Companies are refinancing out of private credit and into the syndicated market at roughly three times the rate the traffic runs the other way. The trade press has read this as banks clawing back ground they lost. Having sat on the lending side of these deals, I’d argue the headline mistakes a handover for a defeat, and that the real mechanism is both duller and far more reassuring than a turf war.
By Scott Townsley · Investment Director · 9 min read
The bottom line, up front
- The data: in the US second quarter, banks refinanced nine deals worth $4.5 billion out of private credit; only three worth $1.3 billion travelled the other way. In Europe, direct lending hit a record while its pricing premium over banks thinned. On the surface, banks are winning.
- The better reading: most of this isn’t banks growing braver. It’s loans written in 2022 and 2023, when banks couldn’t or wouldn’t lend, that have since seasoned, de-levered and proven their thesis, and have now fallen back into ordinary bank appetite. That isn’t the model breaking. That’s the model working as designed.
- For you: if you’re paying a private credit coupon today, the right question isn’t “why so expensive”. It’s “what does this business need to prove to graduate to cheaper capital, and when”. Priced that way, a high coupon is a bridge, not a penalty.
Pricing always matters to a borrower; no one signs a term sheet without scrutinising the margin. But in every financing negotiation I’ve ever sat in, that concern is often superseded by a more pressing one: who else would say yes, and how fast. For most of the last five years across the UK and continental Europe, the honest answer for a whole category of businesses was: not many, and not quickly. That gap is precisely what direct lenders were paid to fill. This summer’s run of refinancing stories has been read as evidence that the gap is closing and the direct lenders are losing. I think that gets the mechanism almost exactly backwards.
Term to know: the liquidity premium
The extra yield a private credit fund charges over a comparable bank or bond loan, in exchange for certainty of funds and speed of execution. The interesting question isn’t whether that premium is compressing, it plainly is. It’s why. If it’s because borrowers got genuinely safer, the market is working. If it’s because too much capital is chasing too few deals, discipline is slipping. This month, both are true at once, and telling them apart is the whole game.
What the headline gets right, and what it hides
Start with the facts, because they aren’t in dispute. Bloomberg’s reporting on the US second quarter is stark: companies carrying private debt refinanced into the syndicated market about three times as often as syndicated borrowers moved into private credit. Baker Tilly had lined up Deutsche Bank to move north of $2 billion out of a private credit facility and into a syndicated loan before shelving the plan in recent weeks; Fidelis Partnership is running a similar roughly $2 billion trade with Morgan Stanley; KKR-backed Accuris was preparing a $700 million leveraged loan to take out its direct lenders entirely. Even the high yield bond market has muscled in: Ancestry.com’s $2 billion refinancing grew its bond tranche to $950 million from $450 million while shrinking its loan portion.
The lazy interpretation writes itself. Banks retreated after the financial crisis, private credit ate their lunch for a decade, and now the banks are eating it back. Tidy, adversarial, and mostly wrong. It treats a loan as a fixed object that simply changes owners, when the more important thing happening is that the loan itself has changed.
A credit that was unbankable at origination isn’t unbankable forever. If the business does what it promised, it graduates.
The graduation thesis
Consider what a typical direct lending borrower from 2022 actually looked like at the point of financing. Perhaps mid-integration on an acquisition. Perhaps pre-profitability, or profitable but with a few aggressive EBITDA add-backs in the bridge. Perhaps in a sector the bank credit committees had quietly redlined. Perhaps requiring a structure, stretched senior, unitranche, a turn of mezz or PIK, that a deposit-funded balance sheet simply cannot hold at a sensible regulatory capital charge under Basel III. None of these in isolation necessarily make the business a bad credit. It made it a bad fit for a bank at that moment. That distinction is the entire reason private credit exists.
