Can EPR deliver source reduction and prevention?¶
Prevention sits at the top of every waste hierarchy, and packaging EPR names it among its objectives — yet no operating scheme demonstrably delivers it, and the deepest reason is neither bad faith nor weak fees. It is that prevented waste leaves no trace to count: a package never made produces no tonne to weigh, no gate record, no receipt, and a policy architecture built on measured tonnage is structurally blind to the outcome it ranks highest. This article sets out the measurement problem precisely; traces its consequences through EPR's incentive and reporting machinery; examines the two legislative attempts to define prevention into observability — the EU's state-level reduction targets and California's producer-level dual-metric mandate, the first of its kind; and closes with the decoupling argument: the case that prevention should never have been assigned to a financing instrument at all.
1. The measurement problem, stated precisely¶
Recycling produces a tonne that can be weighed at a sorting plant. Prevention produces nothing — and the academic literature treats this as a definitional rather than practical difficulty. The organisation scholar Hervé Corvellec put it sharpest: prevention "produces no waste to count," which means it cannot be measured by the instruments waste systems are built from; analyses of European waste-prevention programmes have characterised them as aspirational documents precisely because their central objective resists quantification.
Four specific difficulties follow, each fatal to casual measurement:
No counterfactual. A prevention claim asserts what would have been placed on the market absent the intervention — a baseline that is unobservable and contestable by anyone with an interest in contesting it.
Perverse denominators. Recycling rates are reported as a share of what was placed on the market. Successful prevention shrinks the denominator — so a system that prevents waste can show a worse recycling rate and a higher cost per tonne, penalising itself in its own performance reporting. Jurisdictions that set recycling and prevention targets simultaneously, without adjusting the denominator, have built systems in which success on one metric mechanically damages the other. Several have.
Confounding. Packaging tonnage moves with GDP, e-commerce penetration, household size and retail format. Distinguishing a prevention effect from a recession requires econometric work nobody funds.
Gameable metrics. Weight-only measures reward lightweighting — including into thinner multilayer formats that are harder to recycle. Unit-only measures reward concentration and bulking. Either alone can be satisfied without environmental gain.
2. What this does to EPR¶
The consequence is a structural bias, not a moral one. A fee-funded system can report only what it collects, so its incentives, its reporting and its political case all cluster in the middle of the hierarchy — collection and recycling — while the top tier goes unmeasured, unfunded and unrewarded. Fees levied per tonne placed on the market do create a small, genuine incentive to place less; but at 0.4–2% of product price (the fee arithmetic), the per-tonne fee is a weak reduction signal, and the modulation apparatus layered on top rewards recyclability, not reduction, in almost every scheme (the modulation record). France's early 8% prevention bonus and California's statutory source-reduction credits are the exceptions, and neither has produced documented reduction.
The arithmetic of the per-tonne incentive is worth making concrete, because it explains the econometric record before the econometrics arrive. A producer that reduces its packaging tonnage by 10% saves 10% of its fee — and since the fee is 0.4–2% of product price, the saving is 0.04–0.2% of revenue. Set that against the costs reduction typically carries — reformulation, line changes, transport and damage risk from lighter formats — and the fee's reduction incentive is real, positive and roughly two orders of magnitude below the decision threshold of any product manager. Prevention bonuses have been tried and illustrate the same arithmetic from the other side: France's original 2012 schedule included an 8% bonus for producers running prevention actions, and no documented reduction followed — 8% of a small fee being a very small number. The modulation literature's advocacy analyses have flagged the same structural gap: prevention criteria are the least developed basis in every modulation framework, because the attribute being rewarded is an absence.
The econometric record matches the design logic. Eugénie Joltreau's 25-country panel — the baseline study of this literature — found the fee incentive produced "very little (though statistically significant) packaging reduction": the reduction margin exists and is economically marginal, exactly what a sub-1% price signal predicts. The result is a policy that names prevention as its highest priority and then measures, funds and rewards recycling — through instrument design, not hypocrisy.
3. First fix: state-level targets¶
The EU's packaging regulation (Regulation 2025/40) sets the first binding prevention quantities in packaging law: per-capita packaging-waste reduction of 5% by 2030, 10% by 2035 and 15% by 2040, against a 2018 baseline. Their significance is real — a legislated ceiling on a quantity that has only ever grown — and so is their limitation: these are obligations on member states, not on producers. The fee system still prices nothing at the prevention margin; a government must reach its target through whatever instruments it chooses — reuse mandates, format bans, charges — with EPR fees at most a supporting mechanism. State-level targets, in other words, concede the article's premise: prevention was not deliverable through the EPR machinery, so it was assigned to governments instead.
The same regulation's reuse targets and format prohibitions (the 2030 bans on certain single-use formats) belong to the same concession: where the EU wants less packaging, it legislates less packaging, rather than pricing its way there.
4. Second fix: the producer-level mandate¶
California's SB 54 contains the genuine innovation: the first producer-level prevention mandate in any packaging EPR law. Covered plastic packaging must be reduced 25% by 2032, measured in both units and weight, with a required portion of the reduction coming from elimination or shifts to reuse and refill rather than from lightweighting.
The dual metric is the analytically interesting part, because it closes both gaming routes simultaneously: a producer cannot satisfy a unit target by making packages thinner, nor a weight target by concentrating product. The elimination/reuse floor closes the third route — meeting the target through material substitution that shifts rather than reduces. As far as this library can establish, it is the closest any jurisdiction has come to making prevention administrable at the level of the obligated party.
