As the EU preparatory study on textiles moves from its third to its fourth milestone, the discussion around the future delegated act under the Ecodesign for Sustainable Products Regulation is entering a new phase: how far ambition can be translated into enforceable requirements, how data and verification systems can support that transition, and which methodological choices are most likely to shape the first phase of implementation. The feedback now being collected will help the JRC define its next steps, and registered stakeholders can still submit written comments on the 3rd milestone through the dedicated questionnaire until 30 March. Through this interview with Marina Prados, Director of Policy Hub, we explore the emerging role of recycled content, the relationship between ESPR and Product Environmental Footprint Category Rules (PEFCR), the limits of current approaches to durability and recyclability, and the broader challenge of applying European standards across global textile value chains.
The preparatory study’s work is shifting from mapping and defining technical design options to stress-testing those options through impact assessment and policy scenarios. Many stakeholders are trying to anticipate what the “end state” will be. How do you assess the balance currently being struck between ambition and deliverability?
If I look at the process from the first milestone to where we are now, I would say that the Joint Research Centre has moved toward a more pragmatic perspective. But I do not see that as a retreat. At a certain point, any serious legislative process has to move from ambition to implementation, and that is especially true in textiles. ESPR will not be a marginal piece of legislation for this sector. It will be the first framework that not only sets conditions for placing products on the market, but also requires the generation, reporting and verification of a very large volume of product-related information.
For that reason, I think the first textile delegated act is increasingly likely to function as a transitional instrument. I do not think we are yet at the stage where the full architecture of performance requirements can be imposed across the board. What seems more plausible is a first step centred on data gathering, on understanding what information is actually available, and on building the evidence base that will later support stronger requirements. In that sense, the act is likely to be gradual and phased in. It will still be politically and structurally important, because it will establish the foundations, but it will probably do so more incrementally than many expected at the beginning. That is also why information requirements are becoming so central.
So if I had to characterise the current balance, I would say that it leans more toward deliverability than toward maximal ambition. But I would not describe that as a weakening of ambition itself. The ambition remains embedded in the ESPR. What has changed is the way that ambition is being operationalised. The question is no longer only how far the EU wants to go, but how it can build a framework that is credible, implementable and defensible in practice. That, in my view, is why the JRC has taken a more practical direction and why a phased approach now looks more realistic.
In your view, what would success look like for the textile delegated act?
In the immediate term, success would not simply mean – for example – only inserting a recycled-content requirement into the delegated act. It would mean going far enough into the implementation conditions that make such a requirement real. If recycled content is to become part of the framework, then we also need to define what a credible chain-of-custody system looks like, what kind of traceability is required, how verification will work, and how compliance will actually be assessed by authorities. A delegated act that remains at the level of principle, without addressing those practical conditions, would not be a success.
In the short term, success also depends on whether ESPR is properly connected to the other frameworks that have to support it. Recycled content can only become a meaningful market driver if it is linked to effective and harmonised Extended Producer Responsibility (EPR) schemes and to a functioning secondary raw materials market. Feedstock availability, output quality, waste collection and material recovery are not peripheral issues. They are part of whether the policy can work at all.
In the longer term, success would mean that the first layer of information requirements actually produces credible evidence and that this evidence is then used to strengthen performance requirements over time. If product-level data show, for example, that a certain level of recycled polyester is technically feasible and materially available, then that should inform a future revision of the delegated act and justify a higher requirement. In that sense, long-term success depends on whether data collection becomes part of an iterative regulatory system, rather than a purely administrative exercise. The real test will be whether information can flow back into policymaking in a way that progressively makes products placed on the EU market even more sustainable.
A recurring consultation topic has been how far ESPR should build on existing Product Environmental Footprint Category Rules (PEFCR) work. What do you see as the appropriate relationship between PEFCR-based methods, categorizations and rules, and ESPR ecodesign methodology, categories, and requirements? Why?
My view is that it would be a mistake to ignore the PEFCR work. The apparel and footwear PEFCR process has involved years of highly technical discussion, and that work has generated a great deal of knowledge that remains relevant. It was developed through a multi-stakeholder process, and the lessons that emerged from it should be taken into consideration. To set it aside altogether would not be a wise use of the expertise that has already been built.
At the same time, I do not think the relationship between PEFCR and ESPR should be understood as one of direct transfer. PEFCR was developed in a specific context and for a specific purpose. It was conceived as a voluntary methodology for measuring environmental footprint. ESPR, by contrast, is mandatory and will apply across all products placed on the market that fall within the relevant delegated act (or horizontal ones). That difference changes the regulatory meaning of the exercise. A methodology that can function in a voluntary framework does not automatically function in a mandatory one. The issue, therefore, is what would need to change in order to take something built as a voluntary instrument and insert it into a mandatory regulatory framework without creating distortions or losing scientific credibility.
