The 5% cap is a constant; the queue decides who gets paid

One BlackRock-affiliated private credit fund returned 43 percent of what investors asked to redeem; another returned all of it. The two run in the same asset class, reach investors through the same wealth channel, and operate under the same 5 percent quarterly repurchase cap, yet fifty-seven points separate their fill rates. The credit has nothing to do with it.

Three NAV BDCs reported repurchase results inside two days, PWD's tracking shows, and the two BlackRock-affiliated vehicles in that batch landed on opposite sides of the same line, while the third filing turned on HPS's 5 percent cap—the cap doing the work rather than a cleared queue. Set the three side by side and the lesson is structural: identical ceilings, materially different outcomes.

A capped repurchase window is a rationed exit with a simple rule: requests below the cap are paid in full, and requests above it are prorated. BDEBT's queue came in under the line, so its holders got everything they asked for; HLEND's came in over it, and 43 percent is what came out the other end. Run that backwards and HLEND's requests worked out to roughly 2.3 times what its cap would allow. The oversubscription multiple is the figure that measures pressure; the fill rate is only its residue.

Advisors have treated the 5 percent cap as the dial to watch, but this week makes that reading hard to defend. A cap is a term, and terms are uniform inside a fund family: two vehicles with the same parent and the same ceiling produced fills that differ by more than half the amount requested. The spread came from the length of the line in front of the window, not the paperwork or the loan book.

Brand is the proxy advisors usually reach for when they pick between these vehicles, and the week complicates that instinct. Platform scale, credit team, and distribution reach are all shared between HLEND and BDEBT, and none of it showed up in the result. What showed up instead was the composition of each shareholder base and its appetite for cash, a variable that a due-diligence questionnaire captures badly and a sponsor does not control.

The line is a product of the other holders, which is what makes it so unstable. A capped window makes every investor's exit contingent on everyone else's decision to stay, since a partial fill leaves the remainder in the fund with no claim on the next window's capacity, and that structure manufactures its own crowd. In light demand it looks like a well-behaved product; the moment a few large positions move, the sensible response for everyone else is to move with them, and the fill rate settles wherever the cap forces it.

The five percent figure also does less for an individual holder than it appears to. It is a ceiling on the fund's aggregate outflow, not an entitlement attached to any single position, and the distance between those two readings is the whole of HLEND's 43 percent. An advisor who sold the wrapper on quarterly liquidity up to 5 percent described a feature of the vehicle; clients who heard a promise about the account are the ones now holding a partial fill and a longer wait.

The case for these vehicles rests on a mismatch advisors tolerate in calm quarters: the loans inside do not trade daily, yet the wrapper offers periodic liquidity anyway. BDEBT's clean tender is that promise working; HLEND's 43 percent is the same promise meeting a quarter in which more holders wanted out than the clause could accommodate.

The 57 percent of HLEND requests that went unfilled did not evaporate; those holders still own the position, and the ones who filed because they needed the money will need it again, which is why an oversubscribed window tends to repeat. BDEBT's full tender removed its overhang outright—everyone who wanted out got out, and the next window opens with a shorter line—and that asymmetry is the part of this week's batch a sponsor cannot file its way around.

A quarter is the unit of pain here. A capped window is not a slow payment so much as a full cycle of the calendar before the next attempt. A 43 percent fill costs a holder not 57 percent of their money but three months of access to it, and the second attempt arrives with no guarantee of a better ratio. That is the cost advisors tend to leave out of the pitch, because in ordinary quarters nobody has to price it.

There is a countervailing force worth naming: proration punishes the casual redemption alongside the urgent one, and a holder who filed on a rumor and received 43 percent may decide the exercise is not worth repeating. A fund can, in principle, bleed its way down to a base of holders who have chosen to stay, but that path is real and slow, and the results filed this week do not show it at work.

For an advisor the nearer consequence is a cash-flow problem with a lag: a client who asks for a full redemption and receives 43 percent still has the balance exposed to the loan book, still subject to the next window, and still unavailable for whatever the money was raised for. Portfolios built on the assumption that the quarterly window is a soft form of daily liquidity need that assumption restated—the window is a queue with a ceiling, and the ceiling is where the plan bends.

The reporting question follows from the same asymmetry: if the cap is uniform and the queue is not, the number an advisor wants before committing capital is the oversubscription multiple—requests as a fraction of available capacity—and today that number reaches the market only after the fact, embedded in a fill rate that has to be decoded. A fund reporting requests at four-tenths of its cap would be telling an advisor the exit works; a fund reporting 2.3 times would be telling them to price the wait into whatever they pay for the position. Semi-liquid vehicles publish reams about credit and, for now, one retrospective number about the line outside the door.

The obvious objection is that publishing the multiple would make queues worse, since visibility into a 2.3-times oversubscription reads as an invitation to file before the crowd does. That objection proves more than its proponents would like. The queue already exists and is already being acted on by the investors closest to the filings, while the people who cannot see it are the ones without the arithmetic. A metric whose chief virtue is that holders cannot see it is not a metric the market can price.

Until someone publishes it, the multiple can be rebuilt by hand, and that reconstruction is the due diligence: every fund that reports a fill rate is also reporting, in code, how long its line ran against its cap. Collect four quarters of those and the pattern does more work than any credit commentary—a fund that fills at 100 percent four times has capacity its holders are not using, and a fund that prorates twice in a row has a queue that has learned to arrive early.

The sponsor-level read is a distribution question as much as a portfolio one, because a platform running two vehicles under one brand and one wholesale force now owns two different answers to the same question. Wholesalers who have spent years explaining the 5 percent cap as a feature will spend the next several explaining why it bound at one fund and not the other, which is the harder half of the pitch.

The next quarterly window will print another fill rate for both funds, and the spread between them will matter more than either number on its own, because if BDEBT clears again while HLEND sits in the forties the question stops being demand for one fund's paper and becomes which vehicles in the wealth channel still let people out. Advisors can get ahead of that by assembling the fill history of every capped product they own and asking what each rate implies about the line behind it; HLEND's implies a line running at more than twice the cap.