Who is the economic operator? The scope questions the destruction ban leaves open

Who is the economic operator? The scope questions the destruction ban leaves open

fashion.business Team
21 July 2026
5 min read

The EU destruction ban attaches to whoever discards, not to the brand. How scope works across wholesale, franchise, concession and 3PL stock.

Two pallets, both belonging to the same brand.

The first sits in a distribution centre in Northampton, apparel made for a European spring range that was cut back after a poor pre-season order book. Made for the EU, coded for the EU, forecast into EU markets. It never left the United Kingdom.

The second sits behind a concession in a Düsseldorf department store. The stock is on the brand's own ledger, replenished from the brand's own warehouse, sold through the store's till under a commission arrangement. Commercially the brand thinks of it as the department store's business.

Under Article 25 of the Ecodesign for Sustainable Products Regulation, the first pallet is outside the destruction ban and the second is inside it. Most brands have it precisely the other way round, and the reason is not carelessness. It is that the regulation attaches its obligations to a role rather than to a company, and almost everybody reads it as though it attaches to them.

The question is never whether a business is in scope. It is which economic operator decided to discard a particular unit, whether that operator is large, and whether the unit had been placed on the Union market before the decision was taken. Answer those three and the edge cases mostly answer themselves. Ask the first question instead and you will build a programme that captures stock which was never in scope while missing stock you no longer sell but still control.

A note on how this piece is organised. Some of what follows is what the instruments state. Some is what practitioners advise where the instruments permit more than one approach. Some is territory nobody has addressed, where the legislation is silent and a defensible position has to be chosen and documented rather than looked up. Each section says which of the three it is.


Who the economic operator actually is

What the instrument states. Article 24 attaches the disclosure obligation to economic operators that discard unsold consumer products, or have them discarded on their behalf. Not to the manufacturer. Not to the brand whose label is in the collar. To whoever made the decision to discard, or procured somebody else to execute it.

The term reaches wide. Economic operator under the ESPR covers manufacturers, importers, distributors, dealers, fulfilment service providers and providers of online marketplaces. What it does not do is re-allocate anything by brand identity. Being the name on the garment confers no obligation and removes none.

That sentence governs everything else here, and it does two things at once. It follows the goods rather than the brand, so a unit sold on carries the obligation to its new owner. And it reaches through intermediaries, so a brand cannot shed the obligation by having somebody else operate the skip.

The second half is where the money is. A third-party logistics provider destroying goods on a brand's written instruction has not become the economic operator for them. It is discarding them on the brand's behalf, which is the express language of the obligation. Article 33 imposes separate duties on fulfilment service providers and Article 35 does the equivalent for online marketplaces, but neither converts those parties into the primary Article 24 or 25 operator for stock the brand still owns. The 3PL's duties are its own. They are not a transfer of yours.

Section 4.4 of the destruction ban report sets out the practical consequence, which is that the outcome data you must publish sits inside a supplier's systems and your contract almost certainly does not require them to send it to you.


Which enterprise size test applies, and why headcount decides it

What the instrument states. This has been genuinely contested in practice, with some commentary and some national guidance treating Directive 2013/34/EU, the Accounting Directive, as the governing instrument for enterprise size under the ESPR. It is not.

The ESPR's definitions article imports the definitions of SMEs, small enterprises and microenterprises from Article 2(1), (2) and (3) of Annex I to Commission Recommendation 2003/361/EC. The cross-reference is explicit and it is to the Recommendation. Freshfields reads it the same way in its analysis of the draft acts, describing the medium-sized threshold by reference to the Recommendation.

The distinction is not academic, because the two instruments treat headcount differently. Under the Recommendation an enterprise is an SME if it employs fewer than 250 persons and has either turnover of no more than EUR 50 million or a balance sheet total of no more than EUR 43 million. Headcount is a hard gate: cross 250 staff and you are large whatever the financial position, stay below it and you remain an SME even at substantial turnover provided one financial ceiling is met. Under the Accounting Directive, headcount is one of three criteria and a high-turnover business with a small payroll is large on the financial tests alone.

The consequence is an asymmetry that sections 4.1 and 4.1a of the report develop in full. A brand with a store estate carries hundreds of sales staff and crosses the headcount gate early, often years before its turnover would suggest. A pure-play of similar or greater revenue, running on a warehouse and a marketing team, may sit below 250 heads for a long time. Two commercially comparable businesses can land four years apart on the same obligation.

