Managers returned 38 cents of every requested dollar in the second quarter. The sharper story is dispersion: Goldman's and Fidelity's funds honored every request while the biggest vehicles queued theirs.

What happened. Investors asked semi-liquid private credit funds for $15.6 billion back in the second quarter of 2026, up from $13.9 billion in the first quarter, according to Robert A. Stanger data first reported by the Wall Street Journal. Managers returned $5.9 billion, down from $7.4 billion the prior quarter, leaving roughly $9.7 billion of requests unmet. New fundraising fell to about $500 million in May, the smallest monthly inflow in at least 18 months.¹

Who is gating and who is not. Blackstone's BCRED honored all first-quarter requests, then capped second-quarter payouts at its 5% quarterly limit.¹ Blue Owl received about $4.7 billion of requests across two funds, 18.8% of shares at its credit income fund and 38.1% at its technology income fund, both limited to 5%.³² Apollo curbed withdrawals after requests near 17%.⁴ At the other end, Goldman Sachs Private Credit Corp. received requests of just 3.24% of shares and fulfilled them in full, and Fidelity Private Credit Fund fulfilled all requests at roughly 2.9% while posting net inflows.⁵⁶

Why an alt-lending desk should care, and the limit of it. These funds lend to corporate borrowers, not to the merchants an MCA, factoring or equipment finance shop serves. The bridge is the capital stack: semi-liquid credit vehicles and their institutional cousins are part of the money that buys non-bank lending paper, funds leverage facilities and sets the risk appetite of every layer below. The Financial Stability Board sizes private credit at $1.5 trillion to $2 trillion globally and warns that redemption features wrapped around illiquid loans are exactly where stress concentrates.⁷

The part to keep in view. A gate is not a default. Every fund named here is operating inside repurchase terms investors signed. The open question is why investors want out, and the reported answer includes doubt about marks on software borrowers whose revenue models are being reshaped by AI.¹ Until marks are tested by exits, the queue itself is the market's opinion.

_{Sources
1 investingLive | Investors Demand $15.6 Billion From Private Credit, Get Back Just $5.9 Billion (Stanger data via WSJ)
2 Seoul Economic Daily | Blue Owl Private Credit Funds Face Q2 Redemptions, Cap Withdrawals at 5%
3 Benzinga | Blue Owl Records $4.7 Billion in Redemption Requests
4 CNBC | Apollo Curbs Private Credit Fund Withdrawals Amid 17% Redemption Wave
5 SEC EDGAR | Goldman Sachs Private Credit Corp. Q2 2026 Tender Offer Update
6 SEC EDGAR | Fidelity Private Credit Fund Q2 2026 Repurchase Notice
7 Financial Stability Board | Leverage in Non-Bank Financial Intermediation: Private Credit
8 Fitch Ratings | US Private Credit Default Rate Remains at Record High 6.0% in May 2026
9 Proskauer | Private Credit Default Index Reveals Rate of 2.73% for Q1 2026
10 Houlihan Lokey | Private Performing Credit Index, Q1 2026
11 Yahoo Finance (Bloomberg) | BlackRock's Tseng to Exit as CEO of Troubled Private Credit Fund
12 citybiz | BlackRock TCP Capital CEO Phil Tseng to Step Down
13 ABL Advisor | Non-Bank Lender Confidence Climbs to Three-Year High Amid Shifting Market Dynamics, SFNet
14 New York Attorney General | AG James Sues Arbitration Service Tied to Merchant Cash Advance Industry
15 ICLG | State of New York Sues Arbitration Provider}

Two numbers moved in opposite directions in the second quarter, and the spread between them is the story. Requests rose from $13.9 billion to $15.6 billion while payouts fell from $7.4 billion to $5.9 billion.¹ A 5% quarterly repurchase cap works as designed when requests run near 5%. When requests run at 10%, 17% or 38% of shares, the cap converts an exit into a multi-quarter queue, and each quarter of pro rata fulfillment rolls unmet demand into the next window.

The inflow side is the quieter half of the squeeze. May's roughly $500 million of new fundraising was the smallest monthly figure in at least 18 months.¹ Semi-liquid vehicles were built on the assumption that new subscriptions would help fund old exits. When inflows fade at the same time exit requests build, managers must either sell loans into a soft market or hold the gate down longer. The reported choice so far is the gate.

