Private credit redemptions in Q3 2026 eased at the margin, and the managers want you to notice. Apollo Global Management told Apollo Debt Solutions BDC (ADS) shareholders on September 22 that investors who asked to leave during 2026 have received about 75% of the capital they requested. Cliffwater put its own figure at about 78%. We rebuilt both numbers from each fund’s quarterly proration rates, and they hold up, but only for an investor who first asked in Q1 and resubmitted every single quarter since. A client who asked for the first time this quarter gets 34 cents on the dollar from Apollo. At the Q3 fill rate, getting 90% of that request out takes six quarterly windows.
Key Takeaways
- Apollo’s “about 75% returned” matches the math exactly (74.5%) for a Q1 requester who re-tendered in Q2 and Q3. A first-time Q2 requester is at 53.7%, and a first-time Q3 requester at 34%.
- Cliffwater’s “about 78%” is only consistent with a Q2 fill near 36%, a figure the fund has not published. That is our back-solve from its Q1 and Q3 fills.
- At each fund’s Q3 fill rate, recovering 90% of a new request takes 4 windows at Blackstone’s BCRED and Morgan Stanley’s North Haven, 5 at HPS Corporate Lending (HLEND), 6 at ADS, and 7 at Cliffwater Corporate Lending Fund.
- The unmet queue shrank at Apollo (down 18%) and HLEND (down 22%) but barely moved at BCRED (about $2.3 billion to about $2.15 billion) and North Haven (down 3%).
- The only live secondary bid for HLEND, $18.40 against a $24.53 NAV, prices liquidity at roughly 65% annualized against the fund’s expected payout clock.
What Did the Five Big Gated Funds Report for Q3?
Five of the largest non-traded private credit vehicles have now reported third-quarter tender results, and all five hit their caps again. The headline trend is down, which is why Evercore analyst Glenn Schorr called the BlackRock numbers “another positive signal” for wealth demand in direct lending. We keep the fund-level detail in the Trading Market Signals Private Credit Redemption Monitor.
| Fund | Q2 requests | Q3 requests | Q3 cap | Q3 fill rate | Reported |
|---|---|---|---|---|---|
| Apollo Debt Solutions BDC (ADS), ~$25.9B | 16.8% | 14.7% | 5% | 34.0% | Sep 22 |
| Cliffwater Corporate Lending Fund | ~17% | ~16% | 5% | ~31% | Sep 3 |
| HPS Corporate Lending Fund (HLEND), $23.1B | 13.3% | 11.5% | 5% | 43.5% | September |
| Morgan Stanley North Haven Private Income Fund | 11.6% | 11.4% | 5% | 43.8% | Sep 18 |
| Blackstone Private Credit Fund (BCRED), ~$77B | ~10% ($4.5B) | ~10% ($4.3B) | 5% | ~50% | Sep 3 |
Sources: Bloomberg (Sep 3 and Sep 22, 2026), Reuters (Sep 18, 2026), AltsWire, Private Equity Wire (Sep 23, 2026), PitchBook. Fill rate = 5% cap divided by requests. As of September 24, 2026.
The smaller vehicles moved further. BlackRock Private Credit Fund fell to 4.58% from about 5.3%, HPS Corporate Capital Solutions Fund to 1.9% from 4.7%, and TPG’s Twin Brook Capital Income Fund to 1.2% from 2.1%, according to Private Equity Wire. KKR’s offshore KKR-Income Trust I saw requests of about 2.5% of NAV after hitting its 5% cap in Q2, per Bloomberg.
So the large flagships are still asking for two to three times what they can pay, while the second tier has dropped under its caps. That split matters more than the average.
Does Apollo’s “75% Back” Hold Up?
It does, and the arithmetic shows exactly who it describes. Apollo filled 45% of Q1 requests, when shareholders asked for 11.2% of shares (CNBC, March 23). It filled about 29.8% in Q2 (5% of 16.8%) and 34.0% in Q3 (5% of 14.7%).
Chain those three rates for an investor who asked in January and resubmitted every quarter after. After Q1, 55% of the request is still in the fund. After Q2, 55% times 70.2% leaves 38.6%. After Q3, 38.6% times 66.0% leaves 25.5%. That investor has received 74.5% of the original request. Apollo’s figure is correct.
