Investing in Independent Watchmakers

On December 6, 2025, at Phillips in New York, a wristwatch that had never been sold at retail made $10.8M. It was the FFC prototype: an F.P. Journe with no hands and no dial in the usual sense, showing the hour with a mechanical hand whose fingers fold and unfold, built after a conversation between François-Paul Journe and Francis Ford Coppola, and consigned by Coppola himself. The figure beat every price ever paid for a watch by a living independent maker, and it beat most Pateks.

Four years earlier, on November 5, 2021, Philippe Dufour’s Grande et Petite Sonnerie No. 1 had made $5,182,109 at Phillips in Geneva. Dufour was then 73 and had made eight of them across his career. That sale set the independents’ auction record and held it until Coppola’s Journe — the qualifier matters, because a later example of the same model, No. 3, changed hands privately through the London dealer A Collected Man at a reported $7.3M, a price no saleroom ever published. The two auction results bracket the whole thesis of this guide: an object made by one identifiable person, in a quantity you can count on your fingers, can be priced by a room full of collectors at a level no marketing budget has ever produced.

What the two records do not tell you is how narrow the road to them is. Between them, the makers this guide covers ship perhaps two thousand watches a year, and the mechanism that sets their prices is not advertising or even scarcity in the abstract; it is a waiting list controlled by a person who chooses who gets one.

The hub’s flagship guide, Investing in Luxury Goods, covers the boutique economy at large. The sister guide, Investing in Luxury Watches, covers the wider watch market, its indices and its auction machinery. Investing in Rolex and Investing in Hermès cover the two houses that industrialised the waiting list, and Investing in Signed and Vintage Jewellery covers the other object whose value sits in a maker’s signature. This guide is about the makers who never industrialised anything, what they have actually returned, and what the round trip costs.

An independent watchmaker, in the sense this market uses the word, is a house that is not owned by Swatch Group, Richemont, LVMH or Kering, whose annual output is measured in dozens or low hundreds, and whose movements are finished by hand to a standard a trained eye can see. Ownership alone is not the test. Rolex is independently owned by a Geneva charitable foundation and Patek Philippe is family-owned, but both are industrial manufacturers, and neither trades in this market. Richard Mille is independent by any structural definition and sells at six figures, but it is a marketing-led brand with a very different, more fashion-exposed price behaviour, which is why the sister guide treats it separately.

The working definition collectors use has three parts, and it is worth internalising because it is the only durable filter in a field full of new names. First, a named person whose hand is in the product and whose reputation is the asset. Second, finishing that is done with a rotating tool and a human wrist rather than a machine, which is what a black-polished steel bevel or an anglage with sharp interior angles actually signals. Third, output small enough that the maker cannot meet demand at the list price, which is what creates the second market where the returns in this guide are earned or lost.

By that filter the investable set is short. The names that clear reliably at auction are F.P. Journe, Philippe Dufour, De Bethune, Rexhep Rexhepi (whose house is Akrivia), Kari Voutilainen, MB&F, H. Moser & Cie and Greubel Forsey. Behind them sits a frontier tier that is genuinely new and genuinely unproven, covered in section nine. The formal association most of these makers belong to or grew out of, the Académie Horlogère des Créateurs Indépendants, was founded in Geneva by Svend Andersen and Vincent Calabrese in 1985 — some accounts date the constitution to 1984 — to give one-man workshops a stand at the Basel fair; the fact that a trade body was needed at all tells you how marginal this business was for its first two decades.

Scale is the first thing to get right, because almost every mistake made in this market is a liquidity mistake dressed up as a valuation mistake. Swiss watch exports were CHF 25.5B in 2025, down 1.7% in a second consecutive annual decline, on volumes of 14.6 million units (Federation of the Swiss Watch Industry, January 2026); the first half of 2026 ran at CHF 12.8B, down a further 0.7% (FH, July 2026). Against that, Morgan Stanley and LuxeConsult’s 2025 industry report, published in March 2026, put F.P. Journe, H. Moser and MB&F each somewhere in a CHF 50–125M band — that is, the three most commercially successful purist independents together are roughly one percent of the industry, and Journe alone is perhaps a hundredth of Rolex.

The unit numbers are smaller than the revenue numbers suggest, and every one of them is an estimate rather than an audited disclosure: none of these houses publishes accounts. On Morgan Stanley’s estimate F.P. Journe stays under a thousand mechanical watches a year, with the quartz Élégante line taking total output to roughly 1,900 pieces in 2025, while MB&F runs at about 300 and has said it is working towards 500. De Bethune’s own manufacture describes up to 350 against trade estimates of 150–250; Greubel Forsey, which once targeted 500, runs at about 200 on its chief executive’s own account; Voutilainen makes roughly 60 and Akrivia fewer than 40.

