How it works
A company wants to borrow more money than any single lender sensibly wants to hold. So it hires a bank — the arranger — which underwrites the loan, then goes out and sells participations to a crowd: CLO vehicles, loan mutual funds, insurers, hedge funds, other banks. The borrower signs one credit agreement and makes one set of payments. Behind that agreement sit a great many lenders, each owning a piece and each free to sell it to someone else.
An illustration with invented round numbers: a company borrows $800m, the arranger commits to the whole amount, then places $750m with roughly forty accounts at 99 cents on and keeps $50m. The buyers now own tradeable paper. Six months later that paper might quote at 97, or at 88, and everybody who owns it can see it.
These are almost always floating-rate loans, priced as a spread over , senior and secured, and in the leveraged corner of the market they finance buyouts. During the easy years the packages got progressively thinner, to the point where "cov-lite" stopped being a description and became the default.
Syndicated loans trade, and that secondary market is the part that matters for everything below. Dealers quote them, pricing services publish those quotes, and holders their books to them. Nobody has to believe the borrower's own view of how it is doing. This is exactly the discipline that was invented to escape — a fund holds the loan to maturity and values it using a model, which is more comfortable for everyone right up until it isn't.
The two markets have therefore always been understood as a hierarchy. Syndicated is cheaper, more liquid and more standardised, but it is a market, and markets have moods. When the syndicated bid closes — too much leverage, an unloved sector, a nervous quarter — the borrower goes to private credit, pays a premium, and gets certainty of execution from a lender that does not need to sell the paper to anyone.
Why it matters to this crash
Because the traffic just reversed, and the direction of travel is the whole story.
-backed Sophos needed to refinance or extend more than $2bn of loans. It spent months trying to line up private credit funds to take the paper. Those efforts faltered, and the company went back to its existing leveraged-loan lenders, offering a higher coupon, amortisation payments and a tighter to get it done. We wrote it up on 19 August and again later that day, and the Bloomberg report is worth reading in full.
Read the hierarchy again. Private credit's pitch is that it is the buyer of last resort: it lends where the syndicated market won't, on terms the syndicated market won't write, at a price. If the last resort declines and the borrower has to go back to the public market — the one place its debt gets every day, and the one place lenders can extract amortisation and covenants as the price of staying — then the marginal leveraged borrower has one fewer door than it thought.
This is not a default and we did not report it as one. Lenders demanding amortisation and covenants is lenders doing their job, and it is the healthy version of repricing. But it is the third Thoma Bravo software credit in three months to need something: the firm made major concessions in July to get a $5bn Proofpoint refinancing over the line, and in June creditors took control of Medallia outright. Set that beside the median rate at the twenty largest listed reaching 2.8% of cost in Q2 2026, the highest since 2017, and the pattern is consistent: private credit is getting choosier, and the syndicated market is absorbing what it turns away.
What would make this dangerous
The observable escalation is a queue. One sponsor-backed borrower rejected by private credit and taken back by its syndicated lenders is a repricing. Four or five in a quarter, all paying up in coupon and giving back covenants, is the private credit bid withdrawing from the refinancing function it was built to perform.
Watch for the next rung down: a borrower that private credit declines and the syndicated market also fails to clear — a deal pulled, or an arranger left holding paper it cannot place. That is the point at which there is no door, and the alternative is a rather than a refinancing.
Watch the secondary marks on the loans that do get done. If the reopened syndicated deals trade below their issue price within weeks, the concessions were not enough and the next borrower will need bigger ones.
And watch the sponsors. Thoma Bravo told lenders it will not inject fresh capital into Sophos, rebuffing a request from investors worried about what AI does to a cybersecurity software business. A declining to defend its equity means any loss stays with creditors. If that becomes the standard sponsor response rather than the exception, the syndicated market will price accordingly, and quickly, because it is a market and that is what markets do.