Fashion, beauty & personal care

Fashion, beauty & personal care

Industry ID: 23 | Slug: fashion-beauty | Researched: 2026-09-15 | Analyst: agent

Research-capability note. The shared WebSearch budget was exhausted before this sector began (200/200 calls). One search was attempted and refused. All findings below were obtained by direct WebFetch on primary sources — company investor releases and results PDFs, EUR-Lex, the French Assemblée nationale, BLS, USDA ERS, FDA and SEC EDGAR. Tier-A verification is consequently strong; discovery is narrow, and four significant primary sources were blocked outright (Kering, Inditex, OTEXA, the Swiss watch federation). See §13 and the source registry.

1. Definition and boundaries

In scope. Brand and category economics for: luxury goods (soft luxury, hard luxury, leather goods, watches, jewellery); apparel and footwear brands including sportswear and activewear; fast fashion and ultra-fast fashion as brands and business models; beauty and personal care across prestige, mass, dermocosmetics, professional and fragrance; resale and secondhand as a business model and brand strategy; textiles, fibres and materials; and the supply-chain, product and sustainability regulation attached to all of the above — EU extended producer responsibility and ecodesign, French fast-fashion law, US MoCRA, tariff and trade rules as they bear on landed cost and sourcing.

Explicitly out of scope, and who owns it.

• Retail channel economics — store formats, e-commerce logistics, marketplace take rates, returns processing, retail media, de minimis volumes as a channel phenomenon — belongs to sector 12 (retail & e-commerce). This dossier deliberately does not restate sector 12's figures on Target/Walmart refunds, PDD/Temu, or ThredUp's resale-market sizing.
• Pharmaceuticals including GLP-1 receptor agonists themselves — sector 06 (healthcare & biotech). Only the claimed downstream apparel and beauty effect is assessed here, and it is assessed as unproven (T-23-19).
• Advertising and marketing platform economics — sector 16. Generative engine optimisation appears here only as a beauty-brand deployment (T-23-13), not as an ad-tech market.
• Warehouse robotics as a technology market — sector 13. Zalando's robotic picking is cited only as deployment evidence.

Boundary disputes worth naming.

1. Resale. Sector 12 covers resale as a retail channel and inventory-sourcing problem. This dossier covers resale platform unit economics and profitability (T-23-07), because the question "does any resale platform make money" is a business-model question about this sector's brands, not a channel question. The two treatments should be read together.
2. Beauty versus healthcare. Dermocosmetics sits on a boundary that is regulatory, not commercial: La Roche-Posay and CeraVe are cosmetics in law and pharmaceuticals in distribution. The fastest-growing part of beauty is the part closest to that line.
3. Luxury versus retail. Travel retail (DFS, duty free) is a channel, but when LVMH sells its Greater China travel-retail business the transaction is a luxury-portfolio decision. It is treated here as portfolio strategy (T-23-06) and left to sector 12 as channel economics.
4. Ultra-fast fashion. Shein and Temu are simultaneously brands, marketplaces and logistics arbitrages. Their pricing and regulatory exposure as apparel businesses are here; their marketplace mechanics and parcel volumes are sector 12's.

2. Subcategories

3. Market structure

Concentration is barbelled and differs by subindustry.

• Luxury is an oligopoly with one dominant firm. LVMH's H1 2026 revenue of €38.6bn is larger than Richemont's entire FY2026 revenue of €22.42bn and nearly five times Hermès's H1 of €8.163bn. Entry barriers are heritage, retail real estate, vertically integrated ateliers and access to hard-luxury supply. Pricing power is genuine but category-specific: Hermès can hold a 41.0% recurring operating margin while Richemont's Specialist Watchmakers sit at 3.4%.
• Beauty is a concentrated oligopoly with a fast-moving fringe. L'Oréal's H1 2026 sales of €23.776bn exceed Estée Lauder's entire fiscal-year $15.049bn. But e.l.f. grew 36% in a single quarter, so the fringe is not decorative: the barrier is not R&D, it is retail shelf and regulatory compliance cost, and MoCRA raises the latter.
• Sportswear is a duopoly under attack. NIKE at $46.4bn and adidas growing 14% currency-neutral, with PUMA loss-making and challengers taking share.
• Fast fashion is a two-firm race plus insurgents. Inditex and H&M, with Shein and Temu competing on a cost base that de minimis closure directly attacks.
• Resale is fragmented and structurally unprofitable except at the peer-to-peer end.

Where margin actually sits. In luxury, at the maison: Richemont's Jewellery Maisons earn a 30.5% operating margin while its Fashion & Accessories arm loses money, inside the same group with the same retail network. In beauty, margin has moved toward formulation plus channel control: L'Oréal's 21.3% operating margin and 74.8% gross margin come from scale manufacturing plus pharmacy and professional channels that are harder to disintermediate than department stores. In sportswear, margin is currently not where the reported numbers suggest — see §8. In resale, margin sits with whoever does not take custody of the inventory.

Pricing power, tested. US apparel CPI rose 3.6% in the twelve months to August 2026 and footwear 3.6%, against core CPI of 2.4% — so the sector as a whole passed through cost. But lululemon raised price into a −12% Americas comparable-sales decline, which is the definition of pricing power failing. Pricing power in 2026 is brand-specific, not sector-wide.

On market size. This dossier deliberately quotes no total market-size figure. Every readily available "global fashion market $X trillion by 2031" number encountered during research came from a content-farm or vendor-sponsored source with no disclosed methodology, and the contract forbids laundering those. The honest scale anchor is issuer-disclosed revenue: the ten listed groups covered here with retrievable 2026 disclosures reported, on their own most recent published periods, roughly €150bn-equivalent of combined revenue — a bottom-up sum of primary disclosures, not a market estimate, and explicitly not a measure of the sector's total size. See §13.