Two refinements matter to anyone close to the line, and both follow from the Recommendation governing rather than the Accounting Directive. Headcount is measured in annual work units rather than in bodies, so part-time staff count in proportion to their hours and seasonal workers are averaged across the financial year. A retailer with 300 people on the payroll, heavily weighted to part-time shop floor and Christmas temporary contracts, may sit well below 250 annual work units. Nobody should assume the payroll report answers the question. And the Recommendation's annex distinguishes autonomous enterprises from partner enterprises, held between 25% and 50%, and linked enterprises, held above 50%, requiring proportional aggregation for the first and full aggregation for the second. A majority-owned subsidiary is counted with its parent. Decision 1 of the report puts the assessment at group level for that reason.

Neither refinement is stated in the ESPR. Both come from the Recommendation and its user guide, which is precisely why the choice of instrument decides more than it appears to.


Placed on the Union market, and when

What the instrument states, with one inference marked. Article 25(1) prohibits the destruction of unsold consumer products listed in Annex VII that have been placed on the Union market. Article 2(40) defines placing on the market as the first making available of a product on the Union market. Goods never placed are outside the ban entirely, and three cases follow.

Stock in a non-EU distribution centre. The Northampton pallet was intended for EU sale and never supplied there. Not placed, and Article 25 does not reach it. The caution is evidential rather than legal: a brand relying on this must be able to show the disposition happened before any EU placement, which means the record establishes location and sequence rather than intent. Retrospective assertions about where stock was when it was destroyed are exactly what an authority will test against shipping records.

One asymmetry is worth noting, because it is easy to miss. The placing precondition appears expressly in Article 25(1). It does not appear in the same explicit form in the disclosure obligation. The working position, that Article 24 likewise does not reach destruction occurring wholly outside the EU, rests on Commission and member state guidance describing the duty as one about discards on the EU market rather than on a limb of the text. That is a sound reading and it is what firms are advising. It is not the same thing as the express words that carry the Article 25 position.

Stock in a bonded or customs warehouse. Goods in an EU customs warehouse before release for free circulation have arguably not been placed. The Commission's own ESPR FAQ of September 2024 states that where products are presented to customs and declared for release for free circulation, they can generally be considered to be placed on the EU market, which is the reasoning the position rests on. Two qualifications belong with it. That FAQ predates both secondary instruments, so its horizontal content on placing remains good while anything it says about the destruction regime does not. And it does not address bonded warehouses directly, so this is an inference from the general rule rather than a settled answer. The counter-argument is available: an authority could contend the commercial arrangements meant the goods were effectively supplied for EU distribution, whatever their customs status. Anyone structuring around this reading should hold the reasoning in writing and expect to defend it.

Goods listed online by a non-EU seller. For distance selling, a product is made available when the supply contract is concluded rather than when it is listed. Unsold units sitting in a non-EU warehouse against an EU-facing website have not been placed. Units already sold have been, and the same Commission FAQ gives this as an express example: for online sales by operators located outside the EU, placing occurs before the product arrives at EU customs. Those units, if they come back under the right of withdrawal, re-enter the regime as returns within Article 2(37).

One point sits underneath all three and is most likely to be misread inside a finance function. The regulation attaches to physical disposition, not to accounting treatment. Writing stock down to zero is not a regulated act. Destroying it is. Align the two anyway, which is section 4.4 of the report arriving from the opposite direction: not because the ledger entry triggers an obligation, but because it is the only place most businesses record that a decision was taken at all.


Wholesale, franchise, concession and marketplace

What the instrument states, applied. Where title has passed, the operator that discards bears Articles 24 and 25, and the enterprise size test is applied to that operator rather than to the brand that supplied it. A wholesale customer, a franchisee that buys stock outright, and a marketplace running a wholesale model are each acting as importer and distributor in their own right. If they destroy unsold goods, the obligation is theirs. If they are small or micro, they are exempt from the ban, though not from the Article 23 prevention duty, which has applied to every economic operator of any size since 18 July 2024.

Two qualifications attach to that. The first is that the brand's position reverts the moment it takes the goods back. Sale or return, a season-end recall from a franchisee, a buy-back to protect a market: in each case the stock is the brand's again and so is the obligation, even though the units spent a year on somebody else's balance sheet. The second is contractual rather than statutory. Nothing in the ESPR requires a supply agreement to allocate compliance responsibility, and nothing prevents it. Firms are advising that B2B terms say plainly which party carries the Article 24 and 25 duties on unsold stock, because the default allocation follows title and title is not always where either party assumes it is. Decision 5 of the report puts the equivalent clauses into logistics contracts; the same exercise is owed to the wholesale book.