Now run the clock forward three years. The acquisition integrated. Leverage has reduced as the business either ramps up earnings or pays down debt and grows into its capital structure. The scary adjustment washed through and the numbers are now simply the numbers. What was a complicated, sub-investment-grade story in 2022 is, in 2026, a more conventional senior credit that any bank would be glad to write, at a bank’s price. The loan didn’t move into bank territory because the bank stretched to reach it. It moved because the borrower walked into the bank’s territory on its own two feet.
This is not a novel or contrarian idea on the lending side of the table. It’s the expected life cycle of a well-underwritten credit, and it’s exactly what a good direct lender tells a borrower at the outset: this coupon reflects where your business is today, not where it’s going. Execute the plan, and you should refinance us out with something cheaper. When that happens, the direct lender hasn’t been beaten by a bank. It has been repaid, on schedule, having earned a premium for taking the early risk nobody else would. That is a successful outcome dressed up by the trade press as a defeat.
The same loan, three years apart
2022 · Origination
Mid-integration or pre-profitability, aggressive structure, redlined sector, or all three. Banks constrained by capital rules and risk appetite. Direct lender steps in, prices the uncertainty, holds the whole facility.
2023–25 · Seasoning
Thesis executes. Leverage falls, cash flows prove out, the messy adjustments clear. The credit quietly de-risks while the coupon stays fixed at its origination level.
2026 · Graduation
Now an ordinary senior credit. A bank will write it at a bank’s price; scale and lower cost of capital do the rest. The direct lender is refinanced out, repaid in full, premium earned.
Illustrative life cycle of a well-underwritten mid-market credit, drawn from the author’s origination experience. Not every loan follows this path; distressed and underperforming credits do the opposite.
Why I won’t oversell my own argument
Here is where most commentary, including the version of it that flatters my own argument, starts cherry-picking. So let me be straight about the limits of the graduation thesis, because at least two other things are happening at the same time, and honest analysis means holding all three at once.
The first is plain spread-shopping. Some of these refinancings aren’t graduations at all; they’re borrowers exploiting a reopened syndicated market to lock in a lower rate as spreads compress, regardless of whether their underlying credit improved. Octus, which has branded 2026 “the year of maturity” as the 2021 vintage comes due, is explicit that falling spreads alone are pulling refinancings forward. That’s not the credit cycle working; rather a reflection of liquidity in the market generating cheaper money on offer, and it would reverse the moment syndicated capacity tightened again.
The second is genuine: banks are consciously trying to reclaim leveraged finance ground, and some funds are meeting them head-on, underpricing risk on “regular way” credits to keep capital deployed. White & Case’s read of the reopening US syndicated market describes exactly this, banks strategising to win back LBO financing they’d ceded. Where that’s the driver, spread compression really is a competition story, and a mild warning sign, not a seasoning story.
My read on the split is roughly this: a large and under-reported share of the refinancing wave is credits graduating as designed; a meaningful share is spread-shopping that says nothing about credit quality; and a smaller but real share is banks and funds genuinely competing on price in a way that should make a disciplined lender uneasy. Anyone selling you a single clean narrative, mine included, is editing the data to do it.
What this means for you
If you’re a borrower, this distinction is worth real money. If your loan is up for refinancing because your business genuinely de-risked, you hold the leverage and should run a full process, banks, syndicated and direct lenders in competition, rather than defaulting to an amend-and-extend with your incumbent. But if the only thing that’s changed is the market mood, be honest with yourself about how durable that window is before you restructure around it.
The European picture, and why it’s quieter
Europe tends to run six to eighteen months behind the US on these shifts, so these American stories are best read as a preview rather than a mirror. The domestic numbers point the same way but with far less drama. European direct lending volume hit a record €41 billion in 2025 (roughly £35 billion), and total leveraged loan issuance reached an all-time high of €141 billion (about £120 billion), so this is emphatically not a shrinking market. But within LBO financing specifically, direct lending of around €22 billion (some £19 billion) is now running close to broadly syndicated bank loans, and the pricing gap between the two has thinned.