4. Who matters

Leading companies. LVMH (https://www.lvmh.com) · Hermès International (https://finance.hermes.com) · Kering (https://www.kering.com) · Compagnie Financière Richemont (https://www.richemont.com) · L'Oréal (https://www.loreal-finance.com) · The Estée Lauder Companies (https://www.elcompanies.com) · Coty (https://www.coty.com) · e.l.f. Beauty (https://www.elfbeauty.com) · NIKE (https://about.nike.com) · adidas AG (https://www.adidas-group.com) · PUMA SE (https://about.puma.com) · lululemon athletica (https://corporate.lululemon.com) · H&M Group (https://hmgroup.com) · Inditex (https://www.inditex.com) · Fast Retailing (https://www.fastretailing.com) · Zalando SE (https://corporate.zalando.com) · ThredUp (https://www.thredup.com) · The RealReal (https://www.therealreal.com) · G-III Apparel Group (https://www.g-iii.com) · China Tourism Group Duty Free (https://www.ctgdutyfree.com.cn)

Notable private and startup. Vinted (https://company.vinted.com) — the only scaled resale platform disclosing a net profit · Shein (https://www.sheingroup.com) — no verified financials · Vestiaire Collective (https://www.vestiairecollective.com) · rhode (https://www.rhodeskin.com), acquired by e.l.f. with a revenue-linked earnout · Profound (https://www.tryprofound.com) — generative engine optimisation · Nomagic (https://www.nomagic.ai) — robotic picking, named by Zalando.

Active investors and acquirers. WHP Global (https://www.whp-global.com) — brand-management platform acquiring Marc Jacobs · G-III Apparel Group — co-owner and operator · China Tourism Group Duty Free — acquirer of DFS Greater China · L Catterton (https://www.lcatterton.com) — consumer-focused private equity affiliated with LVMH. Venture funding is not a meaningful driver in this sector and sector-level totals are not reported here; the capital that matters is corporate M&A and licensing.

Platforms and standards bodies. Zalando's SCAYLE and ZEOS · the EU Digital Product Passport framework under ESPR · Refashion (https://refashion.fr), the French textile producer responsibility organisation and operational template for EU-wide EPR · Textile Exchange (https://textileexchange.org) · Better Cotton (https://bettercotton.org).

Regulators. European Commission DG Environment (https://environment.ec.europa.eu) — Waste Framework Directive and textile EPR · European Commission ESPR framework (https://commission.europa.eu) — ecodesign, unsold-goods destruction ban, Digital Product Passport · US FDA Office of Cosmetics and Colors (https://www.fda.gov/cosmetics) — MoCRA · US Consumer Product Safety Commission (https://www.cpsc.gov) — children's apparel and product safety · US Customs and Border Protection (https://www.cbp.gov) — de minimis and tariff collection · Autorité des marchés financiers (https://www.amf-france.org) and CNMV (https://www.cnmv.es) — issuer disclosure for the French and Spanish groups.

Research institutions and statistical agencies. US Bureau of Labor Statistics (https://www.bls.gov/cpi/) · USDA Economic Research Service (https://www.ers.usda.gov) · Eurostat (https://ec.europa.eu/eurostat) · OTEXA, US Department of Commerce (https://otexa.trade.gov) · European Environment Agency (https://www.eea.europa.eu).

Trade organisations. EURATEX (https://euratex.eu) · Refashion (https://refashion.fr) · Federation of the Swiss Watch Industry FH (https://www.fhs.swiss) · Personal Care Products Council (https://www.personalcarecouncil.org) · American Apparel & Footwear Association (https://www.aafaglobal.org) · Cosmetics Europe (https://cosmeticseurope.eu).

Consumer and civil-society groups. Clean Clothes Campaign (https://cleanclothes.org) · Fashion Revolution (https://www.fashionrevolution.org) · Changing Markets Foundation (https://changingmarkets.org) · BEUC, the European Consumer Organisation (https://www.beuc.eu) — the complainant behind several EU consumer-protection actions against low-value import platforms.

Influential named people. Bernard Arnault (Chairman and CEO, LVMH) · Axel Dumas (Executive Chairman, Hermès) · Johann Rupert (Chairman, Richemont) · Nicolas Hieronimus (CEO, L'Oréal) · Stéphane de La Faverie (President and CEO, The Estée Lauder Companies) · Björn Gulden (CEO, adidas) · Heidi O'Neill (CEO, lululemon, named 2026-04-22) · Thomas Plantenga (CEO, Vinted) · Tarang Amin (Chairman and CEO, e.l.f. Beauty) · Aude Gandon (Global Chief Digital and Marketing Officer, The Estée Lauder Companies) · James Reinhart (CEO, ThredUp).

5. Products, business models, technologies, customers

Major products. Leather goods and handbags; high jewellery and fine jewellery; mechanical watches; ready-to-wear and footwear; performance athletic footwear and apparel; technical activewear; mass-market apparel; prestige and mass skin care, makeup and fragrance; dermocosmetics; salon-professional hair care; and secondhand versions of all of the above.

How money is actually made today.

• Luxury: vertically integrated retail at 60–70%+ gross margin, with scarcity managed on the supply side. Hermès converts this into a 41.0% recurring operating margin; LVMH into 22.5%; Richemont into 20.0% at group level but 30.5% in jewellery and 3.4% in watches. The spread within Richemont is the single most instructive number in luxury.
• Beauty: formulation scale plus channel position. L'Oréal's gross margin is 74.8%, Estée Lauder's 75.5%. The money is made on mix and on advertising efficiency, not on manufacturing cost.
• Licensing, and its return. The largest structural change of 2026 is the re-emergence of the long-dated beauty licence: Kering handed Gucci beauty to L'Oréal for 50 years from 2027-07-01. A 50-year licence trades long-duration brand margin for royalty certainty, which is what a group under balance-sheet pressure does.
• Sportswear: wholesale plus direct. NIKE's multi-year shift to direct has reversed — NIKE Direct fell 8% currency-neutral in fiscal 2026 while wholesale grew in Q4, which is the inverse of the strategy the industry copied for five years.
• Resale: two incompatible models. Peer-to-peer (Vinted) pushes photography, listing and shipping onto the seller and monetises through buyer fees, advertising and a logistics arm — €62m net profit on €1.1bn revenue. Managed consignment (ThredUp) takes custody and cannot cover operating cost even at a 79.9% gross margin — a $5.9m net loss on record revenue.
• Platform B2B: Zalando now sells its own operating stack. B2B revenue grew 27.6% to €334.7m at a 12.2% adjusted EBIT margin, up from 4.3% — twice the margin of its consumer business.

