How Textile Recycling Actually Works

Textile recycling is frequently described as though it were a single process. It is at least three, they accept different inputs, and the reason so little clothing is recycled into clothing has more to do with sorting than with any of them.

The three routes

Each route trades input tolerance against output quality, and the trade is close to a straight line: the more forgiving a process is about what it accepts, the less valuable what it produces.

Mechanical shredding is by far the most widely deployed and the least valuable. It breaks fabric back to fibre, shortening the fibre each time, which is why mechanically recycled cotton is typically blended with virgin fibre to be spinnable at all.

The chemical routes produce genuinely equivalent output and are correspondingly fussier. Depolymerisation of polyester can yield monomer indistinguishable from virgin feedstock, provided the input is polyester and not a polyester blend with several per cent of something else.

Sorting is the actual bottleneck

Every route above begins with knowing what the input is. That knowledge has to be produced at collection, on garments arriving mixed, at a cost per item that the recovered material can support.

The available methods each fail in a different way. Manual sorting by care label is fast and unreliable, because labels are removed, faded or wrong. Near-infrared spectroscopy is accurate and capital-intensive, and struggles with dark colours and multi-layer constructions.

This is the specific point at which the passport intervenes. A garment carrying its own composition, dye class and trim inventory arrives pre-sorted. The most expensive step in the chain has already happened, at the point of manufacture, where the information was free.

What makes a garment hard to recycle

Recyclability is decided at the design stage, largely by decisions taken for reasons that had nothing to do with end of life.

• Elastane — a few per cent transforms handle and comfort, and it blocks or complicates almost every route.
• Fibre blends — polycotton is the archetype, requiring both a polyester route and a cellulose route to run on the same item.
• Coatings and laminates — waterproof membranes bond layers that then cannot be separated.
• Trims — metal zips, rivets and prints must be removed before processing, usually by hand.
• Dark and complex dyes — some routes cannot strip colour, so output is constrained by input colour.

None of these are avoidable in every product. The point of recording them is not to eliminate them but to let a recycler route the garment correctly and to let a designer see the end-of-life consequence of a specification choice at the moment they make it.

What happens to collected textiles now

It is worth being accurate about the current baseline, because the gap between collection and fibre-to-fibre recycling is where most of the public misunderstanding sits.

Reuse is genuinely the best outcome for a wearable garment and is not a failure of recycling. The problem is the last box, and its size is a function of how cheaply items can be assessed rather than of how much processing capacity exists.

How EPR fees change the calculation

Extended producer responsibility for textiles is rolling out across EU member states, and where fees are modulated by recyclability the design consequences become financial rather than reputational.

A brand paying a lower fee for a mono-material garment than for a blend has a direct incentive to record composition accurately and to design toward the cheaper category. The passport is what makes that modulation administrable at all, since a fee that depends on composition needs composition to be reliably declared.

This is the first mechanism that connects a design decision made in a studio to a cost line that appears on a producer responsibility invoice years later. Until now the person specifying a stretch content and the person paying the disposal cost worked in different organisations and never exchanged a number.

Whether modulation is set at a level that actually changes behaviour is a separate question, and it varies considerably between member states. What is no longer in doubt is that the data needed to operate it will exist, because the passport requires it for other reasons entirely.