentially a consumer loan that is paid in installments, which doesn’t sound revolutionary at all. To further illustrate the point, Wikipedia suggests that some forms of BNPL trace back to the 19th century, when consumers paid for large purchases over time. My personal opinion is that people didn’t wait until the 19th century to come up with the brilliant idea of paying in installments, and likely have been doing it since some form of goods-for-goods exchange existed, but that’s just me.

In its modern fintech form, though, everything points to Sweden in 2005, with the birth of Klarna. It was later joined by the other major US players: Affirm in 2012, Afterpay in 2014, which was acquired in 2022 by Block (fka Square), and PayPal, which entered the market in 2020. Their respective US market shares are below.

BNPL’s core product has been the so-called “pay in 4,” which means you pay 25% of a product’s purchase price at checkout, and the remaining 75% is paid in three equal installments every two weeks. The appeal to consumers is not only the ability to split the payment into four installments, but also the fact that there’s zero interest. Additionally, just like most fintechs, it’s a very simple, one-click type of transaction. Lastly, unlike credit cards, which typically rely on FICO scores and affect your credit score when you apply, BNPL providers use a combination of public and proprietary data to assess your creditworthiness. As a result, they can often approve you without affecting your credit score.

I’m sure you are now wondering how fintechs that offer 0% APR loans actually make money. They do so by solving two of the biggest problems facing retailers, particularly e-commerce companies where BNPL first took off: (i) getting consumers to buy more expensive products and (ii) reducing cart abandonment.

The first problem is self-explanatory. Both psychologically and mathematically, it’s easier to pull the trigger on a purchase when the amount looks smaller and you pay no interest. The second problem relates to the first, because e-commerce in the US experiences roughly 70% cart abandonment, meaning people add items to their carts but never complete the purchase. That is a gigantic problem, costing retailers an estimated $260B in lost sales every year. Anything that reduces friction during the final steps of the purchase, whether by breaking the price into installments or making checkout a one-click transaction, can significantly improve conversion rates. Because BNPL helped solve both of those problems, providers were able to charge retailers a 2%-7% commission on every sale. So, yes, consumers paid 0% interest, but they also ended up buying more than they had originally planned, making it a win-win for both the BNPL provider and the retailer. Separately, some BNPL providers also generated revenue by charging late payment fees, so there was that too. Here you have it, and when Klarna says “protecting consumers while expanding access to smart credit,” hear it the same way you hear alts talking about "democratizing alternatives for the masses" when they are pitching private assets for 401(k)s.

We are getting closer to the private credit angle, but there's one last thing to note. While BNPL became popular by offering 0% financing, many providers have expanded into longer-term installment loans that charge traditional interest. The extent of that shift varies widely by company. The chart below shows how much of each major BNPL provider's loan book consists of APR-bearing (interest-bearing) loans versus the classic 0% installment loans.

BNPL is in the business of lending, and lenders need capital to lend. Banks fund loans with deposits, private credit invests its LPs’ capital, but what about BNPL providers, which are neither (except Klarna, which has owned a bank since 2017)?

Like most startups, BNPL providers initially relied on venture capital to fund growth. But using equity to generate debt-like returns is not a sustainable business model. As they matured, they adopted the same funding model used by traditional specialty lenders (mortgage, auto, etc.): first, warehouse facilities from banks that are repaid as borrowers repay their loans, and later, securitizations that package those consumer receivables into standard asset-backed securities (ABS). As with most ABS deals, the senior tranches are sold to institutional investors, while the most junior tranche is typically retained as the first-loss piece (well, to be precise, it might be retained by the SPV created for that specific ABS rather than the BNPL provider itself, but that's information for nerds). Here is an example of Affirm’s ABS.

Another source of funding is so-called Forward Flow loan sales to institutional investors, but it is the agreements with private credit firms that make the news, creating the impression that they are the primary buyers (the Bloomberg article, for example, framed it as a private credit story). Below is an excerpt from Affirm's investor letter. You don't need to read it, but it covers, at a high level, much of what we just discussed.

Details around the forward flow agreements are not widely available, but we can piece together a rough outline of what the deals typically look like. Although keep in mind, they are all bespoke (highly customized).

BNPL enters into a 2-3 year agreement with a private credit firm that commits to acquiring up to, say, $2B of consumer loans on an ongoing basis. It operates as a revolving line, so as consumer loans are repaid (and, if you recall, they are all short-term, ranging from a few weeks to a couple of years), they are replaced with new loans. The private credit firm and the BNPL provider negotiate the eligibility criteria for loans included in this revolving pool, such as tenor, yield, and perhaps FICO score. As loans mature, the BNPL provider replaces them only with new loans that satisfy those criteria, while continuing to service the portfolio.

Similar to an ABS, the BNPL provider retains the first-loss risk, although the mechanics may differ. Affirm's 10-K states that, for certain loan buyers, it has risk-sharing agreements requiring it to compensate investors if portfolio performance falls below agreed levels. Economically, this serves the same purpose as first-loss protection but without a traditional first-loss tranche.

Bloomberg cites Stone Ridge as one of the funds buying BNPL loans, and a short excerpt from the multi-thousand-page report gives a pretty good picture of what the portfolio looks like. You can see that loan amounts start as low as $15, with yields up to 35% and a 12-month tenor. Not bad!

The article also includes a chart highlighting notable forward flow deals, featuring the usual suspects: private credit firms (Blue Owl), institutional investors (PGIM), and hedge funds (Elliott).

Of course, it mentions the problems that have been taking place in private credit (even though those were direct lending issues, while this article is discussing ABF). Besides that, it suggests that persistent inflation negatively impacts consumers, and that more borrowers are failing to keep up with BNPL payments. Of course, the main point of the article is to suggest that losses are coming for private credit, and I fully expected that conclusion. But then I came across this paragraph:

“If there is a significant disruption in private credit, then we may see a scaling back of the backing private credit has for BNPL loans,” said Nick Maynard, vice president of research at Juniper Research.