Technologies that matter, ranked by disclosed evidence.

1. Generative AI in content operations — the only AI application in this sector with hard published numbers. Zalando's SCAYLE STUDIOS reached 100+ brands in 2.5 months with content production time down >95% and cost down ~90%; generative image processing handles ~6,000 articles daily; up to 85% of articles go live in under three days.
2. Conversational commerce assistants — Zalando Assistant at ~10m users year to date versus 6m in 2025, with a 63% increase in high-value interactions in H1 2026.
3. Warehouse robotics — 2m automated picks per month at Zalando with Nomagic.
4. Generative engine optimisation — Estée Lauder deploying Profound across 20+ brands and ~150 markets (2026-09-14), with no disclosed outcome metrics yet.
5. Digital product passports — a legal framework under ESPR, not yet a deployed technology for textiles; no delegated act verified as adopted.
6. AI-generated design and virtual try-on — no disclosed revenue, margin or return-rate benefit from any issuer reviewed. See T-23-20.

Customer segments and what they buy on. The aspirational luxury consumer buys on entry price and has been squeezed out — this is why soft luxury is flat while high jewellery grows. The ultra-high-net-worth consumer buys on scarcity and is insensitive to price, which is why Richemont's jewellery grew 24% in a quarter. The beauty consumer increasingly buys on efficacy claims and ingredient literacy, which is why dermocosmetics outgrows prestige makeup. The mass apparel consumer buys on price into a market where apparel CPI is running 1.2 points above core — which is why resale supply and demand both rise.

6. Geography

Production concentrates in Asia. USDA's August 2026 data shows the top five suppliers of US cotton products — China, India, Bangladesh, Vietnam, Pakistan — at 69% of imports, with China's share down to 20%, India's to 13% and Bangladesh's up to 12.5%. Total US cotton-product imports fell 6% to 8.4m bale-equivalents in January–June 2026: volumes falling and mix shifting simultaneously. European luxury production remains concentrated in France, Italy and Switzerland by regulatory origin rules and by craft-labour supply.

Capital and corporate control concentrate in France and Switzerland. LVMH, Hermès, Kering and L'Oréal are French; Richemont and the watch industry Swiss; adidas, PUMA and Zalando German; H&M Swedish; Inditex Spanish; Vinted Lithuanian. This is why the sector's disclosure backbone is euro-denominated half-year reporting rather than SEC filings.

Demand has rotated away from the narrative centre. On the most recent disclosed periods: Richemont's Americas +27% and Japan +36% at constant rates in the June 2026 quarter, with Asia Pacific +21% and Europe +11%; Hermès's Americas +15.3% against Asia excluding Japan +2.4%. Japan's apparent boom is substantially currency: Richemont's Japan grew 36% at constant rates but only 20% at actual rates, a 16-point wedge that reverses if the yen strengthens.

Regulation concentrates in the EU, and this is the single most important geographic asymmetry in the sector. Directive (EU) 2025/1892 requires every Member State to establish textile extended producer responsibility by 17 April 2028. ESPR bans destruction of unsold apparel and footwear and creates the Digital Product Passport. France went further and alone: Loi n. 2026-602 du 8 juillet 2026 bans advertising for ultra-fast-fashion products and brands. Nothing comparable exists in the US, where regulation of this sector runs through trade law (tariffs, de minimis, forced-labour import bans) and through MoCRA for cosmetics.

Required non-US, non-French, non-Italian regional sources used here. Vinted (Lithuania) — FY2025 results, the only disclosed profitable resale platform; Richemont (Switzerland) — FY2026 and Q1 FY2027 ad hoc announcements under SIX Art. 53 LR; H&M (Sweden) — six-month report to 31 May 2026; adidas and Zalando (Germany) — Q2 2026. Vinted in particular is the sector's most valuable under-used regional disclosure: a €10.8bn-GMV European platform publishing audited-style annual figures voluntarily, with no equivalent US comparator.

Mainland China, specifically. Company-disclosed data contradicts the prevailing narrative. Estée Lauder's mainland China organic net sales grew 9% in fiscal 2026 (Q4 +7%) with value share gains. Richemont's China, Hong Kong and Macau grew low single digits across FY2026 and returned to double digits in the June 2026 quarter. L'Oréal's North Asia grew 4.6% adjusted like-for-like with management describing progressive improvement and roughly 15% growth for its Luxe and Dermatological Beauty divisions in China. LVMH described strong growth in Asia excluding Japan in both Q1 and H1. Hermès is the outlier at +2.4%, and that is a supply constraint, not a demand signal. A direct Chinese statistical source (National Bureau of Statistics monthly retail sales by category) could not be retrieved and is recorded as a gap.

7. Historical trend patterns

Over 10–25 years this sector has cycled with unusual regularity, and it has produced a long list of confident false positives. Being specific about them is the point.

What has genuinely cycled.

• The Chinese luxury cycle, roughly four times. 2008–2012 boom; 2013–2015 anti-corruption bust (gifting collapse); 2016–2019 recovery and repatriation of spend to the mainland; 2020–2021 pandemic super-cycle as travel spend came home; 2022–2024 property-driven bust; 2025–2026 stabilisation. Each downturn was declared structural at the time. None has been yet.
• The watch cycle. Swiss watch exports have gone through repeated two-to-three-year contractions. Richemont's own language for the period ending FY2026 — "a challenging 24-month period for the watch market" — is the fifth such episode in twenty-five years.
• Fast-fashion expansion and contraction. Two decades of store-count growth to roughly 2016, then a decade of developed-market closure. H&M's estate fell to 4,038 from 4,166 in a year.
• Beauty's counter-cyclicality. The "lipstick effect" is real but has migrated: in 2026 the resilient category is not lipstick — makeup was flat at Estée Lauder — it is efficacy-led skin care and fragrance.

