The Yieldstreet Fund Sold to Ordinary Investors Just Ceased to Exist. Its Last Exit Filled 11.8% of Requests, and Fees in the Replacement Are 75% Higher

What actually ended, and why it is not just another fund closure

Funds are reorganised constantly. Most of it is plumbing and none of it is news.

This one is different for a single reason: the Yieldstreet Prism Fund was the door. Yieldstreet built a brand on the idea that ordinary people should get the investments rich people get, and almost everything on the platform required accreditation anyway. Prism was the exception. A filed advertisement from November 2020 puts it in the company's own words: the fund "is Open to Non-Accredited Investors", and the opportunity to invest "was previously limited to accredited investors."

That is the product that just ceased to exist. Not a strategy wind-down, not a share class merger. On August 28, 2026 it filed a Form NPORT-P with the box isFinalFiling set to Y, which is the regulatory equivalent of turning off the lights. We have written before about what a filing that ends disclosure tells you, and the principle holds here: the last filing an issuer makes is the last chance anyone gets to look.

The six-year record, from the audited highlights

Here is every year of the fund's life, taken from the financial highlights in its annual reports rather than from any marketing page.

Three things in that table deserve to be said plainly.

The share price never recovered its issue price. Six years in, NAV is $9.03 against a $10.00 launch. That is not fatal on its own, because an income fund is supposed to pay you along the way. But it means every dollar of the return had to come from distributions, and it means anyone who bought at $10.00 and sold at a tender was crystallising a loss on principal in every single quarter of the fund's existence.

Return of capital ran through every year. Across six years the fund distributed $4.19 per share and $1.22 of that, 29.1%, was return of capital. Return of capital is not automatically sinister; funds with lumpy income use it, and it is tax-deferred rather than tax-free. But it is not yield. It is your own money coming back with your cost basis reduced, and a distribution rate quoted without that breakdown flatters the product. Nearly three dollars in ten here were that.

The fund's own since-inception number is 6.02% a year. That is the figure in the proxy, on a NAV basis, to December 31, 2025. Chain-linking the six annual returns in the table above gives +40.4% cumulative, which back-solves to 6.02% over 5.81 years — so the disclosed figure is internally consistent. It is a real number honestly reported. It is also, for a vehicle marketed on access to private-market returns, roughly what an investor could have had from a diversified bond fund over the same span with daily liquidity.

The part that matters most: the exit was rationed for eleven straight quarters

A closed-end fund like this has no secondary market. The only way out is the issuer's periodic tender offer, and the offer is capped. The fund offered to buy back roughly 3% of shares a quarter, occasionally 5%.

If shareholders tender more than the cap, everyone gets cut back pro rata. That is disclosed and legal. What is not obvious from any marketing page is how consistently it happened here, and how steadily it got worse. Each row below comes from the final amendment to that quarter's Schedule TO, which is required to report shares tendered and shares accepted.

Read the two end columns together. The number of shares accepted barely moves — it sits between roughly 400,000 and 460,000 every quarter, because it is a fixed percentage of a fund whose size was roughly stable. The number tendered nearly triples, from 1.32 million to 3.86 million. The queue grew; the door did not.

By the last offer, a shareholder asking to leave got 11.8% of the request filled. In practical terms, someone who wanted $10,000 out received about $1,180 and kept waiting for the rest.

Then the waiting stopped being an option. The March 2026 tender documents told shareholders that "it is anticipated that the current Offer will be the last opportunity for you to tender your Shares prior to the Asset Acquisition", and that the company "has suspended the offering of its shares for sale." The proxy confirms the board "determined to suspend repurchases pursuant to AIF's share repurchase program in connection with the proposed transaction."

So the arithmetic of that final quarter is stark. 3,404,518 shares were tendered and refused — about $31.1 million at the $9.1395 price. Every one of those holders had asked to leave, been told no, and then found the exit closed behind them. They are SOFIX shareholders now whether they wanted to be or not.

What they were moved into

The pitch to shareholders leaned on liquidity. AIF, the argument went, offered "no more than 5% (typically 3%)" and the board could suspend it, whereas SOFIX offers "at least 5%" under a fundamental policy changeable only by shareholder vote.

That is accurate, and the fundamental-policy protection is a genuine structural improvement worth stating clearly. But SOFIX's own financial statements say its repurchase offers were being rationed too. Its annual report for the year ended September 30, 2025 states that "three of the quarterly repurchase offers were oversubscribed such that pro-ration was required." Its semi-annual report for the six months ended March 31, 2026 goes further: "each of the quarterly repurchase offers were oversubscribed such that pro-ration was required."

In other words, in the two quarters immediately before shareholders voted on a deal sold partly on better liquidity, every one of the acquiring fund's repurchase offers was rationed as well. A bigger stated cap is not the same as a shorter queue, and if you are wondering why liquidity terms are the thing to read first in this whole category, this is the cleanest example we have found.

The fees

This is the proxy's own table, not our estimate. It is annualised, as a percentage of net assets.