Do you think the JRC followed the same thought process? And why do you think the 3rd milestone’s categories are different from, and fewer than, those used in PEFCR?
I think this is precisely where the discussion becomes not only technical, but also political. It is important not to present the textile file as if it were purely a scientific design exercise. It is also a legislative and political process, and that shapes categorisation choices. In the PEFCR context, one of the recurring difficulties was the relationship between synthetic and natural fibres and the problem of how to address mismatches between them. My impression is that some of the simplifications proposed by the JRC are, at least in part, a way of navigating those politically sensitive tensions by making certain mismatches less visible. But in doing so, there is a risk of moving away from what would make the most sense from a scientific and environmental perspective.
This is why we at Policy Hub have raised concerns about the use, in the preparatory study’s milestones, of very broad representative categories like denim, knit and woven as the basis for requirements such as recycled content or environmental footprint. Those categories are simply not homogeneous enough. Denim products can differ greatly from one another, and the same counts for knit products and woven products. If the category is too broad, then the representative product stops being truly representative, and the environmental analysis can become misleading.
That is also why we have argued that the representative-product logic developed in PEFCR remains useful. It is closer to what actually happens in the market and in product diversity. If, by contrast, one uses a simplified categorisation that is too coarse, one can end up assimilating garments that are not environmentally comparable. A shirt and a pair of trousers may then end up being treated through assumptions that are much too similar, which does not make sense if the objective is to build technically credible ecodesign requirements.
So yes, I think the JRC is trying to respond to some of the same issues, but I do not think it is doing so in exactly the same conceptual way. And that is why the current phase is so important. We are now in the moment when stakeholders have to do more than point out mismatches. We also have to propose alternatives. That is what makes this stage so intense. It is not enough to say that the proposed categories are not right; one also has to offer a regulatory alternative that remains implementable.
Across the current design options – a robustness score, a recyclability score, recycled-content requirements, and an environmental or carbon-footprint “excellence” approach focused on the manufacturing stage – where do you think the most boundary effects could arise and why?
Among the options currently being discussed, I think recycled content is the one most likely to have the strongest consequences, and therefore also the one most likely to generate the most significant boundary effects. But I say that with caution, because it is also the option that raises the greatest number of unresolved implementation questions. Its impact will depend heavily on how the requirement is framed, how compliance data are verified, what the competent authorities will actually control, and where the thresholds are ultimately set.
Some of the thresholds discussed so far do not yet seem fully coherent or fully justified. In the case of denim, for example, a 20 per cent recycled-cotton threshold may be easier to reach in certain products, especially indigo-dyed ones, than in others. If the same logic is not carefully adapted to product realities, the rule risks becoming either ineffective or distortive.
What makes recycled content even more important is that it is the only option that already has the shape of a genuine performance requirement. The other options are, for the moment, still predominantly informational. That distinction matters, because performance requirements can directly change sourcing, design and market behaviour, whereas information requirements only have strong effects if they are embedded in a larger governance mechanism.
This is where I become more cautious about robustness and recyclability scores. I understand the logic behind them, and I understand what the JRC is trying to achieve by simplifying the information into a score. But I am not convinced that a simplified score, if presented directly to consumers, will automatically produce meaningful change. One has to ask whether a consumer will truly understand what, for example, a recyclability score of eight actually means in practice. That is far from obvious. In my view, such information may become more useful if it is not treated first and foremost as a consumer label, but as a regulatory data instrument. It could then feed future performance requirements, or perhaps more importantly, support eco-modulation mechanisms under the Waste Framework Directive, where better-performing products could be rewarded through lower fees. In that case the score would contribute to market incentives indirectly, through governance, rather than directly through consumer interpretation.
The environmental-footprint option is more complex still. I do think that an environmental-footprint information requirement could have a meaningful role, but if this design option is to have real value, it should not be restricted to the manufacturing stage alone. The current JRC focus on manufacturing is too limited. The discussion we are having in Policy Hub is rather about how to include a wider logic spanning raw materials, manufacturing and distribution. Only then can the information begin to influence not just process efficiency, but also material selection and broader product strategy. The question of whether such an approach should remain voluntary is one of the most delicate issues we are discussing. If it remains voluntary, then one has to ask how it will interact with national initiatives, such as the French Eco Score or other schemes that may emerge. The worst possible outcome would be a fragmented market in which multiple methodologies coexist, forcing companies to navigate parallel and potentially inconsistent systems. That would increase burden without increasing clarity or environmental impact. So if the EU wants to move toward an environmental-footprint information requirement, it also needs to confront what would be required to make it mandatory in a coherent way. That means data support, methodological revision, possibly additional impact indicators, and a serious effort to build an accessible and credible database for all relevant actors, from farmers and manufacturers to brands.