Philippe Dufour is effectively out of series production altogether: the Simplicity, his best-known watch, ran to 204 pieces across some seventeen years and stopped. Treat all of these as orders of magnitude rather than measurements, because every one is a third-party estimate or an interview remark and not a filing. The order of magnitude is the point.

What matters about a market this size is that the price you read is thin. A steel Rolex Submariner has thousands of comparable sales a year and a published index; a Voutilainen has perhaps two or three public sales a season, and the price of the one you own is the price of the last one somebody else chose to sell. When a maker is in favour the thinness works for you, because a single strong result reprices everything. When it turns, the same thinness means there is no bid at all for months, which is the risk that does not show up in any return series.

The one structural tailwind is real and worth naming. The very top of the Swiss market has been taking share from the middle for years: watches priced above CHF 50,000 were 1.4% of export volume and 37.3% of export value in 2025, and accounted for the large majority of what growth there was (Morgan Stanley × LuxeConsult, March 2026, via trade coverage). Independents sit almost entirely inside that 1.4%. The money has been moving up the price curve, and that is the current the independents have been swimming with.

The honest way to look at an independent’s record is at auction, because that is the only place where prices are public, contested and dated. The chart below sets the two independent records against the watches that top the all-time list, and the comparison is the useful part: the FFC at $10.8M in December 2025 sits a whisker below the steel Patek 1518 that made $11,136,642 in 2016, well below the Grandmaster Chime’s $31.19M (Christie’s, Only Watch, November 9, 2019) and the Henry Graves Supercomplication’s $24.0M (Sotheby’s, November 2014), and just ahead of the ref. 2523 that made $10.2M in Geneva in May 2026. Since December 2025 one living maker has been inside the top ten. Ten years earlier none was.

Underneath the trophies, the volume story is the auction houses’ own. Phillips, whose watch department has been run by Aurel Bacs since 2015, reported $370M of watch sales in 2025 including private treaty, of which $290,463,315 was hammer-plus-premium at auction, a fifth consecutive year above $200M; its Geneva Watch Auction XXIII on May 9–10, 2026 made $96.3M (CHF 74.8M) with 224 of 225 lots sold, the largest watch auction ever held. Sotheby’s watch sales were $193.6M in 2025, up 22% (IDEX, December 2025), and Christie’s luxury division as a whole passed $1B (WatchPro, 2025), though it does not publish a watches-only total. Those three houses are effectively the entire public market for an independent watch.

Below the two records sits the number that is actually useful to an owner — the level a living independent’s serial production reaches — and Rexhep Rexhepi is the clearest case. His unique Chronomètre Antimagnétique made CHF 2.1M (about $2.3M) at the Only Watch charity sale in Geneva on May 10, 2024. In the same Geneva week a platinum Chronomètre Contemporain from 2019 — a serial watch, one of twenty-five — made $1,274,852 at Phillips, a record for the reference, having made US$924,000 in pink gold at Phillips Hong Kong the year before (Bloomberg and SJX, May 2024).

Dufour’s serial work prints at the same order of magnitude: a steel Simplicity made HK$9,461,500, about US$1.2M, at Phillips Hong Kong in March 2026. Hold those two alongside the $10.8M headline and the shape of the market becomes clear. A million dollars, not ten, is what a good example by a celebrated living independent is worth; the eight-figure prices belong to unique pieces with a story attached, and section four explains why you should not price your own watch off them.

The arc these figures describe is a decade long and it is genuinely unusual. In 2015 an independent maker’s watch was a connoisseur’s object with a thin, largely private market. By 2021 a Dufour was a five-million-dollar lot. By the end of 2025 a living maker held a top-ten record, and Morgan Stanley was reporting that the three largest purist independents grew revenue in a year when the Swiss industry shrank. There is no comparable re-rating anywhere else in collecting over the same window.

Every figure in the previous section is a sale that happened, which is precisely what makes the set unrepresentative. There is no index of independent watchmakers. Knight Frank’s Luxury Investment Index carries a watches line, and it is a good long-run reference — +125.1% over the ten years to end-2024, +1.7% in 2024 and +5.1% in 2025 (Knight Frank Wealth Report, 2025 and 2026 editions) — but it is built from WatchCharts data dominated by Rolex, Patek Philippe and Audemars Piguet, the three brands with enough transaction volume to index. The independents are too thin to appear. When somebody tells you independents returned a number, ask which watches were in the denominator, and you will usually find the answer is “the ones that sold”.

Four biases work in the same direction, and they are worth naming because they apply to every collectible market and are unusually severe here.

