Direct Lending Against Broadly Syndicated Loans: Who Holds the Paper

Two lenders can look at the exact same company, offer a nearly identical facility, and end up in completely different relationships with the borrower for the life of the loan. One sells its exposure within weeks to dozens of buyers it will never meet. The other keeps the loan on its own book until it is repaid, sometimes years later, and is still the first call when something goes wrong.

Who actually holds the paper is the spine of this whole comparison. A direct loan and a broadly syndicated loan (BSL) can look almost identical on the term sheet and behave completely differently once it is signed. A candidate who can argue both sides of that, not just define the two, has understood more about this asset class than one who has only memorised the vocabulary.

90%+

of broadly syndicated deals are cov-lite

Proskauer, November 2025

125-300bps

direct lending's spread premium today

down from 220-350bps historically, Brookfield, March 2026

99% vs 24%

of loans carrying a live market price

BSL vs private-credit CLOs, Moody's, August 2024

$4.3bn

moved from direct lending into the syndicated market

four deals, early 2026

Who holds the loan, and why almost everything else follows from it

A broadly syndicated loan starts life the same way a direct loan does: a bank or a lender agrees to finance a company, usually one a private equity sponsor already owns. What happens next is where the two markets split.

A BSL is underwritten and then distributed, sold down piece by piece to a syndicate of buyers, chiefly collateralised loan obligations, alongside loan mutual funds and other institutional accounts. A direct loan skips that step entirely. It is negotiated bilaterally and stays with the lender that wrote it, or a small aligned club, all the way to maturity.

That single choice, distribute or hold, shows up in the data. Average deal size in August 2024 ran $471 million for broadly syndicated CLO collateral against $553 million for SME and private-credit CLO collateral, with wider variance on the private-credit side. Par outstanding told a scale story of its own: $626 billion in broadly syndicated CLOs against $61 billion in SME and private-credit CLOs the same month, a gap of roughly ten times.

The collateral quality trade-off runs in an interesting direction, too. SME and private-credit CLO collateral carries higher exposure to lower-rated names than a broadly syndicated pool, but it is generally backed by a larger overcollateralization cushion, while broadly syndicated CLOs show stronger weighted-average rating factors and recovery rates on the names inside them. Neither structure is simply the safer one; each compensates for its own weaker spot in a different place.

A collateralised loan obligation is not a competing strategy to either of these. It is a financing wrapper that can hold BSL paper or direct loans, and the choice of what sits inside it is downstream of everything in the table above, not a separate decision. The mechanics of that wrapper, and how covenant-lite creep has moved inside direct lending itself, are covered in full in private credit strategies.

The premium, and what it is actually buying

Direct lending has historically priced wider than a comparable broadly syndicated loan, and the gap is real enough to measure. First-lien direct lending margins ran around 500 basis points in September 2025, against roughly 332 basis points for a comparable syndicated B2/B3 loan. That spread has not been static. The historical premium of roughly 220-350 basis points has narrowed to something closer to 125-300 basis points today, depending on credit quality and tenor.

What that premium buys is not a lower expected loss. It buys illiquidity, execution speed and certainty. A direct lending agreement can often close in weeks; a public or syndicated issuance can take months once rating-agency review and syndication are added in. A borrower negotiates with one counterparty instead of dozens, and keeps a relationship it can return to eighteen months later for an amendment or an add-on, rather than reopening a whole syndicate.

Proskauer's own comparison of the two markets is blunt about the direction of travel: private credit terms have historically been more lender-favourable, and that remains the case. The flip side is equally direct — broadly syndicated deals often carry more borrower-favourable pricing and looser covenants than a private credit alternative, precisely because the price of that flexibility is spread across a whole syndicate rather than concentrated in one balance sheet.

Why the gap keeps narrowing

Competition, not generosity, is doing the narrowing. When the broadly syndicated market seized up in 2022 and secondary prices fell to 92 cents on the dollar, direct lending grew 188% that year, proving it could fund deals banks temporarily could not. That crisis premium has been competed away ever since, as more capital has chased the same borrowers and private credit funds have had to sharpen pricing to win deals a bank syndicate is once again able to underwrite.

Covenants: the one boundary that still mostly holds

The covenant question is where the two markets diverge most visibly, and it deserves only a market-level answer here — the ratios, the headroom conventions and the breach mechanics belong to private credit covenants explained, which owns that ground in depth.

At the market level, the split is stark. Over 90% of broadly syndicated deals were covenant-lite as of late 2025, and that share keeps climbing. Direct lending has built its identity on the opposite choice.

• Broadly syndicated: cov-lite as the default. A covenant only bites if the borrower takes a specific action, so a business can miss its plan badly and still be in perfect technical compliance.
• Direct lending: maintenance as the default. A leverage test runs on a fixed schedule regardless of what the borrower does, which is what buys the lender an early seat at the table.

