Case 17 · Luxury · From The Strat, episode 17
EPA: RMS
Hermès
A sixth-generation family-controlled French house that makes leather goods by hand in its own French workshops, sells them almost exclusively through its own stores, and deliberately produces fewer of them than people want to buy.
- Founded
- 1837
- Founders
- Thierry Hermès, Jean-Louis Dumas
- Headquarters
- 24 rue du Faubourg Saint-Honoré, Paris
- Moat
- Wide · Brand
“Every other company asks how much it can sell. Hermès asks how much it can make, and then makes slightly less.”
Listen first — The Strat 17 · 10 min
Let the artisans set the pace, and turn a production constraint into the most valuable waiting list in retail.
Notes on the episodeOperating margin
41.0%
The highest sustained margin in physical consumer goods
Artisans per bag
1
Start to finish, signed
Net cash
€12.8B
No debt, no acquisitions
Years under family control
189
Six generations, unbroken since 1837
§01 — The business model
Hermès is the only large luxury company whose binding constraint is production rather than demand. Leather Goods and Saddlery — €7.07 billion in 2025, 44% of revenue — is made in twenty-five French workshops by artisans who each build a bag from start to finish and sign it. Training takes around eighteen months before an artisan is trusted with a saleable piece. Capacity therefore grows at roughly the rate at which Hermès can train people, which is high single digits a year, and never faster.
That constraint is the business model, not a problem with it. Because supply is fixed by craft, Hermès never needs a promotion, never carries aged inventory, and never has to decide whether to protect margin or volume. Leather goods grew 13.1% at constant currency in 2025 while the wider luxury sector was contracting — not because Hermès marketed harder, but because it opened a twenty-fourth workshop.
Around leather sits a deliberately broad house: ready-to-wear and accessories (28%), silk, jewellery and home (13%), watches, perfume. Some of these are structurally lower-margin and Hermès keeps them anyway, because the silk scarf and the €200 bracelet are how a client enters. Roughly nine-tenths of revenue comes through Hermès's own stores — about 300 of them, a network the company spent the 1980s and 1990s buying back from distributors and has never let go.
Where the revenue comes from
Leather Goods & Saddlery
44%
€7.07bn in 2025, +13.1% at constant currency. Made in France, capacity-constrained, and the source of nearly all the brand's pricing power.
Ready-to-wear & Accessories
28%
€4.53bn. The second engine, and the category that has absorbed the most of Hermès's recent growth in store space.
Other Hermès sectors — jewellery and home
13%
€2.06bn, +11.2% at constant currency. The fastest-scaling adjacency and the one that extends the price ceiling upward.
Silk & Textiles
6%
€964m. The carré, first sold in 1937. Small in revenue, disproportionate in brand identity, and often the client's first purchase.
Watches
3%
€549m, down 1.5% at constant currency. A genuinely competitive category where Hermès is a challenger rather than an incumbent.
Perfume & Beauty
3%
€489m, down 7.6% at constant currency. Notably, Hermès refuses to run this as a mass-volume licence business — which is exactly why it is small.
Unit economics — One Birkin 30, Togo leather
The bag is one of very few consumer products that reliably resells above its original price. That single fact does more work than any advertising campaign: it converts the purchase from consumption into acquisition, and it means the client's downside is not the price — it is not being offered one.
§02 — The moat
Hermès's moat is the rarest kind, because it is a moat made of a bottleneck the company chose to keep.
The brand layer is obvious: 189 years, a saddler's origin story that is literally true, and two products — Kelly and Birkin — that have become financial assets. But the durable layer underneath is process. Hermès makes its own leather goods in its own French workshops, owns tanneries, trains its own artisans in its own schools, and refuses to subcontract the core. That means the constraint on growth is a physical, verifiable thing: the number of trained hands. It cannot be bought, it cannot be outsourced, and — crucially — it cannot be faked by a competitor either, because a rival cannot suddenly produce twenty years of trained French leatherworkers.
Distribution completes it. Nine-tenths of sales run through Hermès's own doors, so the company controls price, allocation and who gets offered what. The allocation system is the most under-discussed strategic asset in luxury: it makes the client want to buy other categories in order to be eligible, and it converts scarcity into a relationship rather than a transaction.