• Selection. An owner consigns when they expect a good price and holds when they do not. The auction record is therefore a record of the moments owners felt strong, not of what the average watch is worth on an average Tuesday.
• Survivorship. The makers in section one are the ones who lasted. The market of the 1990s and 2000s contained plenty of one-man workshops that folded, and their watches do not appear in anybody’s return series. Picking the survivors twenty years early is the entire skill, and the record does not show you the attempts that failed.
• Provenance premiums. The two records that define this market both carried something other than the watch: Coppola’s name on the FFC, and the fact that the Dufour was number one of eight. Neither premium is available to a buyer of a normal example, and both are quoted in trade coverage as though they set a level for the maker. They do not.
• The costs the record omits. An auction result is the buyer’s all-in price. It is not the seller’s proceeds, which are net of commission, and neither figure is net of the tax that section twelve prices. A headline that says a watch “doubled” is describing gross prices at two moments, with a friction gap between them that section thirteen measures at roughly three-quarters of the original hammer.

The wider watch market gives you the honest cyclical picture that the independents’ own results hide. The Bloomberg Subdial Watch Index of the fifty most-traded models was 41% below its spring-2022 peak as of January 8, 2026, and rose about 8% during calendar 2025; the WatchCharts Rolex Market Index bottomed in February 2025 at around 30% below its March 2022 peak, and had recovered to roughly +7.9% year on year by March 2026, with the broad market up 8.2%. Independents did not sit that out. They are less liquid than the references those indices track, which in a drawdown means they mark down later and clear more slowly, not that they do not mark down.

The right mental model is that you are buying a single-name equity in a private company whose only product is the founder’s hands, in a market with no index, no continuous quote and a settlement cycle measured in seasons. That can be an excellent investment. It is not an asset class, and anyone selling it to you as one is selling you the survivors.

Two prices exist for every independent watch and the gap between them is the whole market. The first is the list price, set by the maker, paid by whoever the maker allocates a watch to, and not discounted. The second is the secondary price, set by whoever wants one and cannot get an allocation. Because output is fixed by hand-finishing capacity rather than by a production plan, the maker cannot expand supply to close the gap even when it wants to; a workshop that finishes forty movements a year finishes forty movements a year whether the waiting list is two years or ten. The premium is the price of skipping the queue, and it is the volatile part of what you own.

Three forces move that premium, and they move on different clocks. Maker output is the slowest: a house can add a watchmaker or two, which shifts capacity by single-digit percentages a year. List price is next, and independents have used it aggressively — repeated increases at F.P. Journe through the first half of the 2020s narrowed the gap to the secondary market from the bottom, which is the same lever Rolex pulled and the Rolex guide documents in detail. Demand is the fastest and the only one that can halve in a quarter, because the marginal buyer of a $150,000 watch with a five-year wait is discretionary in a way the marginal buyer of a Submariner is not.

The consequence is an asymmetry you should hold in your head whenever you read a premium figure. List moves a few percent a year. The premium moves tens of percent in a cycle, and can move in a day when a reference is discontinued or a maker announces a successor model. If you buy at list you own the watch and a free option on the premium. If you buy on the secondary market you own the watch and a liability: the premium has to hold for you merely to break even, and it has to grow for you to make money after the costs in section eleven.

Price discovery happens in three places. The auction room is public, dated and contested, and it is where the marks that matter are made — one Geneva result reprices a maker’s back catalogue for a season. The specialist dealer market is private and faster: a dealer buys your watch this week at a discount to where they think it clears, typically ten to twenty percent on trade consensus the sister guide flags as unverified, and that spread is the price of immediacy. Collector-to-collector sales are the cheapest and slowest, and carry the counterparty risk section ten describes.

Our tape

Invest Alternative runs its own daily watch series, and it is days old rather than years old, so we present it with its limits stated. Our radar stores the median asking price across a Watchfinder basket once a day; between September 1 and September 8, 2026 it moved from 8,726 to 9,110, a rise of 4.4% across seven observations.

The limits are severe and they matter more than the number. That is one dealer, asking prices rather than sales, and a broad pre-owned basket in which no independent maker appears in size; it is a first reading on the liquid end of the market, not a signal about independents, and watches remain an “awaiting” category in our composite — a 2.3% target weight, no live weight — until the series has history. The IA Composite itself stood at 100.271 on September 8, 2026, up 5.74% over thirty days and 0.29% over a year, on a provisional basis. That figure is ours and is not a watch-market number.

Allocation is the only reliable profit in this market, and it is the one thing money cannot buy directly. A maker producing forty watches a year with four hundred names on a list is running a rationing system, and every rationing system has a currency. At Rolex and Hermès, as the sister guides document, that currency is purchase history at an authorised dealer. At an independent it is something closer to a relationship with a person, which is both more human and, for an outsider, much harder to fake.

