What is the asymmetry?
The lender's best outcome is that it gets its money back plus the coupon. Nothing about a business tripling in value improves that outcome by a single basis point. Its worst outcome is that it loses the principal. So the whole distribution of possible results, for the party writing the largest cheque in most transactions, runs from a modest fixed gain to a total loss.
Put current numbers on it. A senior lender to a borrower below $10m of EBITDA earns 7.15% to 7.90% a year if everything goes right. A junior lender in the same structure earns 13.00% to 16.00% (SPP Capital Partners, Market At A Glance, July 2026). Those are the ceilings. There is no version of the transaction in which either of them does better.
Equity is the mirror image, which is why the two parties in the same meeting appear to be discussing different companies. An equity investor's downside is also the whole investment, but its upside is unbounded, and a portfolio is built on the arithmetic that a small number of large outcomes pay for a larger number of write-offs. That model requires the investor to spend its diligence on how large the good case can get.
A lender cannot run that arithmetic, because it has no large outcomes to fund the losses with. It has to be right almost every time. At a coupon of roughly 9%, a single total loss consumes the annual return on more than ten performing loans of the same size, before any consideration of what the fund promised its own investors. That is the entire explanation for behaviour that borrowers routinely read as pessimism about their business.
“Owners hear the downside questions as scepticism about the plan. They are not about the plan. A lender who believed every word of the forecast would still ask them, because the forecast is not what it is being paid to have a view on.”
What does that do to the diligence?
It reorders it. An equity diligence process is built around what could go right and how large it could get: market size, pricing power, the credibility of the growth plan, what a strategic acquirer might pay in five years. A credit diligence process is built around what has to keep happening for the interest to be paid.
The questions follow from that. How much of the revenue recurs, contractually, without a new sales decision. Which customers could leave, how quickly, and what happens to fixed cost if they do. Which suppliers are single-sourced. How much of the EBITDA in the model is an adjustment rather than a receipt. What the working capital cycle does in a bad quarter. What the business looks like at 80% of the base case, which is the number that actually gets modelled inside the credit committee.
It also explains the apparent obsession with the trailing period. A lender underwrites last twelve months EBITDA because it is the only figure that has already happened, and it discounts forecast improvement not because it doubts management but because a forecast is a claim on the upside that the lender does not participate in. The improvement raises the equity value. It does not raise the coupon.
The corollary is worth stating for any owner about to sit through this. The lender is not deciding whether your business is good. It is deciding what it is worth in the state of the world where things did not work, and how quickly it could get there. Those are different questions, and answering the first one enthusiastically does not advance the second.
What does the loss experience actually look like?
It is contested, and both readings are current, which is itself informative about how thin the evidence base is. Restructurings observed at business development companies have recovered around 50 cents on the dollar (Octus, 11 May 2026), while the rated privately monitored cohort shows recoveries of 70% to 90% with minimal realized losses (Fitch Ratings, 6 March 2026). Those measure different populations and should never be averaged into a single number.
The broader loan market print sits between them. The twelve-month first-lien par-weighted emergence bid level was 74% in June 2026, up from 65% (Fitch and LSTA, July 2026). A recovery in the seventies on a first-lien claim is a good outcome by historical standards and still means a quarter of the principal did not come back.
The sensitivity of a credit fund to that is the number worth carrying out of this article. A stress test of a 10% cumulative default rate at 65% recovery would reduce unlevered fund returns from around 9.0% to around 8.3% (Lincoln International, published 7 May 2026). Seven tenths of a point. That is the whole buffer: the difference between a portfolio behaving and a portfolio in significant distress is most of the manager's excess return, which is why the underwriting is written the way it is.
The default rate the lender uses to run that test is itself unsettled. The same market currently prints roughly 1.5% by volume on a payment-default basis, 2.51% on a documentation basis, 3.1% on a covenant basis and 6.0% on a basis that counts maturity extensions, and the global regulator reaches the same conclusion independently, putting outright defaults near 1% and rising to around 5% once selective defaults are counted (KBRA, May 2026; Proskauer Private Credit Default Index, 28 July 2026; Lincoln International, as of 31 March 2026; Fitch Ratings, 30 July 2026; Financial Stability Board, 6 May 2026).
| Outcome | Reading | Source and as-of date |
|---|---|---|
| Best case, senior bank cash flow below $10m EBITDA | 7.15% to 7.90% a year | SPP Capital Partners, July 2026 |
| Best case, junior capital and mezzanine | 13.00% to 16.00% a year | SPP Capital Partners, July 2026 |
| Default rate, depending on definition | 1.5% to 6.0% | KBRA May 2026 to Fitch Q2 2026 |
| Recovery, restructurings observed at business development companies | About 50 cents | Octus, 11 May 2026 |
| Recovery, rated privately monitored cohort | 70% to 90% | Fitch Ratings, 6 March 2026 |
| First-lien emergence bid, twelve months to June 2026 | 74%, from 65% | Fitch and LSTA, July 2026 |
| Unlevered fund return at a 10% default rate and 65% recovery | About 8.3%, from about 9.0% | Lincoln International, 7 May 2026 |
Why do a lender and a sponsor value the same business differently?