The harder case is the one in the Düsseldorf department store. Under a concession or agency arrangement the brand retains title to the stock. Destruction of that stock happens on the brand's instruction or with its consent, which places it squarely within the language of having goods discarded on its behalf. The obligation stays with the brand. Article 35 imposes its own duties on online marketplace providers, and Article 33 does the same for fulfilment service providers, meaning parties that receive, warehouse, package and dispatch goods on behalf of another operator. Neither provision displaces the brand as the primary operator for concession stock it still owns. A 3PL's duty is to cooperate and to ensure that products under its control meet ESPR requirements. It is not to inherit your disclosure.

The instruction that falls out of this is unglamorous and worth writing down. Scope by title and by decision rights, not by channel. The compliance boundary runs along the line where ownership passes, and in a business with wholesale, franchise, concession and direct channels that line corresponds to no reporting structure the business already has.


Consolidated waste loads, where nobody has an answer

What nobody has addressed. A 3PL serving eight brands consolidates textile waste into a single load. Each brand owes a disclosure stating units and weight. No rule requires a waste operator to break a load down by brand, no published commentary addresses attribution, and the legislation is silent.

Note the framing of the handover obligation, because it is suggestive. Article 4 of the delegated regulation requires the statement identifying the applicable derogation to be given by the delivering economic operator to the receiving waste treatment operator. It is framed per operator rather than per consignment, which implies that a mixed load carries several statements rather than one, and by extension that the units in it remain attributable to their respective operators throughout.

What practitioners advise. Attribute at the point of instruction rather than at collection, because after mixing the information does not exist. Weigh and record per brand before consolidation where practicable, at pallet or SKU level. Require a manifest allocating units and weights per brand and per derogation ground, and have each brand's Article 4 statement travel with its own line rather than with the load. Where precise weighing is impracticable, a documented allocation method is what the market is relying on, typically an average weight per garment type applied to a per-brand unit count, read against the 10% tolerance that Annex III of Commission Implementing Regulation (EU) 2026/2 treats as compliant. That tolerance makes a reasoned allocation defensible. It is not a licence to estimate loosely, and the method itself should be written down before it is used rather than reconstructed when it is questioned.

None of that is a legal requirement. It is a defensible position, and defensible positions in silent territory have to be chosen and written down before an authority asks, not after. Decision 5 of the report puts the clauses that implement it into the next 3PL contract renewal.


Which entity in a group publishes

What the instrument states. Each subsidiary that discards is in principle an economic operator with its own Article 24 obligation. Article 24 requires the information to appear clearly and visibly on at least an easily accessible page of the operator's website, which does not require a separate page per legal entity provided a shared corporate site attributes the data clearly. Operators may place the information inside sustainability reporting under Article 19a or 29a of the Accounting Directive, and the Implementing Regulation permits either a dedicated report in the Annex I format or a link to a report containing the information with a clear indication of where it sits. Article 29a covers consolidated sustainability reporting by parent undertakings, so group-level publication is plainly permitted.

What nobody has addressed. Permitted is not the same as discharging. Whether a parent's consolidated report satisfies a subsidiary's own Article 24 obligation, such that an authority could not enforce against that subsidiary directly, is not stated. The legislation does not say and no published commentary resolves it.

For a group headquartered outside the EU and therefore outside CSRD, the practical route firms are taking is a central group-level page that states explicitly which subsidiaries and which operations the figures cover. That is a reasonable reading of the website requirement. It is not a confirmed one.

In practice this produces a split most groups land on by default rather than by decision: reporting is structured centrally, because that is where the sustainability report is produced, while the supporting records sit at entity level, because that is where the write-offs happen. The work lands on whoever can reconcile the two, and there is frequently nobody whose job that is. Naming that person is Decision 8 of the report, and it is cheaper now than during an assurance process.


When a return becomes an unsold consumer product

What the instrument states. Article 2(37) expressly includes within the definition of an unsold consumer product those products returned by a consumer under the right of withdrawal in Article 9 of Directive 2011/83/EU, which gives 14 days from delivery on distance and off-premises contracts, or during any longer withdrawal period the trader offers. A brand running a 60-day returns policy has extended the window in which its returns carry this status, by its own choice.