3.53% to 6.17% is an increase of 264 basis points, or about 75%. The single largest new line is an incentive fee of 1.22%, which AIF holders did not pay at all — a performance charge arriving on a portfolio that has just been handed over.

To the filing's credit, it does not hide this. The proxy tells shareholders in plain words that they "should expect to bear significantly higher total annual expenses than you currently bear." That sentence is in the document. It is simply not the sentence that led the marketing email, which instead emphasised that SOFIX "has delivered stronger historical returns after fees and expenses (even though they are higher than those you pay in AIF)."

On that performance claim, one caution that the headline numbers obscure. SOFIX's since-inception return of 14.39% looks commanding next to AIF's 6.02%, but its audited fiscal-year returns are 1.20% (stub), 30.31% in FY2023, 7.32% in FY2024 and 9.76% in FY2025. The 30.31% year was earned while the fund grew from roughly $2.3 million to $37.9 million in net assets — a micro-cap fund riding the 2022-23 credit recovery. The fair comparison is the three-year: 7.51% for AIF against 10.83% for SOFIX. SOFIX is ahead on that measure. It is a real edge and a much smaller one than the inception figure implies.

The minimum

SOFIX offers a single share class, Class I, and its prospectus states a minimum initial investment of $1,000,000. The fund or the adviser may waive it at their discretion, and advisers may aggregate client accounts to meet it.

Former AIF holders were carried across regardless of size, so nobody was thrown out. But it is worth sitting with the shape of this: a fund explicitly marketed to non-accredited investors at a $10,000 minimum has been folded into one whose stated entry price is a hundred times larger. We could not find any statement granting former AIF holders a blanket waiver to add to their positions, so whether a $10,000 investor can put in another $5,000 is, on the public record, unclear.

Who paid for the transaction

Reorganisations have costs and somebody bears them. In this one they are itemised.

• AIF shareholders bore approximately $1.1 million of the deal expenses, which the proxy says "will reduce the net assets of the Acquired Fund and, accordingly, the value of the consideration to be received by Acquired Fund Shareholders." On roughly $134 million of net assets, that is about 0.8% of NAV.
• SOFIX bore approximately $2.5 million, for a combined total near $3.6 million.
• Willow Wealth was paid up to $5,000,000 by Mount Logan under a Transition Services Agreement: $2 million cash at closing, $1 million in Mount Logan stock, and up to $2 million more over two years.
• The Special Committee was paid $30,000 for the chair and $25,000 per member, half in January 2026, "without regard to the outcome."

None of that is improper and all of it is disclosed. The reason to line it up is simply that the ledger reads in one direction: the shareholders whose exit had been rationed for eleven quarters contributed about 0.8% of their remaining net asset value to the cost of being moved, and the manager leaving the business collected up to $5 million for the handover.

The rebrand nobody outside the filings would notice

There is a detail here that is easy to miss and hard to unsee once you have.

In January 2026 a prospectus supplement recorded that YieldStreet Management, LLC changed its name to Willow Asset Management LLC, that Yieldstreet Inc. changed its name to Willow Wealth Inc., and that all references to yieldstreet.com were replaced with willowwealth.com.

And then: "The Fund's name and its website ... remain unchanged."

The corporate group scrubbed the Yieldstreet name off the parent and the adviser, and left it on the product that still held retail investors' money. The fund filed its final report eight months later still called the Yieldstreet Alternative Income Fund.

The rename followed a rough stretch of press. CNBC reported in December 2025 that Yieldstreet investors had lost a cumulative $208 million across the platform, and had reported in August and September of that year on losses in its real estate bets and on $89 million of marine loans backed by ships lost in international waters. Those are CNBC's figures, and we cite them as CNBC's rather than adopting them as our own.

Behind the platform sits a 2023 SEC settlement that names the same adviser entity that ran this fund. On September 12, 2023 the SEC issued a settled order against YieldStreet Inc. and YieldStreet Management, LLC — now Willow Asset Management — finding violations of Securities Act Sections 17(a)(2) and 17(a)(3) and Advisers Act Section 206(4) and Rules 206(4)-7 and 206(4)-8. The adviser was censured; total monetary relief was $1,939,220.41, including a $1,000,000 civil penalty, with a Fair Fund created for investors. The conduct at issue was a $14.5 million 2019 offering to finance ship-deconstruction loans, where the company learned that vessels backing earlier loans "had already been broken up without repayment of the debt" and proceeded without telling investors. The order was settled without admitting or denying the findings.

The adviser's stated reason for leaving the fund business, incidentally, is in the proxy and is candid: Willow plans to "continue its evolution into a platform that connects individual investors with leading institutional managers across private markets rather than serving as a portfolio manager", and proposed the reorganisation in connection with plans "to end its closed-end fund portfolio manager business."

Nobody covered this

We looked hard for consumer or personal-finance coverage of the wind-down and found none.

There are three genuine articles about the transaction. One is a trade rewrite of the press release in Alternative Credit Investor headlined around Mount Logan doubling its interval fund. One is an AltsWire piece framed the same way. The third is the deal counsel's own announcement. Everything else — Yahoo, StockTitan, TipRanks, TradingView and the rest — is syndication of Mount Logan's two press releases.

