Blackstone Isn't the Only One: Blue Owl vs J.P. Morgan vs Morgan Stanley vs Apollo on Fee, Coverage and Redemptions (Q2 2026 Filings)

Why these four and not BREIT

Blackstone's BREIT gets written about constantly, and we have our own forensic review of it. The other four institutional vehicles in this category get written about almost never — when we checked the search results before writing the underlying reviews, two of them had no independent review in the top ten at all, only issuer pages, sec.gov and directory listings.

That is odd, because between them they hold well over $13 billion of investor money, and all four filed a quarterly report in the last three weeks that says exactly what they charge and exactly whether the cheque they are sending you is being funded by the buildings or by something else.

Everything below comes from those four filings. Nothing comes from a fact sheet, a directory or a press release.

Fees: a 2.5x spread on products that look the same

The headline is that the same category, in the same quarter, prices between 0.50% and 1.25% of NAV. On a $100,000 position that is the difference between $500 and $1,250 a year, forever, before performance fees. Nothing in the marketing of these vehicles prepares you for that spread, because each one presents its own fee as though it were the market rate.

Two qualifications, both of which cut against reading that table too simply.

First, the cheap classes are not always the ones you can buy. Morgan Stanley's 0.50% applies to Class F-S and F-I. J.P. Morgan's 0.75% applies to Class X and Y. These are typically the classes sold through fee-based advisory relationships, where you are already paying the adviser separately. If you buy through a commission-based broker you land in the 1.00-1.25% class, and — as we set out in what a financial adviser actually costs a real estate investor — which class you end up in is decided by how your adviser is paid, not by how much you invest.

Second, Apollo's 0.85% has an expiry date, and it is close. The filing is explicit: the reduction runs "for the period from April 1, 2023 through January 2, 2027." Class A-III is 59.2% of the entire fund, $1,034,747,000 at July 31. When that reduction lapses in four months, the fee on the largest class goes from 0.85% to 1.00% — roughly $1,552,000 a year more, on money already invested. It is properly disclosed and nobody is hiding it. It is also the kind of thing that never makes it into a comparison table, which is why it is in this one.

Coverage: the number that actually separates them

A management fee is a known cost. Whether the distribution is being funded by the properties is a question about whether the thing works.

Three of these four say the same sentence and only one of them means something different by it.

ORENT's is the case worth understanding, because it is not a lie and it is not an error. Its Sources of Distributions table does read 100% funded from operating activities. Footnote (1) then defines the measure: it is inception-to-date, not the period, and it is as adjusted — including, among other things, rent from build-to-suit properties "for which rent has not commenced." So the 100% is arithmetically defensible against the definition the footnote gives, while the plain period comparison directly above it reads $252,061,000 of operating cash flow against $283,965,000 distributed. The label and the number are measuring different quantities, and only the label is on the page in bold. Coverage on the plain basis also fell from 97.4% a year earlier.

That is not a reason to write ORENT off. It is much the largest of the four, it did not gate anyone, and in the same quarter it took a listed REIT private in a $2.45 billion transaction — a deal it funded on its credit facility, while it kept up the largest private real estate equity raise we have measured this year: $1.92 billion across seventeen closings in 2026, $580.1 million of it after the deal closed. It is a reason to read the footnote before you quote the percentage.

North Haven's is the opposite problem in miniature. Its 100% is honest, period-basis, GAAP — and the margin is $136,000 on $42,514,000, which is 0.32%. One soft quarter and the answer changes. Saying "fully covered" and stopping there would technically be accurate and would badly misdescribe the position.

Redemptions: the test the category keeps failing

The whole promise of a NAV REIT is that you can get out at net asset value, monthly or quarterly, subject to a cap. It is the promise that has broken most often. Fundrise suspended Equity REIT redemptions in October 2025; RealtyMogul suspended its share repurchase programme and DRIP in April 2026. We keep the running list in the NAV REIT redemption status tracker.

Against that backdrop, two of these four put a clean sentence in writing:

• ARIS: repurchased $35,000,000 in the half, against $9,786,000 in the first half of 2025, and "had no unfulfilled repurchase requests as of June 30, 2026."
• ORENT: no unfulfilled repurchase requests in either H1 2026 or H1 2025.

That is a real distinction and it deserves to be said plainly, because the honest reading of this category is that the institutional vehicles have so far kept the liquidity promise the retail platforms broke. Whether that survives a genuine redemption wave is untested at all four — none of them has been through one.

There is a governance wrinkle on the other side of the ledger. North Haven's arrangement with Morgan Stanley explicitly puts the sponsor last in the queue: the Company repurchases Class E shares from Morgan Stanley only after first satisfying repurchase requests from every other common shareholder that quarter. That is the right way round and worth crediting.

J.P. Morgan's is the opposite. JPMREIT's adviser seed of $100.0m was fully repurchased as of June 30, 2026, under a clause taking 80% of net monthly offering proceeds. The half's cash flow shows $61,070,000 going out to the adviser against $7,747,000 of stock repurchased from shareholders — a ratio of 7.9 to 1. The mechanism was disclosed in advance and the adviser is entitled to it. It still means that in a period when very little new money was arriving, most of what did arrive went to redeem the sponsor.