The mechanisms in practice, roughly in order of how open they are to a newcomer:

1. The retailer route. Most independents sell through a short list of retail partners and their own boutiques. Buying a piece from the maker’s more available references, at list, through a retailer who knows you, is the standard way onto a list for the scarcer ones. It is slow, it costs the spread on the first watch, and it works.
2. The maker directly. Several of these houses will talk to a collector who shows up at Geneva Watch Days, Watches and Wonders in April or Dubai Watch Week and who can hold a conversation about finishing. The smaller the maker, the more this matters, because the founder is doing the allocating.
3. Deposits and the queue. Some makers take orders years ahead, and the queue can outrun the workshop entirely. Kari Voutilainen closed his order book after waits on the orders already placed ran past a decade, reported at up to eleven years; Akrivia’s list is counted in years against fewer than forty watches a year. A closed book is the end state of this mechanism, and it is worth understanding what it does to you: when the queue shuts, the only way in is the secondary market at whatever the premium is that day. While a book is open, understand what a deposit actually is — an unsecured claim on a small private company, usually with no delivery date and no interest.
4. Charity auctions and prizes. Only Watch, the biennial charity sale that produced the $31.19M Grandmaster Chime at Christie’s in 2019, was postponed from November 2023 over governance questions and eventually held on May 10, 2024 at Palexpo, run again by Christie’s and raising more than CHF 28M. It is one of the few places where a unique independent piece is openly for sale to the highest bidder — Rexhepi’s CHF 2.1M Chronomètre Antimagnétique came out of that 2024 session. The Grand Prix d’Horlogerie de Genève each November — its 25th edition was held on November 13, 2025 — does not sell anything, but it reliably lifts the makers it honours.
5. The secondary market. Buy the watch from somebody who already has one, at auction or through a dealer, and pay the premium. This is the only route that is open to everybody on day one, and it is the route that carries every cost in section eleven.

Two warnings about the allocation route specifically. Flipping an allocated watch is usually detected and usually ends the relationship; the maker sees the auction catalogue, and small houses keep long memories. And the arithmetic in section thirteen shows that even a 60% day-one paper premium converts to about 27% after commission and tax, which is a very good outcome but not the multiple the paper premium implies.

The independents’ public market runs on a calendar you can put in a diary, and it is short. Phillips, whose watch department Aurel Bacs joined in 2015 after leaving Christie’s at the end of 2013, holds its Geneva Watch Auction in May and November and New York sales in June and December; it has been the house of record for independents since the Dufour sonnerie in 2021, and its Geneva XXIII in May 2026 was the largest watch auction ever held at $96.3M. Christie’s and Sotheby’s run their own Geneva sales in the same weeks and their own New York and Hong Kong sessions; Sotheby’s did $193.6M of watches in 2025, up 22%, and Christie’s luxury division passed $1B. Outside the salerooms, Watches and Wonders at Palexpo in April (66 exhibitors in 2026), Geneva Watch Days, Dubai Watch Week and the GPHG in November are where the makers are seen rather than traded.

The fee structure is the largest single cost in this guide, and all three houses publish it. The buyer’s premium is added to the hammer price and paid by the buyer. Sotheby’s has charged 28% up to $2M, then 22% to $8M and 15% above, since February 13, 2026, and Christie’s moved to the same 28% / 22% / 15% structure with effect from September 1, 2026. Phillips runs a schedule of its own: 27% up to $1M, 21% from $1M to $6M and 14.5% above, effective April 12, 2026, with the Geneva tiers set in francs at CHF 1.6M and CHF 7M. Read those brackets and note where the watches in this guide sit: every one of them falls in the bottom band of every schedule, which is also the most expensive band. Assume 27% over hammer at Phillips and 28% at the other two, plus any applicable sales, use or import tax.

The seller’s commission is the other half and it is negotiable in a way the buyer’s premium is not. Nominal rates in the region of 10% are common and are routinely waived or cut on desirable consignments, sometimes to zero and occasionally to an “enhanced hammer” arrangement in which the seller takes a share of the buyer’s premium. Insurance, photography and lotting charges of a percent or two usually remain. The worked example in section thirteen assumes 6% all-in, and shows what a negotiated zero does to the break-even.

The mechanics matter as much as the rates. A reserve is the confidential minimum below which the lot will not sell, and it may not exceed the low estimate; a lot that fails to reach it is “bought in” and is publicly burned for a season or more, which is why consigning into a weak market is worse than not selling. A guarantee, whether from the house or a third party, removes your downside and gives away most of your upside above a threshold. And estimates are marketing: a low estimate set deliberately light draws bidders, and a maker whose pieces routinely hammer at three times low estimate is not being valued accurately by the house, it is being merchandised.

F.P. Journe is the only independent with enough public trading history to describe a cycle, and the cycle is instructive precisely because it looks like the wider watch market rather than like an exception. Journe is the largest of the purist independents, the most widely held, and therefore the most liquid — which in 2021 and 2022 meant it was the most flipped. Watches that had been bought at list by collectors appeared at auction within months, prices ran far above list on the secondary market, and the maker’s name entered the same speculative conversation as the steel Nautilus and the ceramic Daytona.

