Investing in Art Through Fractional Shares and Funds

On June 17, 2026, Masterworks sent a letter. Its recipient was North Capital Private Securities, the broker-dealer whose alternative trading system, PPEX, had since January 27, 2025 been the only place an ordinary investor could sell a Masterworks share without waiting for the painting behind it to be sold. The letter terminated the agreements under which that trading ran, effective on or about December 14, 2026. It was disclosed, as Regulation A requires, on Form 1-U current reports filed with the SEC in July 2026 by the Delaware companies that hold the paintings, Masterworks Vault 14, LLC among them, documents read mostly by lawyers and by us.

The filings say the company intends to arrange an alternative, which may take the form of a bulletin board, a matching service or another mechanism, and that no assurance can be given that any will exist on the day PPEX closes. No provider, mechanism or launch date was named. Somewhere above a million registered members of a platform pitched as a way to own a Basquiat for the price of a dinner hold shares in single-painting LLCs whose only exit, from mid-December 2026, is the one the manager chooses.

Six years earlier the same platform had shown the other face of the trade. In October 2019 it offered investors Banksy’s Mona Lisa at $1,039,000; in October 2020 it sold the work for $1.5 million, and investors booked a 32% net annualized return on the platform’s first exit. Fox Business ran the story. That number, and a handful like it, built a business that by June 2026 had deployed roughly $1.2 billion into more than 500 paintings and signed up 1,061,245 members. Both facts are true at once: the wrapper can produce a spectacular realised return on one painting, and it can leave a million people with no market for the other 470.

This guide is about the wrapper: the single-painting LLC and the pooled fund, who runs them and who is on the other side of the trade, the fee stack, the exits and how to read them, the secondary market and its December 2026 cliff, the conflicts the SEC made Masterworks spell out, the funds that failed, the 28% look-through tax, a worked example in dollars, and the rules for evaluating any offering that lands in your inbox. The flagship guide, Investing in Fine Art, covers the market itself, the auction houses and their premiums, the segments, the records, forgery, storage and how a collector actually buys; the two figures you need from it, art’s real return and its round-trip cost, are restated here.

Every fractional pitch begins with an index, so the first thing to fix is what the underlying asset has actually returned, because a wrapper cannot create return, only subtract from it. The longest-window academic estimate, Dimson and Spaenjers’ study of art and other collectibles since 1900, puts art’s real return at roughly 2.4% a year through 2012, before any transaction cost. Over the same window equities returned about 5.2% real. Art beat bonds, bills and gold, which is why it appears in the wealth-manager decks, and lost to shares by nearly three points a year, which is why it does not appear in pension portfolios. That figure is the honest baseline: a low, positive, real return for an asset you also have to insure, store and sell through a 28% buyer’s premium.

The indices the platforms quote are higher, and the gap has a name. A repeat-sales index sees only the works that came back to auction, and works come back when they have appreciated; the losers stay on the wall, or go quietly through a dealer, and never enter the data. The Mei Moses index, the most cited of them, showed about 8.2% real from 1950 to 1999 on exactly that construction.

When Korteweg, Kräussl and Verwijmeren corrected for the selection in the Review of Financial Studies in 2016, using repeat-sales data on 32,928 paintings from 1960 to 2013 (the window in the working paper; a second version of it reports 20,538 paintings from 1972 to 2010, and neither changes the result), the annual return fell from 8.7% to 6.3%, the Sharpe ratio from 0.27 to 0.11, and the correlation with equities to roughly zero. Their conclusion, in plain terms, is that a broad passive portfolio of paintings is not an attractive investment. That is the number to hold in your head when a platform shows you a chart: the uncorrected index is 8.7%, the corrected one is 6.3%, and both are nominal and before fees.

The print market gives a third reading, and it matters here because a print is the cheapest way to own blue-chip art outright. Because prints are multiples, editions of 50 or 100 or more, they produce far more repeat sales than unique paintings, and James Pesando used that in 1993 to build a repeat-sales index of modern prints at auction from 1977 to 1992. The real return of the aggregate print portfolio was 1.51% a year; Picasso’s prints did slightly better at 2.10%. A print by a canonical artist, bought at auction and held for a decade, is roughly the same asset class as a share in a Basquiat, with the same low real return, no manager, no carry and a wall to hang it on.

The most recent year fits the pattern. Artprice’s Artprice100, an index of the hundred most-traded blue-chip artists rebalanced each January, rose 11.2% in 2025 and still trailed the S&P 500’s 17% over the same twelve months (Artprice, January 2026); Artprice’s own analysis puts the two series’ correlation from 2000 to 2026 at 0.88, the wrong number for a diversifier. Our tape stores the published year-end levels of that index, 91.7 on December 31, 2024 and 101.97 on December 31, 2025, and uses them later as a benchmark for the platform exits, never as a market-wide figure.

A wrapper is any legal structure that stands between you and a painting, and the structure decides who chooses the work, who sets its price, who can sell it, who pays the fees and who gets the gain. Four kinds are sold to outsiders, and the rest of the guide takes them in turn. The first is the single-painting vehicle: a company that owns one work and sells shares in it. Masterworks in the United States, ARTEX in Liechtenstein, Mintus in the United Kingdom and Arttrade in Germany all use it. You know exactly what you own, you have no diversification, and the manager decides when it sells.