Wait, for once, they are not fearmongering about potential private credit losses. They are actually concerned that if consumers stop paying, private credit will pull back from buying these loans, which will negatively impact BNPL. That’s fascinating! Listen, I’m pro-capitalism, but I wouldn’t miss BNPL for a second.

Moving on…

3. Credit Musical Chairs

I really liked this report from Allianz, which discusses not only private credit and BSLs, but also High Yield and Investment Grade bonds, and how all four markets coexist in today’s environment. Somehow, this relatively short report manages to summarize an entire year of podcasts I’ve listened to on both public and private credit.

Here is one tidbit on liquid credit:

HY offers a more defensive risk/reward profile than BBB – but largely due to a composition effect, as the weakest names have migrated out to private credit.

Let me break this down for the illiquid credit crowd, since credit ratings are not something we talk about much in our world. Investment grade is AAA, AA, A, and BBB. Everything below that, BB, B, CCC, and lower, is sub-investment grade, aka high yield or junk.

What the quote above is really saying is that, instead of buying BBB-rated investment grade bonds, investors may be better off buying the high yield index, which holds nearly 2,000 bonds, most of them rated BB, B, and CCC. The reason is that, over the years, the makeup of the index has shifted toward higher-quality issuers. BB-rated bonds now make up more than half of the index, the highest share on record, while CCC-rated bonds have fallen to roughly 10%, one of the lowest shares on record. Part of that shift is because many smaller, riskier companies now borrow through the BSL and private credit markets instead of issuing high yield bonds.

The chart below shows how the composition of the index has changed over the years.

There are a bunch of other gems in the report, such as this piece of recent history:

[…] Private credit was built around a structural gap of companies too small or too idiosyncratic to access public syndicated markets, where minimum deal size, rating requirements and investor base breadth create a de facto floor. The median private credit borrower carries EBITDA of USD 30–50mn, roughly ten times smaller than the median HY bond issuer (USD 568mn EBITDA) or leveraged loan borrower (USD 522mn EBITDA). In that respect, the original logic holds: private credit occupies the space below IG and HY by borrower size, not by credit quality, and the analytical frameworks governing public credit markets apply to it at most by analogy. But that clean separation has been eroding since 2022 and at the upper end of the market, it has broken down materially. The 2022 rates shock that froze the BSL market for leveraged buyout (LBO) financing for the better part of 18 months pushed large-cap sponsors into private credit as the only available execution route for transactions that, by borrower size, would previously have been routed to the BSL market. The result was the emergence of “jumbo” direct lending, bilateral or small-club transactions providing debt packages above USD 1bn, a territory previously the exclusive preserve of syndicated finance. That shift has not fully unwound: by Q1 2026, private credit was financing approximately 80% of all LBO deals by count, and the average enterprise value of new third-party buyouts exceeded USD 1bn.

It is wordy, but a perfect summary of how BSL and private credit started going toe-to-toe by refinancing each other’s deals.

One last thing from the report worth highlighting is its argument that private credit loan structures have historically been stronger than those in the BSL market, thanks to tighter covenants, direct relationships with borrowers and sponsors, and first-lien security. However, intense competition has gradually eroded many of these advantages, particularly in the upper end of the market. The chart below shows covenant quality across different segments of the private credit market.

Anyway, I absolutely loved this report. I rarely come across a paper where almost every paragraph is worth discussing here, but this was definitely one of them.

4. PIK and Choose

The title of the Financial Times article, What Medallia’s Faceplant Tells Us About Private Credit, was enough to make me click. We all know the FT‘s stance on private credit, so I expected to roll my eyes every other paragraph, but I was pleasantly surprised by its measured takes.

The article opens with a fact I wasn’t actually aware of (emphasis mine):

Thoma Bravo made headlines a couple of weeks ago after writing off $5.1bn of equity in Medallia, a company selling customer experience analysis software-as-a-service, in the second-largest private equity loss in history.

Apparently, the largest PE loss before Medallia was TXU (an electric utility) that was owned by KKR, TPG, and Goldman Sachs and filed for bankruptcy in 2014. Unlike Medallia, which was acquired with $5B of equity and $1.8B of debt, TXU was bought with $8B of equity and $26B of debt, which helps put Medallia into perspective. BTW, the head of KKR’s Energy Group at the time of the acquisition was... Marc Lipschultz (now the co-founder and co-CEO of Blue Owl).

But anyway, the article walks through the Medallia loan from its inception in 2021, looking at the exposure of different BDCs and how they marked it over time. It then shifts to the composition of Medallia's interest payments, breaking them down into the benchmark rate (LIBOR/SOFR), cash interest, and PIK, which was interesting to see laid out on the chart.

The FT continues:

In its first year the loan paid US dollar Libor (RIP 💀) plus 6.75 per cent. And this 6.75 per cent came in the form of new payment-in-kind debt. There’s a time and a place for PIK debt. And, while it didn’t turn out well in the end, lending Medallia on PIK terms made a certain amount of sense. After all, profitable and fast-growing software companies that can use any spare cash for even faster expansion are precisely the kinds of companies that shouldn’t mind a slightly higher interest rate if it means they get to defer cash payments for a while.

This entire paragraph runs counter to what the media and pundits have been saying, which is why it caught me by surprise (I’m proud of FT). In fact, I recently heard an executive from Oaktree say on a podcast, "There is no Good PIK, only Bad and Worse PIK." That is an interesting statement because Oaktree Specialty Lending has a bunch of first-lien term loans with PIK interest. What it doesn't have, though, is a disclosure in its reporting of whether those loans have bad or worse PIKs. So, something for their compliance department to think about.

Look, if lenders were given a choice between cash interest and PIK, they would choose cash interest. But can we agree that only a minority of loans, including those with PIK, actually blow up? Can we also agree that PIK typically comes with a premium of, say, 100-150 bps over cash interest? So, if you agreed to PIK upfront, charged that premium, and successfully exited the deal, didn't you just generate excess return? So was it a bad PIK or a worse PIK? I'd say it was a pretty darn good PIK.