Now run the same chain for other clients. Someone who first asked in Q2 is at 53.7%. Someone who first asked this quarter is at 34%. And someone who asked in Q1, took the partial fill and did not resubmit got 45% and is still holding the rest. The 75% describes the most persistent seller in the book, not the typical one.
Cliffwater’s 78% needs a closer look because the fund has not disclosed all three fills. Its Q1 fill was about 50% (a 7% repurchase against 14% of requests) and its Q3 fill about 31% (5% against 16%). For a Q1 requester to reach 78% cumulative, the Q2 fill has to be close to 36%, meaning about 6.1% of shares repurchased against 17% requested. That is our back-solve, not a Cliffwater figure. If the actual Q2 repurchase was the standard 5%, the cumulative number would be closer to 75%.
How Many Quarters Does a Gated Dollar Take to Get Out?
The payback claim looks backward. An advisor with a client who wants out today needs the forward version: if fill rates stay where they were in Q3, how long until the money arrives? We call this the exit clock.
The method is simple. At a fill rate f, the share of a request still stuck after n windows is (1 − f) raised to n. Solve for the window count that leaves 10% or less, and for the average wait of a requested dollar, which is 1 divided by f.
| Fund | Q3 fill rate | Windows to recover 90% | Average windows per requested dollar |
|---|---|---|---|
| BCRED | ~50% | 4 | 2.0 |
| North Haven Private Income Fund | 43.8% | 4 | 2.3 |
| HPS Corporate Lending Fund | 43.5% | 5 | 2.3 |
| Apollo Debt Solutions BDC | 34.0% | 6 | 2.9 |
| Cliffwater Corporate Lending Fund | ~31% | 7 | 3.2 |
TMS calculation from Q3 2026 fill rates, holding each rate constant. Windows are quarterly, so 6 windows is roughly 18 months. As of September 24, 2026.
Two caveats. First, fill rates will not stay constant. If requests keep falling, the clock shortens. If a credit event hits the software-heavy loan books that worry investors, it lengthens. Second, the clock assumes the investor resubmits every quarter. Most of these funds require a fresh request each window, and a client who forgets one loses a turn.
The spread is wider than the headline caps suggest. Every fund in the table pays 5% a quarter, yet a Cliffwater requester waits almost twice as long as a BCRED requester for the same 90%. The cap tells you the size of the door. The request level tells you the length of the line.
Is the Queue Actually Getting Shorter?
A lower Q3 request figure does not mean fewer people want out. Part of every Q3 number is Q2 demand that was prorated and came back. Separating the two tells you whether the line is shrinking.
Two funds gave us the split. Morgan Stanley said nearly two-thirds of North Haven’s Q3 requests came from investors who had not been able to fully cash out in the previous two offers. That puts fresh demand at roughly 3.8% of shares, under the 5% cap. For BCRED, TD Cowen estimated that the Q2 backlog made up about half of the $4.3 billion Q3 request, which puts new demand near $2.15 billion, about half of what arrived in Q2.
For Apollo, we can only set bounds. Its unmet Q2 demand was 11.8% of shares. If every one of those investors came back, fresh Q3 demand was just 2.9%. Apollo said the majority of Q3 requests were re-tenders, which caps fresh demand below about 7.35%. The truth sits somewhere between the two.
The cleaner test is the size of the unfilled queue going into Q4:
- Apollo ADS: unmet requests fell from 11.8% of shares after Q2 to 9.7% after Q3, down 18%.
- HPS Corporate Lending Fund: from 8.3% to 6.5%, down 22%.
- North Haven: from 6.6% to 6.4%, down 3%.
- BCRED: from about $2.3 billion to about $2.15 billion, down roughly 7%.
That is our reading of the Q3 data. The two funds that led the Q2 surge are clearing their lines. The two funds with the most moderate requests have stalled lines, because their fresh demand is running close to what the cap can absorb each quarter. BCRED’s flat queue is the number to watch in December. It is the largest fund in the category, and its line has barely moved in a quarter.
What Is a Gated Dollar Worth Today?
The exit clock has a price, and the secondary market has named it. On July 14, Cox Capital Retail Secondaries Fund I bid $18.40 per Class I share for HLEND against a May 31 NAV of $24.53, a discount of 25%. The HLEND board unanimously told shareholders to reject it on July 27, per its Form 8-K. The offer covered about 0.1% of the share class, so treat it as a price signal rather than a market.