Prior hype waves and how they resolved — this sector's false positives.

1. Fashion NFTs and the metaverse wardrobe (2021–2022). Every major house launched something. Commercial residue in 2026: essentially none. No issuer reviewed discloses metaverse revenue.
2. Made-to-measure at scale via body scanning (2014–2018). Repeated pilots, no disclosed return-rate or margin benefit at scale. The current virtual try-on wave is the same claim with better rendering, and still no disclosed numbers — this is why T-23-20 is classified overhyped.
3. Direct-to-consumer as a strategy (2016–2022). Every brand was told to go direct. NIKE executed it hardest and has now reversed: NIKE Direct fell 8% currency-neutral in fiscal 2026 while wholesale grew in Q4. DTC turned out to be a channel choice, not a moat.
4. Blockchain product authentication (2019–2022). Aura and comparable consortia launched with large claims. The idea has survived but only by being absorbed into a regulatory mandate — the ESPR Digital Product Passport — rather than winning commercially on its own.
5. "Sustainable fashion" as a consumer-demand-led premium (2018–2022). Willingness-to-pay surveys promised a green premium that never showed up in issuer revenue. What is actually changing behaviour in 2026 is not consumer preference; it is mandatory producer fees and an advertising ban. The trend was right and the mechanism was wrong, which is a specific and repeatable failure mode.
6. Ultra-fast fashion as unstoppable (2021–2024). Treated as structurally advantaged until a customs rule change removed the advantage. The lesson: a business model whose edge is a tariff exemption is a regulatory position, not a moat.

The pattern worth generalising. In this sector, technology hype waves have almost all failed to produce disclosed economics, while regulatory and trade changes have repeatedly produced large, immediate, measurable effects. A trend platform covering fashion and beauty should weight customs law and EU directives far more heavily than it weights product technology, because that is where the historical base rate of real consequence sits.

8. What is changing now (as of 2026-09-15)

1. The single largest distortion in this sector's 2026 financials is a legal windfall. The macro brief established that court-ordered tariff refunds are inflating 2026 retail margins. In this sector the effect is larger relative to earnings than anywhere else measured:

Strip the refunds and lululemon's gross margin fell roughly 360bp rather than rising 200bp. This is the most important adjustment a reader of 2026 fashion and beauty results can make.

2. The "luxury slowdown" is a category story, not a China story. Richemont Jewellery +24% against Specialist Watchmakers at a 3.4% FY operating margin; LVMH Watches & Jewelry +9% organic against Fashion & Leather Goods −1%; Hermès Leather Goods +9.8% against Ready-to-wear +2.0% and Perfume & Beauty −4.5%. Group-level organic growth conceals a 25-point spread inside single companies.

3. Chinese demand has stabilised and the consensus narrative has not caught up. See §6. Meanwhile the Americas is the fastest-growing luxury region and Japan's growth is 44% currency.

4. Prices are up and the regulator's own index proves it. Apparel +3.6% and footwear +3.6% over twelve months to August 2026, personal care +3.8%, against core CPI 2.4%. Note the index dependency: against core PCE at ~3.3%, apparel's premium disappears.

5. Beauty's outperformance is real but mis-specified. It is not prestige-versus-mass. It is dermocosmetics (+11.3%), professional (+12.6%) and value-innovation mass (e.l.f. +36%) beating classic prestige (Estée Lauder organic +3%, LVMH Perfumes & Cosmetics 0%, Hermès Perfume & Beauty −4.5%). Fragrance (+10%) is the exception inside prestige.

6. The conglomerates have become net sellers. Three transactions in seven months: DFS Greater China to China Tourism Group Duty Free (2026-01-19); Marc Jacobs to WHP Global with G-III (2026-05-14); Gucci beauty licensed to L'Oréal for 50 years (2026-07-07). This is the reverse of the 2010–2022 accumulation pattern.

7. EU textile regulation acquired hard dates, and France jumped ahead. Directive (EU) 2025/1892 in force 2025-10-16, EPR schemes required by 17 April 2028. ESPR's unsold-goods destruction ban and Digital Product Passport apply across the bloc. Loi n. 2026-602 was promulgated on 2026-07-08 — the first national advertising ban on ultra-fast fashion in a major market — but its thresholds and penalties await décrets en Conseil d'État and therefore bind nothing yet.

8. Sourcing is reorganising, not just being taxed. US cotton-product imports down 6% with China's share at 20% and Bangladesh's rising to 12.5% — falling volume and shifting mix.

9. A second cost shock is loading for 2027. USDA forecasts the US upland cotton farm price at 75 cents/lb against 61.5 cents, +22%, with world ending stocks at their lowest since 2011/12 — arriving after refunds stop and after pricing headroom is spent.

10. AI has produced real numbers, in an unglamorous place. Zalando's content-operations disclosures are the only quantified generative-AI economics in this sector. Design and try-on have none.

9. The five lists

Five most important current trends

1. Court-ordered tariff refunds distorting sector margins (T-23-01)
2. Hard luxury pulling away from soft luxury (T-23-02)
3. Chinese demand stabilising while the Americas and Japan carry growth (T-23-03)
4. Dermocosmetics, professional and value-mass beauty outgrowing classic prestige (T-23-05)
5. Violent share redistribution inside sportswear (T-23-08)

Five fastest-growing signals

1. EU textile extended producer responsibility with a 17 April 2028 deadline (T-23-09)
2. France's Loi 2026-602 advertising ban on ultra-fast fashion (T-23-10)
3. Generative AI collapsing product-content production cost (T-23-12)
4. Cotton price shock rebuilding 2027 input costs (T-23-14)
5. Sourcing reallocation away from China under the rebuilt tariff architecture (T-23-15)

Overlap note: T-23-15 is also driven by the same trade-law change as T-23-01, and T-23-09 and T-23-10 are two instruments of one regulatory direction.

