Private Credit

Private Credit's Redemption Caps Are Working - and Moving the Liquidity Risk to Investors

The recent limits do not yet amount to a system-wide credit crisis. They reveal the bargain inside semi-liquid funds: the loan portfolio is protected by making investors wait, find another buyer or accept a discount.

Private Credit's Redemption Caps Are Working - and Moving the Liquidity Risk to Investors

Lara Sabri — Contributing Journalist  |  About the author →

Additional research and verification by UAO Editorial. Research cutoff: 24 July 2026, 09:13 UTC.


In the second quarter, investors sought $4.7 billion from two Blue Owl private-credit vehicles. They asked to redeem 10% of shares in Blackstone's flagship BCRED fund, 17% at Cliffwater's corporate lending fund and 16.8% at Apollo Debt Solutions. Each manager limited repurchases to about 5%, as its documents allowed.

The caps worked. They prevented funds holding hard-to-trade loans from selling assets simply because too many investors wanted cash at once.

They also exposed the other side of the contract. Some investors received only part of the money they requested. The fund avoided a liquidity problem by moving it to the investor.

A quarterly window is not a promise of cash

Private-credit funds hold loans negotiated directly with borrowers. Those loans do not trade like public bonds, and their values are not discovered continuously in a deep market.

To make such assets available beyond closed-end institutional partnerships, managers have built evergreen funds, interval funds and non-traded business development companies, or BDCs. These vehicles may accept new money continuously and offer quarterly repurchases. The usual limit is about 5% of outstanding shares or net assets.

That creates semi-liquidity: more access than a ten-year closed-end fund, but far less than a mutual fund or listed share. When requests exceed the limit, each investor receives a proportional share of the available cash and may have to apply again next quarter.

The sequence in 2026 shows how the mechanism operates.

  • Blackstone BCRED: Investors requested 7.9% of shares in the first quarter. Blackstone raised the normal limit to 7%, and the firm and employees supplied about $400 million so all requests could be met. In the second quarter, requests rose to 10% and BCRED returned to the 5% cap.
  • Cliffwater: Investors asked to redeem 17% of the $31.3 billion Corporate Lending Fund in the second quarter. The fund offered 5%.
  • Apollo: Investors requested about 16.8% of the $26 billion Apollo Debt Solutions fund, up from 11.2% in the previous quarter. The fund offered 5%.
  • Blue Owl: Two funds received $4.7 billion of second-quarter requests. The technology-focused OTIC vehicle received requests equal to 38.1% of shares; the larger OCIC fund received 18.8%. Both retained 5% limits.
  • BlackRock HPS: A $25 billion fund received first-quarter requests equal to 13.3% of assets and offered to repurchase 5%.

Blue Owl also permanently removed redemptions from a separate vehicle during a February restructuring. That event should not be conflated with the quarterly limits at its larger funds. The structures, investor groups and remedies were different.

Why a gate can be prudent

A manager that sells private loans quickly to satisfy every withdrawal can harm the investors who remain. Buyers may demand discounts, the best assets may be the easiest to sell, and those first in line may receive better treatment than those who wait. A cap is a circuit breaker against that first-mover advantage.

Apostolos Thomadakis of the Centre for European Policy Studies has argued that the central problem is not illiquidity itself but the appearance of greater access created by periodic windows. Timothy Chamberlain of Brunel Partners has made the counterpoint that limits can show the structure operating as designed rather than failing. Both propositions can be true.

The cap may protect the fund and still matter greatly to the investor. A pension plan, insurer, endowment or wealth platform may have included the quarterly window in its cash plan. If only part of the request is paid, it must find the balance in public securities, cash reserves or credit lines. Nicholas Lubin of Kade Technologies has described this as the liquidity need moving onto the allocator's own balance sheet.

These observations are paraphrased from attributed commentary supplied with the original article; they are not presented as new interviews conducted during this revision.

