EU Unsold Textiles Destruction Ban Statistics (2026): 43+ Data Points on Company Size Thresholds, Disclosure and 5-Year Recordkeeping, and the Overstock Volume Behind the Rule

Key Takeaways

1. 19 July 2026 — the date the destruction ban began applying to large enterprises; it has been in force ever since (European Commission, T1).
2. 2 March 2027 — the next live deadline: financial years starting on or after this date must use the standardised disclosure format (Commission Implementing Regulation (EU) 2026/2, T1).
3. 10 derogations — permitted exemption categories, enumerated (a)–(j) in Article 2 of the delegated regulation, not the six to nine several summaries report (Commission Delegated Regulation (EU) 2026/296, T1).
4. 250 employees — or more than EUR 50M turnover AND more than EUR 43M balance-sheet total; the test that brings a company inside the ban (Commission Recommendation 2003/361/EC, T1).
5. 5 years — how long derogation evidence must be kept, running from the date of destruction (Commission Delegated Regulation (EU) 2026/296, T1).
6. 30 days — to put that evidence in front of a competent authority in electronic form once it is requested (Commission Delegated Regulation (EU) 2026/296, T1).
7. 12 months — after financial year end to publish the annual unsold-goods disclosure on the company website (ArentFox Schiff, T1).
8. 4–9% — of textile products placed on the EU market are destroyed before use (European Environment Agency, T1).
9. 264,000–594,000 tonnes — of textiles destroyed in Europe every year, carrying up to 5.6 million tonnes CO2e (European Environment Agency, T1).
10. 0.2% — of EU-27 textile and clothing companies are large enough to be caught; 99.8% are micro or SME (EURATEX, T1).
11. Just under 15% — the EU household textile separate-collection rate in 2022, the capacity the ban redirects stock into (European Environment Agency, T1).
12. 88% — of major fashion brands still do not disclose their annual production volumes (Fashion Revolution, T1).

1. The Ban in Force: Scope, Legal Basis, and the Next Live Deadline

The ban has applied since 19 July 2026, and the seven weeks since have been quiet by design. Nothing in the prohibition requires a filing on day one — the first dated obligation is a website disclosure, and the standardised template behind it does not bite until financial years starting on or after 2 March 2027. That gap is the working window, not a grace period.

Read the definition of destruction before assuming a category is safe. Deliberately discarding or physically damaging unsold goods already placed on the EU market is caught, and routing stock to recycling or energy recovery does not automatically put a company outside the rule. Where a derogation applies, disposal still has to follow the waste hierarchy in order: recycling first, then other recovery, then disposal.

Scope is drawn by customs code, not by marketing category: CN 4203, 61 and 62 for apparel and clothing accessories, CN 6401–6405 for footwear. A brand that cannot map its own SKUs to CN codes cannot answer the first question a caught retail partner will ask.

“The prohibition has applied since 19 July 2026. The paperwork that proves compliance arrives on 2 March 2027.”

The 19 July 2030 date extends the same prohibition to medium-sized enterprises. It is written into the regulation, not a forecast.

The destruction ban is one obligation inside a much larger regulation, and it is not the one with the longest lead time. For the dates, product scope and documentation duties that sit around it, see the wider ESPR compliance timeline for garment brands.

2. Who the Threshold Actually Catches

The threshold is an OR test, not a scorecard. 250 or more employees is enough on its own; the financial route needs more than EUR 50 million in turnover and more than EUR 43 million in balance-sheet total. The EUR 25 million balance-sheet figure circulating in compliance write-ups belongs to the 2023 Accounting Directive size-threshold update — a different instrument that ESPR does not reference.

Applied to EU textiles and clothing, that test binds roughly 0.2% of companies. The other 99.8% are micro or SME, and 88.8% employ fewer than ten people.

That statistic gets misread as an exemption. It is closer to the opposite: those SMEs are the suppliers filling the shelves of the retailers who are caught, and the disclosure template a caught buyer has to complete asks for weight and waste-treatment data that only the supply chain can produce. Being outside the ban does not put a brand outside the questionnaire.

“Roughly 0.2% of EU textile and clothing companies are large enough to be caught. The other 99.8% supply the ones that are.”

EURATEX’s size-class breakdown traces to its 2020/2022 Facts & Key Figures editions; the 2026 edition publishes only the headline company, employment and turnover figures. Treat the 99.8 / 88.8 / 0.2 split as directionally current rather than a 2026 measurement.

3. What Must Be Published, and the Five-Year Paper Trail Behind It

Two separate five-year clocks run under this regime, and conflating them misstates the obligation. Derogation evidence is kept for five years from the date the product was destroyed, and must reach a competent authority in electronic form within 30 days of a request. Disclosure documentation runs five years from the disclosure itself, under a different instrument.

The disclosure is more granular than most brands expect. The template asks for units and weight by CN code, the reason for discarding, and a percentage split across reuse, recycling, other recovery, disposal and unknown — calculated on weight, not unit counts.