That is not a clean, permanent line. Roughly 40% of upper-middle-market private credit deals were themselves covenant-lite as of 2024, and the share is rising fastest exactly where direct lending now competes hardest against the syndicated market: the largest borrowers. That erosion, and what it costs a lender in recovery, is the covenant piece's subject, not this one's.

Marks, liquidity and price discovery: the difference that is hardest to see

A broadly syndicated loan trades. Somebody is usually willing to buy or sell it on any given day, which means it carries a real, observable price. A direct loan almost never trades. The lender holding it has to estimate its value using its own credit model, because there is no active market quoting it back.

The Federal Reserve Bank of Boston put the structural fact plainly: after origination, broadly syndicated loans are generally traded in secondary markets, while commercial-and-industrial and private-credit loans remain on lenders' balance sheets. That is not a footnote. It is the reason a broadly syndicated loan's price can move the moment sentiment shifts, and a direct loan's mark does not have to move at all, even if the underlying business is genuinely deteriorating.

The IMF has made this the centre of its own warning about the asset class. Its April 2024 review flagged "stale valuations based on mark-to-model" as a structural vulnerability, alongside hidden layers of leverage and unclear interconnections between participants. Regulators, it added bluntly, simply do not have the data to judge what sits beneath the surface calm.

What a model-based mark actually produces

The clearest illustration is a comparison the IMF's own report leaned on. Between 2019 and 2023, Lincoln International's European Senior Debt Index, tracking private credit loans, returned an 8.2% compound annual growth rate with a standard deviation of just 1.1%. The public Morningstar European Leveraged Loan Index, tracking traded syndicated paper, grew 4.4% over the same period with a standard deviation of 5%.

Read those two numbers side by side and private credit looks dramatically steadier. Read them the way the IMF does, and the steadiness itself becomes the concern: a model-based mark that moves quarterly cannot show the same volatility as a price that trades daily, whether or not the underlying risk has actually changed.

Broadly syndicated CLO collateral with a pricing-service mark

99%

SME and private-credit CLO collateral with a pricing-service mark

24%

Nearly every broadly syndicated loan in a CLO gets a pricing-service quote. Only about a quarter of private-credit collateral does, which is most of the reason its valuations are harder for an outsider to check.

What actually trades, and what does not

Pricing marks are one measure of liquidity. Turnover is another, and it tells the same story from a different angle. Over the six months before Moody's compared the two markets in November 2024, 32.6% of broadly syndicated CLO collateral was sold, against 12.8% of SME and private-credit CLO collateral.

A Federal Reserve Board note on bank lending to private credit vehicles makes the consequence explicit: public disclosures from these vehicles are generally not available. That gap, combined with limited understanding of how private credit connects to the rest of the financial system, makes it genuinely difficult to assess systemic risk from outside.

That is a regulator's problem as much as an investor's, and it is worth having ready as an answer that goes beyond "private loans are illiquid."

None of this means a direct loan is a worse credit. It means the two markets answer the question "is this loan still worth par" in structurally different ways, and only one of those answers updates in real time.

What each is actually good for, from the borrower's side

None of this is abstract for the company doing the borrowing. Each market solves a different problem, and a borrower's choice usually comes down to which problem is most pressing that quarter.

A direct loan suits a borrower that wants

• A term sheet in weeks, not months
• One counterparty for the life of the loan
• Confidentiality — no rating agency, no public book
• A complex story a generalist syndicate might price badly

A broadly syndicated loan suits a borrower that wants

• Dozens of buyers competing on price and terms
• Cov-lite terms once the borrower qualifies for them
• A public-style credit profile it is comfortable disclosing
• Sheer size — the very largest tickets still clear here

Direct lending's sweet spot has traditionally been companies too small to issue public debt, or that do not meet a bank's own lending criteria — broadly, middle-market businesses in the $10 million to $100 million EBITDA range, where a direct lending agreement can close in weeks rather than the months a public or syndicated issuance typically requires once regulatory filings and syndication are added in.

Confidentiality is a real, separate reason to choose the private route, not a footnote to speed. A direct lending transaction is negotiated with a single or limited group of counterparties, which keeps the terms of the deal out of a wide market. A broadly syndicated loan, by contrast, could require input from hundreds of institutional buyers before it closes, and every one of them sees the credit.

Where a borrower outgrows a direct lending club

That $10 million to $100 million range describes the traditional core of the market, not its current ceiling. Private credit has moved decisively upmarket: Ares itself led a $3.3 billion loan to a European insurance broker in the first quarter of 2024, a size that would have gone straight to the syndicated market five years earlier.

There is still a real edge, though, and Proskauer states it without hedging: broadly syndicated deals can beat private credit on size. The market is simply deeper and more liquid, and no single balance sheet, however large, can distribute risk across as many buyers as a syndicate can — which is why the biggest tickets in the market still tend to clear on the bank-arranged side first.