Where it is weaker: watches and perfume are competitive markets where none of this applies, and both shrank in 2025. And 52% of revenue comes from Asia, which is a real concentration risk in a business with no ability to discount its way out of a regional demand shock.
Porter's five forces — 5 ticks means the force is squeezing hard
Competitive rivalry
Chanel, Louis Vuitton and Rolex compete for the same client, but Hermès does not compete for volume and cannot be undercut — its problem is never a rival's price, only a rival's newness. In watches and perfume, rivalry bites much harder.
Threat of new entrants
The barrier is 189 years and a pipeline of trained artisans. A well-funded entrant can replicate the marketing and none of the manufacturing.
Threat of substitutes
The pre-owned Birkin market is technically a substitute, but it trades above retail and therefore reinforces the primary market rather than cannibalising it. Investment-grade alternatives — watches, art, gold — compete for the same discretionary euro.
Buyer power
Inverted. Hermès decides who is offered a Birkin. There is no negotiation, no discount, and a queue. This is as close to zero buyer power as exists in consumer goods.
Supplier power
The one genuine pressure point. Exotic skins, high-grade calfskin and precious stones are scarce, and prices rise. Hermès has responded by buying tanneries and farms, but it does not yet control this input the way it controls the workshop.
§03 — The financials
Revenue quality
Exceptional, and the cleanest in the sector. Revenue of €16.00 billion in 2025, up 8.9% at constant exchange rates and 5.5% as reported — growth in every quarter and every geography while the rest of luxury contracted. Roughly nine-tenths is retail through Hermès-operated stores: full-price, cash-settled, no wholesale receivables, no markdowns, no returns of consequence. There is essentially no revenue in this company that was bought with a discount.
Margin structure
Recurring operating margin of 41.0% in 2025 (€6.57bn), up from 40.5% in 2024 despite adverse currency. To put that in context: LVMH's Fashion & Leather Goods division earned 35.0% and Chanel earned 24.5% in the same year. Hermès holds the highest sustained operating margin of any large physical-goods company in the world, and it does so while manufacturing in France, which is supposed to be impossible.
Cash generation
Adjusted free cash flow of €3.88 billion in 2025 on €5.61 billion of operating cash flow, after €1.16 billion of operating investment — most of that going into new leather workshops and stores. Capital intensity is rising by design, because building workshops is how Hermès buys future revenue. Cash conversion remains very high because inventory does not have to be cleared.
Balance sheet
Fortress, and then some. Restated net cash of €12.77 billion at end-2025 after paying €2.80 billion of dividends, up from €12.04 billion. Equity of €18.84 billion. Hermès carries no meaningful debt and has none of the acquisition leverage that sits on its peers' balance sheets. The cash pile is arguably too large — it earns a low return and management has shown no urgency about it.
Revenue
€16.00B
+8.9% at constant currency, +5.5% reported
FY2025
Recurring operating income
€6.57B
41.0% of sales, up from 40.5%
FY2025
Net profit, group share
€4.52B
€4.86B and +5.5% excluding France's one-off levy on large-company profits
FY2025
Adjusted free cash flow
€3.88B
FY2025
Restated net cash
€12.77B
After €2.80B of dividends paid during the year
31 Dec 2025
Leather Goods & Saddlery
€7.07B
44% of revenue, growing 13.1% at constant currency
FY2025
Asia as share of revenue
52%
€8.29B including Japan. The main concentration risk in an otherwise unusually safe business.
FY2025
Employees
26,494
Up 5.2% — headcount growth is the leading indicator of future capacity
31 Dec 2025
Leather workshops in France
25
The 24th opened in September 2025, the 25th in April 2026; four more announced through 2030
April 2026
§04 — The valuation
P/E (trailing)
~38x
On €4.52bn of 2025 net profit at roughly €173bn of market capitalisation. Expensive in absolute terms and materially cheaper than the ~55x the shares carried at the 2024 peak.