The price of an immediate exit is now visible

The secondary market offers another route. In July, tender offers for shares in several non-traded private-credit vehicles appeared at discounts of 15% to 30% to their May-end net asset values. One buyer offered Apollo fund investors 70 cents on the dollar, BlackRock HPS investors 75 cents and Ares investors 85 cents.

Those prices do not prove that the underlying loans have lost the same amount. The discount includes the price of immediacy, uncertainty, a small buyer pool and the difference between a modelled fund value and a cash bid today. It is nevertheless a market price for leaving the queue.

This is why credit risk and liquidity risk must remain separate. A portfolio can contain performing loans and still receive more repurchase requests than it was designed to meet. Conversely, a low level of defaults does not prove that investors can retrieve their cash at reported net asset value.

The pressure may be peaking - but the structure remains

There is counterevidence to a crisis narrative. Blue Owl's second-quarter requests eased from the first quarter, although they remained high. On 23 July, Blackstone said early third-quarter requests at BCRED had fallen materially. Several large vehicles continued to report positive returns and said they had enough liquidity to meet their tenders without forced loan sales.

That is important. The evidence at the research cutoff supports a period of elevated withdrawal pressure, not a claim that the entire private-credit market is insolvent or entering a generalized default cycle.

But an easing quarter would not remove the structural issue. The funds remain invested in assets that cannot be sold as quickly as their investors may ask to leave. The cap is the mechanism that reconciles those two facts.

Why universal owners should care

Large asset owners are exposed through more than fund units. Banks lend to private-credit vehicles and their portfolio companies. Insurers hold private loans and sometimes provide financing or capital to managers. The same borrower may appear in a private-credit allocation, a public-equity portfolio, a collateralized vehicle and an insurer's balance sheet.

Valuation timing can hide the overlap. Public equity reprices immediately. Private loans are marked less frequently and with model judgment. A borrower problem may hit the liquid book first while the private mark and reported portfolio weights catch up later.

The Financial Stability Board and the Federal Reserve have focused on these connections: leverage, bank funding, insurers, valuation and liquidity transformation. The systemic question is not whether one non-traded fund pays every investor on demand. It is whether several vehicles, borrowers and funding providers respond to the same shock by selling the same liquid assets or withdrawing the same lines.

The governance test

Investment committees should treat quarterly repurchases as conditional liquidity, not cash on hand.

  • Assume the vehicle pays only the contractual minimum during stress.
  • Model repeated proration over several quarters.
  • Track tender prices and listed BDC discounts as estimates of exit cost, while recognizing differences among vehicles.
  • Map overlapping borrower exposure across private debt, equity, insurers and bank counterparties.
  • Identify in advance which liquid assets fund unpaid withdrawals and capital calls.
  • Watch manager support, borrowing used to meet tenders, payment-in-kind interest, restructurings and changes in loan marks.

Private credit was sold on an honest premise: investors can earn more because the capital is harder to retrieve. The governance error is forgetting the second half of the bargain once a quarterly window appears on the factsheet. A cap can work exactly as designed and still be the moment an allocator discovers that its liquidity belonged to someone else.

Sources and methodology

Reuters: Blue Owl retained 5% limits after $4.7 billion of second-quarter requests

Reuters: BCRED received second-quarter requests equal to 10% of shares

Reuters: Cliffwater received requests equal to 17% of shares

Reuters: Apollo received requests equal to 16.8% of shares

Reuters: BlackRock HPS received first-quarter requests equal to 13.3% of assets

Reuters: private-credit tender discounts showed the cost of an immediate exit

Reuters: early third-quarter BCRED requests had fallen materially

Financial Stability Board: private-credit vulnerabilities

Federal Reserve: Financial Stability Report, May 2026

All fund percentages are tied to the stated reporting period. The three expert views are paraphrased because independent on-record confirmation was not supplied with this final revision.

The Daily Brief

The morning briefing for the people who allocate long-horizon capital.

Research, charts, video and podcast analysis for the institutions investing at the scale of the world.

Universal Asset Owners