The second is the pooled art fund, a private-equity-style vehicle that raises committed capital, buys a portfolio over an investment period, holds for five to ten years and returns cash as works are sold. The Fine Art Group, Anthea, Artemundi and Yieldstreet’s Art Equity Funds are examples. Minimums run from about $10,000 on a retail platform to millions at the traditional houses; lockups run three to ten years; fees follow the hedge-fund template of a management charge plus a share of profit.

The third is art-backed credit, where you do not own art at all but lend against it. Athena Art Finance, owned since 2019 by Yieldstreet (which renamed itself Willow Wealth on October 22, 2025 and appears under its old name throughout this guide), lends at roughly half of appraised value to collectors who would rather borrow than sell, and its Diversified Art Debt Portfolio passes the interest to retail investors. The fourth is the one nobody sells you: outright ownership of a print or a modest work, bought at auction or from a dealer, which is the comparator every other wrapper has to beat.

Who is on the other side matters more here than in most markets. When a fractional platform buys a painting it buys from a collector, a dealer or an auction house, and pays the buyer’s premium, which stood at 28% on lots up to $2 million at both Christie’s and Sotheby’s in 2026; it then sells you shares at a price that recovers that premium plus its own sourcing costs. When it sells the painting, it sells to the same collectors and dealers, who know precisely what it paid, because the purchase is on the public record in an SEC filing. The wrapper buys at retail and sells at wholesale, on a public timetable, through a manager paid from the profit. Every one of those facts is a cost or a conflict, and the rest of this guide prices them.

The mechanics of a Masterworks offering follow from one piece of securities law. Since 2015, Regulation A Tier 2 has let a company raise up to $75 million in a twelve-month period from the public, accredited or not, without a full registration, on condition that it files an offering statement (Form 1-A) that the SEC qualifies, an annual report (Form 1-K), a semi-annual report (Form 1-SA) and current reports (Form 1-U) for material events. A non-accredited investor may put no more than 10% of the greater of annual income or net worth into a Tier 2 offering. Masterworks built its business on that rule: each painting is held by its own Delaware limited liability company, which files its own Form 1-A, and the current filers are numbered “Masterworks Vault” entities, running to Vault 20 by mid-2026. What you buy is a Class A share in that LLC, priced at $20.00 a share, a figure the offering circulars say was arbitrarily determined by Masterworks, and the LLC’s only asset is the painting.

The manager’s job is then straightforward to describe. Masterworks’ research team selects artists whose auction records it judges to be rising, buys a work at auction or privately, files the offering, sells the shares to members, stores and insures the painting, and holds it for a target of three to ten years before selling it at auction or privately and distributing the proceeds, less fees, to shareholders. There are no dividends and no income; the return is price appreciation only.

The platform reported 1,061,245 members as of June 10, 2026 in its annual platform review of July 16, 2026, has acquired more than 500 paintings and deployed roughly $1.2 billion (its own SEC filings put the total at more than 450 works and $1.2 billion of purchase price as of March 31, 2025), and employed 217 people in late April 2026 (Tracxn). It has been, since its founding in 2017 by Scott Lynn and three co-founders, the only fractional art platform of any scale in the United States, and the terms in the rest of this section are its terms.

Our own tape reads the same paperwork the SEC does. Since September 1, 2026 we have run a daily full-text search of EDGAR for Masterworks filings, and in the 90 days to September 8, 2026 it logged 69 Form 1-U current reports and 9 new offerings. A 1-U is filed when something material happens to an LLC, a sale, a change of terms, a termination notice, and 69 of them in a quarter is a measure of how much administrative event a portfolio of 470 unsold paintings generates. Nine new offerings in the same window is the pace at which the machine is still buying. Neither is a market-wide figure; both are ours, dated September 8, 2026, and we watch them for the reasons set out in the final section.

Three charges stand between a Masterworks painting’s appreciation and your return, and their order matters because the first one is paid before the painting has moved. The sourcing markup comes first: the offering price of the shares is set above what the LLC paid for the painting, to recover the buyer’s premium, transport, insurance and Masterworks’ own acquisition work.

The offering circulars put a number on it. Each series’ offering size equals the estimated purchase price of the painting plus approximately 11% of that amount, about 10% of the maximum offering, paid to Masterworks upfront as an “Expense Allocation” (Masterworks Vault 2 and Vault 5 offering circulars, 2023–2025). The earlier, pre-Vault vehicles called the same charge a “true-up” and capped it at the lesser of 10% or the painting’s estimated appreciation over the pre-offering period, and third-party reviews in 2026 read the current terms at 10–11%, so the figure has held across two generations of paperwork. In plain terms, $10,000 of shares buys about $9,000 of painting on day one, and the work has to rise 11% before you are level.

The management fee comes second, at 1.5% a year, and it is paid not in cash but in newly issued shares of the LLC. That is a real cost with a soft appearance: your share count stays the same, but the number of shares outstanding grows by 1.5% a year, so your fraction of the painting shrinks by about 7.2% over a five-year hold and 14% over ten. Because the fee is charged on the whole vehicle whether or not the painting appreciates, it is the one charge that is certain.