July 2026
EV / Sales
~10x
Enterprise value net of the €12.8bn cash pile
July 2026
EV / EBIT
~24x
July 2026
Dividend yield
~1.1%
€18.00 per share proposed for 2025, of which €5.00 paid as an interim in February 2026
Peer EV/EBIT — LVMH
~15x
The multiple the market pays for a conglomerate with 22% group margins rather than 41%
What has to be true to justify the price
- 01Leather goods capacity keeps compounding at high single digits. Every announced workshop — Loupes in 2026, Charleville-Mézières in 2027, Colombelles in 2028, Les Andelys by 2030 — is effectively pre-sold revenue, provided the artisans can be trained.
- 02Operating margin holds around 40%. Hermès has been raising French wages and building expensive domestic capacity; the margin has not cracked yet, but it is the number a bear should watch.
- 03Asia, at 52% of revenue, does not deteriorate structurally. Hermès has no discounting lever to pull if it does.
- 04The allocation system survives its own scrutiny. Antitrust complaints in the US and France over tying Birkin access to other purchases have so far gone nowhere, but the practice is the mechanism, and a legal constraint on it would be a genuine change to the business.
Run it yourself
Move the growth rate and the margin and watch the implied value move. Same inputs, live.
§05 — Capital allocation
Hermès allocates capital like a company that expects to exist in 2130, because the people making the decisions expect their grandchildren to own it.
The priority order is unusually explicit and unusually consistent. First, manufacturing capacity: €1.16 billion of operating investment in 2025, most of it into French leather workshops, each of which takes years to reach full output and is announced half a decade before it opens. Second, the store network, which is expanded slowly and almost never through franchise. Third, dividends — €2.80 billion paid in 2025, plus occasional exceptional distributions. Fourth, essentially nothing: no leveraged M&A, no brand roll-up, no diversification into categories with better near-term returns.
The result is a €12.8 billion cash pile that keeps growing, which is the one genuine criticism an analyst can make. That capital earns a low return sitting still, and management's answer — that it is insurance and optionality — is more convincing for a family partnership than it would be for a normal listed company. It is also, in fairness, the same conservatism that let Hermès survive the 2010 LVMH raid without needing anyone's permission.
The minority acquisitions Hermès does make are revealing: tanneries, silk printers, crocodile farms, John Lobb, Puiforcat, Saint-Louis. It buys inputs and savoir-faire. It does not buy brands.
Operating investment
€1.16B
7% of revenue, overwhelmingly into French production capacity
Net cash
€12.77B
Growing every year. The honest weak point in an otherwise excellent record.
Dividends paid
€2.80B
€18.00 per share proposed for 2025
Large M&A
None
Hermès has never made a transformational acquisition in 189 years
Supply-chain ownership
Extensive
Tanneries, silk printing, crystal, silverware, bootmaking — bought to secure inputs, not to add revenue
Debt
Immaterial
The company is structurally unleveraged
§06 — The thesis
Hermès is the closest thing in listed consumer goods to a business with no strategic weaknesses. It has pricing power it does not fully exercise, a moat made of a physical constraint rather than a marketing budget, a 41% operating margin achieved while manufacturing in one of the world's most expensive labour markets, €12.8 billion of net cash, and a partnership structure that makes hostile change of control effectively impossible. In 2025 it grew 8.9% at constant currency while LVMH's fashion division shrank. That is not luck; it is what happens when your constraint is supply and everyone else's is demand.
The argument against is entirely about price. At roughly 38x trailing earnings and 24x EBIT, the market has already understood all of the above. Nothing here is a secret. But the shares have de-rated substantially from their 2024 peak, and a business that can compound revenue at high single digits for decades — because the growth rate is set by workshop construction, not by fashion cycles — is one of the very few where paying a high multiple is defensible rather than merely optimistic.
The realistic bear case is not that Hermès breaks. It is that it stays wonderful and the multiple compresses anyway, and an owner earns 6% a year for a decade instead of 15%. That is a genuine risk and it should be stated plainly rather than argued away.
What would change my mind
If Leather Goods and Saddlery growth at constant currency falls below 5% for two consecutive years while new workshops are still opening, the constraint has moved from supply to demand — and the entire thesis inverts, because a capacity-led model that runs out of buyers is just an expensive factory. A secondary trigger: a binding legal or regulatory ruling against tying Birkin allocation to other purchases would dismantle the mechanism that converts scarcity into cross-category revenue.