Because they are underwriting different denominators as well as different outcomes. Reported leverage in private credit runs 5 to 6 times debt to EBITDA against roughly 4 times in leveraged loans, and stripping out EBITDA adjustments puts true leverage closer to 7 times (Financial Stability Board, Report on Vulnerabilities in Private Credit, 6 May 2026). A full turn of the difference between an owner's view and a lender's view can be created by definitions alone, before either party has expressed an opinion about the business.
The equity side is not being careless when it uses the adjusted figure. An adjustment for a cost action already taken, or for a genuinely non-recurring item, describes the business more accurately than the statutory number does. The problem is that an adjustment is a forecast wearing the clothes of a historical figure, and the party whose upside is capped is the party least willing to lend against a forecast.
The second source of disagreement is time. An equity investor holds for a period it chooses and can wait out a bad year. A lender has a maturity, a fund life and, increasingly, its own investors asking for capital back. Patience is not evenly distributed across the capital structure, and the tranche with the least of it is usually the one with a contractual date attached.
This is also why the same lender behaves differently toward a sponsored and a founder-owned credit. Sponsored borrowers are less likely to progress from delinquency to outright default, because a sponsor can inject liquidity (Financial Stability Board, 6 May 2026). The lender is not expressing a view on management quality. It is counting how many parties have a reason and a means to put more money in.
What should a borrower bring to the meeting?
The downside case, built by the borrower rather than by the credit committee. A modelled 20% revenue decline, the specific cost actions available at each stage, when they can be taken, what they cost to take, and where covenant headroom sits at each point. A borrower who supplies this is answering the only question in the room. A borrower who supplies a base case alone has left the committee to build the downside, and committees build them conservatively.
Then the quality of the earnings, stated plainly. Which portion of EBITDA is cash received, which portion is an adjustment, what each adjustment assumes and by when it must be delivered. Presenting adjustments openly, with their conditions attached, is treated very differently from having them found. Median middle-market interest coverage is 1.6 times across 2,785 borrowers and more than $1.2 trillion of debt (KBRA, twelve months ended 30 June 2026, published 28 July 2026), and a lender starting from that base has no room for surprises in the denominator.
Then the concentration facts, before they are asked for: customers, suppliers, key contracts, renewal dates, and the people whose departure would change the answer. Concentration is not automatically disqualifying. Concentration discovered in week six of a process usually is.
And finally, a view on what the business looks like at a higher base rate, since the forward curve currently prices three-month SOFR at 4.04% at the end of 2027 against 3.76% today (Blue Gamma and CME, 4 August 2026). A borrower who has already run that and knows which covenant binds first has removed the single most common reason a credit process stalls in its final week.
As of August 2026
Sources: SPP Capital Partners, Market At A Glance, July 2026, for senior and junior capital pricing by EBITDA size band; Octus, 11 May 2026, for recoveries on restructurings observed at business development companies; Fitch Ratings, 6 March 2026, for recoveries on the rated privately monitored cohort, and Fitch Ratings, 30 July 2026, for the default rate counting maturity extensions; Fitch and LSTA, July 2026, for the first-lien par-weighted emergence bid level; Lincoln International, published 7 May 2026, for the default and recovery stress test, and as of 31 March 2026 for the covenant default rate; KBRA, May 2026, for the payment-default measure, and Q2 2026 Middle Market Compendium, twelve months ended 30 June 2026, published 28 July 2026, for median interest coverage, borrower count and debt covered; Proskauer Private Credit Default Index, 28 July 2026, for the documentation default measure; Financial Stability Board, Report on Vulnerabilities in Private Credit, 6 May 2026, for reported and adjustment-stripped leverage, the outright and selective default framing and the sponsored delinquency asymmetry; Blue Gamma and CME, 4 August 2026, for the three-month SOFR forward curve and spot rate. The two recovery figures cover different populations and are presented separately rather than combined. A first-lien recovery percentage published by one agency for 2025 sits behind a paywall and is not quoted here in any form.