That single word decides whether a brand can file. Weight per style is a factory-side data point: it comes off the packing list and the fabric consumption sheet, not out of a retail ERP. Brands buying through a trading company routinely never see it, because nobody ever asked the mill or the sewing floor to pass it up the chain.

“The waste-treatment split is calculated on weight. Most brands have never asked their factory for it.”

The first disclosures are already landing in free format. The standardised Annex I template only governs financial years starting on or after 2 March 2027.

Weight by CN code is not a one-off request. It is the same class of factory-held record the next wave of EU rules will ask for on every garment — see the Digital Product Passport data your factory has to hold.

4. The Ten Derogations, and an Enforcement Record That Barely Exists

Ten derogations sound generous until you read what each one costs to prove. The failed-donation route alone requires evidence of an offer rejected by at least three social economy entities, or an unsuccessful eight-week public listing on the company’s own website. Two further derogations exist only for goods that were already donated or already prepared for reuse and found no taker — which is why several legal summaries undercount the list at six to nine.

As of 4 September 2026, no member state has published an ESPR-specific penalty schedule for this obligation and no first enforcement action has surfaced. Germany’s EUR 50,000 figure in the table below is the pre-ESPR Ecodesign ceiling — widely reported, but not an ESPR destruction-ban penalty. The enforcement route through national market surveillance authorities is described consistently across legal commentary rather than confirmed against a pinpoint article of the Regulation.

France is the only useful precedent, and it is a national one. Its DGCCRF campaign figures below are AGEC-law enforcement, not ESPR enforcement: 113 textile establishments inspected in 2024, with roughly 5% of all inspections producing any follow-up. That is what a well-resourced national regulator’s throughput looks like on a rule that has been live since 2022 — a useful calibration for anyone assuming the EU ban will be policed line-by-line in its first year.

“Ten derogations exist. Proving one takes three refused donation offers or an eight-week public listing.”

Tier 3-flagged in this section: Germany’s EUR 50,000 pre-ESPR ceiling, kept because it is the only published penalty figure adjacent to this obligation and the documented absence of an ESPR schedule is itself the finding.

5. The Volume the Ban Redirects

264,000 to 594,000 tonnes of textiles are destroyed in Europe every year before anyone wears them — 4–9% of everything placed on the market, carrying up to 5.6 million tonnes of CO2e, roughly what a million petrol cars emit in a year.

Set that against 6.94 million tonnes of EU-27 textile waste in 2022 and the destroyed share looks marginal. Set it against the 4.6 kg per person that gets separately collected and it does not. Destroyed stock is the most avoidable slice of the whole figure precisely because it never entered a use phase: no wear, no washing, no end-of-life decision — just a production run that met no buyer.

One caveat travels with these numbers. The EEA’s destruction estimates rest on 2020 market data and have not been updated since the March 2024 briefing, so they describe the problem the ban was designed against, not its first months of application.

One figure in the table below needs reading with care. France’s ~EUR 630 million of unsold products destroyed each year is an unfootnoted European Commission estimate — it appears in the Commission’s own press release with no citation to an underlying study. Treat it as an order-of-magnitude illustration of one member state’s exposure, not as a measured national total.

“Between 264,000 and 594,000 tonnes a year — clothing that was made, shipped, warehoused, and never worn.”

France’s EUR 630 million figure is an unfootnoted European Commission estimate: it appears in the Commission’s own press release with no citation to an underlying study. Treat it as an order-of-magnitude illustration.

The destroyed tonnage starts as a purchase order. If you want the numbers behind that decision rather than the aftermath, we published MOQ floors by garment type and fabric route.

6. The Alternatives, Sized — and Where the Overstock Starts

The alternatives the ban points to are real but thin at the bottom of the market. Just under 15% of EU household textile waste was captured separately in 2022, and only 1.7 kg per person was actually reused. Resale is the healthiest route — the global secondhand market is on a path to $393 billion by 2030 — but resale absorbs a leftover 200-piece run at a fraction of wholesale, and preparation-for-reuse operators are built for post-consumer volume, not sealed cartons of one colorway in one size curve.

So the cheapest tonne is the one never cut. Overstock is a purchase-order decision before it is a warehouse problem, and colorway count multiplies it faster than style count does. A 1,000-piece dyed-lot minimum across four colorways commits a brand to 4,000 units before a single one has sold — and that 1,000-piece floor is real: it is our own minimum whenever a style needs custom-dyed fabric or yarn, because a dye house will not run a lot smaller than that.

The way around it is fabric selection, not negotiation. On stock fabric we start at 100 pieces per style for dresses, sportswear and plus-size (300 for knitwear on stock yarn). The same four-colorway story becomes a 400-unit buy, and bulk repeats run about 30 days after PP sample approval, so a brand can watch which colorway moves and reorder it instead of guessing across the whole range up front.