Then the same thing happened to Journe that happened to everything else. Rates rose through 2022, the crypto and equity wealth effect reversed, and the buyers who had bought to sell became sellers. The Subdial 50 ended up 41% below its spring-2022 peak (Bloomberg, January 8, 2026), the Rolex index about 30% below its own, bottoming in February 2025 before recovering to about +7.9% year on year by March 2026 (WatchCharts). Journe compressed with it.

No Journe index is published, so the way to measure that compression is to follow one reference through the cycle, and the entry point to the range is the one with the deepest trade. The Chronomètre Bleu lists at about $37,400 and most commonly changes hands on the secondary market at $75,000–90,000 in 2026: still roughly double retail, but materially below the 2022–23 peak and a long way from the five-to-seven-times-retail multiples the trade reported on some references in 2019–21. Those are dealer and marketplace observations rather than index readings, and we use them as such — one well-traded reference watched across a cycle is the closest thing this market has to a price series. The maker met the compression from the other side, by raising list prices repeatedly.

That last move is the mechanism worth carrying away, because it is the one an independent controls. Raising list narrows the gap between the two prices without the secondary price having to fall further, transfers the premium from flippers to the maker, and makes the allocation less profitable to abuse. It also means that a buyer who paid a 2022 secondary price is holding a watch whose list price has moved up underneath it, which cushions the mark but does not restore the loss.

The top of the market went the other way entirely. Morgan Stanley’s 2025 industry estimates have F.P. Journe, H. Moser and MB&F all growing revenue in a year when Swiss exports fell 1.7%, each in the CHF 50–125M band, and Journe’s auction results ran well above estimate. In December 2025 the FFC prototype made $10.8M. Both things were true at once: the speculative middle of the Journe market deflated, and the connoisseur top of it set a record. That is the normal shape of a collectibles cycle, and it is why the two questions to ask about any maker are different questions — is the flipping premium gone, and is the top-end bid intact?

The frontier tier is where the returns of the next decade will be made and where most of the money will be lost. The names the trade watches include Rexhep Rexhepi, whose house Akrivia has moved from frontier to proven inside a decade; Kari Voutilainen, who is proven on any measure but whose output is small enough to behave like a frontier name; and behind them a genuinely new cohort — Petermann Bédat, Berneron, Kudoke, Simon Brette, Sylvain Pinaud, Raúl Pagès and a handful of one-man workshops that have appeared since 2020.

Some of that cohort already carries the marks of a real career. Petermann Bédat, founded in 2017 by two watchmakers out of A. Lange & Söhne, took the GPHG Horological Revelation prize in 2020 and the Chronograph prize in 2023, and its pieces have since entered the Phillips Geneva catalogues. What none of the new cohort has is a decade of dated public results, and no reliable output or list-price figures are published for any of them; treat any number you read about the newest names, including in the enthusiast press, as unverified.

The base rate is the thing to reason from. Making a beautiful watch and building a durable secondary market are different achievements, and the second one takes about a decade and depends on things a founder does not control: whether the workshop survives an illness, whether a retail partner fails, whether the aesthetic that looks fresh in year two looks dated in year twelve. For every Rexhepi there are several makers who produced fine work, sold their annual output, and never developed a bid that clears above list once the initial cohort of buyers has what it wants. Those makers are not frauds and their watches are not bad; they are simply consumption.

What separates the two, on the evidence available, is boring and checkable:

• Public results with dates. Pieces that have sold at Phillips, Christie’s or Sotheby’s, in more than one season, at prices you can look up. One strong result is a data point; a decade of them is a market.
• A serviceable movement. In-house calibres finished by hand are the point of the exercise, but a movement that only the founder can service is a different asset from one a trained watchmaker can open. Section ten prices that difference.
• Output discipline. A maker who responds to a long waiting list by tripling output is telling you the scarcity was a choice rather than a constraint, and the premium will follow the supply.
• A second buyer. Ask who buys the watch from the person who bought it new. If the honest answer is “the same fifty collectors”, the market is a club, and clubs do not clear at scale.

The risk that ends you in this market is not a fake watch; it is a watch nobody can open. A Rolex service costs $800–1,600 depending on model and can be done in any major city (Bob’s Watches, 2026); a Patek Philippe overhaul runs from about $2,000 on a time-only model to well above $50,000 on a grand complication, and is quoted case by case.

An independent’s movement is a different proposition. Parts are made in batches of dozens, sometimes by the founder; finishing that took a hundred hours to do takes comparable skill to redo; and for the smallest makers the set of people qualified to work on the calibre is measured in single digits and includes the founder. Ask, before you buy, who services this watch in 2040, and treat a vague answer as a price.

The larger independents have built real answers to this. Houses with boutique networks run their own service centres in the cities where they sell, which is one of the practical arguments for the bigger names over the smallest ones. The smallest workshops are, structurally, single points of failure, and a founder in his seventies with no successor and no parts inventory is a serviceability risk that should be reflected in what you pay — not because the watch will stop, but because the next buyer will ask the same question you should have asked.

