The Borrower Has to Qualify for the Exit
The borrower, not the market headline, determines whether private credit and leveraged loans are real substitutes. This chapter compares the firms and loans served by the two markets, then turns those differences into an access test: can a company that loses private-credit capacity plausibly present itself to the syndicated market? The evidence supports a conditional judgment, not a universal eligibility rule.
Private credit and leveraged loans overlap because both finance highly leveraged, below-investment-grade middle-market companies. They also finance many of the same activities. But the overlap is broad enough to permit some switching, not so complete that every borrower can move between the markets.
The overlap is real, but the borrower populations are different
The Federal Reserve’s comparison of 2025 loan issuances gives the clearest starting point. The median private-credit borrower had revenue of $223 million, while the median leveraged-loan borrower had revenue of $902 million. The interquartile ranges also point in different directions: $35 million to $700 million for PC borrowers and $264 million to $2.747 billion for LL borrowers. Those ranges overlap, but the typical borrower in the syndicated market is substantially larger.
The difference is not only scale. Median debt-to-EBITDA was 5.0x for PC borrowers and 3.2x for LL borrowers. Average interest coverage—EBITDA divided by cash interest—was 2.2x for PC borrowers versus 3.7x for LL borrowers. A firm with more debt relative to operating earnings and less room to cover interest may value a lender willing to tailor the structure around its cash generation. That same profile can be difficult to distribute to a broad investor base, particularly when the borrower has limited public information or limited collateral.
Loan size reinforces the distinction. The median PC loan in the table was $20 million, compared with $275 million for an LL loan. A smaller loan can fit a bilateral relationship in which a lender is prepared to do detailed underwriting and hold the exposure. A larger syndicated loan can spread risk across many investors, but it must be packaged in a form those investors can evaluate and buy. The two loan sizes are not just different price points; they are evidence of different financing interfaces.
Credit information is another dividing line. Most PC borrowers lack public credit ratings. The Federal Reserve therefore reports estimated credit categories for PC borrowers, while the LL figures use public ratings of loans. Among LL borrowers, 28% were rated BB or above, 63% single-B and 9% CCC or below. The estimated PC distribution was 0% BB or above, 85% single-B and 15% CCC or below. These categories are not perfectly comparable because the methods differ, but they indicate why PC can be important for borrowers that are riskier, less transparent or harder to classify in a standardized market.
The two markets still finance similar corporate purposes. LBO and M&A accounted for 38% of LL issuance and 32% of PC issuance in the Federal Reserve table. General-purpose or refinancing loans were 52% of LL issuance and 63% of PC issuance. Dividend recapitalizations represented smaller shares in both. Both markets therefore meet borrowers at familiar points in the corporate-finance cycle. The difference is not that PC funds one kind of activity and LL funds an unrelated kind. The difference is which borrowers can bring that activity to each market on acceptable terms.
Maturity is an example of a similarity that can conceal a larger difference: the median maturity was five years in both markets. A borrower may therefore see two loans with a similar term and assume the financing is portable. But the five-year claim is supported by different underwriting, distribution and monitoring arrangements. The price also differs in the sample: the median spread over SOFR was 500 basis points for PC and 400 for LL. That comparison is informative but not a universal ranking, because the borrowers are not the same and the spread compensates for different risks and structures.
The numbers should be read as a pattern of selection, not as a two-column scorecard. Lower interest coverage can make a PC loan more valuable to a borrower because a relationship lender may spend more time understanding the company’s cash flows and negotiating protections around them. It can also make the credit less attractive to a dispersed investor base that has less ability to investigate the borrower. A higher PC spread may compensate the lender for risk, illiquidity and monitoring effort, but it may also reflect the fact that the PC sample contains smaller or more leveraged firms. Comparing spreads without comparing borrowers would mistake a market price for a pure measure of lender efficiency.
The same caution applies to ratings. Public ratings are designed to travel across investors, whereas an internal or third-party estimate can incorporate information available to a direct lender but not to the market as a whole. The Federal Reserve notes that most PC borrowers lack public ratings and that the PC categories are estimated for borrowers with estimates less than one year old. This does not make the estimates useless. It means that “single-B” in the two columns should be treated as an indicator of relative credit distribution, not as a perfectly matched experiment.
Transaction purpose matters in a similar way. A refinancing may be straightforward for a lender that already knows the borrower, while the same refinancing may require a new investor audience to reassess the credit from the beginning. An acquisition may justify a larger syndicated facility for one company but require a more tailored structure for another. The fact that both markets fund LBOs, M&A and general corporate purposes establishes potential overlap. It does not establish that a borrower can move without changing the size, documentation, pricing or speed of the transaction.