The one thing only Morgan Stanley tells you

Every one of these vehicles will concede, if you ask, that the retail share class costs more than the adviser class. Only North Haven converts that into a return number and prints it.

Same buildings, same quarter, same manager: 9.00% to the house and 5.90% to the customer, and the retail total return falls from 7.62% to 5.7% once the sales commission is counted — roughly 192 basis points a year. Morgan Stanley disclosed that itself, in footnote (3) of its own quarterly report. We have not found another sponsor in this category that publishes the commission drag as a return figure, and the reason is presumably that it does not read well.

The governance item that should not be buried

On August 7, 2026, JPMREIT filed an 8-K under Item 5.07 reporting that its 2026 annual meeting was adjourned because there was not a sufficient number of stockholders present in person or by proxy to achieve a quorum. No directors were elected. No auditor was ratified. The incumbent directors continue as holdover directors.

The context makes it less sinister and more revealing: the second public offering, opened February 4, 2026 with a $3.8 billion primary target, has sold about $95.1 million, and no Class S share has ever been sold. A vehicle with very few shareholders will struggle to raise a quorum. But the fair way to describe the position is that this is a fund which is not being bought, and the failed meeting is a symptom of that rather than a scandal in itself.

To be fair to it on the numbers: JPMREIT has the best distribution coverage of the four and a fee below the category standard. It is the cheapest well-covered option here. It is also the smallest and the least liquid by virtue of being tiny.

Update, September 21, 2026: the August 31 marks, and which of the four will show you one

Everything above reads the four Q2 2026 Form 10-Qs, filed in August. The next round of 10-Qs is not due until the middle of November, so a reader arriving in September is looking at a three-month-old page unless somebody goes and gets the marks these funds have published since. We went and got them. All four numbers below are as of August 31, 2026 and come from a filing made between September 2 and September 18.

That last row is the finding, and it is a difference in kind rather than degree. ARIS and JPMREIT are registered public offerings, so each month they file a prospectus supplement that prints the NAV per share for every class and the full component table underneath it — investments, cash, debt, accruals — so a holder can see what moved. Blue Owl's ORENT is a Regulation D placement and files no prospectus supplement, but it publishes the same two tables voluntarily in a Form 8-K. North Haven, also Regulation D, publishes neither. Its September filings declare distributions per class and report share sales to a feeder vehicle, and that is all. The only way to infer its August 31 NAV from the public record is to divide: it sold 216,664 Class I shares for approximately $4,530,000 and 13,137 Class F-I shares for approximately $276,000, both "based on the net asset value per share as of August 31, 2026." That implies roughly $20.91 and $21.01 a share — our arithmetic on figures the filing itself rounds, so treat it as an order of magnitude, not a mark.

For a page whose whole argument is that these four products differ under identical-looking covers, this belongs next to the fee table: the cheapest fee in the group comes with the least monthly disclosure in the group.

What the August marks do and do not tell you. ORENT's total NAV is $9.97 billion against the $9.67 billion in its Q2 filing, and ARIS's is $1.78 billion against $1.75 billion. Neither move is a return: these funds are still raising, and new money raises total NAV without moving NAV per share. ORENT's own September 18 filing shows the mechanism in the same document — it sold 2,318,790 Class I shares to feeder vehicles for approximately $25.0 million, priced at the August 31 NAV. That figure is a useful check rather than a headline: 2,318,790 shares at the stated Class I NAV of $10.7695 is $24,972,209, which is the "approximately $25.0 million" the filing reports, so the sale did happen at the published mark.

The one per-share number worth watching is JPMREIT's, because it is the only one of the four that prints a total NAV and a single share/unit count in the same table: $1,140,485 thousand over 100,917 thousand shares and units is a blended $11.30, while its class transaction prices run from $10.56 to $11.56. The spread between those is the share-class structure, not performance — the same mechanism we take apart in the NAV REIT share-class fee drag.

Coverage, fees and redemptions are unchanged from the Q2 reading above: none of those figures is republished monthly, and we have not re-derived them from anything newer. The next real movement in this comparison is the Q3 Form 10-Q round in mid-November.

So which one

We are not going to pretend a table produces a recommendation, and none of these is available to most readers of this site anyway — ORENT and North Haven are accredited-only Regulation D placements, and all four are sold through advisers rather than direct. What the filings support is narrower and more useful:

The one-line version

Morgan Stanley is the cheapest and the most candid. J.P. Morgan covers its distributions best and cannot fill a shareholder meeting. Apollo is well covered and about to get more expensive. Blue Owl is by far the biggest, charges the most, and has the coverage figure that most needs its footnote read.

Individual forensic reviews, each built from the same filings: Blue Owl ORENT · J.P. Morgan JPMREIT · North Haven Net REIT · Apollo ARIS. If you want to know how Apollo's vehicle is actually put together underneath, the 36-subsidiary structure is broken down here.