Put that bid next to the exit clock. At HLEND’s Q3 fill of 43.5%, the average requested dollar comes out after 2.3 windows, or about 0.58 years. Accepting a 25% haircut to skip a 0.58-year wait works out to an annualized cost of about 65%. That compares with the 9.9% annualized net return HLEND reported through July, and it ignores the income the client would keep earning while in the queue.
The only way the Cox bid makes sense for a seller is if NAV is about to fall by something close to a quarter before the fund pays out. The data do not point there yet. Across 44 US BDCs, fair value stood at $92.88 billion against a cost of $95.19 billion on June 30, a gap of 2.4%, per the Reuters roundup of September 4. BCRED’s NAV per share did slip from $24.68 to $23.64 between January and July, a 4.2% decline, which is real but a long way from 25%.
Does the Fed Hike Change the Math?
The Federal Reserve raised the federal funds target by 25 basis points to 3.75%-4.00% on September 16, its first increase since 2023, and the median dot shows one more hike in 2026 (CNBC). For investors waiting in the queue, a higher base rate lifts the floating-rate coupon on the loans they still own, so the income side of the wait improves.
The other side is the borrower. Software and services companies are the concentration that has made investors nervous since spring, and higher floating rates reduce their interest coverage directly. The second-quarter losses in the category were concentrated in a minority of over-levered software borrowers rather than spread across portfolios, per the September 4 Reuters roundup. A rate hike makes that minority larger at the edges. For the exit clock, the hike is a small positive on income and a potential negative on NAV, and the NAV side is the one that could turn a 2.4% fair-value gap into something the Cox bids would start to look reasonable against.
Fundraising is the offset worth tracking. Global private credit fundraising reached $33 billion in Q3 through August 25, tracking toward the $45 billion raised a year earlier, according to the same Reuters roundup. Apollo reported about $200 million of ADS gross subscriptions through September 1 against about $700 million of Q3 repurchases. New money is coming in, but at the flagship level it is covering well under a third of what leaves.
What Should Advisors Do With the Q3 Numbers?
Treat manager payback figures as a best case. When a sponsor says investors have received 75% or 78% of requested capital, ask which cohort the figure describes. Then run the chain for your own clients by the quarter they first asked. The math takes two minutes and the answer can differ by 40 points.
Put the exit clock in the plan. For a client with a known liquidity need, a gated position is worth its NAV only if the need falls beyond the clock. Six windows at ADS or seven at Cliffwater means about a year and a half of scheduled resubmissions. Build the calendar, and make sure nobody misses a window.
Price the secondary bid against the clock, not against NAV. A 25% discount looks like a loss on the statement. Against a wait of under seven months, it is a 65% annualized cost of liquidity. Most clients with a genuine emergency still should not accept it without exhausting every other source.
Watch BCRED’s Q4 queue. If fresh demand stays near the 5% cap, the line will not clear by itself, and Blackstone faces the same choice it made in Q1, when it lifted the repurchase to 7.9% with about $400 million of firm and employee capital. We covered that backstop in our Q2 redemption analysis. Whether the manager writes that check again is the most useful single signal for Q4.
For allocation decisions going forward, the June case for resetting private credit sizing in model portfolios still holds, and the Q3 data sharpen it: size the position to the exit clock, not to the stated cap. Interval funds with a mandated minimum repurchase have different mechanics, and the difference is worth explaining to clients before the next window rather than after.
Three Questions for Your Investment Committee
- For each gated private credit position we hold, which quarter did each client first request, and what is that client’s actual cumulative fill, as opposed to the sponsor’s headline figure?
- At the Q3 fill rate, how many windows until each pending request is 90% paid, and does any client have a liquidity need that lands inside that window?
- If BCRED’s queue is still flat after the December tender, what is our policy on new private credit subscriptions, and who decides it?
Figures as of September 24, 2026. Fill rates and exit-clock estimates are Trading Market Signals calculations from publicly reported tender results and assume constant fill rates. This is not investment advice.
About Me
Founder and Chief Research Analyst at Trading Market Signals. Abdelali El Khadmaoui specializes in AI Wealth Intelligence, Wealth Management, Registered Investment Advisors (RIAs), Family Offices, Retirement Planning and Private Credit. He publishes in-depth research and data-driven analysis for financial professionals.
