Private credit's retail vehicles gated redemptions across 2026 not because they failed but because they worked as designed. The 5% quarterly tender cap is the mechanism that lets these funds hold illiquid loans at all. This piece explains how tender windows work, what Q1 and Q2 2026 actually showed, why the forced asset sales matter more than the gates, and the three questions to ask before buying any semi-liquid fund.
In the first quarter of 2026, investors in non-traded business development companies asked for $13.9 billion of their money back. Sponsors returned $7.4 billion. More than $4.6 billion sat behind gates. In the second quarter it got heavier: across the sixteen perpetual non-traded BDCs that Fitch Ratings tracks, redemption requests averaged 10.3% of shares outstanding, up from 9.7% in Q1, against a quarterly cap that is typically 5%. Ten of those sixteen funds breached their caps and had to allocate pro rata.
Nothing broke. No fund defaulted. No sponsor went under. The structure did precisely what its prospectus said it would do — and that is the uncomfortable part, because it means the thing that surprised investors was not a failure. It was the product.
Private credit's retail pitch rested on one implicit promise: the returns of a loan book without the volatility of a bond fund. No daily marks bouncing around on your statement. No drawdown to explain to a client in a bad quarter. Steady quarterly distributions from senior secured floating-rate loans. It was, on the numbers, a genuinely attractive proposition, and the growth of the category reflected that.
But the absence of daily marks was never evidence of stability. It was evidence of illiquidity. A loan that nobody prices every day does not have a smooth price; it has an unobserved one. Those are different things that look identical on a quarterly statement, and the distinction stayed academic for as long as nobody needed their money back in a hurry.
The pitch was always describing illiquidity, not stability. In 2026 investors asked for their money, and the difference stopped being academic.
Almost no coverage explains the mechanism, and the mechanism is the whole thing. A perpetual non-traded BDC or interval fund does not trade on an exchange. Instead it offers a share repurchase programme: once a quarter, the fund makes a tender offer to buy back a limited quantity of shares at net asset value. That quantity is capped, conventionally at around 5% of shares outstanding per quarter — roughly 20% a year.
The cap is not a crisis measure and it is not hidden. It is disclosed in the prospectus, it has been there since the fund launched, and it is the specific feature that allows the vehicle to hold illiquid, privately negotiated loans at all. A fund obliged to return capital on demand cannot own assets that take months to sell. The cap is the trade: you accept a queue, and in exchange the fund can own the assets that generate the yield you came for.
Which means the cap is invisible right up until demand exceeds it. Below 5%, everyone who asks gets paid in full and nobody thinks about the mechanism. Above 5%, the fund accepts requests pro rata — if 10% of shares are tendered against a 5% cap, every investor gets roughly half of what they asked for and rejoins the queue for the rest. There is no discretion involved and no judgement being made about individual investors. It is arithmetic.
Fitch's Q2 sample shows an average request rate of 10.3% of shares, with a range running from 1.3% at the lightest to 38.1% at the heaviest — Blue Owl Technology Income Corp posted the highest rate in the group. That dispersion matters: this is not a uniform stampede out of the category, it is concentrated pressure on specific vehicles, with others barely touched.
The largest single instance came on 4 June 2026, when Blackstone limited redemptions on BCRED, its flagship non-traded private credit fund, described in coverage as roughly $79 billion in size. Investors requested about $4.4 billion in the quarter; they received the 5% the documents allow. It was the first time BCRED had hit that limit in its history.
The flow picture underneath is the cleaner sentence. Non-traded BDC fundraising reached roughly $11.9 billion through May 2026, down about 55% year on year, against roughly $12.9 billion in redemptions — a net outflow of about $1 billion for a category that had, until this year, only ever grown. Stanger's data has quarterly outflows exceeding quarterly inflows for the first time in the sector's history. The gates are the visible symptom; the fundraising number is the diagnosis.
The second thread is the one with teeth, and it gets far less attention than the gating headlines. When a fund with illiquid holdings must generate cash on someone else's timetable, it has to sell something. And when it sells, the marks that had been comfortably model-derived get tested against an actual bid.