European direct lending: growing market, thinner premium
| Metric | ~2024 | 2026 |
|---|---|---|
| Direct lending premium over syndicated | ~150bps | ~125bps |
| Margins on the best-quality assets | 500bps+ | sub-500bps |
| Unitranche loans with cov-lite terms | Lower | ~65% |
| Syndicated loans with cov-lite terms | Lower | ~40% |
Source: R.W. Baird, “Looking Through the Noise: Q2 2026 Trends Across the European Debt Market”; DD Talks, “Unitranche Financing: European Market Developments 2026.” Sterling equivalents are approximate, at illustrative rates.
The compression is not evenly spread, and that unevenness is the tell. Margins are thinnest exactly where competition for the best assets is fiercest, on clean, larger credits, while storied or AI-exposed names are paying premiums of 50 to 100 basis points more than they would have late last year, if they can get financed at all. A market genuinely re-rating on improved credit quality would compress fairly uniformly. A market where the graduating top end is being fought over while the difficult middle is left behind looks exactly like this. Which brings us to the borrowers this article is really for.
The borrowers the good-news story forgets
Read the trade press right now and you’d think every private credit borrower just got a pay rise in negotiating leverage. Look at the default data by size and that story falls apart. I won’t relitigate the headline default debate here, I set out in Private Credit’s Default Reckoning why the numbers read so differently depending on methodology; the short version is that defaults are ticking up modestly, but nothing like as dramatically as the headline Fitch figure implies. What matters for this piece isn’t the average. It’s the spread around it. Businesses in the small mid-market with EBITDAs of $25 million to $50 million have seen defaults increase from 3.1% to 3.4%.
Businesses in this band are the ones with no syndicated or bond alternative to graduate towards. The refinancing window reopening at the top of the market simply does not extend to them. So as the largest, cleanest, most profitable mandates migrate back to banks and bonds, direct lenders will need to work the assets left behind harder to hit their return targets, which points to tighter terms, not looser ones, for exactly the businesses least able to shop around. If you run a company in that band, the dominant “private credit is repricing in your favour” narrative was not written for you, and mistaking it for your reality is a genuine risk.
Even the general partners are recycling
One data point has nothing to do with borrowers and everything to do with where the market’s own operators think value now sits. Bridgepoint, a well-regarded manager and no distressed seller, is exploring the sale of more than €1 billion (around £850 million) of private credit stakes through a secondaries deal structured as a continuation vehicle: existing investors cash out, new capital rolls in. The obvious reading is defensive, a liquidity mechanism for an asset class that markets itself on illiquidity, and there is truth in that. Distributions have been slow across the industry and continuation vehicles relieve the pressure.
But the defensive reading is only half of it, and the smaller half. Ares and BNP Paribas both frame these structures explicitly as capital recycling, not just liquidity: a way for a manager to free up capital from a mature, performing book and redeploy it into new deals struck at today’s pricing and structure, while diluting the risk in the existing portfolio. That this is now a proactive play rather than a distress signal is borne out by the shift in who initiates them; in 2025, manager-led secondaries volume overtook allocator-led for the first time. In other words, it isn’t primarily investors demanding an exit. It’s GPs choosing to reset. So read Bridgepoint less as “even the insiders want out” and more as “even the insiders are repositioning towards the deals that fit this market”, which is, tellingly, the same instinct the graduation thesis asks borrowers to adopt.
One number worth watching: the regulator
On 6 May 2026 the Financial Stability Board published its first formal report on vulnerabilities in private credit, sizing the market at $1.5 to $2.0 trillion and flagging the interlinkages practitioners have quietly tracked for years: bank exposure to fund financing lines, insurer participation, leverage and liquidity mismatches. The conclusion is procedural, not alarmist, better data, harmonised definitions. But a global regulator rarely publishes a vulnerabilities report as the end of a conversation. UK and EU supervisors have historically followed within a cycle. Expect more disclosure requests from your lender over the next 12 to 18 months, not because you did anything wrong, but because their supervisors are asking harder questions upstream.
So: defeat, or the model working?