Whether it works is genuinely open, and the difficulties are the ones Section 1 predicts. Hundreds of producers must establish 2023 baselines — the unobservable counterfactual, now with a filing deadline; the measurement disputes are foreseeable; and the mandate must survive implementation and litigation in a program whose regulations were already rewritten once under cost pressure. But the design stakes are large: if California produces verified reduction by 2032, it will be the first demonstration that prevention can be regulated directly rather than aspired to — and the two-goal problem of Theme 1 will have acquired its first counter-example by mandate rather than by price. If it fails, the decoupling argument below is effectively complete.
5. The decoupling argument¶
Standing against both fixes is the position that prevention should never have been assigned to EPR at all. Its most concrete form is the proposal by the NGO coalition Zero Waste Europe to split EPR budgets into a waste-management budget and a protected waste-reduction budget — an admission, from the advocacy side, that an unprotected budget governed by fee-payers will always flow to managing waste rather than preventing it (the governance logic). The stronger version generalises: let EPR be the financing instrument it demonstrably is, and assign prevention to instruments that can carry it — reuse mandates, format bans, charges on problematic formats, and the state-level targets of Section 3.
The record assembled across this theme supports the decouplers more than the patchers on current evidence: three decades of fee incentives produced marginal reduction; modulation rewards recyclability rather than reduction; the EU's own architecture routes prevention through member-state obligations and bans rather than fees; and the one direct producer mandate was written into an EPR law but operates beside its fee machinery, not through it. The honest statement is that every serious prevention instrument now in force works around the fee system rather than through it — which is either an indictment of the fee system or a sensible division of labour, depending on what one thought fees were for.
The instrument landscape, arranged by where the obligation sits:
| Instrument | Obligated party | Measurement basis | Status |
|---|---|---|---|
| Per-tonne fee | Producer | Reported supply | Universal; reduction incentive ~0.04–0.2% of revenue |
| Prevention bonuses | Producer (voluntary claim) | Declared actions | Tried (France 2012); no documented effect |
| Per-capita reduction targets | Member state | National packaging-waste statistics | EU: 5%/10%/15% by 2030/35/40 |
| Producer reduction mandate | Producer | Dual metric (units and weight), 2023 baseline | California only; assessed to 2032 |
| Format bans and reuse mandates | Market-wide | Prohibition (no measurement needed) | EU 2030 formats; various national |
The table's last column is the measurement argument in miniature: the instruments that work do so either by prohibiting (no counterfactual required) or by legislating the counterfactual into a filed baseline — never by pricing the absence.
6. Reading prevention claims: three habits¶
For any reader of scheme reports and policy claims, three habits follow from this article.
Ask what the denominator is. A rising recycling rate can mean better collection, a moved calculation point, or more packaging placed on the market. The EU's own 2019 methodology change — shifting measurement from input-to-sorting to input-to-final-recycling — reduced reported plastic packaging rates across Europe with no physical change whatsoever; denominators move for reasons that have nothing to do with performance.
Treat prevention claims as unverified by default. Not false — unverified. Every prevention claim embeds a constructed counterfactual; ask who constructed it and what would falsify it.
Watch for the self-penalising design. Where a jurisdiction holds recycling-rate targets and prevention targets simultaneously on an unadjusted denominator, its own metrics are in conflict, and scheme behaviour will follow the funded metric — which is never prevention.
7. Where the argument stands¶
The prevention question admits a cleaner verdict than most in this library, because the mechanism is structural. EPR as constituted cannot demonstrate prevention, because its measurement apparatus counts what is collected and prevention is the absence of something to collect. The per-tonne fee provides a real but arithmetically weak reduction incentive; the modulation layer points elsewhere; and the reporting layer is denominator-blind. The two legislative fixes take opposite routes around the machinery — the EU assigning prevention to states, California defining it into producer-level observability with a dual metric and an elimination floor — and both are, in effect, concessions that the fee system could not carry the objective.
What would change the verdict is specifiable: California's 2032 accounting producing verified, litigation-surviving reduction; or any scheme funding the econometric work that would detect a fee-driven reduction effect against economic confounds. Until one arrives, the defensible position is the one the evidence has supported since Lindhqvist's own retrospective (Theme 1): financing delivered, prevention not — and the instruments that may finally deliver it are mandates and bans that EPR's fee machinery hosts but does not power.
References¶
- Corvellec, H. (2016) — waste prevention as producing "no waste to count"; Johansson & Corvellec on European prevention programmes as aspirational documents.
- Joltreau, E. (2022). "Extended Producer Responsibility, Packaging Waste Reduction and Eco-design." Environmental and Resource Economics 83(3), 527–578 — the "very little (though statistically significant) packaging reduction" finding.
- Regulation (EU) 2025/40 — the 5%/10%/15% per-capita reduction targets (member-state obligations), reuse targets and format prohibitions.
- California SB 54 (PRC, Chapter 3) — the 25%-by-2032 source-reduction requirement, dual unit-and-weight metric, and elimination/reuse floor; the revised CalRecycle regulations (2025).
- Commission Implementing Decision (EU) 2019/1004 — the calculation-point change and its effect on reported European plastic packaging rates.
- Zero Waste Europe — the split-budget (protected waste-reduction budget) proposal.
- Lakhan, C. (2026). Ecomodulation of Extended Producer Responsibility Fees: A Literature Review. Working paper, Circular Innovation Hub, York University — the prevention-criteria gap in modulation frameworks.
- Lindhqvist, T. — retrospective interviews (Packaging Europe; Resource/Ecoveritas, 2023), via the Theme 1 treatment.
Verification note: the prevention-measurement literature is characterised from the sources cited; the claim that SB 54 contains the first producer-level prevention mandate in packaging EPR is an absence finding from this library's comparative review, stated as such. See Sources and method.