The situation with Medallia needs to be viewed from a more nuanced perspective. Think about it: at the time of the original deal, upfront PIKs were slowly becoming the norm for software deals and were being accepted by lenders because they had seen the structure work. So, you had Thoma Bravo, one of the pioneers of software LBOs, putting $5.1B of equity into the deal and asking lenders to put $1.8B ahead of it in the capital structure, providing roughly a 3x equity cushion beneath your exposure. Who, if not Thoma, knew how to value software companies? And when the other lenders included Apollo, HPS, Blackstone, and KKR, how many people would have said, "Well, all these funds don't know what they are doing"? I'm not saying this is the logic you need to apply when investing, especially with the benefit of hindsight, but going against these types of players all the time would put you out of business much faster.

To be fair, I believe the leverage on Medallia was around 4x ARR, which is well above the more typical 2.5-3.0x seen even at that time. Also, Medallia's software doesn't sound very sticky to me, although I don't know its retention figures. But if there was ever a deal where you'd consider stepping outside your usual lending box, this was probably it. I can easily imagine a credit committee saying, "Okay, whatever, let's do it."

Alright, from there the Financial Times shifts to BDC fees, specifically incentive fees.

Some readers familiar with the old hedge fund two-and-twenty fee model might assume that BDC incentive fees follow this pattern, where performance fees are charged on total returns above some hurdle rate. Such readers would be wrong. The bulk of BDC incentive fees are instead levied on income — presumably on the basis that their punting is concentrated in senior secured private credit loans marked at, or close to, par that pay fat incomes. Importantly, no distinction is made between cash income and PIK income.

Well, I think they asked a loaded question, but then answered it accurately. Hedge funds earn a 20% performance fee on NAV appreciation because they invest in assets where capital appreciation is the primary objective. BDCs, on the other hand, are built to generate income, so their NAV generally doesn’t move much in either direction.

If you start from the premise that PIK income is “fake,” but BDC incentive fees are based on income generation, then it’s a logical conclusion that those incentive fees are overstated and undeserved. That said, I separate PIK into good and bad, so it’s less of an issue for me. I also barely invest in BDCs, so I don’t really care one way or the other.

You might, though, so…

5. AI, Depreciated

We didn’t talk about AI for, like, 20 seconds, so I thought I’d sprinkle a little bit of it into this letter, as I came across a couple of curious pieces. Bloomberg’s AI Sales Start to Justify Data-Center Spending Boom, discusses a fairly bullish AI report by Exponential View.

Revenue from artificial intelligence has reached a tipping point, showing that the hundreds of billions of dollars tech companies are spending on it may be economically sustainable. Global AI sales, excluding China, reached $25 billion in the first quarter of 2026, exceeding the industry’s estimated $21 billion in depreciation costs tied to investments in data centers and chips for the second consecutive quarter. While the milestone suggests that AI companies are beginning to cover the cost of their capital spending, the margins are thin.

Yeah…that covering cost of “capital spending” is an interesting way of spinning it, right? But it comes with a cool chart, so let’s give it a pass for a moment.

Back to the cost of capital spending, where revenue is finally covering depreciation. I can just say, “Hey Claude, change depreciation from 5 years to 6 years,” and this chart will look much better, right?

That said, this might be a good first step, but I look at this as a lender, and I’m concerned. When it comes to debt coverage... well, actually, forget debt coverage. We are three steps away from that because this is an industry with gross margins of just 30-50% (depending on the player), before even accounting for opex. Even sticking to this depreciation metric, it is a lagging indicator because of the exponential growth of capex over the past 2 years. To illustrate the point, below is the chart from the same AI report. Assume for a second that it is all AI spend because we just need to calculate this directionally.

If you sum capex from 2021 through 2025 and assume a 5-year depreciation schedule (or whatever), you get $59B of quarterly depreciation. If you then calculate the same for the period from 2022 through 2026, because capex accelerated so dramatically in 2025 and 2026, your quarterly depreciation in Q1’27 will jump to $95B. That depreciation will accelerate by 60% YoY, which means revenue should keep up (it likely will). Hopefully capex slows down and the line flattens because the numbers start looking too crazy to pronounce, let alone to keep up with, and the whole cash flow discussion becomes an illusion.

The good news is that lenders found a way to finance this whole thing through ABS deals backed by chips (GPUs, TPUs, whatever-U) as collateral, so the cash flow of the entire project can wait a little longer, as long as senior tranches are getting paid.

Per 9fin:

Investors of chip-backed ABS, for example, demand structural protections that include over-collateralization, rapid amortization schedules, strict limits on loan-to-value, dedicated debt service reserve accounts, and hard triggers if the underlying collateral loses value.

Now, the issue with this type of financing is that there is no consensus on how long these chips actually last. 9fin says the optimists think older chips can still be repurposed for less demanding tasks, like running chat responses. The skeptics point to the relentless pace of chip development, and chipmakers churn out next generation GPUs.

The funny thing is, there seems to be convincing evidence for both sides if you look hard enough. So…

6. Deals on the Block

Download below: the complete Excel file covering deals completed in 3Q26 (plus the prior three quarters).

CLICK to Download 3Q26 Details

KIÉR LIÓR – Providing outsourced private and leveraged loan underwriting and portfolio management to institutional investors, family offices, and direct lenders.

www.KierLior.com

7. Bulletin Board

• WSJ Online is offering 65% off for subscribers ($4 per week vs $11) LINK.

• Bloomberg is offering 60% off its annual subscription ($180 vs $399). If you are in Business, Bloomberg is a MUST, with no real close alternative (LINK).

• Shortcut.AI, an AI tool for Excel that lets you build models by simply typing requests (similar to ChatGPT). It is free to use up to a certain number of tasks per day (LINK). If you hit your limit before finishing your project, simply upgrade to the Pro plan ($20/month) and enter promo code DEBTSERIOUS (all capital letters) to get 50% off.

• LoanEdge, a BDC research tool that lets you look into BDC loan compositions, individual loans, which BDCs own a specific loan, and more, is kindly providing DEBT SERIOUS readers with a free 3‑month trial (the regular price is ~$3k per year, and it’s higher for institutions). Here is a LINK to the tool. Send an email to Sadaf Khan at sadaf@theloanedge.com and let her know you are from the DEBT SERIOUS community.

