The EU bans destroying unsold clothes. Overproducing them stays cheap.

From 19 July 2026, large brands can no longer destroy, or even recycle, unsold apparel and footwear in the EU. The ban closes the cheapest way to clear surplus. The economics that create the surplus are left in place, and the measure that would price volume itself does not arrive until 2028.

On 19 July 2026 the largest fashion companies selling into the EU lose the cheapest way they have to clear clothes that do not sell. The Ecodesign for Sustainable Products Regulation bans the destruction of unsold apparel, clothing accessories, and footwear, and its definition of destruction is wide enough to close recycling as an exit too. The rule reaches the end of the line, where the surplus is thrown away. It does not reach the start of it, where making too much is still cheaper per unit than making too little.

That gap is the subject of this piece. The ban makes disposal expensive, and the measure that would make overproduction expensive, textile extended producer responsibility, does not arrive until 2028. For the two years in between, a brand that overproduces still pays less to clear the surplus than it loses by making too little, so the question is whether a ban on the cheapest exit, backed by a disclosure rule, changes what brands order, or only how they get rid of what does not sell.

What the ban prohibits

The ban sits in Article 25 of the Ecodesign for Sustainable Products Regulation, Regulation (EU) 2024/1781, and covers the categories listed in Annex VII: footwear, apparel, and clothing accessories such as hats, belts, ties, and scarves. The word destruction does more work than it appears. In EU law it covers disposal, other recovery including energy recovery, and recycling. So for unsold in-scope products, shredding a never-worn coat into fibre is destroying it, the same as landfilling or burning it. The compliant exits narrow to four: resale, reuse, repair and remanufacture, and donation. The table below sets out where unsold stock can still go, and what each route costs the brand that used to incinerate.

The unsold category is broad. It covers surplus stock, excess inventory, and deadstock, and it reaches goods a customer returns inside the 14-day cooling-off window, the returns that fill a warehouse after every online season.

The rule and who it binds

The duty attaches to economic operators, a chain that runs through manufacturers, importers, distributors, dealers, and fulfilment providers, and it follows the product wherever it was made, because what counts is that it was placed on the EU market. From 19 July 2026 the ban binds large companies. A company counts as large if it has 250 or more employees, or if it exceeds both 50 million euros of turnover and 43 million euros on its balance sheet, so a 300-person brand is bound whatever its revenue. Companies below those lines are bound from 19 July 2030, and micro and small companies are exempt from the ban. A separate transparency duty runs ahead of it: since the first financial year after the regulation took effect, large companies have had to publish, within twelve months of year-end, the number and weight of unsold products they discard. Most are reporting on their 2025 financial year during 2026. The panel below shows who is in and from when.

The calendar

Two acts adopted on 9 February 2026 turned the ban from principle into procedure. A delegated regulation, in effect from 12 May 2026, sets the cases where destruction stays legal. An implementing regulation fixes the disclosure format. The dates stack across four years, and the timeline below separates them: the general duty to prevent destruction has applied since the regulation took effect on 18 July 2024; the ban reaches large companies on 19 July 2026; the mandatory disclosure format applies for financial years starting on or after 2 March 2027; textile extended producer responsibility fees arrive by 17 April 2028 at the latest; and the ban and the disclosure reach medium-sized companies on 19 July 2030.

What can still be destroyed

The delegated regulation is where the ban meets reality, because it lists the cases in which destruction stays legal. There are five, and the table below sets out each one and what it demands. A product can be destroyed if it is dangerous under the General Product Safety Regulation; if it is non-compliant for another reason, such as being made with forced labour, or unfit for purpose through a defect that cannot be repaired; if it is damaged, contaminated, or a hygiene risk and repair would cost more than replacement; if it carries intellectual-property restrictions, including labels or logos that cannot be removed; or if donation has failed, meaning the goods were offered to at least three social-economy organisations and refused, or listed for donation for at least eight weeks with no taker. Each route carries a documentation burden: the reasoning is handed to the waste operator and kept for five years, and even then the waste hierarchy applies, so recycling is preferred over recovery and disposal. The eight-week donation test is the one to watch, because it turns giving stock away from a gesture into a documented precondition for getting rid of it.