Nor is the risk confined to the one-man shops. At Greubel Forsey, a house of some two hundred watches a year, co-founder Stephen Forsey lost his operational role at the end of 2025 and had his contract terminated with effect from May 31, 2026 after a strategy dispute with the chief executive, while remaining a shareholder (WatchPro, 2026). Founders are a governance risk as well as a watchmaking one.

Authenticity risk in independents is different in shape from the rest of the watch market. Counterfeiting a Submariner is a volume business supported by a supply of “super clones” that vendor pricing in 2026 put at well under $2,000; counterfeiting a hand-finished independent is barely worth attempting, because the finishing is the product and a specialist can see it under a loupe. The real exposures are three, and they run in ascending order of what they cost you:

• Married and rebuilt pieces, where a genuine movement sits in a replaced case or carries a replaced dial. On a maker whose entire value is originality, a swapped component is a large, permanent haircut, and the sister guide’s figures for redialled and over-polished watches — directional dealer heuristics rather than measured data — run from a fifth to half of value.
• Provenance failure. Because these markets are small, the paperwork matters more than usual: original certificate, the maker’s own records, and an unbroken chain of ownership. A watch whose history has a gap in a small market is a watch with a permanent discount.
• Counterparty and custody failure, which is the one that actually destroys capital. Consignment dealers hold other people’s watches and other people’s money at the same time, and when one fails the owners are unsecured creditors of a business with nothing left in it.

The third of those deserves a name and a date, because the mechanism is ordinary and it will happen again. Anthony Farrer traded as The Timepiece Gentleman from Beverly Hills, taking clients’ watches on consignment for a 5% fee. He kept the proceeds, failed to return the unsold pieces, and in some instances sent buyers watches belonging to other clients. He abandoned the store in August 2023, has been in federal custody since November 2023, pleaded guilty in October 2024 to one count of wire fraud and one of mail fraud covering more than 40 victims and at least $5.6M, and was sentenced to 70 months (US Attorney’s Office, Central District of California, and IRS Criminal Investigation). Not one watch in that case was a fake. The defence is procedural rather than forensic: never let a dealer hold both your watch and the buyer’s money, get the consignment terms in writing, and prefer an auction house’s escrow to a handshake.

Insurance is the cheap part of the answer, and section eleven prices it. The thing to get right is the basis: schedule the watch at replacement value rather than purchase price, and re-appraise after any large market move, because an under-insured watch in a market this small is very hard to replace at all.

The cost stack on an independent watch is heavier than on almost any other collectible of comparable value, because you pay a high transaction cost twice and a carrying cost throughout. Take them in the order you meet them.

Acquisition

At auction, the buyer’s premium — 27% at Phillips below $1M, 28% at Sotheby’s and Christie’s below $2M — is added to the hammer before anything else. On top of that sits whatever tax your jurisdiction levies on the purchase: US state sales or use tax where the watch is delivered, UK VAT at 20% or the effective 5% that section twelve explains, and, for a watch imported into the United States from Switzerland, a tariff. That last one has moved repeatedly and is the fastest-changing number in this guide: Swiss watches faced a 10% US tariff from April 5, 2025, 39% from August 7, 2025, and 15% from November 14, 2025; the Supreme Court struck down the IEEPA tariffs in February 2026 and the Section 122 authority lapsed on July 24, 2026, leaving an all-in Section 301 rate the trade put at roughly 12.5% as of August 2026. Check the rate on the day; do not assume this paragraph is current.

Carry

Insurance runs at 0.5–1.5% of insured value a year through a specialist underwriter, on a scheduled personal-articles policy, and at 1–2% through a general broker — a band we carry from the sister guide and have not independently re-verified. A safe-deposit box or a rated safe, a few hundred dollars a year. And service: a hand-finished movement wants attention every five to seven years, and no independent house publishes a tariff — F.P. Journe’s own service page quotes case by case after the watchmakers assess the piece, and we found no published price list at any of them. The benchmark to reason from is Patek Philippe, where an overhaul starts around $2,000 on a time-only calibre and runs far higher with complications; an independent’s will not be cheaper. Budget $2,500–4,000 — our estimate, not a quoted rate — and expect the watch to be away for months rather than weeks.

Exit

Exit costs you either a seller’s commission of nominally around 10%, negotiable and often reduced to a low single digit or waived on a good consignment, plus a percent or two of insurance and photography charges; or a dealer’s spread, which trade consensus puts at 10–20% and which the sister guide flags as unverified. Marketplace fees sit in between: Chrono24 charges a private seller a flat 6.5%, and professional dealers a dynamic commission generally in the 9–12% range plus a monthly package fee.

The friction you cannot see

Time is a cost here in a way it is not in a liquid market. A Geneva sale happens twice a year; consignment deadlines run months ahead of the sale; settlement follows the sale by weeks. From deciding to sell to having the money is realistically six to nine months if you want the best venue, and the ability to shorten that to a fortnight through a dealer is exactly what the dealer’s spread buys.