As a reminder, I don’t personally benefit from either of these offers. I am just saving you money.

Lastly, our community is over 4,200 strong and includes a broad cross-section of mid- to senior-level professionals across private credit, private equity, LPs, and debt advisory. Please use the Bulletin Board at the end of each weekly letter to share events, discounts, or deals that benefit the community. I will continue to feature these opportunities and personally connect interested parties at no cost.

That’s the bell — round over. See you in the next.

Aznaur

Aznaur.Midov@KierLior.com

Weekly Debt Capital Markets Digest

News

1. Sovereign wealth funds are shifting from public to private markets for two reasons: (1) S&P 500 concentration in its top 10 companies and higher equity-bond correlation; and (2) AI exposure, primarily through private credit and infrastructure (Financial Times)

2. Blue Owl’s BDCs received $3.6B (19% of NAV) and $1.1B (38% of NAV) in Q2 redemption requests, down from 22% and 40% last quarter. Both honored 5% (Semafor)

3. Goldman Sachs BDC received redemption requests totaling 3.24% of NAV. Like Oaktree’s 4.5%, the figure is below the typical 5% quarterly cap (ACI)

4. CEO of BlackRock TCP is leaving the firm. The BDC has been in the news for its 19% NAV cut in one quarter and a subsequent DOJ investigation (Bloomberg)

5. Non-traded BDC investors may wait years, not quarters, to redeem due to a rolling backlog. The product was not designed for 15%–40% quarterly redemption requests. Interview with Mark Goldberg (PitchBook)

6. BNPL and forward-flow financing provided by private credit firms are raising concerns among experts because of the scale at which they operate (Bloomberg)

7. Software loans weakened before the SaaSpocalypse: 6.1% traded below 80 cents on the dollar in September 2025, versus a five-year average of 4.4%. (WSJ)

8. Medallia loan: piecing together what went wrong and how lenders managed it (Financial Times)

9. LPs and opportunistic credit funds are taking preferred equity stakes in PE funds as sponsors seek new ways to raise liquidity amid a slowdown in exits (Financial Times)

10. Europe’s private credit market has avoided the US panic, thanks to lower software exposure and a smaller retail investor base, with no semi-liquid fund redemption headlines (Private Debt Investor)

11. Australia's A$200B ($140B) private credit market, where real estate makes up about half of lending, is going through a tightening of loan standards driven by tougher market conditions and heightened regulatory scrutiny (IFR)

12. Emerging market bond sales grew 15% y-o-y, reaching a record $450B YTD. The 157bps spread on EM debt is the lowest in 20 years, fueling the wave of issuance (Bloomberg)

13. Largest bond managers focus on five-year Treasuries in the new Warsh era, as they shield from short-term volatility and appear to be a bargain relative to two- and 30-year paper (Bloomberg)

14. Pimco says differences in NAV marks across private credit managers lead investors to apply varying discounts to BDCs and drive different levels of redemptions (Bloomberg)

15. Pimco has increased its private placement activity, including deals with sovereign governments (Qatar, Kuwait, Colombia) and private credit transactions, such as AI data centres and Blue Owl’s bonds (Financial Times)

16. Private placement bonds are experiencing a renaissance, driven by life insurers' annuity sales and the need to deploy capital. AI financing leads demand, with other industries also tapping the market (Bloomberg)

17. Bank-capital rules proposed by the Fed penalize JPM- & BofA-type commercial banks while benefiting Goldman & Morgan Stanley-type IB/trading/prime brokerage-heavy banks (Bloomberg)

18. Saks Global emerged from bankruptcy after selling half of its stores and slashing its debt by 75% ($1.2B), wiping out shareholders and giving control to senior lenders. It was also renamed Exemplar Luxury Group (Reuters)

19. EchoStar’s Dish DBS and Dish Wireless filed for bankruptcy largely driven by inability to repay $2B of 7.75% senior notes. Full breakdown (Bondoro)

20. Japan's weak currency drove a 30% year-over-year increase in bankruptcies, to 45 firms. The YTD total is the highest since 2022, when currency-related bankruptcies were first tracked (Bloomberg)

Transactions

1. Morgan Stanley Direct Lending Fund raised $350m in 5-year notes at 2.20% above Treasuries (Bloomberg)

2. SoftBank is negotiating a $10B margin loan backed by its stake in OpenAI. After lenders pushed back on using the OpenAI stake as the sole collateral, SoftBank proposed providing a repayment guarantee (BigGo Finance)

3. Apollo-owned Atlas SP provided a $400m warehouse facility to Ireland-based Aquila Air Capital to expand its engine and commercial aircraft portfolios (AOL)

4. Sixth Street and KKR provided $1 billion of preferred equity to BridgeBio to accelerate the launch of genetic disease therapies. The investment carries a 7% step-up dividend that increases to 17% (Bloomberg)

5. Royal Bank of Canada is working on a $2B significant risk transfer (SRT) transaction to offload exposure to a portfolio of AI-linked loans (Bloomberg)

6. Standard Chartered arranged a $50m green bond for African Frontier Capital, guaranteed by Green Guarantee Co, raising funds from UK and US investors and paving the way for more such issuances in Africa (Bloomberg)

7. Hercules Capital provided a $200m facility to 4D Molecular Therapeutics as the company advances into Phase 3 development and pre-commercial planning for its single-dose gene therapy (Biospace)

Chart of the Week

Strong market returns and higher interest rates eliminated funding gaps for corporate defined benefit pension plans for the first time in nearly two decades.

1. Economists debate AI’s effects on jobs

2. Surprising driver of BDC redemptions

3. SEC approves JPMorgan's unique interval fund

4. Fitch’s report on defaults

5. Yet another weekly letter

6. Hedge fund shorts insurance

7. State Street’s report on private markets

8. Global banks target India’s ABS

9. Notable deals

10. Bullet board

Disclaimer: there are 4,200+ of us here discussing private credit and leveraged loans, which is not a big world. At some point, I will probably throw punches at your company. I am blunt and sarcastic and, as friends like to remind me, too mouthy to live to an old age, so my take likely won’t please you. But don’t worry, next week it will be someone else in the spotlight and you will quietly nod along in satisfaction.