The profit share comes third, at 20% of the gain when the painting is sold, calculated after the management-fee dilution. It is the carried interest of a private-equity fund applied to a single asset, with the difference that a fund’s carry is usually paid only above a hurdle rate and only after the whole portfolio has returned capital; a Masterworks LLC has one painting, so there is no cross-netting of a loser against a winner and, in the circulars, no hurdle: Masterworks holds Class B shares carrying a 20% profits interest from the first dollar of gain.

Add the three and the arithmetic is the whole argument of this guide. Over a five-year hold the markup costs about 2.2% a year, the management fee 1.5%, and the eventual selling commission, if the painting goes back through an auction house at a 10% seller’s rate, another 2%. That is roughly 5.7% a year before the 20% carry, on an asset whose honest real return is 2.4% and whose selection-corrected nominal index return is 6.3%. The stack does not merely reduce art’s return; at the index rate it consumes nearly all of it, and the profit share takes a fifth of whatever is left. Longer holds dilute the fixed charges, which is why the platform’s ten-year horizon is not a courtesy to the painting but a necessity for the fee model, and why a forced early sale is the worst outcome the structure allows.

A fractional platform’s track record is its list of paintings sold, and the way to read that list is to ask who chose which paintings to sell. Masterworks’ first exit set the template. It offered Banksy’s Mona Lisa to members in October 2019 at $1,039,000, sold the work in October 2020 for $1.5 million, and reported a 32% net annualized return; the second exit, George Condo’s Staring Into Space, sold for $2.9 million with an estimated 31.7% net IRR (company release, January 21, 2022). Those two numbers did more for the platform’s growth than any index chart, and both are real. The question is what they tell you about the other paintings.

By June 2026 the platform had sold 29 paintings out of more than 500 acquired and advertised a 17.0% median net annualized return across those exits, with individual results ranging from 4.1% (a Warhol held under a year) to 77.3% (a Cecily Brown sold after 259 days) annualized; its annual platform review of July 16, 2026 adds a 21.6% average and describes every exit as profitable. The platform presents these as net of fees; at least one 2026 review reads them as before the 20% profit share and the dilution, and we could not settle the definition from this desk, so treat the 17% as the platform’s number.

Three things follow. First, 29 of 500 is under 6% of the portfolio; the other 94% is marked, not sold, and the marks are the manager’s. Second, the exits are chosen by a manager who earns 20% of the profit on each painting separately, so it has every reason to sell a winner as soon as one appears and to hold anything that would print a loss; a median built from those choices is a survivorship statistic, not a forecast. Third, “net annualized” flatters short holds: a painting sold after twelve months at 32% has not compounded at 32% for a decade, and the 4.1% floor of the range is closer to what a long hold at index-level appreciation looks like after the fee stack.

Our tape carries a narrower reading of the same record: our own median exit IRR, computed on the exits the platform reports rather than on an independent count of the filings. A single read on July 16, 2026, the day the platform published its annual review, counted 8 exits in the preceding 400 days, and the median IRR on those eight stood at 11.5%. That is our construction, dated, and not a market-wide figure. It says only that the realised pace over the latest year was about seven paintings, and that the median on the most recent exits was around two-thirds of the advertised all-time median. Set the three numbers side by side, the advertised 17%, our 11.5%, and the selection-corrected index at 6.3%, and the honest planning figure for a new offering sits between the last two, before you subtract your own tax.

A share in a single-painting LLC has no natural buyer, and that fact governs everything about selling one. The painting itself has a market: collectors, dealers and two auction houses will bid for it, at their own price and on their own timetable. A $20 membership interest in the company that owns it has no such market, because nobody who wants a Condo wants one four-hundredth of a Condo they cannot see, and because the only people who know what the share is worth are the manager who appraised it and the other members. Masterworks’ own disclosures say its secondary market “frequently lacks liquidity”, which in practice means you can list shares and find no buyer at any price you would accept for months. That is not a defect of the platform; it is what a single-asset vehicle sold to retail is.

The venue that existed was borrowed, and it was already the second one. Until about January 22, 2025 the shares traded on an alternative trading system run by Templum Markets; from January 27, 2025 they moved to PPEX, an alternative trading system run by North Capital Private Securities, a registered broker-dealer, which gave members a regulated place to post bids and offers and clear a trade (Form 1-U, late 2024). On June 17, 2026 Masterworks delivered written notice terminating those agreements, and the arrangement ends on or about December 14, 2026. Covered shares stop trading through PPEX on that date.

The same filings say the company intends to arrange an alternative means of secondary liquidity, which may take the form of a bulletin board, a matching service or another trading mechanism, and that no assurance can be given that any alternative will be in place on or after that date or will match the liquidity PPEX offered. As of September 10, 2026 no provider, mechanism or launch date had been confirmed. What a member holds on December 15 is, until then, a position whose exit is the manager’s decision to sell the painting.

