Insight, Circular Economy
The Commission reviewed 120 Horizon projects to find out why circular technologies fail to scale. Not one of the recurring bottlenecks it identified is a recycling technology. They are all institutions for verifying a claim that a buyer cannot check by looking.
The European Commission has drawn together findings from 120 Horizon 2020 and Horizon Europe projects to inform the forthcoming Circular Economy Act, spanning construction, textiles, electronics, plastics, mobility and bio-based industries. The recurring bottlenecks it identifies concern procurement, producer responsibility, markets for recycled materials, regional cooperation and the recovery of critical raw materials.
Read that list again. Not one item on it is a recycling technology.
After a decade and a half of demonstration funding, the obstacles Europe has identified are all about whether a recovered material can be sold, to whom, on what terms, and with what assurance. That is not a materials science problem. It is an information problem, and it has a well developed theory that the circularity literature reaches for surprisingly rarely.
A manufacturer buying virgin polymer buys a specification backed by a producer with a reputation, a batch record and a plant that made the same thing yesterday.
A manufacturer buying recyclate buys a promise about a material whose entire history is unknown. What was it before. What additives, stabilisers, pigments and flame retardants are in it. How many heat cycles has it been through. What contaminated the collection stream in the week it was gathered. What is the variance between this batch and the next one.
Almost none of that is verifiable by inspection, and the properties that matter most, long term ageing and fatigue behaviour, cannot be tested quickly at all. The buyer is not assessing a material. They are assessing a claim.
George Akerlof described what happens to a market in that condition. In conditions of quality uncertainty where the buyer cannot distinguish good from bad, the buyer rationally pays no more than the expected value across the distribution of quality on offer. Sellers whose material is genuinely better than average cannot obtain a price that reflects it, so they withdraw, downgrade, or sell into applications where the difference does not matter. What remains in the market is material worth roughly the average price, and the average falls.
Now put that next to the thing everyone in the sector complains about.
Downcycling is usually presented as a technical limitation: polymer chains shorten, fibres get shorter, alloys accumulate tramp elements. All true. But a substantial part of downcycling is the Akerlof mechanism operating in a materials market. Good recyclate ends up in park benches and drainage pipe not only because its properties have degraded, but because there is no way for its producer to prove they have not degraded enough to matter, and therefore no way to be paid for the difference.
That distinction matters because the two diagnoses imply different investments. If downcycling is purely technical, the answer is better sorting and better reprocessing, which is what fifteen years of demonstration funding has largely bought. If a meaningful part of it is informational, the answer is testing infrastructure, standards, liability and traceability, which is what the Commission's own bottleneck list describes.
Batches of recyclate vary in quality but the buyer cannot distinguish them before purchase, so pays a single price reflecting average expected quality.
The remedy is not a better sorting machine. It is a cheap, trusted way to distinguish batches before the material is bought, and somebody liable if the answer is wrong.
The 1970 paper does not stop at the failure. It identifies the institutions that arise to counteract quality uncertainty: guarantees, brand names, chains, and licensing. Translate those into secondary materials and the list is specific.
Standards that specify performance rather than composition. A recyclate specified as post-consumer content of a particular polymer tells a buyer nothing about whether it will survive their process. A specification written against the properties that matter for an application is what allows a premium to be paid for meeting it.
Batch testing infrastructure that is cheap and fast. The economics of verification decide whether a market exists. If testing a batch costs more than the price differential it establishes, nobody tests and everybody pays the average.
Liability that sits with somebody. This is the one Europe has least of. If a recycled input causes a product failure three years later, who carries it. Until that question has an answer that a procurement director can live with, recycled content in anything load bearing, safety critical or warranted will remain marginal regardless of price.
Long term supply agreements with a named counterparty. Reputation only functions where there is repeat dealing with an identifiable seller. Spot purchasing of anonymous recyclate is precisely the structure in which reputation cannot form.