Also, none of this is investment advice.

1. Economists Walk Into a Bar

I typically like to start my letters on a light-hearted note, so I picked the WSJ article with an appropriate title: Is an AI Jobs Apocalypse Coming?

It's a panel discussion featuring three economists from MIT, the University of Virginia (UVA), and Yale (I'll refer to them by their universities), sharing their views on the long-term impact of AI on labor. Here are a few highlights:

WSJ: What might slow AI’s ascendancy?

Yale: Anything involving human contact. We’ve been able to fully automate piano players for more than a century, but we haven’t done so. That’s not what people want. I understand all the arguments about robotics, but I don’t care how good the robot is. I’m not having a robot raise my 1½-year-old.

WSJ: Can some of us opt out, in favor of a slower-paced, more artisanal life?

MIT: You are asking about what I call “the Etsy economy,” which is based on the notion that we value certain things because they are done by people. I might want a handmade basket from Etsy even though there’s a machine-made one on Amazon that costs a 10th as much. But I don’t think such sentimentality can drive a sufficient amount of economic activity to employ most of us.

WSJ: Whose livelihood is most at risk, if we’re heading toward an AI-centered economy?

UVA: It almost certainly looks like it will be white-collar work, or what some people call “the laptop professions.” That doesn’t mean that all the jobs in fields such as law, finance, accounting and consulting will disappear overnight. But there will be strong competitive pressures to automate them.

Ignore the UVA professor. He has an Eastern European name, and we’re generally a gloomy bunch.

I think the most important question you should ask yourself is this: if AI is smart enough to do the job of an associate and consistently cover all the key points the investment committee cares about, then shouldn’t it also be smart enough to replace the investment committee? Now, the investment committee consists of executives who ultimately decide whether or not to deploy AI to replace people. So, is that really going to happen anytime soon, if they know AI would eventually come for their jobs too?

If you are an associate reading this and nodding along, I'm not saying you won't be replaced. I'm saying you'd better start working your way onto the investment committee, because that's where the safe jobs are.

Also, below is an exchange I had with Claude today (really). My impression is that these AI things are basically good-hearted and aren't looking to put people out of work...

Claude is such a softie...

2. Private Credit: Gates. Redemptions. Arbitrage.

Last week was once again dominated by headlines about BDC redemptions, with redemption requests continuing to increase quarter over quarter. As expected, virtually every firm capped redemptions at 5%. Both the increase in redemption requests and the widespread adoption of 5% redemption gates were exactly what we expected, as discussed in previous posts.

The Financial Times published a great chart tracking redemption requests over the past several quarters. A few things stood out to me.

First, despite Apollo’s thought leadership and vocal criticism of the industry, it received one of the highest levels of redemption requests.

Second, despite BlackRock’s TCP fund making headlines for the wrong reasons (trading at a 50% discount to NAV, being investigated by the DOJ, etc.), its sister fund, BDEBT, saw redemption requests of only a little over 5%.

Lastly, despite Oaktree’s portfolio composition being similar to that of other lenders, it received redemption requests of just 4.5% this quarter. I’m surprised, to be honest. My expectation was that, unlike last quarter, when it honored the full 8.5% of redemption requests with support from its parent, Brookfield, it would gate redemptions at 5% this quarter like everyone else. Hey, good for them.

Generally speaking, there are common reasons why investor redemption requests increase. Some investors are reacting to the headlines and are generally worried they won’t be able to get their money back. Others expect to get their money back but have liquidity needs. However, knowing that BDCs will gate redemptions and redeem investors pro rata, they submit redemption requests that far exceed their actual liquidity needs, driving the numbers higher.

But there is a third category that people online have been talking about, although it initially sounded too rational to be true. Bloomberg just published a piece, Private Credit Arbitrage Trade Gets Momentum as Redemptions Rise, which I think captures the idea well. Here is the quote from the article that elaborates further:

“When the same credit manager is running a non-traded BDC at its net asset value and a listed sibling fund at a 24% to 27% discount, that’s not a philosophical debate, that’s a math problem,” John Cole Scott, CIO at CEF Advisors, said in an interview. Investors who are patient enough to sell the non-traded vehicle and rotate into a listed one “are effectively buying the same portfolio at a meaningful discount,” he said.

Because non-traded BDCs are redeemed at par, this arbitrage trade makes a lot of sense. The only catch is that you need to make sure the sibling funds hold essentially the same portfolio, which is not always the case, although they are typically similar.

I would speculate that only a minority of redemption requests are driven by this arbitrage trade, though. But maybe, once the data becomes available, we’ll discover that the entire redemption wave originated on WallStreetBets, with everyone trying to pull off this trade, and we all got played by a bunch of Roaring Kitties.

Time Out

When I was a kid, I dreamed of becoming an astronaut. I’d close my eyes and imagine myself exploring the vast universe, floating through space, traveling from one planet to another.

That was before the internet (I’m dating myself)... but if the internet and Buyside Hub had existed back then, my dream would have been to land a Senior Associate role at a private equity firm, pulling in $550k a year (see the BH screenshot below). I’d have closed my eyes and imagined myself building Excel models, coming up with the best projections in the universe, linking formulas, and dealing with the occasional “#REF!”. And if the deal still blew up a few years later, I’d collect my $550k anyway, because you can’t exactly pin the blame on the Senior Associate…go after the MDs and Principals!

Anyway, below is the word from our sponsor.

Buyside Hub is a compensation analytics platform heavily used by Private Credit and Leveraged Finance professionals.

It is the one-stop shop for a detailed compensation breakdown for Buyside Credit professionals.

And it’s 100% free for those who join and contribute to the platform. You can join 15,000+ other Buyside Hub community members here:

www.buysidehub.com/welcome

3. SEC Said Yes

While non-traded BDCs typically offer to repurchase up to 5% of their NAV each quarter, the JPMorgan Public and Private Credit Fund requested and received approval from the SEC to offer monthly redemptions of up to 2% of NAV. There’s been a lot of excitement about this, which I find a bit puzzling.