How to test whether a borrower has an exit
A useful first test is historical access. The Federal Reserve classifies a firm as PC-only or LL-only when it has borrowed exclusively from that market up to the relevant date. Once it has switched from one market to the other, it is classified as having accessed both markets. The classification covers firms that issued debt between the beginning of 2020 and the end of 2025. It describes borrowing history, not a permanent corporate identity, a legal eligibility category or the amount of credit each firm received.
Across the full sample, 38.9% of borrowers were PC-only, 29.0% LL-only and 32.1% had accessed both markets. The last category is the most important for the switching question, but it should not be read as proof that every firm in it can switch at any moment. A previous loan shows that an entry was possible under some earlier conditions. It does not guarantee that the same lender, investor demand, leverage capacity or transaction size will be available at the next refinancing.
The size split makes that qualification central. Among larger firms, 40.9% had accessed both markets, 43.2% were LL-only and 15.9% PC-only. Among smaller firms, only 25.3% had accessed both, while 56.9% were PC-only and 17.9% LL-only. The Federal Reserve uses revenue above $250 million to classify a firm as larger and below that threshold as smaller. The cutoff is an analytical definition, not a universal market rule. Its importance is that it separates groups with different observed histories of access.
Why might size matter? A larger company is more likely to support a loan large enough to attract a syndicated investor base, to maintain formal reporting and financial controls, and to have sponsor, bank or adviser relationships that make a market transition possible. It may also have a more diversified business and a longer record that investors can compare with other credits. A smaller company may have strong local knowledge or reliable cash generation, but still lack the reporting scale, transaction size or distribution relationships that the LL market requires. Size is therefore best treated as a bundle of capabilities, not as a standalone cause.
The reverse mistake is also possible: assuming that PC access proves LL ineligibility. Private credit has expanded into larger borrowers, and some companies use both markets over time. A borrower may choose PC because it needs speed, a bespoke covenant or a lender willing to understand an unusual business, even when it could technically approach an LL arranger. “PC-only” in the Federal Reserve’s historical classification means no observed LL borrowing in the specified history; it does not reveal whether the firm was rejected, never applied, preferred PC, or faced a different transaction need.
An analyst judging the outside option should therefore ask several questions in sequence. First, has the firm actually borrowed in both markets, or is the alternative merely assumed because the industry appears in both? Second, does the proposed transaction resemble the firm’s prior borrowing in size, purpose, leverage and maturity? Third, is the borrower’s financial information suitable for a broad investor audience, or does it depend on the deeper context of a relationship lender? Fourth, are banks willing to underwrite and investors willing to buy comparable LL risk at the time the financing is needed? Finally, would the firm accept the costs of switching—possibly less customization, a different reporting burden or a higher risk premium in a stressed market?
Each question filters a different kind of false confidence. Industry overlap tests whether the borrower is in the same broad universe, but it says little about underwriting access. Prior borrowing history is stronger evidence because it shows that a market entry occurred, yet it can age: a prior transaction may have been smaller, supported by a different sponsor, or completed under easier conditions. Loan size and purpose test whether the proposed financing can fit the distribution channel rather than merely whether the firm has ever borrowed. Leverage and coverage indicate how much room the company has to absorb a higher spread or stricter covenant package, but they do not reveal all of the information a lender will use.
The market-side questions are just as important. A firm can be well prepared for a syndicated loan and still arrive when banks are unwilling to warehouse the risk or institutional investors are reducing exposure to leveraged loans. Conversely, a PC fund may have a strong relationship with the borrower but lack fresh capital or borrowing capacity. The access test must therefore be performed at the time of the financing need. It is not enough to mark the firm “both markets” based on an old loan and stop there.
Suppose, as an analytical example, that a smaller PC borrower has a five-year loan coming due. Its industry may appear in LL issuance statistics, and its prior lender may have been willing to negotiate bespoke covenants. That does not establish a syndicated exit. The proposed refinancing would need to be large enough and documented clearly enough for broad distribution, while the borrower’s leverage and interest coverage would need to be acceptable to investors at the prevailing risk premium. If any of those conditions fail, the borrower may have to seek another PC lender, renegotiate with existing creditors, reduce the financing need or accept a more expensive structure. Those are possible responses, not outcomes demonstrated for every PC-only firm.
These questions do not reproduce an underwriter’s decision. They prevent an analyst from using the existence of a large LL market as evidence that a particular PC borrower can enter it. The relevant conclusion is narrower: a borrower with a demonstrated both-market history and the scale and information to repeat that transaction has a more credible exit than a PC-only borrower. Even then, the alternative remains conditional on market capacity and approval.
The distinction between a historical profile and a guaranteed option is essential. The percentages describe firms that issued debt, not all firms that wanted debt, and they describe borrower shares, not loan-dollar shares. A PC-only firm may be especially exposed to a tightening, but the data do not say that every such firm will lose financing or that every both-market firm will be protected. They show where the outside option has been more common—and where the evidence gives less reason to assume it.