US tax treatment is the single most under-modelled cost in collecting, and for watches it is unambiguous. A wristwatch is a collectible under IRC §408(m), so a long-term gain is taxed at the collectibles rate of up to 28% under IRC §1(h)(4) rather than the 20% top rate that applies to stock (IRS Topic 409). Add the 3.8% net investment income tax under §1411, which applies above modified adjusted gross income of $200,000 for a single filer and $250,000 for a married couple filing jointly, and the top federal rate on a watch gain is 31.8%, before state tax. Short-term gains — a watch held a year or less, which is what a flip is — are taxed as ordinary income.

Four asymmetries make it worse than the headline rate suggests, and all four are structural rather than seasonal.

• Losses on personal-use property are not deductible. Under IRC §165(c) an individual deducts losses only if the property was held for investment or in a trade or business. A watch you wore is personal-use property to the IRS, so you pay 28% on the winners and deduct nothing on the losers. Holding a collection strictly for investment — unworn, insured as investment property, documented — is a defensible position, and it is one to take with a tax adviser before you buy, not after you sell.
• No like-kind exchange. Section 1031 has been limited to real property since the 2017 Act took effect in 2018. You cannot roll a Dufour into a Journe without settling the tax.
• Basis is what you actually paid. Hammer plus buyer’s premium plus sales or use tax plus import duty all go into basis, which is the one piece of good news; carrying costs such as insurance and service generally do not.
• The wash-sale rule does not help you. Section 1091 applies to securities only, which cuts both ways: there is no harvesting discipline to lean on and no restriction to work around.

Non-US readers face a different but not lighter arithmetic. UK import VAT is charged at the standard 20% on the customs value plus duty, though a watch that qualifies as an antique or collectors’ item is taxed on a reduced value at import for an effective 5%, with the full 20% falling due when it is later sold; capital gains treatment varies; and several European jurisdictions treat a watch held for personal use more favourably than the US does. Whatever the jurisdiction, model the tax before you model the return, because on the numbers in the next section the tax is comparable in size to the buyer’s premium.

Here is the whole thing in dollars, with every assumption stated so you can change one and re-run it. A US buyer acquires an independent watch at Phillips New York for $100,000 all-in: a hammer price of $78,740 plus a 27% buyer’s premium of $21,260. State sales or use tax at 8% on the invoice adds $8,000, and all of it goes into basis, so the cost basis is $108,000. The watch is held for eight years, insured at about 1% of value a year ($9,600 over the period), serviced once ($3,000), and kept in a safe-deposit box ($250 a year, $2,000). Total cash out over eight years: $122,600.

Now sell it, back at auction, on a 6% all-in seller’s commission. To get the $122,600 back after commission and tax, the watch has to hammer at about $137,700: proceeds of $129,438 after commission, a gain of $21,438 over the $108,000 basis, tax of $6,817 at 31.8%, and $122,621 in the hand. That is a required rise in the hammer price of 75% over eight years, or 7.2% a year, simply to break even. The gap is not mysterious — it is the sum of the frictions in the chart below, which comes to about $58,900, almost exactly the $59,000 the hammer has to climb.

Push it further, because break-even is not the goal. Suppose the watch does very well and the hammer price rises to $160,000, up 103% over the eight years, which is 9.3% a year in gross terms — a result that would put the maker among the better performers of any decade. Proceeds after commission are $150,400, the gain over basis is $42,400, tax at 31.8% is $13,483, and $136,917 lands in the account against $122,600 of cash out. The net is $14,317, or 1.4% a year. A gross compound rate of 9.3% became a net compound rate of 1.4%. To match a portfolio compounding at a plain 7% a year, the hammer would have had to reach about $275,000 — nearly 3.5 times what you paid for it, or 16.9% a year, every year, for eight years.

Two variations change the answer materially, and both are decisions you make rather than markets you wait for. If you negotiate the seller’s commission to zero, which a strong consignment can, the break-even hammer falls from $137,700 to about $129,400, a 64% rise rather than 75%, or 6.4% a year.

The second variation is where you buy. Take the same $100,000 watch at the Geneva sale in May rather than the New York sale in December and you pay two levies instead of one, because the tariff does not replace the state tax but stacks on top of it. Importing into the United States at the roughly 12.5% all-in rate that applied in August 2026 — a 5% most-favoured-nation duty plus a 7.5% Section 301 duty — adds $12,500, and the same $8,000 of state use tax still falls due on delivery, so the basis becomes $120,500 rather than $108,000. Cash out over the eight years rises to $135,100 and the break-even hammer to about $151,000, a 92% rise, or 8.5% a year. Where you buy is worth more than a full percentage point of annual return.

