Is EPR a "polluter pays" policy? What the economic theory actually supports¶

Extended producer responsibility is routinely justified in three economic phrases — "polluter pays," "internalising externalities," and "market-based instrument" — and all three are doing less work than their users assume. This article assembles the economic foundations: where the polluter-pays principle actually came from and what it licenses; why no operating EPR fee is a Pigouvian tax; what the welfare-economics literature on waste policy actually ranks as efficient; and what is known about the size of the externality EPR is nominally internalising. The conclusion is uncomfortable for both advocates and critics: EPR's genuine achievements rest on foundations quite different from the ones printed on its label.

1. The polluter-pays principle: a trade rule that grew¶

The polluter-pays principle did not begin as environmental ethics; it began as a subsidy discipline. The Organisation for Economic Co-operation and Development (OECD) introduced the principle in a 1972 Council Recommendation (C(72)128) framing it thus: the polluter should bear the expenses of pollution-control measures "decided by public authorities to ensure that the environment is in an acceptable state." The operative concern was trade. Governments should not subsidise their industries' abatement costs, so that environmental costs enter product prices without distorting international competition. A 1974 extension (C(74)223) made the no-subsidy rule explicit, with narrow, time-limited exceptions.

In its original form, then, the principle says nothing about damages, victims, or full internalisation of environmental harm. It is a cost-allocation rule about who pays for required measures — and the OECD's own 1992 retrospective documents its gradual widening over two decades toward administrative costs, residual damages, and accidental pollution.

The principle subsequently hardened into law with characteristic hedging. It entered European primary law through the Single European Act (1987), and now appears in Article 191(2) of the Treaty on the Functioning of the European Union. Principle 16 of the 1992 Rio Declaration carries a triple hedge worth noticing: national authorities "should endeavour" to promote internalisation, "taking into account the approach that the polluter should, in principle, bear the cost of pollution," and "without distorting international trade."

For waste specifically, two decisions of the Court of Justice of the European Union frame the field. In Commune de Mesquer v Total (Case C-188/07, 2008), the Grand Chamber established that the principle can reach upstream to the producer of the product from which waste came — the doctrinal bridge to EPR — though with an operative condition frequently omitted in citation: such a producer cannot be made to bear the cost "unless he has contributed by his conduct to the risk that the pollution... will occur." In Futura Immobiliare (Case C-254/08, 2009), the Court held that waste charges need not be calibrated to the waste actually generated; capacity-based proxies satisfy the principle, provided cost allocation is not "manifestly disproportionate." Read together, European law permits assigning waste costs to producers, and permits those costs to be rough. That is precisely the legal space EPR occupies.

Is EPR "really" a polluter-pays instrument? The OECD's own working papers of the late 1990s confronted the obvious objection — the consumer who discards a package is at least as much the "polluter" as the firm that made it — and resolved it functionally: financial responsibility is assigned to producers because they control design, not because they are uniquely culpable, and the costs are expected to pass into product prices in any event. Legal scholarship (the standard treatment is by Kleoniki Pouikli, in ERA Forum, 2020) formalises this as a cheapest cost-avoider rationale rather than a fault rationale. The familiar critique — that near-complete pass-through makes EPR a "consumer pays" charge with polluter-pays language as decoration — is correct about incidence but misses the design logic: the point of routing the charge through producers is the decision rights it attaches to, not the wallet it ultimately lands in. Whether those decision rights are actually exercised is the subject of the preceding article.

2. Pigou versus the service charge: what an EPR fee actually is¶

The textbook framing treats EPR fees as Pigouvian — named for the economist Arthur Pigou, whose corrective tax equals the marginal external damage of an activity, letting markets adjust to the true social cost. Operating EPR systems are nothing of the sort, and the mismatch is documented at every level.

Margaret Walls of Resources for the Future, the Washington research institution whose waste-economics program produced much of the relevant literature, found in a 2006 synthesis that no EPR fee anywhere is calibrated to marginal external cost, or even to marginal disposal cost. Fees are weight-based averages set to balance a producer responsibility organisation's budget. A 2014 study for the European Commission by BIO Intelligence Service and Deloitte found that fees "seldom reflect the true management costs" of the products concerned. And the OECD's 2021 working paper on modulated fees treats precisely this misalignment as the explanation for the design-incentive failure reviewed elsewhere in this theme.

The law then codifies the non-Pigouvian reading. Article 8a(4)(c) of the EU Waste Framework Directive caps producer contributions at the costs necessary to provide waste management services "in a cost-efficient way" — a ceiling defined by system cost, not by damage. And France's highest administrative court, the Conseil d'État, ruled squarely in December 2017 (n° 408425) that the French éco-contribution is not a tax or fiscal levy, because it constitutes "the direct counterpart of the service" the producer organisation renders on the producer's behalf. The fee is, legally, consideration for a service — definitionally cost-based, with damage internalisation at most incidental. The consequences of that classification are treated in the following article.