The pooled funds are worse on paper and more honest about it. A traditional art fund takes committed capital for a fixed term, typically five to ten years, with an investment period during which money is drawn and a harvest period during which works are sold and cash is returned. There is no secondary market at all, only the occasional negotiated transfer of a limited-partner interest at whatever discount a buyer demands. Yieldstreet’s art funds and debt portfolios are similar: you are told the target term, and you wait. The difference between the fund and the fractional share is that the fund never told you there was a way out, so nobody plans around one.

The structure of a single-painting vehicle puts the manager on every side of the trade, and the disclosure fight over that is on the public record. Masterworks’ entities do not hire an independent appraiser to value the paintings between purchase and sale; the appraisals are performed by employees of the Administrator, which is Masterworks itself, and the offering circulars acknowledge that the resulting fair-market-value figure may differ from what an independent third party would produce. That figure is what a member sees as the value of a holding, what a secondary-market buyer prices off, and what the platform’s unrealised return claims rest on. The same firm sources the painting, sets the markup, marks the value, decides the sale date and takes 20% of the result.

On August 8, 2023 SEC staff wrote to Masterworks’ counsel, Joshua B. Goldstein, on the Masterworks Vault 2 offering, and asked the company to revise its risk-factor section to discuss the conflicts of interest in greater detail; the company answered on August 14, 2023 by adding a risk factor in its fourth post-qualification amendment, and the correspondence is filed on EDGAR. The substance of the exchange is on the record: the staff wanted the reliance on an affiliate to appraise the art described as the conflict it is, not as a potential one. The circulars filed since 2023 carry that language.

None of it is an allegation of wrongdoing, and the platform has operated since 2017 under continuous SEC review, which is more than any traditional art fund can say. It is simply the honest description of a business in which the person telling you what your share is worth is paid from the difference.

The corporate history is thinner in the public record but not blank: ARTnews reported in December 2022 (updated March 2023), from interviews with more than 20 current and former employees, on internal rifts, alleged recklessness and staff cuts at the company, and recorded that by December 2022 the platform had sold 11 of 118 works, with realised returns of 10.4% to 39.3% after fees; Tracxn’s count put headcount at 217 in late April 2026. A platform whose revenue is a fee on a fixed pool of paintings, carry on exits it controls and a markup on new offerings needs to keep buying to keep growing, and our tape’s nine new offerings in the 90 days to September 8, 2026 say that it is. The conflict to price is not fraud, of which there is no evidence, but incentive: the manager is paid to acquire, to hold and to sell winners, and none of those is the same as being paid to make you money.

The traditional art fund predates the fractional share by three decades, and its economics are the private-equity template with a painting in the portfolio. The Fine Art Group, founded in London in 2001 by Philip Hoffman, formerly of Christie’s, is the reference case and the largest survivor. Between 2001 and 2021 it ran nine art investment funds on five- to ten-year terms, and it states that its blue-chip funds achieved an average 15% compound annualized return over that period, a self-reported figure that we could not audit and that, in an earlier form, Hoffman put differently: in 2015 he told CNBC that, based on assets sold plus the estimated value of works not yet sold, the funds had produced an average return of about 9% before fees. Both numbers can be true of different funds and different dates; the gap between a realised-plus-marked 9% gross and a marketed 15% net is the reason to ask which.

Fees at the traditional houses, per the same CNBC reporting, run 1–3% a year plus 20% of profit, with The Fine Art Group taking its share only after investors have earned a 6% hurdle. Minimums are in the millions and lockups three to ten years (The Art Newspaper, June 23, 2025). One unusual term is real and worth the price to some: investors may borrow works from the fund and hang them at home.

The continental funds are smaller and the record shorter. Anthea Art Investments launched the Anthea Contemporary Art Investment Fund, a Luxembourg SICAV-FIS, in April 2013 to buy Post-War and Contemporary work “whose potential growth in value is still unexpressed”; by September 2014 it held 36 works by 13 artists, 35% of them emerging, and reported a 23.4% return since launch with its best position up 404.3% and its worst down 16%. A successor vehicle, Anthea CAIO, followed; we found no audited performance for either after 2015. Artemundi, Swiss-based, runs traditional funds with high minimums and, with Sygnum Bank, has tokenised single works into interests priced from CHF 1,000 a token (the 2021 Picasso carried a CHF 5,000 minimum subscription). Arte Collectum was launching a new fund in 2025 (The Art Newspaper, June 2025); its terms were not verifiable for this piece.

The canonical success is older than all of them: the British Rail Pension Fund put about £40 million, roughly 3% of assets, into some 2,400 works from 1974 and earned about 11% a year nominal on them, which was about 4% real because it rode the high-inflation 1970s, while its equity portfolio compounded faster over the same span (Britannica). Even the famous art-fund success underperformed shares.

The industry’s own barometer records the shape of the demand. Deloitte Private and ArtTactic’s Art & Finance Report 2025 finds that 51% of wealth managers offer art-related services, against a quarter in 2011, and estimates that $992 billion of art and collectibles will change hands in the coming decade’s wealth transfer; the same report puts art-secured lending at an estimated $34–40 billion outstanding at loan-to-value ratios of 50–60%. Those are the numbers that launch funds. What they do not show is the assets the funds have actually gathered, which the report’s earlier editions put at a peak of a little over $2 billion in 2012, about 70% of it Chinese art trusts, falling nearly 40% to about $1.3 billion by 2014 as Beijing tightened its shadow-banking rules (Deloitte & ArtTactic, via Artnet News, 2014; the 2017 edition records the Chinese decline continuing through 2016). Funds launch in every up-cycle and close in every down-cycle; the next section is about the closings.