Five trends most likely to affect businesses

1. Tariff refunds and their 2027 disappearance (T-23-01) — also in list 1
2. EU textile EPR and eco-modulated producer fees (T-23-09) — also in list 2
3. Cotton at +22% into 2027 costings (T-23-14) — also in list 2
4. Luxury conglomerate divestment and licensing-out (T-23-06)
5. Generative AI in content operations resetting the minimum viable assortment (T-23-12) — also in list 2

Five trends most likely to affect consumers

1. Apparel and personal-care prices running above core inflation (T-23-04)
2. Peer-to-peer resale as a mainstream substitute for new goods (T-23-07)
3. France's advertising ban changing what ultra-fast fashion is allowed to say (T-23-10) — also in list 2
4. Premium activewear discounting as the category reverses (T-23-16)
5. Mass-fashion store closures removing physical access in secondary locations (T-23-17)

10. Overhyped / overlooked / cooling / reversing

Most overhyped

1. GLP-1 as a reshaper of apparel sizing and beauty demand (T-23-19). Not one of the eleven brand-level results documents reviewed for this dossier — LVMH, Hermès, Richemont, L'Oréal, Estée Lauder, e.l.f., NIKE, adidas, PUMA, lululemon, H&M — quantifies a GLP-1 effect on units, size mix, returns or category revenue. The mechanism is plausible; the evidence is a set of consultancy models. Every circulating figure must be labelled estimate with the modeller named.

2. AI-generated design and virtual try-on as a near-term revenue driver (T-23-20). The evidence that exists points elsewhere. Zalando's quantified AI outcomes are all operational — content production time −95%, cost −90%, 6,000 articles/day, 2m robotic picks/month — and none concern design or return rates. Estée Lauder's flagship AI announcement is about answer-engine visibility. Virtual try-on has been promised since roughly 2015 without a single disclosed return-rate improvement at scale.

3. "Luxury is collapsing because of China." Overhyped as a framing. Richemont's China, Hong Kong and Macau returned to double-digit growth; Estée Lauder's mainland China grew 9%. The real weakness is in soft luxury across all regions, including Europe and the aspirational US consumer.

Most overlooked

1. The 2027 cotton shock (T-23-14). A forecast 22% increase in the US upland farm price, with world ending stocks at a fifteen-year low, landing exactly when tariff refunds stop and after apparel prices have already risen 3.6%. Fashion trade press barely covers fibre markets, which is why this is not priced.

2. Richemont's watch division as an earnings option (T-23-18). A division at a 3.4% operating margin returning to +8% growth has more operating leverage than anything else in European luxury. Nobody models it because the recovery is being led by the Americas and Japan rather than by the China demand base that drove the last watch cycle.

3. Zalando's B2B arm. B2B adjusted EBIT margin of 12.2%, up from 4.3%, on revenue growing 27.6% — twice the margin of the consumer marketplace. A European fashion platform quietly became a software and logistics business and is still valued as a retailer.

4. That the EU's textile rules bind in 2028, not now. Compliance vendors are selling Digital Product Passport readiness against dates that do not exist yet — no textile ESPR delegated act was verified as adopted at this research date. The real binding date, 17 April 2028, is far enough away that scope and fee levels can still be diluted in transposition.

5. Vinted as the best disclosure in European consumer internet. A €10.8bn-GMV platform publishing revenue, net profit, EBITDA and free cash flow voluntarily, while its listed US comparators disclose less useful economics.

Trends that appear to be cooling — with the indicator that turned

• Premium activewear (T-23-16). Indicator: lululemon Americas comparable sales −12% and full-year revenue guided down 5–7%. Underlying gross margin down ~360bp once refunds are removed. A category defined by pricing power now guiding to a revenue decline.
• Mass-fashion physical footprint (T-23-17). Indicator: H&M at 4,038 stores versus 4,166, with ~170 net closures planned for 2026 and SEK 679m of restructuring costs — while margin rose, meaning profit recovery is coming from sourcing, not from customers.
• Direct-to-consumer as sportswear orthodoxy. Indicator: NIKE Direct −8% currency-neutral for fiscal 2026 while Q4 wholesale grew 1% currency-neutral. The strategy the whole sector copied has been reversed by the firm that proved it.

Trends that may reverse — and the mechanism

• The tariff-refund windfall reverses mechanically in 2027 simply by not recurring. Every company above faces a gross-margin comparison it cannot repeat. Mechanism: arithmetic.
• The Swiss watch downturn is already reversing (T-23-18): +8% at constant rates in the June 2026 quarter after 24 months of contraction. Mechanism: operating leverage off a 3.4% margin base, led by the Americas and Japan. Watch for whether Chinese watch demand, still declining, joins.
• De minimis closure is reversible by executive or judicial action. The US suspension rests on a Court of International Trade ruling of 2026-08-13 on a privilege-revocation theory. A reversal restores the ultra-fast-fashion cost advantage within a quarter. Mechanism: the same courts that removed it.
• France's advertising ban could be neutered without being repealed. If the décrets en Conseil d'État set thresholds that catch only two or three non-EU sellers — or are simply never published, as the earlier eco-modulation decree was repeatedly delayed — the law binds nobody. Mechanism: administrative inaction, the same failure mode as the EU AI Act high-risk deferral.
• Japanese luxury growth reverses on currency. Richemont's Japan grew 36% at constant rates but 20% at actual rates. Mechanism: yen appreciation turns a 16-point tailwind into a headwind.

11. Risks and major uncertainties

Sector-specific risks.