Every one of them is written from the acquirer's side, counting earnings accretion and assets gained. Not one frames it as a fund sold to non-accredited retail investors ceasing to exist, and none mentions the rationed redemptions, the fee increase, or the $31 million of refused tenders.

Two details sharpen the point. InvestmentNews covered Willow Wealth twice after the deal was announced, including a CEO interview in January 2026 and C-suite hires in June, without mentioning it. And on August 26, 2026 — two days after the merger closed on August 24 — Forbes published a piece titled "Private Credit Comes For The Retail Investor, And The Fund Structure Is The Problem." We read it in full. It does not mention Yieldstreet, Willow, Prism, Mount Logan or SOFIX. It is the thesis without the case study that had just completed.

We checked our own negative before asserting it, because a null result is only worth anything if the method could have found a positive. Running the same searches against the 2023 SEC settlement returned four outlets immediately. The silence around the wind-down is real, not an artefact of the search. What we cannot speak to is coverage behind paywalls and crawler blocks — Ignites and FundFire in particular cover fund reorganisations closely, and we could not read them.

What this means if you held it

If you owned the Yieldstreet Prism Fund or the Alternative Income Fund, some concrete things are now true whether or not you were following the paperwork.

1. You own SOFIX shares. Each AIF share became 0.80 SOFIX shares at NAVs struck May 22, 2026. The dollar value was intended to carry across; the share count changed, so a smaller number in your account is not by itself a loss.
2. Your annual costs went up by roughly 75%, from 3.53% to about 6.17% after waivers, including an incentive fee you did not previously pay.
3. Your exit is still a quarterly, capped, potentially rationed tender — now under a fundamental policy that requires a shareholder vote to weaken, which is better, but SOFIX's own filings show its recent offers were prorated too.
4. The reorganisation was intended to be tax-free, so the swap itself should not have triggered a taxable event. Your basis carries over. Any tender you did get filled in an earlier quarter was a sale and is taxable in that year.
5. Check your cost basis before you compute your return. With 29.1% of distributions classified as return of capital, your basis is meaningfully lower than what you put in, and a naive "sale price minus purchase price" will understate your taxable gain. Our note on how return of capital and basis interact covers the mechanics.

The general lesson, which is not about Yieldstreet

The thing worth carrying away is structural, and it applies to every vehicle in this category.

A capped redemption programme is a promise about the numerator, not the queue. "Up to 5% quarterly" tells you the most the fund will buy. It tells you nothing about how many people will be asking. Those are independent numbers, and the second one is the one that determines whether you can actually get out.

The eleven-quarter table above is the clearest illustration we have. Nothing in it was hidden — every row comes from a filing the fund was required to make. But you would only see the pattern by pulling eleven separate Schedule TO amendments and putting them side by side, which is precisely the work nobody does, and which is why a shareholder could sit through eleven consecutive rationed quarters experiencing each one as an individual disappointment rather than as a trend.

Before you buy anything with a periodic repurchase programme — a non-traded REIT, an interval fund, a tender-offer fund — the question is not "what is the cap?" It is "what fraction of requests has this fund actually filled, quarter by quarter, for the last three years?" That number exists in the filings for every fund that has one. It is the single most predictive thing you can read, and almost nobody reads it.

Sources

Every figure above comes from a filing on SEC EDGAR. The load-bearing ones:

• Final Form NPORT-P, Yieldstreet Alternative Income Fund Inc., filed August 28, 2026, accession 0001049169-26-002339, period ended June 30, 2026, isFinalFiling = Y
• Form 425, filed June 29, 2026, accession 0001104659-26-078850 — the Transition Services Agreement, the sub-advisory agreement, and the investor email and FAQ
• Combined proxy statement and prospectus (Form 424B3), filed June 30, 2026 — the fee table, the performance table, the exchange ratio, the expense allocation and the Special Committee terms
• Schedule TO-I/A final amendments, one per quarterly tender offer, October 2023 through April 2026 — shares tendered and shares accepted
• Form N-CSR annual reports for fiscal 2022 and fiscal 2025 — the audited financial highlights
• Form 497AD, November 2020 — the non-accredited eligibility and the then-current $5,000 minimum
• Form 424B3, January 28, 2026 — the Willow Asset Management and Willow Wealth name changes
• SOFIX prospectus (Form 486BPOS), January 28, 2026, and its Form N-CSR and N-CSRS — the $1,000,000 minimum, the expense ratios and the proration disclosures
• SEC administrative order, September 12, 2023, Securities Act Release No. 11230 / Advisers Act Release No. 6414, File No. 3-21651

What we could not verify, stated plainly. The shareholder vote passed with "a majority of the outstanding shares"; no tallies were disclosed and closed-end funds file no results report, so we do not know the margin. SOFIX's proration percentages are not disclosed — its filings confirm oversubscription without giving the tendered totals. We found no statement on whether former AIF holders may add to their positions below the $1,000,000 minimum. And we could not read Ignites, FundFire, Reddit or several paywalled outlets, so our finding that the wind-down went uncovered applies to the press we could actually reach.