§07 — How it happened
- 1837
A harness workshop near the Grands Boulevards
Thierry Hermès, born in Krefeld to a family that fled to France, opens a workshop making harnesses and bridles for carriage horses. The saddle stitch he uses — two needles, one thread, each stitch independent so a break does not unravel the seam — is still the company's core technique.
- 1922
Buying the zipperThe fork
Émile-Maurice Hermès sees a fastener on a Canadian car in North America and secures exclusive French rights to it for leather goods and clothing. For years the zip is known in France as the fermeture Hermès. The first Hermès handbag follows. The saddler has noticed that the carriage trade is ending.
- 1956
Grace Kelly holds a bag in front of a camera
The Princess of Monaco is photographed for Life using the sac à dépêches to shield her pregnancy from photographers. The bag is renamed the Kelly. Hermès did not plan it, did not pay for it, and has never needed a better advertisement.
- 1978
Jean-Louis Dumas takes a shrinking companyThe fork
The fifth generation inherits a house selling saddles and silk to a customer base that is dying. Dumas rebuilds it — reclaiming distribution from third-party retailers, taking manufacturing back in-house, buying suppliers, and insisting that Hermès make things by hand at a moment when every competitor is industrialising. The company Hermès is today is his design, not Thierry's.
- 1984
A conversation on a flight to London
Dumas sits beside Jane Birkin, who complains that no bag is big enough for a mother's life. He sketches one on a sick bag. The Birkin becomes the most profitable single object in the history of fashion, and the most instructive: it was designed for utility, and became a store of value.
- 1993
Listing 27% without giving up controlThe fork
Hermès floats on the Paris Bourse to settle a fragmented family shareholder base, and the offer is oversubscribed roughly thirty-four times. It is structured as a société en commandite par actions — a partnership in which the family holding company appoints management and cannot be outvoted. The structure looks like a technicality for seventeen years.
- 2010
LVMH appears on the register with 17%
Bernard Arnault discloses a stake accumulated quietly through cash-settled equity swaps, later rising above 23%. The families respond by locking a majority block into a new holding company, H51. France's market regulator fines LVMH €8 million in 2013 for disclosure failures; in 2014 a court-brokered settlement forces LVMH to distribute the shares to its own shareholders and stand still. The 1993 structure was the reason Hermès survived.
- 2025
Growing while the sector shrinks
Revenue reaches €16.0 billion, up 8.9% at constant currency, with a 41.0% operating margin — in the same year LVMH's fashion division declines and Chanel grows 1.8%. Hermès opens its twenty-fourth leather workshop and announces three more. The capacity, not the marketing, is the growth plan.
§08 — Your turn
Case 17 — Hermès · Jean-Louis Dumas · 1984
Demand exceeds what your artisans can make. Do you scale production to meet it?
Hermès makes leather goods by hand. A single artisan makes an entire bag, start to finish, and takes years to train. Demand for your handbags now exceeds what your workshops can produce. Every consultant in Paris can show you the model: expand capacity, license the name, and capture the demand while it is there.
Choose before you scroll. The answer is hidden until you commit.
§09 — Around this case
The founders
Thierry Hermès
“He built a company around a technique rather than a product. The product became obsolete within fifty years; the technique is still what the company sells.”
Jean-Louis Dumas
“He inherited a company whose growth was limited by demand and deliberately rebuilt it into one whose growth is limited by supply. Every advantage Hermès now has descends from that inversion.”
The episode
17- The Rise of Hermes
Episode 17 · 10 min
Let the artisans set the pace, and turn a production constraint into the most valuable waiting list in retail.
What to listen forSources
- Hermès International — 2025 full-year results (12 February 2026)
- Hermès International — 2024 full-year results
- Hermès International — Universal Registration Document
- AMF decision on LVMH disclosure of its Hermès stake (July 2013)
- The Hermès Saga / contemporaneous FT and Reuters coverage of the LVMH stake-building, 2010–2014
- The Strat, Episode 17
Patterns
§10 — Read next
These cases share the most patterns with Hermès. That overlap is computed from the tags, not chosen by hand.