1. Margin cliff in 2027. Refunds stop, cotton rises ~22%, and pricing headroom is spent. Three negatives arriving in one planning cycle.
2. Aspirational-consumer attrition in luxury. Soft luxury is flat while hard luxury grows, which means the entry-price customer has left. Brands that raised prices to defend margin may find the funnel has been cut below them.
3. Regulatory cost asymmetry between the EU and the US. EU producers face EPR fees, ecodesign and destruction bans; US-facing sellers face tariffs and forced-labour import rules. A global brand now runs two incompatible compliance models.
4. Concentration of luxury earnings in one group. LVMH's half-year revenue exceeds Richemont's full year. Sector aggregates are an LVMH proxy, and LVMH's recurring profit fell 4%.
5. Resale profitability is model-dependent, not scale-dependent. ThredUp's 79.9% gross margin cannot cover operating cost. Scaling a custody-based resale business does not fix it.
6. Labour and supply-chain exposure. Sourcing reallocation toward Bangladesh raises both concentration risk and audit exposure at a time when EU due-diligence rules are tightening.
7. Public-market disclosure is thinning. Shein is unmeasurable; Kering and Inditex were unreachable in this research. The sector's observability is declining.

Genuine unknowns — and the distinction that matters.

Things we do not know but could find out: Kering's H1 2026 brand-level performance (published, blocked to us); Inditex's H1 2026 (published, blocked); Swiss watch exports by destination (published, robots-blocked); OTEXA's full textile and apparel import series (published, blocked); The RealReal's Q2 2026 (published, not located). All are retrieval failures, not information gaps, and are the top priority for a re-run.

Things nobody can currently know: whether the décrets implementing Loi 2026-602 will set thresholds that catch two companies or two hundred; whether the US de minimis suspension survives the next election cycle; whether Chinese luxury stabilisation is a durable recovery or a repatriation artefact from Hong Kong and Macau; whether GLP-1 adoption has any measurable apparel effect at all, since no issuer is obliged to attribute demand to a drug class and none has; whether a 50-year beauty licence proves to be value creation or value transfer, since neither party disclosed terms.

12. Scenarios to 2030

Base — bifurcation hardens. Hard luxury and efficacy-led beauty compound; soft luxury and classic prestige stagnate; sportswear share continues to redistribute. 2027 margins fall as refunds lapse and cotton bites, then stabilise. EU textile EPR lands roughly on time in 2028 at modest fee levels. Resale grows but only the peer-to-peer model earns money. Falsifiable early indicator: Richemont's Jewellery-to-Specialist-Watchmakers growth gap narrows below 10 points for two consecutive quarters — that would say the bifurcation is closing.

Upside — synchronised recovery. Chinese demand recovery broadens from jewellery and beauty into leather goods and ready-to-wear; the watch cycle completes its turn and Richemont's watch margin returns toward the mid-teens; cotton yields recover in 2027/28 and the price forecast unwinds. Soft luxury returns to mid-single-digit organic growth. Indicator: LVMH Fashion & Leather Goods posts two consecutive quarters of organic growth above +3%, having managed only +1% in Q2 2026.

Downside — the 2027 margin cliff bites hard. Refunds lapse, cotton rises 22%, apparel prices have no further headroom above a core CPI already at 2.4%, and volume falls. Premium activewear's reversal spreads to premium apparel generally. Store closures accelerate and mid-market brands consolidate or fail. Indicator: apparel CPI turns negative month-on-month for three consecutive months while unit volumes also decline — price cutting into falling demand, the classic signature.

Disruption — AI content economics reset the competitive floor. If a 90% reduction in content production cost generalises beyond Zalando, the minimum viable assortment collapses and thousands of micro-brands become economically viable, fragmenting mass-market share away from incumbents. Conversely, it may simply hand scale platforms a further cost advantage. Indicator: a second, independent issuer discloses content-production savings of comparable magnitude. Until that happens this remains single-source and is scored accordingly.

Regulatory — the EU becomes the global rule-setter, or does not. Upside case: textile EPR lands on 17 April 2028 with meaningful eco-modulation, ESPR textile delegated acts arrive, and France's decrees set real thresholds — producing a genuine cost differential for low-durability goods that other jurisdictions copy. Downside case: transposition dilutes scope, ESPR textile acts slip, and the French decrees never publish. Indicator: publication of the first décret en Conseil d'État under Loi 2026-602 defining the fast-fashion threshold. If it has not appeared within twelve months of promulgation, assume the diluted branch. This is the sector's version of the AI Act high-risk deferral precedent.

Failure — trade law reverses and the model breaks. US de minimis is restored by a court or by statute, tariffs are further unwound, and ultra-fast fashion's cost advantage returns in full while EU producers carry new EPR fees. European mass fashion is caught between a regulated cost base and an unregulated competitor. H&M's store closures accelerate well beyond 170 a year. Indicator: any judicial or legislative action restoring the de minimis exemption, or a successful challenge to Loi 2026-602 on EU free-movement grounds.

13. Data gaps and limitations

Retrieval failures — published data we could not reach. These are the most serious limitations in this dossier and all are fixable.

1. Kering. kering.com returned a hard site block to every automated fetch. Kering's H1 2026 results — brand-level revenue for Gucci, Saint Laurent and Bottega Veneta — are the single most important missing dataset in this dossier. Kering's position is inferred here only from the L'Oréal Gucci beauty licence. Alternative route for Phase 2: info-financiere.gouv.fr or the AMF.
2. Inditex. Every inditex.com path returned robots.txt fetch failures, so the largest fast-fashion group contributed no 2026 data. Alternative route: the Spanish regulator CNMV.
3. OTEXA (US Department of Commerce textile and apparel import data) — robots.txt connect timeout. Sourcing-shift evidence had to be taken from the narrower USDA cotton-product import series instead. This is the highest-value unexercised dataset for the sector.
4. Federation of the Swiss Watch Industry — robots.txt failure. The watch-cycle reversal is therefore corroborated only by issuer disclosure, with no independent industry series.
5. The RealReal Q2 2026 — investor site is a JavaScript shell and no corresponding 8-K was located in EDGAR for July–September 2026. Resale economics rest on Vinted and ThredUp alone.
6. Coty fiscal 2026 — investor site rendered as a JavaScript shell. Fragrance and mass-beauty coverage is thinner than it should be.
7. China National Bureau of Statistics monthly retail sales by category (cosmetics; garments, footwear and knitwear; gold, silver and jewellery) — the page served 2024 content. The Chinese demand assessment therefore rests entirely on foreign issuers' disclosure of Chinese revenue, with no domestic statistical cross-check.
8. Eurostat extra-EU textile trade — both attempted statistics-explained URLs 404'd.
9. Direct Shein and Temu financials do not exist. Shein is private with no verified financials; Temu is not broken out by PDD. The ultra-fast-fashion effect of de minimis closure cannot be measured directly and is inferred from customs aggregates and price indices.

