What moved in the second quarter?
The list grew and the book shrank. Across the ten largest publicly traded business development companies, debt on non-accrual rose to 3.95 percent of total debt at cost at the end of June 2026, from 3.75 percent three months earlier, according to PitchBook LCD's analysis published on 19 August 2026.
The dollar movement was small. The non-accrual balance rose by 89 million dollars to 3.3 billion. The ratio moved more than the balance because the denominator fell: total debt at cost across the ten contracted by 2.3 percent in the quarter, to 83.6 billion dollars. Repayments and sales are running ahead of new lending, so the same troubled dollars occupy a larger share of a smaller book.
The count of names moved more than either. The number of distinct borrowers with at least one tranche on non-accrual at those ten lenders rose by eleven in the quarter, to 101. That is the figure a borrower should hold onto, because a lender manages a watch list by name, not by tranche.
PitchBook LCD then re-cuts the same books by borrower rather than by loan. Counting every tranche owed by any borrower that has at least one tranche on non-accrual, performing or not, the exposure reaches 5.0 billion dollars, or 5.95 percent of total debt, up 54 basis points on the quarter. The reported figure and the borrower-level figure are two percentage points apart on the same portfolio.
| Measure | Ten largest public BDCs, Q2 2026 | All 213 BDCs, Q1 2026 | Change on prior quarter |
|---|---|---|---|
| Reported non-accruals, share of debt at cost | 3.95% | 1.9% | +20 bps / +52 bps |
| Reported non-accrual balance | $3.3bn | ~$10bn | +$89m / not stated |
| Borrower-level non-accruals, share of debt at cost | 5.95% | 3.3% | +54 bps / +116 bps |
| Borrower-level non-accrual balance | $5.0bn | $17.3bn | not stated |
| Distinct borrowers with a tranche on non-accrual | 101 | 4.69% of all borrowers | +11 / +43 bps on the year |
| Total debt at cost in sample | $83.6bn | $516bn | -2.3% / not stated |
Why are there two numbers for the same loans?
Because non-accrual is an accounting decision taken tranche by tranche, and a borrower is a single credit. A lender stops accruing interest on a specific instrument when it no longer expects to collect it. A second lien held by the same fund, or a first lien held by a different fund, can stay on accrual for a quarter or two after the first instrument goes.
PitchBook LCD's adjusted view treats a borrower's entire debt as non-accrual whenever at least one business development company reports at least one of that borrower's instruments in non-accrual status. It is a blunter test, and it is closer to how a credit committee behaves. Once one lender has stopped accruing on a name, every lender to that name knows it, and the performing tranches are performing on borrowed time.
The gap between the two measures is therefore a measure of lag rather than of disagreement. At the ten largest lenders the reported rate was 3.95 percent and the borrower-level rate 5.95 percent at the end of June. The two percentage points between them are loans that are still paying, to companies that have already stopped paying somebody else.
Part of what sits in that gap is paid in kind. PitchBook LCD notes that much of the debt owed by non-accrual borrowers is payment-in-kind and so does not appear in cash interest at all. A loan that capitalizes its coupon cannot miss a cash payment, which keeps it off the non-accrual line for longer than the underlying business would justify.
“Lenders do not carry a watch list of tranches. They carry a list of companies, and a company is added to it the day any one of its lenders stops accruing. The performing loans to that company keep paying for a while, which is why the reported rate lags. What it does not do is protect that borrower from being treated as a workout at the next amendment.”
Is this a big-fund problem or a market problem?
A market problem, on the wider sample, though with a lower level and a faster rate of change. PitchBook LCD's full universe covers 213 business development companies run by 109 managers with an aggregate debt portfolio of 516 billion dollars, and the latest quarter for which all of them have filed is the first quarter of 2026.
Across that universe, reported non-accruals stood at 1.9 percent of debt at cost at the end of March, up 52 basis points in a single quarter, or roughly 10 billion dollars. The borrower-level view was 3.3 percent, up 116 basis points on the quarter, or 17.3 billion dollars. The wide sample moved further in one quarter than the ten largest funds moved in the next.
Measured by names, 4.69 percent of all borrowers across the universe had at least one tranche on non-accrual at the end of March 2026, against 4.26 percent a year earlier and 3.69 percent in the first quarter of 2023. That share has risen in every year of the series.
Two further figures give the wide sample its weight. The business development companies in aggregate carry 772 million dollars of interest that PitchBook LCD describes as at risk of imminent default, being 204 basis points of total cash interest. And the tranches already on non-accrual were carried at 55.8 percent of cost at the end of March, down 144 basis points on the quarter. The loans that have already failed are being marked lower, not stabilizing.