The way an art fund fails is rarely that the paintings fall; it is that the manager does, and the cases are specific enough to learn from. Fernwood Art Investments was founded in 2003 (2002 by some accounts) by Bruce Taub, a twenty-year Merrill Lynch veteran, as an institutional-grade art fund manager, complete with a Harvard Business School case study; its first funds were due to launch in February 2005. By spring 2006, according to the complaint later filed against him, investors who pressed Taub for financial information discovered that all their money had been spent without Fernwood having launched a single fund. It shut down in the fall of 2006. On February 28, 2007, 28 of its backers sued Taub in the Southern District of New York, alleging he had embezzled $8 million of company funds to promote himself and his wife in the art world and to pay personal expenses (ARTnews, April 17, 2007). The firm’s paper was excellent. Its bank account was the problem.

Castlestone Management’s Collection of Modern Art fund launched in 2009, at the bottom of the financial crisis, with a promoter, Angus Murray, telling the press that art prices were set to appreciate, and a strategy of buying museum-quality Post-War work; it reported $28 million under management by the end of May 2009.

The UK Financial Services Authority raided Castlestone’s London and Chichester offices in July 2011. On August 15, 2011 the British Virgin Islands Financial Services Commission issued a directive to the funds under Castlestone Management Incorporated, Collection of Modern Art Inc. among them, and brought in Grant Thornton to investigate. From October 2011 the UK manager, Castlestone Management Limited, was in liquidation and administration, with administrators reporting that it could not repay its creditors, the art fund among them, and the paintings that had hung in the London office were removed by people claiming to act for the BVI entity (IFA Singapore). Investors who had bought a diversified art portfolio held, in the end, a creditor’s claim in the liquidation of the fund’s own manager.

The Artist Pension Trust, launched in 2004 by MutualArt as a pension scheme in which artists contributed works to pooled trusts, made its first cash distributions, of $200 to $1,700 each to more than 400 artists, only in 2016, and a 2021 New York Times investigation detailed multiple failures of the scheme’s promise of diversified income. The concept was sound; the money did not arrive.

The market itself has done the rest in every cycle. Time magazine’s “Great Massacre of 1990” recorded the end of the contemporary auction wave that crested in late 1989, in which Sotheby’s all but stopped lending to buyers after Alan Bond defaulted on Van Gogh’s Irises, bought in 1987 for $53.9 million with a $27 million loan from the auction house. Funds raised at that top, and at the 2007 top, and at the 2021 top, bought at the peak of a market whose corrected index returns 6.3% a year across the cycle and whose drawdowns run for years. The three lessons are the same in each case: the custody of the money and the custody of the art must both be independent of the manager; the track record that counts is audited and realised, not claimed and marked; and a fund launched because the market is rising is a fund that bought high.

Outside Masterworks the wrappers are smaller, differently regulated, and each has one fact that decides whether it belongs in a portfolio. Yieldstreet, which renamed itself Willow Wealth on October 22, 2025 (its legal entities and SEC filings keep the old name, and so does this guide), is the platform with the most art money behind it and the most public regulatory history. It bought Athena Art Finance, a specialty lender against blue-chip art founded in 2015 with about $280 million of backing led by The Carlyle Group, for $170 million in April 2019 (Artnet News; The Art Newspaper), and says it has funded more than $500 million of art financing since.

Its retail products are two. The Art Equity Funds, launched from November 2021 with a 15–18% target annualized return, hold Post-War and Contemporary works by artists such as Keith Haring, George Condo and Kenny Scharf, and the launch materials supported the target with Athena’s proprietary database showing those artists’ average compound growth since 2015 at 14.3%, 36.1% and 29.4% respectively, which is vendor data about past prices, not a return anyone earned net of fees.

The Diversified Art Debt Portfolio pays monthly interest at a 9.5% target yield from a pool of loans secured, per the offering page, on 156 works by 68 artists (the collateral count changes as loans roll; the pool held 106 works by 82 artists at the end of 2022). We could not verify realised results for either as of September 2026, and the rebranded site removed the platform’s historical track record at the name change, against a cumulative investor loss across its products that Crowdfund Insider put at more than $200 million (January 2026); treat both targets as targets.

The regulatory record is the fact to weigh. In September 2023 the SEC ordered Yieldstreet to pay $1.9 million for failing to tell investors in a $14.5 million September 2019 offering, which financed the transport and scrapping of a retired ship, that ships securing its other loans had been reported as already deconstructed or could not be located because their tracking systems were off; the firm settled without admitting or denying, and the money went to investors. A class settlement with investors in three of its funds, preliminarily approved at up to $9.95 million including a $6.2 million cash fund and $2.75 million of waived fees, followed; InvestmentNews reported separately that the firm kept about $5 million from a ship-scrapping settlement to cover its legal fees. None of this concerns art. All of it concerns the diligence of the platform selling the art, and the ship loan is why this guide asks, of every wrapper, who checked the collateral.