But anyway, Bloomberg says the following (emphasis mine):

The US Securities and Exchange Commission said Tuesday that the JPMorgan Public and Private Credit Fund would be exempt from a rule requiring withdrawals on a quarterly basis. The vehicle’s prospectus from March said it would look to buy back at least 2% of shares each month.

See what I mean? It’s like celebrating being exempt from a rule requiring you to file your tax return once a year, only to be permitted to file it every six months instead and pay more in taxes. Well, congrats, I guess.

But what’s the deal, right? The deal is that JPM is launching two interval funds, the JPMorgan Public and Private Credit Fund and the JPMorgan Tax Aware Opportunities Fund (a muni bond fund), and my guess is that, given all the noise around redemptions, it wanted to differentiate itself by offering monthly repurchases.

The reality, though, is that it can’t just do that because interval funds have to comply with Rule 23c-3 under the Investment Company Act of 1940, which permits repurchase offers only at quarterly, semi-annual, or annual intervals, with each offer capped at between 5% and 25% of a fund’s outstanding shares (so yeah, technically, it could be just 5% per year).

But because the rule says nothing about monthly repurchases, JPM had to get an exemption, even though its proposal would result in at least 6% quarterly redemptions. That’s just how the law works.

If you look at the prospectus for the Public and Private Credit Fund, which was filed before the exemption was granted, JPM is really pushing the extra liquidity. It basically says, “Hey, we’ve applied for 2% monthly repurchases, but if that doesn’t work, we’ll target quarterly repurchases of 7.5% instead.”

4. Defaults With Footnotes

In a couple of separate posts, Fitch Ratings noted that the US private credit default rate remained at a record-high 6% in May, in line with April. It also provided a table showing default rates by sectors (below).

If you looked at that chart and didn’t think, “Wait a minute... what?”, you need to look at it again. Software shows surprisingly low default rates, at least relative to the amount of attention it has received from the media. I’m pretty sure it’s just a timing issue, though. Software subscription revenue keeps the boats afloat, so even if the recent Claude upgrades wreck their product, existing clients are still locked into their contracts for at least a year. See? That’s why everyone loved those recurring-revenue loans for the past 15 years, and this sudden “hindsight is 20/20” criticism from experts who, 20 minutes ago, were raising software-focused debt funds is annoying.

On the flip side, if we are already at 6% now, and AI is going to be as disruptive to software as predicted, things are going to get ugly once software goes through its renewal cycle and more clients churn.

So far, industrial and manufacturing companies have been defaulting at higher rates. Without more detail from Fitch, I would attribute it to weaker transactional revenue driven by the macro environment, which has hurt revenue and, in turn, EBITDA and cash generation. Lower cash generation, of course, makes it harder for companies to service their debt.

Fitch did provide a paragraph breaking down the different types of defaults that have occurred:

Of the 14 private credit default events, eight were new unique defaulters and six were serial defaulters. Serial defaulters are issuers that have defaulted multiple times within a TTM period but count only once in the reported TTM default rate. Seven of the 14 default events involved maturity extensions under stress, five involved the introduction of PIK interest in lieu of cash interest, and two involved uncured payment defaults. This continued the prior month trend of maturity extensions under stress outpacing all other default scenarios. Five of the seven maturity extensions pushed loan maturities out by one to two years from their original maturity dates, while one extended the maturity by seven months and another extended it by one month.

Well, maybe the five PIK defaults had top-line issues, so I wouldn’t completely rule out my speculation. That said, I was surprised that roughly half of the defaults were maturity extensions. In other words, the borrowers couldn’t repay their loan, but refinancing with a new lender into a new 5-7-year loan wouldn’t make much sense if the sponsor planned to exit the company soon. Why pay higher fees and potential early repayment penalties if you’re planning to sell the company within a year? So they paid fees, accepted higher pricing, and agreed to tighter documentation in exchange for the extension.

To be fair, it’s also possible that nobody wanted to refinance the loan, including the existing lenders, who may not have been comfortable lending for another 5-7 years. That’s probably the first explanation that comes to mind, but keep in mind that it’s not always that straightforward, or as bad as it first appears.

By the way, the company that extended its maturity by seven months was likely looking for more time to refinance because of the recent Middle East conflict and the resulting market uncertainty. The one extending by only one month was likely in the final stages of a refinancing but didn’t complete it before maturity. I’m clarifying this for readers who might think the lenders expected the companies to go under in a few months and still granted them an extension for no apparent reason. That’s not how things typically work.

5. WEEKENDER

Every week, I go through hundreds of business and finance articles, many of them focused on private credit and leveraged lending. Most are a waste of time. Some are genuinely interesting, and I have an opinion or a different angle on what’s being discussed, so I pick 5 or 6 of those and include them in this weekly letter.

But plenty of other articles are worth reading even if I don’t have much to add. So I decided to put all of the credit-related articles I found interesting (including liquid credit) into a separate newsletter and send it to you every weekend. Each article comes with a one- or two-sentence summary (not AI slop) and a link to the original source if you want to read the full piece. More importantly, it’s short. In five minutes, you’ll be up to speed on everything that happened in the credit world over the past week.

I called it WEEKENDER, and you'll start receiving it this weekend. A few of the articles in each edition will be discussed in more detail in the following week's edition of DEBT SERIOUS. Here is the first WEEKENDER – I didn't want to send it without a heads-up, so take a look… and brace yourself for even more of me.

6. Next Target: Insurance

Up until recently, we saw article after article claiming that banks’ exposure to private credit, through the back leverage they provide to private credit funds, put them at risk. Then common sense prevailed. People realized those loans are some of the safest assets on banks’ balance sheets, and the wave of articles warning about banks largely disappeared.

Now the focus has shifted to insurance. This isn’t a new debate, but I’ve been seeing more and more articles claiming that private credit is a problem for insurers. Heck, I even covered one just last week.

Bloomberg has an interesting piece, Investor Who Scored 900% Win in 2008 Crisis Has New Big Short Bet, about a London-based hedge fund manager who made a fortune during the GFC and is now betting against insurers because of their exposure to private credit.