Two things follow, and they operate at different moments. The status attaches on return. It does not wait for the inspection, it does not depend on whether the unit passes, and it does not turn on whether the unit is restocked. But the disclosure and derogation obligations bite only when the unit is actually discarded. A returned garment that is cleaned, re-tagged and resold never appears in a disclosure at all, despite having been an unsold consumer product the whole time.

Map that onto how a returns operation actually runs and the seams show. A returns process is built around SLA-driven stage gates: booked in within so many hours, inspected within so many days, dispositioned by a cut-off that exists to keep the bay clear. Those gates are timed to cost and space. The regulation's two moments are timed to legal status and physical destruction. They come apart at the disposition gate, where a unit is routed to a disposal stream on a clock rather than on an evidenced ground. Article A of this series follows one consignment through that gate, including the arithmetic that decides most of it.


The gap the regime leaves, and the signature that would expose it

Obligations that attach to operators rather than to units have a structural consequence. A unit carries its owner's status. Sell it and it acquires a new one.

The regime is not naive about this. Article 25(2) prohibits economic operators not themselves subject to the ban from destroying unsold consumer products supplied to them with the purpose of circumventing it. That is a live prohibition rather than a reserve power, worth stating plainly because the point has been reported inconsistently. Separately, Article 25(5) empowers the Commission to provide by delegated act that the ban or the Article 24 disclosure applies to micro and small enterprises where there is sufficient evidence they could be used to circumvent them. Two mechanisms, different jobs.

Both share a weakness: they are backward-looking. Article 25(2) turns on purpose, and purpose is an intent test, the hardest thing in commercial regulation to evidence, because ordinary clearance trading and circumvention produce identical paperwork. The Article 25(5) power requires evidence of circumvention to exist before it can be used, so the pattern has to run for a while before the instrument that would stop it becomes available. A boundary drawn at the operator attracts pressure at exactly that boundary, and the regime's answer arrives late by design.

COSH!, the Belgian sustainable fashion platform, made the point publicly in March 2026. Its founder Niki de Schryver argued that the micro and small exemption could create an incentive for multinationals to route unsold stock through micro entities, while acknowledging that the ESPR recognises the risk and permits the Commission to extend obligations if evidence emerges. The same piece raises a second concern bearing directly on the derogation regime: if cost-effectiveness comes to be read as cheaper than repairing the item right now, and if extended producer responsibility schemes continue to subsidise recycling, repair loses on price for structural rather than technical reasons. Section 4.2 of the report sets out why the delegated regulation's definition does not support that reading. Whether enforcement holds the line is a different question.

What would expose a pattern of this kind is data, not inspection. The disclosure regime produces a public, comparable, annual number for every large operator, and Annex III already names implausibly low figures as a risk indicator. A brand whose reported discard volume falls sharply while the volume it places on the market does not is describing something, and it is visible without anybody visiting a warehouse. Concentration is the other signal: clearance volume moving consistently to a single counterparty shows up in commercial records and, over time, in the shape of the published figures.

The regime has also built the instrument that makes pattern detection feasible across the market rather than one brand at a time. Article 25 requires the Commission to publish consolidated information on the destruction of unsold consumer products on its website by 19 July 2027 and every 36 months thereafter. That is the first moment at which the aggregate picture becomes public, and it is the moment at which an anomaly stops being a brand's private business and becomes a visible outlier in a published dataset.

This piece describes a weakness in the regime because a compliance lead should understand where the boundary of their obligation sits and why it will be tested. It does not describe how to sit on the wrong side of it, and anybody reading it for that purpose should attend to the second half of this section rather than the first.


Back to the two pallets

The Northampton pallet is outside the ban because the goods were never placed on the Union market. The brand's obligation there is evidential: hold a record showing where the stock was and when the disposition happened, because the claim will be tested against shipping data rather than accepted on assertion.

The Düsseldorf concession stock is inside the ban because the brand still owns it, and its destruction happens on the brand's behalf whatever the commission agreement says about who runs the till. That stock needs a derogation ground, an evidence pack and an Article 4 statement, and the department store will produce none of them.

Neither answer comes from asking whether the brand is in scope. Both come from asking who decided, who owned it, and whether it had been placed. A programme built on the first question will be busy and wrong. One built on the other three will be smaller than the business expects, and pointed at the right stock.


References


Share this report

LinkedIn