ARTEX took the opposite route: a regulated stock exchange for single works, an MTF licensed in Liechtenstein with Euroclear settlement. Its first and, as far as we could find, only listing was Francis Bacon’s Three Studies for Portrait of George Dyer, 1963, valued at $55 million against the $51.7 million it made at Christie’s in 2017, offered at €92.20 (about $100) a Class B share, with trading from March 8, 2024. A second listing, a 1972 Warhol Mao at a €39.5 million valuation through a vehicle called Art Share 004, offering up to 400,000 Class B shares by private placement to professional investors, was announced as in preparation; we found no record of it trading by September 2026.

Mintus, the first company authorised by the UK Financial Conduct Authority to fractionalise investment-grade contemporary art, launched in 2022, distributed through Amicorp Wealth Management as an actively managed certificate with a single ISIN, and has since repositioned itself as an AI platform for banks seeking to grow assets under management; we found no published exit. Arttrade, founded in Düsseldorf in 2021 by Svenja Heyer, David Riemer and Julian Kutzim, sells fractional interests from €1,000 in works by Richter, Baselitz and Uecker as regulated electronic securities with their own ISIN, with a secondary market planned but not yet available (Crowdinform, 2026), and had five employees on January 31, 2026 (Tracxn). Each is a serious attempt at the structure; none has a realised track record an outsider can check, and a single-listing exchange is a listing, not a market.

A share in a fractional art vehicle is taxed as if you owned the painting, and that is the single most expensive fact in the structure after the fees. Each Masterworks LLC is taxed as a partnership, so every year you hold shares in a painting, whether or not it sold, you receive a Schedule K-1 (Form 1065) for that painting; ten paintings mean ten K-1s, often arriving late enough to force an extension. When the painting sells after a hold of more than a year, the gain passes through to you as collectibles gain under IRC §1(h)(4)–(5), taxed at your ordinary rate capped at 28%, not the 15–20% that applies to a share of an index fund. Masterworks’ own guidance says it plainly: an investor whose ordinary rate is 18% pays 18% on the gain, one whose rate is 32% pays 28%; and an investor who owns 10% or more of a single painting vehicle is taxed at the ordinary rate without the cap. Gains on a hold of a year or less are ordinary income at any rate up to 37%.

The look-through follows you into the secondary market. Treasury Regulation §1.1(h)-1 provides that when you sell an interest in a partnership held for more than a year, the part of your gain attributable to the partnership’s collectibles is treated as collectibles gain and taxed at 28%, computed as if the partnership had sold the painting for fair value the moment before you sold your share. Selling shares on PPEX, then, was never a way to convert art into a 20% asset. On top of the 28% sits the 3.8% net investment income tax under §1411 for single filers above $200,000 and joint filers above $250,000 of modified adjusted gross income, for an effective top federal rate of 31.8%, before any state tax. The 2017 tax law removed art from §1031 like-kind exchange, so a gain cannot be rolled into another work, and the One Big Beautiful Bill Act, signed July 4, 2025, made the 2017 rates permanent and left the 28% cap, the NIIT and its thresholds untouched (P.L. 119-21).

The offshore wrappers add a layer rather than remove one. A Luxembourg SICAV, a Liechtenstein share or a German electronic security held by a US taxpayer is a foreign entity holding collectibles, and the passive foreign investment company rules can convert a long-term gain into ordinary income plus an interest charge under the default excess-distribution regime, with a Form 8621 due each year and elections (QEF, mark-to-market) that change the result only if made in time; take advice before buying any non-US wrapper.

Two planning facts that work for a collector do not work for a shareholder. Donating appreciated art to a museum can yield a fair-value deduction and avoid the gain; donating a K-1 interest in an LLC is a different and worse transaction. Holding until death gives heirs a stepped-up basis on a painting on the wall and, in principle, on a partnership interest, but a fractional position with a ten-year manager-controlled exit is an odd thing to plan an estate around. None of this is tax advice; the rates change with your situation, and the point is that the wrapper does not change them.

One rule does follow from the paperwork alone: a $500 stake that generates a K-1 costing $50 to prepare carries a 10% annual charge before the platform has taken anything, so below about $5,000 per painting the administration makes the position uneconomic whatever the painting does, and the diversification across many works that the platforms advertise is, for a small investor, a diversification across tax forms.

A $10,000 subscription at Masterworks’ terms buys $9,009 of painting at the price the LLC paid, because the sourcing markup of about 11% comes off before anything else; the arithmetic below follows the rest of that $10,000 through the dilution, the carry and the tax, with every assumption stated so you can change it. The management fee of 1.5% a year is paid in new shares, so after five years the shares outstanding have grown by a factor of 1.015 to the fifth power and your fraction of the painting is 92.8% of what it was. When the painting sells, assume it goes back through an auction house at a 10% seller’s commission, the standard assumption across this hub; Masterworks also sells privately, and if it negotiates a lower cost the numbers improve by that amount. The profit share is 20% of whatever comes back above your $10,000. Federal tax is the 28% collectibles rate plus 3.8% NIIT, no state tax.