Today, [Lee Robinson of Altana] sees a new opportunity as risks bubble up around private credit. But instead of betting directly against the sector, he’s focused on potential second-order effects and is shorting some of the $1.8 trillion market’s biggest backers: insurers.

Robinson is ramping up bearish wagers on firms from Lincoln National to MetLife and even Berkshire Hathaway through the use of credit default swaps, derivative contracts designed to protect investors against a default.

Lincoln and MetLife are life insurance companies, and life insurers’ exposure to private credit has been discussed extensively. Betting against Berkshire (BRK), though, is an interesting twist because it’s a P&C insurer, and because of its reputation and financial strength.

My speculation is that if Robinson’s thesis is correct and sentiment toward insurance companies turns extremely negative, Berkshire will get caught up in it by association. You can bet Berkshire’s CDS are among the cheapest in the industry, so a change in sentiment could drive their prices up dramatically.

It’s similar to what happened in The Big Short. Unlike Burry, who shorted the lower tranches of MBS, the guys from Cornwall Capital bought CDS on the AA tranches, where protection cost pennies on the dollar, and made a killing.

Per Bloomberg:

[Robinson] sees as an inevitable downturn in private credit, a cooling-off in AI hype, and the impact of declining liquidity on corporate valuations. He says there are parallels between the general calm that prevailed in the subprime mortgage market before the blowup of Lehman Brothers and markets today, where corporate yield premiums remain at historically low levels. That investor confidence — or overconfidence, as Robinson sees it — persists even as concerns simmer about private credit’s exposure to software borrowers under threat from artificial intelligence, and as warning signs flash from a couple of corporate blowups.

All that said, he doesn’t expect the insurers to blow up. He just thinks they’re mispriced relative to the potential trouble with their private credit exposure.

I’ll be interested to see how this plays out, and I definitely plan to keep an eye on the trade.

7. Question Marks

I came across State Street's Private Markets Study 2026, which explores market sentiment among GPs, LPs, and other participants in the private assets ecosystem. The report covers several themes, including investor appetite for private assets, the ongoing "democratization" of private assets, and operational challenges. Personally, I was most interested in the first theme, particularly as it relates to private credit.

The first chart was exactly what I expected. Despite the headlines, only 8% of respondents plan to reduce their allocation to private credit, while 53% plan to keep their allocation unchanged and 39% plan to increase it.

Frankly, at this point, you have no choice but to admire the remarkable resilience of the asset class. I was fairly confident I knew why LPs continue to like private assets, but the next chart saved me the trouble of speculating.

I mean, I know there are many skeptics of private credit among my readers, and I myself have issues with some practices within the asset class. But that doesn’t matter. What matters is that the top three reasons above come from the fact that there is no market price discovery for these assets. As such, there is room to delay negative valuation adjustments (if not inflate valuations), which takes care of the returns, volatility, and yield enhancement.

“So, Aznaur, you support that?”

No, but I understand why most GPs and LPs wouldn’t fight those practices too much. GPs’ incentives are self-explanatory: fees and carried interest. And LPs just want peace of mind. They know that some asset prices might be inflated, and sometimes they raise an eyebrow when talking among themselves. But they also know there are third parties that review those valuations. They know those parties are not exactly independent and need to win business from GPs, but that’s still some validation that the prices are not entirely made up.

Put aside your idealistic views and answer honestly: do you think most people would rather sit in front of a committee explaining why the liquid credit fund they invested in has underperformed quarter after quarter, or present an illiquid private credit fund that keeps posting strong returns? That was obviously a rhetorical question.

Which is why this whole quest to make private credit more liquid is going to be long and painful. And when we finally get there, there will be a significant number of LPs asking whether there is something less liquid, with no mark-to-market, available.

8. Road to India

Asset-based financing continues to grow, including expanding into foreign markets like India. Bloomberg discusses how foreign banks' share of total issuance in India grew by 35% to 1.5T rupees ($16B), following growth of 30% and 28% in the prior two years.

Banks like JPM, Citi, and Barclays are among the active investors in the market and seem to be capitalizing on India's fast-growing credit market and, of course, the yield pickup.

Non-banking entities typically securitize loan portfolios with remaining maturities of one to 2.5 years, according to two foreign banks involved in such transactions. The average coupons range between 7.3% and 11.5% for securities rated AAA, AA and A.

That coupon pickup is pretty sizable, as the equivalent range in the U.S. would likely be around 4.50%–6.50%.

For Indian borrowers, ABS is cheaper than issuing in the local bond market, reducing borrowing costs by 75bps-100bps. That’s because the ABS structure creates tranches with different levels of seniority, allowing the senior tranches to receive higher credit ratings than the originating borrower itself.

India's private credit market is growing rapidly and is currently estimated at roughly $30B. As I noted in Round #45, there is significant interest in the market from both the world's largest banks and leading alts, many of which have been expanding aggressively in India. It's a market worth getting familiar with.

9. Deals on the Block

Download below: the complete Excel file covering 60+ deals completed in 2Q26 (plus the prior two quarters).

CLICK to Download 2Q26 Deals

10. Bulletin Board

• WSJ Online is offering 65% off its subscribers ($4 per week vs $11) LINK.

• Bloomberg is offering 60% off its annual subscription ($180 vs $399). If you are in Business, Bloomberg is a MUST, with no real close alternative (LINK).

• Shortcut.AI, an AI tool for Excel that lets you build models by simply typing requests (similar to ChatGPT). It is free to use up to a certain number of tasks per day (LINK). If you hit your limit before finishing your project, simply upgrade to the Pro plan ($20/month) and enter promo code DEBTSERIOUS (all capital letters) to get 50% off.

• LoanEdge, a BDC research tool that lets you look into BDC loan compositions, individual loans, which BDCs own a specific loan, and more, is kindly providing DEBT SERIOUS readers with a free 3‑month trial (the regular price is ~$3k per year, and it’s higher for institutions). Here is a LINK to the tool. Send an email to Sadaf Khan at sadaf@theloanedge.com and let her know you are from the DEBT SERIOUS community.

As a reminder, I don’t personally benefit from either of these offers. I am just saving you money.