What is your view on textile-to-textile recycling?
I completely understand why textile-to-textile recycling has become such a strong point of reference in the discussion. It captures a very important ambition, which is to move away from more diffuse and less circular forms of recycled content and toward a genuinely closed-loop textile system. In the long term, I do think that textile-to-textile recycling should be the objective.
The difficulty is that the regulatory benchmark has to correspond to actual system conditions. At the moment, I do not think it is realistic to impose a textile-to-textile target across all products placed on the market, because ESPR will be mandatory and the obligation will ultimately rest on brands. If the necessary feedstock is not available in sufficient volume, if separate collection is not strong enough, if end-of-waste criteria remain unclear, and if the broader waste-management architecture is not harmonised, then the legal obligation may become disconnected from material reality.
That is why our position with Policy Hub has been to support a transitional pathway. In the near term, open-loop recycled content should remain possible. But this should not mean that open loop becomes the destination. What the Commission needs to do, in my view, is define a clear timeline for the transition away from open loop and toward textile-to-textile recycling. One of the things that is currently missing from the discussion is precisely that temporal clarity. Without a transition pathway, the debate risks becoming abstract. We need to know when the system will be assessed, how readiness will be judged, and what conditions will trigger the move to a more demanding model.
Given that the European Parliament has called for a holistic approach to durability, including both physical and emotional durability, how should emotional or use-driven durability be reflected in the ecodesign framework?
I think it is very important to acknowledge the importance of emotional durability, but I also think it is too early to regulate it in a binding way. On this point, I broadly understand and share the JRC’s caution. We do not currently have a sufficiently mature methodology to assess how a textile product ages in real-life conditions in a way that is robust enough for regulation. Durability is not only an intrinsic property of the object. It is also affected by extrinsic factors, by user behaviour, by care patterns, by attachment, by fashion dynamics and by many variables that cannot yet be translated into a stable compliance framework.
That is why I find the term “robustness” more workable at this stage, even though it is not perfect. “Robustness” narrows the focus to something that can be operationalised more plausibly. It links the discussion to measurable physical performance and to identifiable failure modes, without pretending that this captures the entire lifespan experience of a garment. So while durability remains the broader and richer concept, robustness is at present the term that better reflects what can realistically be implemented.
This does not mean that emotional durability should be dismissed. On the contrary, it should remain part of the ongoing scientific and policy conversation. The PEFCR process already recognised its importance, but also acknowledged that the field is not mature enough for regulation. There are now some research projects trying to move that conversation forward, but at the moment I would still say that emotional durability belongs to the horizon of future methodological development rather than to the first generation of binding ESPR requirements.
Textile supply chains are global, and many upstream actors sit outside European standard-setting structures. How should EU policy handle the tension between Europe-led standards and the practical need for global adoption and upstream data generation?
This is one of the hardest governance questions, because the textile value chain is deeply global and because the burden of information generation will not stop at the EU border. What ESPR does, legally speaking, is regulate products placed on the European market. But to make those products compliant, brands will have to retrieve and verify information from actors located much further upstream, often outside European standard-setting structures.
For that reason, I think the first principle has to be technical alignment with internationally recognised standards as far as possible. If Europe diverges too much from globally validated standards such as ISO standards for testing, the system becomes more bureaucratic without necessarily becoming more effective. Even a seemingly small technical divergence can create duplication. If Europe requires a different test protocol, or a different sample format, or even a different fabric extension for a given parameter, then companies may end up having to perform multiple tests for different markets. That increases cost, increases complexity and may reduce uptake without delivering a proportional sustainability benefit.
The second principle is interoperability. In practice, the legal addressee of the regulation may be the brand, but the data request travels upstream to suppliers and manufacturers. If those actors receive fragmented, overlapping or non-harmonised requests, the regulatory system risks degenerating into a transfer of administrative burden rather than a tool for sustainability improvement. That is why legislative coherence across EU files is so important, and why standardisation work will be essential. The role of European standardisation bodies, especially CEN and CENELEC, should in my view be to build as much coherence as possible with international standards rather than inventing unnecessary divergence.
So I would say that the only viable route is a combination of European regulatory direction and global technical compatibility. The EU can and should define the policy objective, but it has to do so in a way that can actually be absorbed by global supply chains. Otherwise the result will be procedural fragmentation rather than systemic change.
Read in Italian.
© Riproduzione riservata