Now let the painting appreciate at 6.3% a year, the selection-corrected index rate. After five years it is worth $9,009 × 1.357, or $12,228. The 10% selling commission takes $1,223, leaving $11,005. Your diluted share of that is $10,216. The profit share takes 20% of the $216 gain, $43, and you receive $10,173, a total return of 1.7% over five years, or 0.34% a year before tax. Federal tax at 31.8% on the $173 gain is $55, and you hold $10,118: about 0.2% a year, nominal, on a painting that did exactly what the honest index says paintings do. Had the same $10,000 gone into an S&P 500 index fund that also happened to grow at 6.3% a year, with a 0.03% expense ratio, it would be worth $13,554; tax at 23.8% on the $3,554 gain leaves $12,708, or 4.9% a year. The asset returned the same in both cases; the wrapper took the difference.

The advertised case looks different, and the difference is the manager’s skill, not the structure’s efficiency. If the painting appreciates at 17% a year, the platform’s advertised median, it is worth $19,752 at sale; after the commission, dilution and a $1,300 profit share you receive $15,202, 8.7% a year, and after tax $13,548, or 6.3% a year, which beats the index fund’s 4.9% at the lower growth rate. At 10% appreciation, a good decade for blue-chip art, you receive $11,697 before tax and $11,157 after, 2.2% a year. Work it backward and the hurdle is precise: for the fractional share to match an index fund growing at 6.3%, the painting has to appreciate at about 14.7% a year for five years, gross, before any of the platform’s charges. That is more than double the corrected index, above our tape’s 11.5% exit IRR, and below the platform’s advertised median. Whether a manager can pick paintings that do that, consistently, on a portfolio of 500, is the whole bet.

Two sensitivities belong in your head. Lengthen the hold to ten years at 6.3% and the fixed charges thin out but the dilution deepens: the painting is worth $16,596, you receive $12,297 before tax and $11,567 after, 1.5% a year, still a third of the index fund at the same growth. Cut the selling cost to 5% on a private sale and the five-year after-tax result rises from $10,118 to about $10,430, or 0.8% a year. No assumption inside the platform’s own terms gets a painting growing at the corrected index rate past an index fund growing at the same rate; only the painting outperforming does that.

No wrapper delivers the financial return of art at less than the cost of owning art, and the numbers above say why: every wrapper adds a manager to a round trip that was already 25–40% and an asset whose corrected index return is 6.3% nominal. What a wrapper can deliver depends on what you actually want. The index fund comparison is not close: the S&P 500 returned about +313% with dividends over the ten years to August 31, 2026, roughly 15.2% a year (S&P Dow Jones Indices data, as used across this hub), at a 0.03% expense ratio and a 20% top capital-gains rate. The Artprice100, the blue-chip art index, rose 11.2% in 2025 alone and still trailed the S&P’s 17% that year, and that was before anyone’s fees.

If what you want is a piece of art, the print is the wrapper the market does not advertise. A signed edition print by Warhol, Picasso, Hockney or Banksy can be bought at auction for four or five figures, hung, insured on a household policy and sold at the same houses, and Pesando’s index says the asset returns about 1.5% real over a long hold. The sister guide, Investing in Prints and Editions, covers how to choose, authenticate and price one; this guide only prices it as the comparator.

The print loses to the fractional share on paper over five years, $9,544 against $10,118 in the example, because it pays the whole 28% buyer’s premium up front and the 10% commission at exit, and wins over ten years, $12,014 against $11,567 after tax, because it pays no management fee and no carry. Neither gets near the index fund. The print has one thing the share does not: you choose when to sell, and the pleasure of ownership, which the academic literature identifies as most of art’s return, accrues to you rather than to a vault.

The fractional share and the fund win in one case, and it is a real one: when the manager buys a painting that appreciates far faster than the index, as the Banksy did in 2020 and as the 29 Masterworks exits did at a 17% median. That is a bet on selection skill in a market where the evidence for persistent selection skill is thin and where the person reporting the skill is paid from it. Taken as that bet, sized for ten years of illiquidity, a fractional position is a legitimate speculation. Taken as cheap access to art’s long-run return, it is a way to pay 5.7% a year for a 2.4% asset.

Every offering circular answers eight questions if you know where to look, and the answers rank the offering before you have a view on the painting. Take them in order; the first three are disqualifying.

