How it actually works
The old way of financing a buyout was a layer cake. A senior secured term loan at the bottom, cheap because it gets paid first; a second lien or mezzanine piece above it, expensive because it does not; sometimes a high-yield bond on top of that. Each layer had different lenders with different tolerances, an intercreditor agreement negotiated between them, and a bank in the middle underwriting the whole thing and then selling it into the . If the market moved between commitment and sale, the bank ate the difference.
A unitranche deletes the layers. One credit agreement, one security package, one floating rate over , one lender to call. Purely as illustration: where a deal might once have needed $350m of senior at SOFR+350 and $150m of second lien at SOFR+750, a fund writes a single $500m loan at SOFR+500 and the signs it in a week. The blended rate sits above senior pricing and below the weighted average of the old stack, which is how the structure pays for itself. What the sponsor is actually buying is certainty. No ratings process, no flex language, no risk that the syndication desk comes back on Thursday and says the market has moved.
What people get wrong
The layers do not always disappear so much as behind a curtain. Large unitranches are frequently split internally into a "first-out" piece, often taken by a bank at a lower rate, and a "last-out" piece held by the credit fund at a higher one, with the split governed by an Agreement Among Lenders. The borrower is not a party to it and generally does not see it.
So the is still there, along with all the questions about who controls a waiver and who gets paid in a default. It has simply been relocated into a private contract between lenders, one that has been tested in court far less often than the intercreditor agreements it replaced.
Why it matters to this crash
The unitranche is the mechanism, not just the marketing, behind the shift of leveraged lending out of banks and into . It is very hard to compete with speed and certainty when the alternative involves a ratings agency and a roadshow.
What that shift changed is not mainly the risk of the loans. It is the visibility. A broadly syndicated loan has hundreds of holders and trades every day, so when a borrower gets into trouble the price falls and everyone can see it fall. A unitranche has one holder who intends to keep it until maturity. There is no price. There is a , produced quarterly by the lender's own valuation process and reviewed by a third party the lender hires. When the same manager holds the same loan across a , a drawdown fund and an , the loan can be carried at three slightly different numbers and none of them is a trade.
The structure has a genuine virtue. Unitranches typically keep a maintenance , which most stopped doing years ago, and a single lender can restructure quickly without the coordination fights that produce . You cannot easily prime a lender group of one. The flip side is that a single lender facing a struggling borrower has no one to argue with when it decides that flipping the coupon to and extending the maturity is the commercially sensible answer.
We have not published a figure for the size of the unitranche market and are not going to invent one.
What would make this dangerous
Watch for the club coming back. Unitranches large enough that three or four funds have to share them recreate every coordination problem of a syndicate while keeping none of the price transparency. Watch for first-out tranches being sold onward to banks and financing vehicles, which means the "single lender" whose patience is the whole safety argument is not actually a single lender.
On the individual credits: a rising share of unitranche interest rather than cash, maintenance being loosened or reset rather than tripped, and rates that stay flat while borrower cash flows visibly do not. And most diagnostic of all, the appearance of a real secondary bid. The day unitranches start changing hands at prices meaningfully below where they are carried, every holder of the same paper has a problem, and none of them will have chosen the moment.