Figures we could not verify and have flagged in place.

• The exact calendar date on which the ESPR unsold-goods destruction ban began applying to large enterprises. The regulation confirms apparel, clothing accessories and footwear are covered, that medium-sized enterprises get six years from entry into force (18 July 2024), and that micro and small enterprises are exempt — but the HTML fetch of Regulation (EU) 2024/1781 truncated before Article 25's application date. The widely circulated 19 July 2026 date is recorded here as unconfirmed.
• Whether Loi 2026-602 was notified to the European Commission under the technical-standards procedure. The legislative dossier does not say, and this bears directly on its legal durability.
• MoCRA facility-registration and product-listing compliance deadlines: the FDA overview page states the obligations without dates.
• Transaction values for all three 2026 luxury portfolio transactions (DFS Greater China, Marc Jacobs, Gucci beauty licence) — none disclosed. Value creation cannot be assessed.
• Vinted's member and active-buyer counts; the 2025 results release omits them.
• Profound's funding and scale; not disclosed in the Estée Lauder announcement.
• Leadership for Kering, Coty, PUMA, NIKE, H&M and Inditex could not be confirmed from a primary source within this research and is recorded as unverified in the entity records rather than asserted from memory.

Conflicting figures recorded rather than resolved.

• Chinese/North Asian beauty growth: Estée Lauder mainland China +9% organic versus L'Oréal North Asia +4.6% adjusted like-for-like. Scope difference (prestige-only, mainland-only versus a zone including Japan and Korea and large mass volumes), not a factual dispute.
• Asian luxury growth: Hermès Asia excluding Japan +2.4% versus Richemont Asia Pacific +21% over overlapping mid-2026 periods. Category mix and Hermès's deliberate supply constraint, not disagreement about the market.
• US core inflation: core CPI 2.4% (August 2026) versus core PCE ~3.3%. Apparel's premium over "core" is 1.2 points against one index and roughly nil against the other. Always name the index.
• Resale profitability: Vinted +€62m net profit; ThredUp −$5.9m at a 79.9% gross margin. Different models, not different facts.

Structural limitations.

• No market-size figure is quoted anywhere in this dossier, by choice. Every accessible "global fashion/beauty market size" source was a content farm or a vendor-sponsored report. The scale anchor in §3 is a bottom-up sum of disclosed issuer revenue and is labelled as such.
• Fiscal-year misalignment. NIKE reports to May, Estée Lauder to June, Richemont to March, H&M to November, e.l.f. and lululemon to varying quarters. Cross-company comparison in this dossier is period-approximate and every figure carries its period.
• Currency. European groups report in euros and Swiss and Swedish issuers in euros and SEK; the reported-versus-constant-currency gap reached 16 points for Richemont Japan. Reported growth and constant-rate growth are never used interchangeably here.
• Discovery breadth. With zero search calls available, this dossier found what it could reach by direct URL. Trends visible only through secondary reporting — designer changes, private-brand performance, emerging-market challengers, supply-chain labour incidents — are systematically under-represented. This sector should be prioritised for re-run at full search budget.