The scale of the problem

The volumes behind the rule are larger than the occasional headline. The European Environment Agency estimates that 4 to 9 percent of all textile products put on the EU market are destroyed before they are ever used, between 264,000 and 594,000 tonnes a year, carrying a carbon cost of up to 5.6 million tonnes of CO2 equivalent, close to Sweden's net emissions in 2021. The stat strip below sets out the scale. The destruction is fed from two streams. Returns are the first: the Agency puts the average online clothing return rate near 20 percent and finds that roughly a third of returned clothing is destroyed, in part because processing a single return costs a retailer between 55 and 75 percent of the item's retail price, on figures from the British Fashion Council's Institute of Positive Fashion. Unsold stock is the second: about 21 percent of textile products are never sold, and roughly a fifth of those are destroyed.

Why there is so much to destroy

Destruction lasted for decades because it paid. Burberry disclosed in its 2017/18 annual report that it destroyed 28.6 million pounds of finished goods in a single year, including 10.4 million pounds of beauty stock, and defended it as protection against counterfeiting and grey-market dilution. For a brand with pricing power, moving unsold stock into a discount channel can cost more than writing it off. The deeper reason sits upstream of the warehouse, and the European Environment Agency states it plainly: overproduction makes economic sense. Most clothing is made in low-wage countries, and economies of scale and scope make a larger, more varied order cheaper per unit than a smaller, tighter one. A brand that overproduces carries the cost of unsold stock, while a brand that underproduces loses the sale and the margin on it, and in fast fashion the second mistake is the more expensive one. The line below is the Agency's own.

Article 25 does not touch this calculation. It governs the garment once it exists and has not sold, and says nothing about the order that made it.

What the disclosure rule asks

The part of the regime aimed at the production decision is the disclosure. A company that discards unsold stock must publish, each year, the number and weight by category, the reasons including any derogation used, the share sent to each end-of-life route, and the measures it is taking to prevent destruction in the first place. From the 2027 financial year the format hardens, set out in the table below: a fixed template, figures broken down by commodity code, estimates flagged, waste-operator statements kept for five years, and a risk-based check in which an unusually low number reported against peers is itself a trigger for scrutiny. Companies already reporting under the Corporate Sustainability Reporting Directive can fold the disclosure into that report. The theory is that a number a brand must publish and defend will pull production down. Whether it pulls hard enough to overcome the unit economics of scale is the open question.

Where the ban is soft

Three things the text leaves soft will decide how much the ban bites. The derogations are the first, and the European Environmental Bureau warned during consultation that a wide set of exemptions could let operators keep destroying stock under documented cover. The disclosure is the second: it is self-reported, and the firm format does not arrive until 2027, so the numbers meant to pressure brands into producing less will for now be uneven and hard to compare. Enforcement is the third, handed to national authorities whose capacity and appetite differ across the Union. France is the nearest preview. Its anti-waste law, AGEC, banned the destruction of unsold non-food goods from 2022 and reached textiles by the end of 2023, pushing brands toward donation and tighter forecasting. How much it actually cut destruction has not been published; what is documented is the scale it targets, around 630 million euros of unsold goods destroyed in France each year. Export is the other gap. No reliable figure isolates the share of unsold stock that simply leaves the continent, but the measured outflow of used textiles from the EU runs at about 1.4 million tonnes a year, split roughly evenly between Africa and Asia in 2023, and the Environment Agency notes that much of what reaches Africa ends up in open landfill or burned in the open air, beyond the reach of any EU ban.

The lever more likely to move production sits in the regulation arriving beside this one. From April 2028 textile extended producer responsibility will charge a fee on every product placed on the EU market, with reuse-fit goods exempt, which for the first time puts a recurring price on volume itself. The destruction ban makes getting rid of clothes visible and expensive. Extended producer responsibility is what could make making too many of them the expensive mistake.