Lastly, our community is over 4,200 strong and includes a broad cross-section of mid- to senior-level professionals across private credit, private equity, LPs, and debt advisory. Please use the Bulletin Board at the end of each weekly letter to share events, discounts, or deals that benefit the community. I will continue to feature these opportunities and personally connect interested parties at no cost.

That’s the bell — round over. See you in the next.

Aznaur

Aznaur.Midov@KierLior.com

Weekly Debt Capital Markets Digest

News

1. Lee Robinson of Altana Wealth, who made 900% betting against subprime mortgages is launching a fund to bet against private credit-exposed insurers (Bloomberg)

2. Morgan Stanley’s $7B North Haven Private Income Fund received redemption requests for 11.6% of outstanding units, but honored its 5% quarterly cap (Reuters)

3. US life insurers hold $807B in private credit, representing 20% of their $4T fixed-income portfolio. The allocation is expected to increase (Bloomberg)

4. Damis Holdings files for Chapter 11, with the company’s liabilities estimated at $500m-$1B (Petition)

5. BoE has sent alts a hypothetical stress scenario that includes rising inflation, a 4% contraction in UK GDP, and an increase in interest rates to 7%. It is viewed by participants as too severe (ACI)

6. Apollo’s $26B Apollo Debt Solutions BDC received redemption requests for 17% of its outstanding units (4.3% from onshore investors and 12.5% from offshore investors), but honored its 5% cap (CNBC)

7. PE zombie funds are expected to multiply as M&A activity remains muted. Among LPs, 54% would expect to cut fees, 18% to reset incentives, and 11% to remove managers (Bloomberg)

8. Ares’ $11B Ares Strategic Income Fund, a BDC, received redemption requests for 14.4% of its outstanding units (vs. 11.6% last quarter) but honored its 5% cap (Financial Times)

9. Blackstone REIT’s $343m loan on an office building in Chicago’s financial center has gone into default after being on a watchlist since 2022. Chicago’s office market has struggled to recover from the pandemic (Bloomberg)

10. 25% of mid-sized insurance companies tracked by Clearwater own stakes in private credit funds to which they also lend. Clearwater doesn’t track the largest insurers, where the figure may be higher (WSJ)

11. JPMorgan Public and Private Credit Fund, an interval fund, won approval from the SEC for monthly 2% redemptions instead of quarterly, semi-annual, or annual redemptions, as required by law (Bloomberg)

12. European PE executives increasingly borrow against their carried interest in their funds, as exits remain muted (Financial Times)

13. Indian banks are offering Non-Resident Indians dollar-denominated deposits at up to 7.1%, and lending up to 9x against those deposits. The purpose is to attract dollars back into the country following a sharp fall in forex reserves (Bloomberg)

14. Venezuela hired Centerview Partners to restructure its $240B debt, which is higher than market participants’ prior estimates of $150B-$200B (Financial Times)

15. Sheila Bair, Former Chair of the FDIC, is concerned about bank deregulation, banks’ exposure to private credit, and parallels with the events that led to the 2008 financial crisis (Politico)

16. US IG bond sales hit a record $175B in June, up 60% YoY, driven by the AI capex boom and concerns over potential rate hikes (Bloomberg)

17. 10.6%, or 100 million Chinese consumers are behind on debt payments, which has resulted in $329B of non-performing loans for financial institutions, a 21% increase y-o-y (Bloomberg)

18. SEC is probing Continuation Vehicles (CVs) used by private equity firms, including conflicts of interest, valuations, and whether investor disclosures are sufficient (Reuters)

19. Private-to-public BDC arbitrage is gaining momentum, but there are some differences in portfolio composition (Bloomberg)

20. 144A bond allocations by U.S. corporate treasurers have risen to 2% of portfolios from 0% two years ago, driven by an average 52 bps premium over similarly rated public bonds (Bloomberg)

21. Insurers have increased their allocation to private credit from 4% in 2018 to 9% today and now own an estimated 30%–50% of all private credit assets (AI-CIO)

22. “Credit allocators don’t fully appreciate that we’ve moved into a higher-interest-rate environment that many are not accustomed to.” – Interview with PGIM Credit’s Co-CIO (AI-CIO)

23. SpaceX’s $2.5B bonds saw their spread widen to 112 bps over the risk-free rate from 106 bps just days after issuance, suggesting some softening in investor confidence, which is unusual for such a recently issued bond (MarketWatch)

24. Aggressive LMEs that disadvantaged certain creditors are increasingly derailing prepackaged bankruptcies, turning what were meant to be fast-track restructurings into protracted and contentious Chapter 11 cases (WSJ)

25. India’s ABS market has surged to a record $16B through March (35% y-o-y), as global banks gain exposure to its fast-growing domestic credit market (Bloomberg)

Transactions

1. Brookfield is seeking a $525m loan for La Trobe Financial Corp, an Australian non-bank lender, to fund a payout to investors and refinance existing debt (Bloomberg)

2. Apollo-backed Eolo SpA is seeking a $575m loan from Apollo to refinance its existing debt (Bloomberg)

3. Blackstone-backed data centre operator AirTrunk is in talks to secure a $3B loan to finance a new hyperscale campus in Sydney (Bloomberg)

4. BMO is offering a $775m loan for HIG Capital-backed Pellera Technologies at an 8-cent discount (Bloomberg)

5. Sony ended a 28-year absence from the USD bond market, issuing $1B of 5- and 10-year USD-denominated bonds. The 10-year tranche priced at 60bps over Treasuries (Bloomberg)

6. ByteDance is in talks to raise a $20B, 3- to 5-year loan, which would be its largest-ever offshore borrowing (Bloomberg)

7. China sold €5B ($5.7B) of euro-denominated bonds, its biggest-ever deal in euros. Its 5-, 8-, and 12-year notes were priced at 3 bps, 12 bps, and 22 bps above the mid-swap rate, respectively (Bloomberg)

8. Mexico raised $6.3B in bonds to finance buybacks of existing debt, with the new issuance priced at 185–195 bps over comparable Treasuries, versus 220 bps for the outstanding debt (Bloomberg)

   ---