Traceability from the original bill of materials. This is where the design decision sits. Verification is orders of magnitude cheaper when the original manufacturer recorded what went into the product than when a recycler has to infer it from a shredded mixture.
That last point suggests a reading of digital product passports that is more useful than the usual one. They are generally justified as transparency measures for consumers. Their more consequential function is as quality signalling infrastructure for a secondary market that cannot otherwise price quality. A passport does not make a material more recyclable. It makes a claim about the material verifiable, which is what allows it to be sold for what it is worth.
This is structurally identical to the barrier we described in battery precursor materials, where a cell manufacturer cannot buy on specification and must qualify a supplier over years of testing because the properties that matter are not fully captured by any written specification.
In batteries the industry solved the problem with qualification, which works and is enormously expensive and locks incumbents in. In secondary materials the equivalent institution barely exists, which is why the market clears at the average and why so much technically recoverable material circulates once and stops.
Seen that way, the Commission's bottleneck list is not a list of separate problems. Procurement rules, producer responsibility, market development and regional cooperation are four different attempts to build the same missing institution, which is a credible way of knowing what you are buying.
There is a comparison here that European circularity research is not making, and it is one of the few places where evidence from outside Europe is directly instructive rather than merely interesting.
Repair and reuse markets in East and West African cities move very large volumes of electronics, appliances, vehicle parts and materials with no certification infrastructure whatsoever. They face exactly the Akerlof problem in acute form, and they have solved it, partially, with entirely different institutions: personal reputation, repeat dealing within a physical market where a seller cannot disappear, warranty offered by the trader rather than the manufacturer, and inspection knowledge held by technicians who can assess a component by handling it.
Two things follow, and both need saying.
The first is that reputation-based verification demonstrably scales further than European policy discussion assumes, and it does so at a fraction of the cost of formal certification. That is worth understanding properly rather than dismissing as informality.
The second is that it carries real costs that should not be romanticised: hazardous processing, unsafe products reaching users with no recourse, and no mechanism at all for the properties that cannot be assessed by handling, which are precisely the ones that matter in a regulated European application.
The research question is which parts of that transfer. Where can reputation substitute for certification, at what scale, with what failure rate, and where can it not. Nobody appears to be studying it for that purpose, and it sits exactly where a comparison between European and emerging market evidence produces something neither could produce alone.
Circularity evaluation currently counts tonnes collected and tonnes recycled, which measures the top of the chain and tells you nothing about whether the material circulated.
Rejection rates, and the stated reason for each. The most informative dataset in the sector and one nobody publishes. A buyer who rejects a batch knows exactly why.
The price discount of recycled against virgin, by application. Tracked over time, this is the direct measure of whether the information problem is easing. If the discount is not narrowing, no amount of collection is building a market.
Repeat purchase. Whether a buyer who paid a premium once does so again. Single transactions prove nothing; repeat purchase is evidence that the promise held.
Where in the chain the material stops. Collection, sorting, reprocessing, qualification, purchase, second use. A chain fails at one specific link and the whole system gets blamed for it, usually the recycler. Locating the failure precisely is cheap and almost never done.
Europe has, as it now concedes, demonstrated the technology. What it has not built is the set of institutions that let somebody buy a promise. That is a governance and information problem, it is unglamorous, and it is the whole game.
The Lab works on this in local manufacturing and climate and ecosystems, and the comparison to reuse markets in the places we work is the kind of evidence a European consortium cannot generate from inside Europe. The pattern of demonstrating rather than adopting runs through this whole field.
If you are working on circular material markets and want the buyer side and the informal comparison examined properly, tell us what you need to know.
This is an independent insight piece by Transitions Lab. For the Lab's applied work, see Local Manufacturing & Supply Chains and Climate & Ecosystems. See also The Lock-In Runs Both Ways on the same qualification barrier in battery precursor, and Europe Has Enough Demonstrations on the pattern of demonstration versus adoption running through this whole field. To discuss a study, see Contact.