1. What did the vehicle pay for the painting, and what are you paying for the shares? The purchase price is in the Form 1-A’s use-of-proceeds and description of the artwork; the offering size is on the cover. The ratio is the markup. At 11% you are behind by that amount on day one; above 15% walk away, because the painting must then outrun the corrected index for two years to reach par.
2. How is the management fee paid, and on what base? A fee paid in shares dilutes you silently and compounds; a fee on the appraised value rises with the manager’s own mark. At 1.5% in shares over ten years you give up about 14% of the painting; confirm the number and the mechanism in the “Management fee” section of the circular.
3. What is the profit share, and is there a hurdle? Private-equity carry is paid above a preferred return, typically 6–8%, and after capital is returned across the fund. A single-painting vehicle with a 20% carry from the first dollar and no hurdle is a worse deal than any fund on the same headline terms.
4. Who appraises the painting between purchase and sale? If it is an employee of the manager, the “value” on your dashboard is a conflicted number; the August 8, 2023 SEC correspondence explains why that matters. Independent annual appraisal is rare and worth paying for.
5. What is the holding period, who decides the sale, and can you vote? The circular states a target (3–10 years at Masterworks) and gives the manager sole discretion. Assume the long end.
6. Is there a secondary venue, who runs it, and when does the contract end? The PPEX arrangement had a start date (January 27, 2025) and, from June 17, 2026, an end date (about December 14, 2026). Look for both in the current filings before you assume any exit besides the painting’s sale.
7. What tax form will you receive, and how many? A K-1 per painting per year for a partnership; a 1099 for a fund taxed as a corporation, which has its own costs. Price the preparation.
8. What is the artist’s actual auction record? The Artnet Price Database and Artprice show what comparable works by size and period have made, and How Art Auctions Work, the sister guide on auction mechanics, explains how to read a guaranteed lot. If the vehicle paid above the recent comps, the markup is on top of a premium.

The sequence assumes you have read the rest of this guide, want a manager’s selection of blue-chip art anyway, and can hold ten years.

1. Set the size first. Cap the total across all fractional and fund positions at 2% of investable assets, and put the rest of any art allocation into an index fund or a print, per the IA Take above. Hold nothing you cannot lose entirely without changing a plan.
2. Confirm your investor status and the 10% rule. A non-accredited investor is limited to 10% of the greater of income or net worth per Tier 2 offering; the platform will ask you to self-certify. Accredited status opens Yieldstreet’s funds and the traditional houses, at higher minimums.
3. Read the Form 1-A on EDGAR, not the app. Search SEC full-text search for the vehicle’s name; note the purchase price, offering size, fee mechanism, profit share, appraiser, holding period and secondary-market language. Then answer the eight questions in the previous section on paper.
4. Check the painting against the comps. Pull the artist’s auction history for the same size and period from Artnet or Artprice, and compare with the vehicle’s purchase price. A price above the comps plus an 11% markup is a painting that must beat the index for years to break even.
5. Subscribe to no more than a few vehicles, each above $5,000. The K-1 arithmetic makes small stakes uneconomic; the fee stack makes many stakes no more diversified than few, because every vehicle carries the same charges.
6. Set up the paperwork and the watch. Tell your preparer to expect K-1s and an extension; put the vehicle’s EDGAR filings on an alert so you see every Form 1-U, including the ones that change the terms; and record the platform’s appraised value each year against the Artprice100 so you can tell a mark from a market.
7. Plan the exit as the manager’s, not yours. Note the holding-period end in your records and expect no liquidity before it. If you need the money earlier, the position was too large.

The figures below are the ones that would change the view in this guide, with the reading and the date that would do it, as of September 10, 2026.

• December 14, 2026. The PPEX arrangement ends. The July 2026 filings say a bulletin board, matching service or other mechanism is intended, with no assurance. A replacement venue announced before that date, with a named broker-dealer and regulated ATS, restores the secondary market as it was; a bulletin board without a broker-dealer is a noticeboard, not a market; no announcement by December 15 means every Masterworks share is a hold-to-sale position, and this guide’s sizing rule becomes the platform’s own.
• Masterworks’ exit count. It stood at 29 in June 2026 against 500-plus paintings; 50 is where the advertised median starts to mean something, and the first reported exit at a loss would be the more useful data point, because it would say the manager reports the full sample.
• Our tape. Our median exit IRR on the platform’s reported exits read 11.5% on July 16, 2026, on 8 exits in 400 days; a read below 8% with the same count would put realised returns at the corrected index rate, and the wrapper’s fees would then exceed the whole return. The 1-U count, 69 in the 90 days to September 8, 2026, and the offering count, 9, are the pulse of the machine; offerings falling to zero for two consecutive quarters would mean the platform has stopped growing, which under a fee-in-shares model changes its incentives on the existing portfolio.
• The Artprice100 at year-end 2026. Our tape closed 2025 at 101.97; a year-end print below 100 puts the blue-chip index back under water for the cycle, and every unsold painting in every vehicle with it.
• The Art Basel & UBS 2027 report, March 2027. Global sales were $59.6 billion in 2025, up 4%, with auctions up 9% to $20.7 billion (report of March 12, 2026); a 2026 total above $60 billion would be the first back-to-back growth since 2021–22 and the environment in which funds launch, which is the environment in which to be careful.
• ARTEX’s second listing. A Warhol Mao at €39.5 million was in preparation in 2025; a listing that trades, with a published daily price, would be the first evidence that an exchange for single paintings can be a market rather than an event.
• Realised results from Yieldstreet’s Art Equity Funds. The 2021 target was 15–18%; a first full-cycle realised figure, net of fees and audited, is the number that would let the pooled model be judged on something other than a database of past artist prices.
• The Deloitte & ArtTactic report. Its next estimate of art-fund assets under management, against a peak near $2 billion in 2012, is the cycle indicator for the whole category.

This guide is a description of a set of financial structures, their costs and their record as we could verify them on the dates given.