Strip away the turf-war framing and the picture that remains is more reassuring than the headlines, and more demanding. Private credit is not being beaten by a resurgent banking system. For the most part, it is doing the job it was built for: pricing risk that banks structurally could not hold, carrying that risk while the business proved itself, and being repaid, at a premium, when the credit matured back into cheaper hands. That is a functioning market completing a cycle, not a bubble deflating.
But there are two caveats. Some of this compression is just cheap money and competitive pressure, which will reverse when the window closes, and no lender should mistake that for durable improvement. And the graduation escalator only runs at the top of the market. For the lower mid-market, the segment I spend most of my time in, there is no syndicated exit waiting, the stress is quietly rising, and the good-news narrative is somebody else’s story.
For a borrower, the practical takeaway is refreshingly clear. Don’t read your coupon as a verdict on your business. Read it as a snapshot of where your business sits today, against where your plan says it’s going. If you’re executing, the path to cheaper capital is a milestone to plan for, not a rescue to wait for. And if you’re approaching a maturity, find out which of the three stories you’re in, a genuine graduation, a fleeting market window, or a middle-market credit with nowhere cheaper to go, because the right move is completely different in each case. That is a conversation worth having well before the refinancing clock runs down.
— Scott
Questions I’d expect from this piece
If banks are refinancing loans out of private credit, isn’t private credit losing?
Only if you think being repaid on schedule is losing. When a credit that was too complex for a bank in 2022 seasons into an ordinary bank credit by 2026, the direct lender gets repaid in full having earned a premium for the early risk. That’s the intended outcome, not a defeat. Some refinancings are more opportunistic than that, but the core mechanism is the model working.
So why is my private credit loan so much more expensive than a bank loan?
Because it was priced against your business as it was at origination, often mid-transition, aggressively structured, or in a sector banks were avoiding, rather than as it may become. The useful way to read that coupon is as the cost of capital that was actually available when you needed it, with a clear path to something cheaper once the business proves out.
Should I refinance out of my direct lending facility now?
Worth exploring seriously if your business has genuinely de-risked and you have the scale and credit quality to interest a syndicate desk or bond underwriter. Less compelling if the only thing that’s changed is market sentiment, because that window can close as fast as it opened. Either way, price the alternative properly and mind your prepayment terms before assuming your incumbent is the only option.
I’m in the lower mid-market. Does any of this help me?
Less than the headlines suggest, and it’s important to be clear-eyed about that. The refinancing window is reopening at the top of the market; the €25–50 million EBITDA band has no syndicated or bond alternative to graduate towards, and the default data shows stress rising there. Appetite hasn’t vanished, but it’s more selective, so a genuine competitive process matters more for you, not less.
Will the FSB report lead to new UK or EU rules soon?
Not immediately; it calls for better data and definitions before any rule-making. But it’s a strong signal that supervisors will start asking lenders more probing questions over the next year or two, which tends to flow downstream into what lenders ask of borrowers.
About the author: Scott Townsley originates and structures private credit transactions for lower- and mid-market businesses across the UK and Europe, with a track record of financing sound, sub-investment-grade companies that traditional banks are structurally unable to serve at the point they need capital. He writes PrivateCredit Weekly to help owners, CFOs and boards read a market that moves faster than most advisory relationships keep up with. Get in touch · Subscribe.
Sources: Bloomberg, “Private Credit Squeezed By Bank Refinancings” (Credit Weekly, August 2026); Octus, “Americas Private Credit 2026 Outlook: The Year of Maturity”; White & Case, “The US Syndicated Loan Market Shows Signs of Recovery”; Proskauer Private Credit Default Index (Q1 and Q2 2026); Fitch Ratings; Financial Stability Board, “Report on Vulnerabilities in Private Credit” (6 May 2026); R.W. Baird, “Looking Through the Noise: Q2 2026 Trends Across the European Debt Market”; DD Talks, “Unitranche Financing: European Market Developments 2026.” Sterling equivalents of euro figures are approximate and for indication only. The views expressed are the author’s own and do not constitute investment, legal or financial advice.