That is the lever a factory-direct relationship gives you and a trading company does not: order quantity gets negotiated against real fabric availability in the mills around us in Zhejiang, not against an intermediary’s own minimums. It is also why 88% of major brands still not publishing production volumes matters — that is exactly the number the disclosure template will eventually pull into daylight.

“Four colorways at a 1,000-piece dyed-lot minimum is 4,000 units bought before one has sold.”

Both forecast rows in this table are projections and are labelled as such: the $393 billion 2030 resale figure and the 2033 micro-factory market size.

Repeat runs on stock fabric are the practical version of “do not make what you cannot sell.” See how we work with small brands on repeat runs, from moodboard through PP sample to bulk.

The EU Unsold Textiles Destruction Ban by the Numbers: All 18 Key Data Points (2026)

Methodology and Sources

Every figure here was traced to the instrument or dataset that first measured or enacted it, not to the article that relayed it. The three legal anchors are Commission Delegated Regulation (EU) 2026/296 (the derogations and the five-year retention rule), Commission Implementing Regulation (EU) 2026/2 (the disclosure format), and Commission Recommendation 2003/361/EC (the enterprise-size test ESPR incorporates). EUR-Lex could not be read by automated fetch during this research, so the delegated regulation was verified against the Commission’s own hosted PDF of C(2026) 659 final, and the implementing regulation against its verbatim text as quoted by named legal analyses — principally Generation Impact Global’s ESPR disclosure analysis, which reproduces the Annex I field list and the CN-code delimitation rule in full, and ArentFox Schiff’s fashion-counsel note. Stats are attributed to the instrument, with the relay that supplied the text recorded alongside them.

Two figures that circulate widely were corrected here: the derogation count is 10, not 9 (Article 2 enumerates points (a) through (j)), and the large-enterprise balance-sheet threshold is EUR 43 million, not EUR 25 million — the EUR 25 million figure belongs to the 2023 Accounting Directive size-threshold update, a different instrument. Volume and destruction figures are attributed to the European Environment Agency, which measured them, rather than to the Commission press release that repeats them.

Tier breakdown: 43 data points kept — 36 Tier 1 (83.7%), 2 Tier 2, 4 Tier 3-consensus and 1 Tier 3-flagged (11.6% combined), across 23 unique sources. Tier 1 is primary legal text or the institution that measured the number; Tier 2 is a reputable aggregator with disclosed methodology; Tier 3-consensus is corroborated across independent sources without a pinpoint primary confirmation; Tier 3-flagged carries a visible inline qualifier and is barred from the Key Takeaways.

Documented absences

• No European Commission guidance document, Q&A or FAQ on the destruction ban has been published since it began applying on 19 July 2026. A promising Commission PDF titled “FAQ-UPDATE-4th-Iteration_clean.pdf” turned out to be the unrelated 2025–2030 Ecodesign Working Plan.
• No member state has published an ESPR-specific penalty schedule for the destruction-ban obligation as of 4 September 2026, and no first enforcement action has surfaced.
• No caught retailer has yet published an ESPR unsold-goods disclosure that we could locate. The first free-format disclosures are due by the end of 2026.

Recency notes

• The European Environment Agency’s destruction figures — 4–9%, 264,000–594,000 tonnes and up to 5.6 million tonnes CO2e — rest on 2020 EU apparent-consumption market data, per the EEA’s own methodology note, and were published in a March 2024 briefing. No newer EEA destruction-volume update exists as of 4 September 2026. Every use of these three figures carries that 2020 base year.
• EURATEX’s 99.8% micro/SME, 88.8% micro and ~0.2% large size-class breakdown traces to its 2020 and/or 2022 Facts & Key Figures editions. The current 2026 edition publishes only the headline figures (nearly 200,000 companies, 1.2 million jobs, EUR 166 billion turnover), so the size split may be up to six years old relative to those headlines.
• Fashion Revolution’s 88% non-disclosure figure comes from the Fashion Transparency Index 2023. It is the most recent edition confirmed against Fashion Revolution’s own site at time of writing.
• EU textile waste generation, separate collection and capture-rate figures are 2022 data; the reuse figure is 2021 and covers the EU excluding Bulgaria and Cyprus but including Norway.
• Used-textile export tonnage peaks at almost 1.7 million tonnes in 2019 and stands at approximately 1.4 million tonnes in 2023 — flat to slightly down, not a continued rise. The 2019 peak is not presented here as current.
• Annex VII product scope is stated here as the CN codes confirmed against primary and primary-adjacent text (4203, 61, 62, 6401–6405). Search snippets suggest headgear under CN 6504/6505 may also fall in scope; that was not confirmed against the regulation and is deliberately excluded rather than asserted.
• Two forecast rows are labelled as projections: ThredUp/GlobalData’s $393 billion global resale market by 2030, and Dataintelo’s USD 6.2 billion on-demand micro-factory market by 2033.

Last updated: September 2026. We update this page quarterly.