What should a borrower with more than one lender take from it?
That the unit of judgement is the company, not the facility. A borrower with a term loan from one fund and a delayed draw or a second lien from another should assume that a non-accrual decision by either lender is known to both within the quarter, and that the performing facility is reviewed on that basis at its next test date.
It changes the arithmetic of a partial problem. An owner who is current on the senior loan and negotiating a deferral on the junior one is, in the borrower-level view, already a non-accrual name. The senior lender will price the next amendment accordingly, and the room to argue that the senior facility is a separate matter is narrower than it looks from the borrower's side of the table.
It also changes how payment-in-kind should be read. Capitalizing a coupon relieves cash and keeps the loan off the non-accrual line, which is exactly why lenders are cautious about what it signals. A borrower proposing a switch to payment in kind should expect the lender to treat it as the first step on a list rather than as a technical accommodation, and should bring the evidence that it is not: a covenant model, a dated recovery plan, and a lender-ready view of the cash conversion cycle.
The books are shrinking at the same time as the lists are growing, and that combination sets the tone of the next negotiation. A lender with a contracting portfolio and a rising watch list defends yield on what it keeps. The borrower who arrives with clean quarterly reporting, no surprises between test dates, and a sponsor or owner visibly funding the plan is the one that stays in the performing column while the rest of the list is worked.
What would change the reading?
A quarter in which the borrower count falls while the tranche rate is flat. That would mean names are leaving the list through repayment, sale or restructuring faster than they are joining it, and it is the only combination that would show the lag being worked off rather than building.
The valuation of non-accrual tranches is the second thing to watch. A carrying value at 55.8 percent of cost and falling says that lenders expect to recover less on the loans that have already failed, which tends to make them less patient with the ones that have not.
The third is the shape of the books themselves. The ten largest funds shrank by 2.3 percent in a quarter. If that continues, the same non-accrual balance produces a rising ratio each quarter without a single new default, and a borrower reading only the headline rate will conclude that credit is deteriorating faster than it is. Read the balance and the count alongside the ratio.
PitchBook LCD closes its analysis with the open question of whether the rest of 2026 brings heightened credit risk for these lenders. On the evidence in the piece, the risk that has already arrived is the gap between what is reported and what is known, and that gap is the borrower's problem before it is the lender's.
As of August 2026
Sources: PitchBook LCD, Rising non-accruals signal growing risk in private credit, published 19 August 2026 and read through its syndicated copy on Yahoo Finance, for the universe of 213 business development companies managed by 109 managers with an aggregate debt portfolio of 516 billion dollars at the first quarter of 2026; for reported non-accrual debt of 1.9 percent of total debt at cost at the first quarter of 2026, up 52 basis points from the fourth quarter of 2025, being roughly 10 billion dollars; for the adjusted, borrower-level view of 3.3 percent of total debt at cost, up 116 basis points on the quarter, being 17.3 billion dollars, and for its definition, under which a borrower's entire debt is treated as non-accrual whenever at least one business development company reports at least one of that borrower's instruments in non-accrual status; for 4.69 percent of all borrowers having at least one tranche on non-accrual at the first quarter of 2026 against 4.26 percent a year earlier and 3.69 percent in the first quarter of 2023; for 772 million dollars of interest at risk of imminent default, being 204 basis points of total cash interest; for the valuation of non-accrual tranches at 55.8 percent of cost, down 144 basis points from the fourth quarter of 2025; for the observation that much of the debt owed by non-accrual borrowers is payment-in-kind and does not appear in cash interest income; and, for the ten largest publicly traded business development companies at the second quarter of 2026, for non-accrual debt of 3.95 percent of total debt at cost, up 20 basis points on the quarter, a balance up 89 million dollars to 3.3 billion dollars, a 2.3 percent contraction in the total debt portfolio to 83.6 billion dollars at cost, 101 distinct borrowers with at least one tranche on non-accrual, up eleven on the quarter, and borrower-level exposure of 5.0 billion dollars or 5.95 percent of total debt, up 54 basis points. The 3.75 percent prior-quarter figure and the two percentage point gap between the reported and borrower-level measures are our own arithmetic from the published figures. A companion article on this site covers the share-price discount at which listed credit funds trade, another covers one large fund's falling net asset value with no new non-accruals, and a third covers the leveraged loan default and distress ladder.