The Gating of
Private Credit
Perpetual BDCs record their first-ever negative net flows in Q1 2026, as AI-driven fears over software collateral send redemptions past inflows — and managers scramble to shut the gates.
— PIMCO's Lotfi Karoui, 2026
Periodic gating continues · 2021-vintage write-downs not finished · SDNY inquiry unresolved · Perpetual BDC retail wrapper contracts · Institutional direct lending functions
Software earnings show resilience · SDNY resolves without charges · Fed holds rates · Inflows stabilize · Sector bifurcates between disciplined and stretched
Bank credit lines to top-10 BDCs = 25% of NBFI commitments · If banks withdraw, forced asset sales pressure sector NAVs · Fed 3-quarter runway evaporates
Q1 2026: The First Quarter
That Ran Backwards
Investors pulled $6.9 billion from perpetual BDCs in Q1 2026. New subscriptions came in at $4.9 billion. The gap — −$2.0B — marks the first net outflow on record for the approximately 23 non-traded BDC vehicles tracked by RA Stanger. Year-on-year, inflows fell ~59% while redemptions surged 429%.
Aggregate redemptions in the Cliffwater perpetual BDC index climbed from ~1.6% of NAV in Q3 2025 to ~4.8% in Q4 2025, approaching the 5% quarterly cap before the first gate was invoked. The Federal Reserve's May 2026 Financial Stability Report confirmed the figures. Fitch and iCapital data align.
Perpetual BDCs hold ~$306B in gross assets and $161B net assets. Together with interval funds ($119B gross), semi-liquid private credit vehicles now represent approximately 20% of all private credit AUM in the United States. The structures were designed for steady-state conditions. Q1 2026 was not steady-state.
Four Managers.
Four Responses.
One Signal.
BlackRock's HLEND ($26B vehicle) activated gates for the first time in the fund's history. BlackRock paid out exactly the 5% cap — $620 million — while holding the remaining $580M of a $1.2B request queue. New subscriptions of ~$840M were not enough to fill the gap.
Blackstone's BCRED ($82B vehicle) saw redemption requests of roughly 8% of assets in a single quarter — well above the 5% cap. Rather than simply gate, Blackstone raised the threshold to 7% and injected approximately $400M of corporate and employee capital to satisfy the backlog.
Blue Owl halted regular redemptions at one retail vehicle entirely, sold ~$1.4B in loan assets, and executed a 15.4% fund buyback in January to stabilize NAV. Four vehicles met less than 50% of Q1 redemption requests: Blue Owl Credit Income Corp, Blue Owl Technology Income Corp, Ares Strategic Income Fund, and Apollo Debt Solutions BDC.
Software Became Private Credit's
Default Destination
SaaS lending grew from approximately $8 billion in 2015 to more than $500 billion by end-2025, representing roughly 19% of total direct loans according to the BIS. TCPC carries approximately 30.5% of its portfolio in software loans. HLEND sits at roughly 19%.
The AI disruption narrative is concrete, not theoretical. Generative AI tools now replicate core functionality of legacy SaaS platforms at a fraction of the cost. Software equities sold off ~30% from October 2025 to February 2026 (BIS). BDC equities fell ~10% on average, with heavy SaaS exposure underperforming peers by approximately 5 percentage points.
In Q1 2026, TCPC recorded $11 million in software portfolio markdowns specifically attributable to AI disruption risk — a realized figure from the fund's own filings, not a stress-test estimate. KBRA has characterized the credit risk as "diffuse and manageable," but that finding does not address the valuation lag. Deterioration accumulates for quarters before appearing in marks.
The Architecture
of Deferred Recognition
In January 2026, TCPC issued an off-cycle disclosure: NAV per share was expected to fall from $8.71 to $7.05–$7.09 — a ~19% markdown in a single quarter. The official Q4 figure came in at $7.07, down $1.64. Total damage: $73.9M in realized losses plus $66.5M unrealized. Six portfolio companies drove approximately 67% of the quarter's NAV decline.
91% of the quarter's overall NAV reduction came from investments underwritten in 2021 or earlier. TCPC shares fell 13% on January 26 — the worst single session since March 2020. The board cut the quarterly dividend 32%, from $0.25 to $0.17 per share. Fitch downgraded to BB from BB+ on January 30. As of Q1 2026, TCPC's NAV stands at $6.72 per share, down 27% in twelve months.
PIMCO's Lotfi Karoui has noted that marks for the same loan held across different BDC portfolios diverged by approximately 5 points on average by year-end 2025. Click the entity boxes at left to explore the structural conflict of interest embedded in the valuation chain. The Southern District of New York is now seeking information about TCPC's valuation practices (Bloomberg, May 15, 2026).
The quarterly mark-to-model system creates a one-way ratchet: fees accrue when the mark is high, losses crystallize when it corrects. The investor bears both the timing and the magnitude risk. No contractual mechanism exists to claw back fees paid on value that proved illusory. This is not a TCPC-specific finding — it is sector architecture.
Not All Vehicles
Are Created Equal
The listed BDC discount tells one story. FSK (KKR) trades at 0.51x NAV on $12.3B of net assets, with non-accruals at 3.4% of fair value. KKR has deployed a stabilization package — $150M preferred equity, $150M tender offer, and a four-quarter incentive fee waiver. TCPC (BlackRock) trades at 0.55x NAV, faces a DOJ probe, and carries a BB/Watch Negative rating from Fitch.
MFIC (Apollo) at 0.76x looks healthier on price-to-NAV, but management exit risk is non-trivial — Apollo has been exploring options for the management contract. The sector average for listed BDCs sits at ~0.74x. For context, the historical trading range for well-managed BDCs is 0.85–0.90x.
Toggle to the non-listed view to see BCRED and HLEND through the lens of redemption pressure rather than price-to-NAV. Both exceeded the 5% quarterly gate in Q1 2026. The two metrics measure different failure modes — but both converge on the same underlying diagnosis: the semi-liquid wrapper is under stress across the entire product category.
Three Scenarios for
Private Credit's Next Chapter
The base case — assigned a ~45% probability — is protracted trust repair through 2026–27. Periodic gating episodes continue. NAV discounts do not recover quickly as retail advisors re-examine allocations. The 2021-vintage write-down cycle is not finished. The SDNY inquiry extends without resolution. The perpetual BDC wrapper as a retail distribution mechanism undergoes sustained contraction, even as institutional direct lending continues to function.
The optimistic scenario (~35%) requires AI fears to cool faster than expected as software earnings data shows resilience. It also requires the SDNY inquiry to resolve without charges and the Federal Reserve to hold rates flat. Inflows stabilize at lower levels and the sector bifurcates between disciplined operators and stretched ones — a healthy outcome for the asset class, even if painful for weaker managers.
The tail risk (~20%) centers on bank revolver tightening. The ten largest perpetual BDCs control approximately 80% of sector assets, supported by bank credit lines that represent roughly 25% of all bank loan commitments to non-bank financial intermediaries. The Fed's three-quarter liquidity runway estimate holds only if those lines remain open. If banks tighten, forced asset sales at scale would pressure NAVs sector-wide.
"The product was sold beyond its design limits. Q1 2026 is when the warranty voided." The perpetual BDC structure can survive this stress cycle — but not without structural reform to the redemption architecture, fee alignment, and valuation governance. Absent those reforms, Q1 2026 is not an anomaly. It is a preview.