Private credit / BDC stress signals — the public-price vs private-mark gap

Built 2026-06-15. Structured data + edges: macro-private-credit-bdc-stress.json. Deep companion to macro-private-credit-marks and macro-cre-privatecredit; the third self-marked number in self_marked_value. (WebFetch unavailable; figures from search surfacing of the cited sources. Both directions are live — Moody's/KBRA argue marks may OVERstate stress — so the thesis is graded contested.)

The tells that the manager-set NAVs are too high: listed BDCs trade ~20% below NAV while non-traded BDCs hold marks near par; PIK income is ~12% and rising; defaults climb (Fitch 5.8%) even as reported non-accruals stay optically low (impaired loans get restructured with PIK, not marked); the same loan is marked 9 points apart by two lenders; and retail money is being gated in just as redemptions spike.

1. Discount-to-NAV — the market voting against the marks

• Listed BDCs trade at ~20% average discount to NAV (2025-26), some near ~50%: Blue Owl (OBDC) ~21% ($11.77 vs NAV $14.89, Sept 2025); FS Specialty ~27%; Barings ~23%; even Ares Capital (ARCC) slipped to a slight discount. Non-traded BDCs hold valuations near par — the core public-price/private-mark gap.
• Cross-lender gap: Medallia — HPS marked the loan 69¢ (Feb 2026); Blackstone marked the SAME credit 77.75¢ (Dec 2025) — a ~9-point gap on one loan. A recurring pattern: a senior loan deeply discounted while a second (often PIK) loan to the same borrower is held at par.

2. Managers & PIK

• Largest vehicles: BCRED (Blackstone) is the largest BDC of any kind — $82.2B investments / $47.6B NAV (Dec 2025), perpetual non-traded; ARCC (Ares) the largest listed (~$30.7B); OBDC (Blue Owl) #2; OCIC (Blue Owl) ~$36B; HLEND (HPS); GCRED (Golub ~$8.6B). Ares firm AUM $644.3B.
• PIK + defaults: PIK was ~11.7% of BDC loans (Q2 2024) and rising — a marks-quality red flag (BDCs must distribute 90% of income whether cash or PIK: "covered, not collected"). Fitch private-credit default rate 5.8% (Jan 2026); ~65% of 2025 defaults were distressed restructurings, 7 of 11 introduced PIK in lieu of cash. Reported non-accruals stay <~1.5% precisely because impaired loans get restructured with PIK rather than marked non-accrual.

3. The bank-to-NDFI chain

• Banks held ~$1.32T funded NDFI loans (FDIC, Q3 2025) + ~$987B unfunded commitments (~$2.3T); the segment grew ~50% in 2024-25 (vs ~4% for total bank loans). JPMorgan ~$50B, Wells Fargo ~$36-85B (to funds), Citi ~$22B. The Fed flags the shift toward credit LINES as raising liquidity-call interconnection. (The "$1.97T" figure sits between funded-only and funded-plus-commitments.)

4. Retail access & the gates firing

• EO 14330 (Aug 7 2025) eases PE/private-credit into 401(k)s (DOL rule advancing Apr 2026). The evergreen/interval universe is ~$431-500B — not the ~$644B sometimes cited (that's Ares firm AUM).
• Mismatch: evergreen funds cap redemptions at ~5%/quarter while a target-date fund trades daily. In Q1 2026 the gates fired: Blue Owl OTIC (tech, $6.2B) redemption requests 40.7%, OCIC ($36B) 21.9%, OBDC II 17% (→ wind-down); the firm filled only ~5%. Cliffwater CCLFX ($33B) ~14% (record). Private credit's first real liquidity test.

5. Stress events vs the bull case

• First Brands (Ch.11, Sept 2025): 15 BDCs hold ~$237M (~0.05% of ~$503B BDC AUM) — small direct hit, but UBS ~$500M / Jefferies ~$715M exposed via factoring/funds (macro-consumer-abs-subprime). Tricolor was bank-financed, little BDC exposure.
• The honest other side: Moody's (June 2026) and KBRA (Q1 2026) argue marks may OVERstate stress and non-accruals stay low; Lincoln is launching monthly valuation indices. The "marks are too high" thesis is contested, not fact.

Synthesis

Private credit is the third self-marked number (self_marked_value): the listed-BDC discount is the market pricing the same loans ~20% below where managers carry them; PIK + restructure-don't-mark keeps non-accruals low; retail money is gated in; and the exposure runs back to the GSIBs via ~$2.3T of loans+lines. Whether that is over- or under-stated stress is genuinely contested — but the gap between the public vote and the private mark is the signal the corpus has argued all along.

What is NOT asserted

• No claim private credit is in crisis — non-accruals low; Moody's/KBRA argue marks may overstate stress.
• The ~20% sector-average discount and the general cross-lender pattern are contested; named cases (OBDC, Medallia, Blue Owl gates) are fact.
• Bank NDFI exposure ~$1.32T funded / ~$2.3T with commitments (definition-dependent); the ~$644B "evergreen" is Ares firm AUM.
• Overlay edges are excluded from the proofs.

Sources: Octus — BDCs at discounts / non-accruals; Mercer — public prices, private marks; InvestmentNews — BDC discounts; Benzinga — PIK "covered not collected"; Funds Society — Fitch PCDR 5.8%; FDIC — bank lending to NDFIs; FA-Mag — banks tally $100B+ private-credit loans; White House — EO 14330; Dakota — evergreen markets; CNBC — Blue Owl redemptions; Octus — Medallia marks gap.

← Research index · structured data: macro-private-credit-bdc-stress.json · macro-private-credit-bdc-stress.md