• Under roughly $100 million EBITDA, direct lending is usually the natural fit — speed and confidentiality outweigh the wider pricing.
• In the low billions, both markets now genuinely compete, and borrowers increasingly run a direct-lending process and a broadly syndicated process in parallel for the same financing, a practice known as dual-tracking.
• At the very largest end, the syndicated market's depth still wins on pure distribution capacity, even where private credit can technically write the whole cheque.

The convergence: competing for the same borrowers now

For years the two markets served genuinely different companies. That separation is breaking down, and the clearest evidence is not a survey — it is money actually moving between the two markets in both directions inside the same year.

At least $4.3 billion across four deals moved from direct lending into the syndicated market in early 2026, driven by a real, quantified cost gap. On stressed, below-investment-grade credits, direct lending spreads widened to between 550 and 600 basis points over SOFR, on fears that software-heavy portfolios face AI-driven disruption and that stress is building among mid-sized borrowers.

Public junk-loan spreads, over the same window, averaged 350 to 400 basis points over SOFR — a roughly 200 basis point gap two bankers described directly to Reuters.

Direct lending, stressed software-heavy credits

550-600bps over SOFR

Public / broadly syndicated market, comparable risk

350-400bps over SOFR

On below-investment-grade credit specifically, in early 2026, direct lending priced meaningfully wider than the public market for comparable risk — wide enough that borrowers started refinancing out.

Not every lawyer close to these deals thinks price is the whole story. Sheel Patel of Mayer Brown put it carefully: borrowers may be drawn to the syndicated market by the pricing gap, but the decision is rarely driven by rate alone. Angela Hagerman of Reed Smith described banks responding to the competitive pressure directly, willing to drop pricing and loosen covenants to compete with the traditional bank lender market on its own turf.

The refinancing runs both directions, and so does the capital

This is not simply borrowers abandoning direct lending. In the $5.75 billion loan banks arranged to help finance Electronic Arts' leveraged buyout, some private credit funds that would ordinarily have bought into the syndicate pulled back their orders or cut them. Private credit funds are buyers inside broadly syndicated loans as well as competitors to them, which means the overlap between the two markets now runs through the investor base, not only through which borrowers choose which door.

• Cov-lite creep inside direct lending itself has moved from a rounding error to a meaningful share of large deals, covered in full with the exact figures in private credit covenants explained — and it is concentrated precisely where this convergence is sharpest, among the largest borrowers.
• Dual-tracking is now routine on financings large enough to genuinely qualify for either market, turning what used to be a single decision into a live competition run in parallel.
• The direction of the flow depends on the calendar. A borrower refinances toward the syndicated market when public pricing tightens and away from it when public markets wobble, which is exactly what happened in reverse during the 2022 dislocation described earlier in this piece.

"If public markets are open, and your credit profile is strong, there's a real case for tapping the [syndicated] market. You get liquidity, price discovery, and the ability to refinance down the road."

That is Moody's own read on why a strong borrower still weighs the syndicated route seriously, even after building a relationship inside private credit. It is also close to the best one-paragraph answer available to a candidate asked to argue the case for either market cold.

How to argue this from both sides in the room

An interviewer testing this comparison is rarely checking whether a candidate can define the two markets. Almost everyone can. What separates a strong answer is whether the candidate can argue either side convincingly, because the honest answer keeps changing as the two markets compete.

1. Open with who holds the paper, not with price. Distribution versus retention is the fact everything else derives from, and leading with a spread number skips the mechanism that explains it.
2. Know the premium's direction of travel. It has compressed from roughly 220-350 basis points to something closer to 125-300, and being able to say why, competition, not generosity, shows market awareness rather than memorised trivia.
3. Place the covenant gap at the market level, and be ready to hand off to the mechanics — maintenance versus incurrence, headroom, what a breach actually triggers — rather than trying to fake depth on ratios in the same breath.
4. Explain why a quiet mark is not proof of a quiet loan. A model-based valuation moving less than a traded price is a fact about the pricing method, not necessarily a fact about the underlying risk, and saying so is a stronger answer than reciting "private credit is less volatile."
5. Have one live convergence example ready. A named refinancing flow, a cov-lite statistic, or the fact that private credit funds also buy syndicated paper all show a candidate is reading the market as it exists now, not as a textbook described it two years ago.

The bottom line

A direct loan and a broadly syndicated loan can finance the same company at nearly the same size, and still behave like different asset classes once the ink is dry. One is arranged, sold and traded; the other is negotiated once, held to maturity, and priced by the lender's own judgement rather than by a market. Every other difference here, the covenant norm, the premium, how easily distress shows up from outside, follows from that single structural choice.

What has changed is that the two markets no longer sit in separate lanes waiting for a company to graduate from one to the other. They compete directly for the same large borrowers, refinance deals away from each other in both directions depending on the calendar, and increasingly share the same capital on the investor side too.

Knowing the textbook difference is table stakes. Knowing which way that competition is running this year, and being able to argue either side of it, is what actually separates a prepared candidate in the room.