14. Ranking scorecard

15. Sources

1. Accelerating growth in the second quarter — Solid first-half results — LVMH — https://www.lvmh.com/en/publications/accelerating-growth-in-the-second-quarter---solid-first-half-results — 2026-07-27 — A
2. LVMH continues to achieve organic growth in the first quarter in a global environment impacted by the conflict in the Middle East — LVMH — https://www.lvmh.com/en/publications/lvmh-continues-to-achieve-organic-growth-in-the-first-quarter-in-a-global-environment-impacted-by-the-conflict-in-the-middle-east — 2026-04-13 — A
3. DFS and China Tourism Group Duty Free announce agreement for sale and purchase of DFS Greater China retail business — LVMH — https://www.lvmh.com/en/publications/dfs-and-china-tourism-group-duty-free-announce-agreement-for-sale-and-purchase-of-dfs-greater-china-retail-business — 2026-01-19 — A
4. LVMH and WHP Global announce definitive agreement for the acquisition of Marc Jacobs — LVMH — https://www.lvmh.com/en/publications/lvmh-and-whp-global-announce-definitive-agreement-for-the-acquisition-of-marc-jacobs — 2026-05-14 — A
5. 2026 First Half Results (press release PDF) — Hermès International — https://assets-finance.hermes.com/s3fs-public/node/pdf_file/2026-07/1785260662/hermes_20260729_pr_firsthalfresults_va.pdf — 2026-07-29 — A
6. Ad hoc announcement pursuant to Art. 53 LR — FY26 Annual Results — Compagnie Financière Richemont SA — https://www.richemont.com/media/ovfnkffy/ad-hoc-announcement-pursuant-to-art-53-lr-fy26-annual-results-en.pdf — 2026-05-22 — A
7. Ad hoc announcement pursuant to Art. 53 LR — FY27 Q1 Sales — Compagnie Financière Richemont SA — https://www.richemont.com/media/xikaciqj/ad-hoc-announcement-pursuant-to-art-53-lr-fy27-q1-sales-en.pdf — 2026-07-15 — A
8. Résultats semestriels 2026 — L'Oréal — https://www.loreal-finance.com/fr/communique-de-presse/resultats-semestriels-2026 — 2026-07-29 — A
9. L'Oréal conclut avec Kering un accord de licence exclusive mondiale de 50 ans pour la beauté Gucci — L'Oréal — https://www.loreal-finance.com/fr/communique-de-presse/loreal-conclut-avec-kering-un-accord-de-licence-exclusive-mondiale-de-50-ans — 2026-07-07 — A
10. The Estée Lauder Companies Reports Fiscal 2026 Results — The Estée Lauder Companies Inc. — https://www.elcompanies.com/en/news-and-media/newsroom/press-releases/2026/08-19-2026-110037466 — 2026-08-19 — A
11. The Estée Lauder Companies Announces Partnership with Profound to Expand AI Visibility Across Its Global Brand Portfolio — The Estée Lauder Companies Inc. — https://www.elcompanies.com/en/news-and-media/newsroom/press-releases/2026/09-14-2026-173210983 — 2026-09-14 — A
12. e.l.f. Beauty Announces First Quarter Fiscal 2027 Results — e.l.f. Beauty, Inc. — https://investor.elfbeauty.com/stock-and-financial/press-releases/landing-news/2026/08-05-2026-210618204 — 2026-08-05 — A
13. NIKE, Inc. Reports Fiscal 2026 Fourth Quarter and Full Year Results — NIKE, Inc. — https://about.nike.com/en/newsroom/releases/nike-inc-reports-fiscal-2026-fourth-quarter-and-full-year-results — 2026-06-30 — A
14. adidas AG Q2 2026 Results — adidas AG — https://res.cloudinary.com/confirmed-web/image/upload/v1785388904/adidas-group/investors/financial-publications/2026/Q2/EN/adidasAG_Q2_2026_Results_EN_Final_z7f9t2.pdf — 2026-07-30 — A
15. PUMA SE Q2 2026 Results Release — PUMA SE — https://about.puma.com/sites/default/files/financial-report/2026/puma-q2-2026-release-eng-final.pdf — 2026-07-31 — A
16. lululemon athletica inc. Announces Second Quarter Fiscal 2026 Results — lululemon athletica inc. — https://corporate.lululemon.com/newsroom/press-releases/2026/09-03-2026-210528733 — 2026-09-03 — A
17. H & M Hennes & Mauritz AB Six-month report 2026 (1 Dec 2025 – 31 May 2026) — H&M Group — https://hmgroup.com/wp-content/uploads/2026/06/H-M-Hennes-Mauritz-AB-Six-month-report-2026.pdf — 2026-06-25 — A
18. Zalando delivers another quarter of profitable growth in Q2, accelerates B2B expansion — Zalando SE — https://corporate.zalando.com/en/financials/zalando-q2-2026-results — 2026-08-04 — A
19. Zalando delivers strong Q1 as artificial intelligence and ABOUT YOU integration drive growth — Zalando SE — https://corporate.zalando.com/en/financials/zalando-q1-2026-results — 2026-05-06 — A
20. Financial results 2025: Vinted brings second-hand goods to more people — Vinted — https://company.vinted.com/newsroom/financial-results-2025 — 2026-04-09 — A
21. ThredUp Announces Second Quarter 2026 Results — ThredUp Inc. — https://ir.thredup.com/news-releases/news-release-details/thredup-announces-second-quarter-2026-results — 2026-08-05 — A
22. Directive (EU) 2025/1892 amending Directive 2008/98/EC on waste as regards textiles — Official Journal of the European Union (EUR-Lex) — https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=OJ:L_202501892 — 2025-09-26 — A
23. Regulation (EU) 2024/1781 establishing a framework for the setting of ecodesign requirements for sustainable products (ESPR) — Official Journal of the European Union (EUR-Lex) — https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=OJ:L_202401781 — 2024-06-28 — A
24. Textiles — waste and recycling policy — European Commission, DG Environment — https://environment.ec.europa.eu/topics/waste-and-recycling/textiles_en — 2026-09-15 (retrieved; page undated) — A (flagged undated: true)
25. Dossier législatif: réduire l'impact environnemental de l'industrie textile — Loi n° 2026-602 du 8 juillet 2026 — Assemblée nationale (France) — https://www.assemblee-nationale.fr/dyn/17/dossiers/impact_environnemental_industrie_textile — 2026-07-08 — A
26. Consumer Price Index — August 2026 (USDL-26-1496) — U.S. Bureau of Labor Statistics — https://www.bls.gov/news.release/cpi.nr0.htm — 2026-09-11 — A
27. CPI news release table 2 — detailed expenditure categories — U.S. Bureau of Labor Statistics — https://www.bls.gov/news.release/cpi.t02.htm — 2026-09-11 — A
28. Cotton and Wool Outlook: August 2026 (CWS-26h) — USDA Economic Research Service — https://www.ers.usda.gov/media/29452/cws-26h.pdf — 2026-08-14 — A
29. Modernization of Cosmetics Regulation Act of 2022 (MoCRA) — U.S. Food and Drug Administration — https://www.fda.gov/cosmetics/cosmetics-laws-regulations/modernization-cosmetics-regulation-act-2022-mocra — content current as of 2026-08-10 — A
30. The RealReal, Inc. Form 8-K — 2026 annual meeting results — U.S. Securities and Exchange Commission (EDGAR) — https://www.sec.gov/Archives/edgar/data/1573221/000157322126000048/real-20260610.htm — 2026-06-10 — A
31. SHEIN Group newsroom — Shein — https://www.sheingroup.com/newsroom/ — retrieved 2026-09-15 — B (company self-published; no financial disclosure)

Macro framing throughout draws on the shared Macro Context Brief and its two addenda (2026-09-15) — the Supreme Court IEEPA reversal, the Court of International Trade de minimis ruling of 2026-08-13, the EU €3-per-item handling fee from 2026-07-01, the core CPI versus core PCE discrepancy, and the EU AI Act high-risk deferral precedent. These are treated as context, not re-reported as findings.

More research in this sector · Explore trend records