Blue Owl provides the worked example. In February 2026 the firm permanently closed the redemption window on its $1.6 billion OBDC II fund — not a temporary gate but the elimination of the quarterly liquidity facility retail holders had relied on. Around the same time, OBDC II disclosed entry into loan sale agreements totalling roughly $600 million at 99.7% of par including unfunded commitments, as part of a broader Blue Owl disposal of about $1.4 billion of assets to institutional buyers including public pension plans and insurers.
Read that 99.7%-of-par figure carefully, because it cuts both ways. It is genuinely reassuring: a portfolio sold at essentially par is a portfolio whose marks were approximately right. It is also a single transaction, negotiated, in a specific book, sold to institutional buyers who wanted it. One clean print is not a valuation regime. But it is more information than the category has produced in years, and redemption pressure is currently the only mechanism generating any.
Forced liquidation is the only process that makes private marks observable. That is an argument for paying attention to what gets sold, not to which funds gate.
The case against reading any of this as a crisis is strong and deserves to be stated properly rather than waved at.
Gates protect the investors who stay. A fund that sold assets at any price to satisfy every redemption request would be transferring value from remaining holders to exiting ones — dumping the most liquid, highest-quality positions first and leaving the rest of the book worse than it found it. The queue is not a trap sprung on investors; it is the alternative to a fire sale, and it is the alternative that the people who did not ask for their money back would choose.
Fitch's own conclusion points the same way. Analysing eight rated perpetual non-traded BDCs, it found that current liquidity and asset-coverage cushions should support quarterly redemptions at the maximum 5% tender level for four more quarters even with no new equity raised, with all eight staying below the regulatory 2.0x leverage threshold under stress. The Financial Stability Board's May 2026 report on private credit vulnerabilities frames bank exposure as largely indirect, senior and buffered, with stress concentrated in corporate direct lending to smaller, highly leveraged, sub-investment-grade borrowers. Most of the loans are floating rate. The total market is roughly $1.5–2 trillion and US-dominated.
This is a mis-selling and expectations story, not a 2008 story, and it is weaker if it reaches for that comparison. Nobody is being wiped out. People are being told to wait.
So the useful question is not whether private credit is in trouble. The default numbers will answer that eventually and they have not answered it yet. The useful question is whether the buyer understood what they bought — and that answer is already legible, not in defaults but in the fundraising line. A category that raises half of what it raised last year while paying out more than it takes in is a category where a material number of holders have concluded the product was not what they thought it was.
If you hold one of these funds, or an adviser has put one in front of you, three questions do most of the work — and they work on any semi-liquid vehicle, not just BDCs:
None of those questions require a view on interest rates, credit spreads or the economic cycle. They require reading three documents the fund has already published. The investors who were surprised in 2026 were not surprised by markets. They were surprised by their own paperwork.
A redemption gate is a limit on how many shares a fund will repurchase in a given period — for perpetual non-traded BDCs and interval funds, typically about 5% of shares outstanding per quarter. It is not a refusal and it is not discretionary in the sense of the sponsor picking who gets paid; when requests exceed the cap, everyone is paid pro rata and the remainder rejoins the queue. The cap is disclosed in the prospectus before you invest and is the structural feature that permits the fund to hold illiquid loans at all. So yes, it is legal, because it is what you agreed to.
DrafterDaily takes apart the financial products people are sold before the mechanics become news. One clear read a day, no hype.
Get the daily briefGrad PLUS lent up to the full cost of attendance. Its replacement caps professional students at $50,000 a year — against a medical-school year that averages $61,110. The difference does not disappear.
Nationally, asking prices are down about 2% and inventory is up 3.2% — a picture of orderly rebalancing. Underneath it, Austin's median list price is down 12.2% while San Francisco's inventory has fallen nearly 20%. Those are not variations around a mean.
A composite index of 56.0 is the strongest US reading since April 2022. It is also an average of a services sector at a 20-month high and a factory sector going the other way.