The same quarter that queued billions at the biggest funds was routine at others, and the primary documents show it. Goldman Sachs Private Credit Corp. told shareholders its requests were 3.24% of shares, below its cap for the second consecutive quarter, and that peer requests at the largest non-traded BDC managers ranged from roughly 10% to nearly 17%, with multiple peers hitting their 5% caps and paying pro rata.⁵ Fidelity's fund fulfilled every request at about 2.9% of shares and still took in net new capital.⁶ Oaktree's fund reportedly saw requests fall to 4.5% of shares from 8.5%.¹

Blue Owl's own numbers show the other side. Requests hit 18.8% of shares at its credit income fund and 38.1% at its technology income fund, a combined $4.7 billion against 5% caps, even as the firm noted most requests came from earlier applicants re-entering the queue and that the credit income fund could meet its tender without selling a single private loan.³ Vintage, borrower mix and inflow engines separate the two groups. Funds that grew fastest in the cheap-money window hold the paper investors most want to hand back.

None of this queue data says private credit is defaulting en masse, and the default series should not be blended to imply it. Fitch's US private credit default rate held at a record 6.0% for the 12 months ended May 2026.⁸ Proskauer's senior-secured and unitranche index, a different book measured a different way, showed 2.73% for the first quarter of 2026.⁹ Pricing has not blown out either: Houlihan Lokey's private performing credit index showed a 9.69% weighted average yield in the first quarter.¹⁰

What the queue prices is doubt, not loss. The reported concern behind the exit wave is concentrated software and technology exposure, where AI is rewriting borrower revenue models faster than illiquid marks move.¹ Blue Owl's technology fund drawing a 38.1% request rate against its diversified sibling's 18.8% is that doubt expressed as an allocation decision.³ Investors are not waiting for the marks to answer the question; they are getting in line first.

Write down your capital behavior before it is tested. If your originations depend on a warehouse line, a forward-flow buyer or a fund investor whose own investors can queue for the exits, map what happens to your funding cost and advance rates if that provider spends four quarters honoring redemptions pro rata. The lenders still originating through the 2022 pullback were the ones who knew their capital's behavior in advance.

Watch the funding layer, not just the borrower. The metrics to track are public and quarterly: repurchase request percentages in fund tender filings, inflow-to-repurchase ratios, and whether your capital partners are net raisers or net payers. Goldman put its inflow-to-repurchase ratio at 0.9x, nearly three times the peer average, in an SEC filing anyone can read.⁵ That is the level of disclosure to demand from any capital relationship you depend on.

Expect the appetite shift before the headline. A gated fund does not call its downstream partners to announce reduced risk appetite; it just gets slower, pickier and more expensive. If private credit spends the second half of 2026 rationing exits, the practical effect downstream is tighter advance rates and longer diligence on every non-bank lending relationship that touches institutional capital.

An active state record proves a business exists, but not that the submitted EIN belongs to it, and that gap sits right before payee setup, ACH, and wire release.

Cobalt's TIN/EIN Verification API validates the name and tax ID against IRS records before funds move, returning an IRS code your treasury policy can route.

Schedule a FREE demo call

Bloomberg reports that Phil Tseng, CEO of BlackRock TCP Capital, is preparing to leave the firm; the timing is not final. The BDC marked down asset values 19% in January and 5% in May, and prosecutors at the Manhattan US Attorney's office have questioned executives as part of a probe into the fund's valuation practices.¹¹¹² BlackRock declined to comment, as did Tseng through a spokesperson, and no allegation of individual wrongdoing has been made public. The operator signal: valuation credibility is becoming a supervisory question, not just an investor one.

While the fund layer rations exits, the Secured Finance Network's Q1 2026 survey shows non-bank asset-based lenders posting their most confident reading in more than three years as borrowers draw harder on facilities and banks stay selective.¹³ Both stories can be true at once: secured lenders with collateral discipline and durable funding are gaining share precisely because capital upstream is getting choosier.

Attorney General Letitia James filed suit against Rapid Ruling and its founders, alleging the arbitration platform presented itself as neutral while a merchant cash advance company drafted its rules, and that 97% of its roughly 3,000 arbitrations concluded without any appearance by the small business involved.¹⁴¹⁵ These are allegations in a civil complaint, not findings. The operator signal for every MCA and revenue-based funder: collection and dispute infrastructure is now squarely inside the enforcement perimeter, and contracts that lean on it are being read end to end.