Now the allocation, which is the other side of the same arithmetic. Suppose you are allocated a watch at a $60,000 list price, pay 8% tax for a $64,800 basis, and the maker’s comparable pieces clear at $96,000 at auction — a 60% paper premium on day one. Sell into that immediately and the proceeds are $90,240 after a 6% commission, the gain is $25,440, the tax is $8,090, and you net $17,350, or 26.8% on your outlay. Through a dealer paying $80,000 instead, the net is $10,366, or 16.0%. Both are excellent returns for a single transaction. Both are also less than half the paper premium, and both will usually cost you the allocation — which, as section six argued, was the asset.

There is a sequence that works, and it is slower than most new buyers want it to be. It is built to spend the first year buying information rather than watches.

1. Learn the finishing before you learn the names. Spend the first months with a loupe and other people’s watches: at a retailer who carries two or three of these makers, at Watches and Wonders in April, at Geneva Watch Days, at a collector dinner. The entire premium in this market is paid for work you should be able to recognise. If you cannot yet see the difference between a hand-applied anglage and a machined one, you are not ready to pay for it.
2. Read two seasons of catalogues. Phillips, Christie’s and Sotheby’s publish estimates before each sale and results afterwards, both free. For every maker you care about, track the low estimate, the hammer and the ratio between them. Two seasons of that is the best education this market offers.
3. Buy one proven watch first, at auction, in the middle of the range. Not the trophy and not the frontier name: a piece from a maker with a decade of results, bought where the price is public. You will pay the 27% and you will learn what settlement, shipping, insurance and a first service actually involve, on a position you can afford to be wrong about.
4. Then open a relationship. Buy your second watch at list from a retailer or the maker, accept that it may not be the reference you most want, and be a good client. This is the step that takes years and is the only one that leads to an allocation.
5. Insure, document and store from day one. A scheduled personal-articles policy at replacement value, the certificate and invoice scanned, the service history kept, and a decision — taken with a tax adviser — about whether this is investment property or personal-use property under §165(c). That decision is much harder to make credibly after the fact.
6. Size it honestly. Independents are illiquid single-name positions in a market with no index. Our own rule is that makers with less than a decade of public results are capped at 20% of watch exposure and any single frontier name at 5%, and that the whole watch allocation is small enough that a six-month wait to sell is an inconvenience rather than a problem.

If none of that appeals, the honest alternative is to not do it. There is no fund, no ETF and no fractional wrapper for independent watchmaking that we would put money into; the liquid ways to own the theme are shares in the listed luxury groups, which own none of these makers and behave like consumer equities, and that is a different investment with a different thesis.

These are the readings that would change the view, each with the level to watch and the reason it matters. All figures are as of September 10, 2026 unless stated.

• The November and May Geneva sales, hammer against low estimate. The single best real-time gauge of whether the independents’ bid is intact. A maker whose pieces average above 1.5× low estimate has a live market; two consecutive seasons averaging below 1.0× is a bid that has gone.
• Swiss export value in the CHF 50,000-plus band. It was 1.4% of volume and 37.3% of value in 2025 (Morgan Stanley × LuxeConsult, March 2026), and the top of the market has been carrying the industry. If that share stops rising while total exports fall, the current that has been favouring independents has slackened. The Federation of the Swiss Watch Industry publishes monthly; the January release covers the prior year.
• Total Swiss exports. CHF 25.5B in 2025, down 1.7% in a second consecutive decline; H1 2026 CHF 12.8B, down 0.7% (FH). A third consecutive annual decline would be the longest contraction since the financial crisis and would test the assumption that the high end is insulated.
• The broad secondary indices, as a proxy for risk appetite. The Bloomberg Subdial 50 was 41% below its spring-2022 peak in January 2026 after rising about 8% in calendar 2025; the WatchCharts Rolex index bottomed about 30% below its March 2022 peak in February 2025 and was up roughly 7.9% year on year by March 2026. Independents do not track these, but they do not escape them either. A renewed drawdown of 15% or more in the liquid indices is a reliable signal to stop bidding.
• The US tariff on Swiss watches. Roughly 12.5% all-in as of August 2026, after a run that went 10%, 39% and 15% in turn. Every point of tariff is a point of return on a cross-border purchase, and the rate changed four times in the eighteen months to August 2026. Confirm it on the day you bid.
• List price and output at each maker. Rising list narrows the premium from below and is usually healthy; a maker that stops raising list while its secondary premium erodes has lost pricing power. Any material expansion of capacity — a new atelier, a headcount jump, a step-change in pieces per year — is a coming compression in the premium, not good news.
• Succession and service. For every maker in the collection, the answer to “who services this in 2040” is a live variable. A founder’s retirement, a sale to a group, or the loss of a key watchmaker changes the asset, and it shows up in the price a season later than in the news. Greubel Forsey is the case to study: Stephen Forsey’s operational role ended on December 31, 2025 and his contract was terminated with effect from May 31, 2026 after a strategy dispute, at a house that had already cut its output ambition from 500 watches a year to about 200.