The synthesis for this article's question: an instrument whose charge is legally capped at service cost, empirically uncalibrated to harm, and largely passed through to consumers is not a Pigouvian tax, and describing EPR as "internalising externalities" is at best a loose metaphor. What EPR internalises is the financial cost of waste management — a real accomplishment, but a different one.

3. The efficiency canon: what welfare economics actually recommends¶

A compact economics literature from the 1990s settled the instrument-ranking question for waste policy, and its results are cited constantly and read rarely. The essentials are worth setting out, because they explain both why economists remain lukewarm about EPR-as-practised and where its defensible economic role lies.

The benchmark and the dumping problem. The first-best policy is a properly priced disposal margin: charge households the marginal social cost of what they discard. The immediate objection — priced disposal invites illegal dumping — generates the entire upstream literature. Terry Dinan, in a 1993 article in the Journal of Environmental Economics and Management, showed that a combination of a producer-level tax and a reuse/recycling subsidy replicates the disposal fee's allocation without creating a dumping margin. Karen Palmer and Margaret Walls (1997) formalised the point: a product tax plus recycling subsidy is a deposit-refund system implemented at the producer level, attaining the social optimum without pricing household disposal at all.

The magnitudes. The reference numbers come from Palmer, Hilary Sigman and Walls, in a 1997 article in the same journal: to achieve a 10% reduction in municipal solid waste, a deposit-refund system costs about $45 per ton, an advance disposal fee $85, and a recycling subsidy $98. The intuition generalises beyond the calibration. The deposit-refund works both margins — less consumption and more recycling. Each single-margin instrument works one margin and wastes effort: the disposal fee suppresses consumption but shrinks recycling feedstock; the subsidy boosts recycling while cheapening output. Single-margin instruments are roughly twice as expensive per unit of waste reduced.

The design margin. Don Fullerton and Wenbo Wu (1998) extended the analysis to packaging attributes: when households face "free" garbage collection, prices cannot transmit design incentives, and the corrective is a producer-level tax on packaging plus a subsidy to recyclable design — government must directly price the attribute the market cannot see.

The transaction-cost turn. The most under-used result in EPR discussions is by Paul Calcott and Walls (2000, 2005). Why don't recyclers simply pay households more for recyclable packaging, generating design incentives through market prices? Because per-item, per-attribute payments carry transaction costs that swamp the values involved — "missing prices" are the natural state of recycling markets. The first-best would require product-specific taxes varying continuously with recyclability, which is implausible; the constrained optimum is a common deposit-refund on products plus a modest disposal fee deliberately set below the Pigouvian level to avoid dumping. In other words, the theoretically correct waste policy in a transaction-cost world looks like a deposit-refund system plus cheap-but-not-free disposal — and notably not like uniform collective EPR fees, which price neither the disposal margin nor the design margin.

Instrument	Cost per ton (10% waste reduction)	Margins engaged
Deposit-refund	~$45	Consumption and recycling
Advance disposal fee	~$85	Consumption only
Recycling subsidy	~$98	Recycling only
Uniform collective EPR fee	Not ranked in the canon	Neither margin priced at the individual-producer level

Source: Palmer, Sigman & Walls (1997); final row is this library's characterisation based on Walls (2006).

Where does this leave EPR? The honest reading: EPR-as-practised is not an instrument the efficiency literature recommends; it is an institutional arrangement for financing collection, onto which economists keep trying to graft the instruments the literature does recommend — deposit-refund elements, and modulated fees approximating design subsidies. Those grafts are the subject of later themes in this library.

4. How large is the externality? Awkward evidence in both directions¶

The final foundation is the size of the externality EPR is nominally internalising, and the literature here embarrasses both camps.

Against the pro-recycling side: the disposal externality is modest. Thomas Kinnaman, an economist at Bucknell University and the leading empirical researcher on household waste, argued in a 2006 "Policy Watch" essay in the Journal of Economic Perspectives for replacing several US state recycling mandates with a moderate landfill tax — a formulation implying a residual un-internalised externality, but a modest one, for modern lined, methane-capturing landfills. With Takayoshi Shinkuma and Masashi Yamamoto, Kinnaman pushed further in a 2014 article in the Journal of Environmental Economics and Management using full social-cost accounting for Japan — including municipal costs and household sorting effort, offset by measured willingness-to-pay for recycling — and concluded that average social costs are minimised "with recycling rates well below observed and mandated levels in Japan." The widely circulated gloss that the socially optimal recycling rate is around 10% is real but originates in Kinnaman's own popular restatement of the finding, not in the article's text, and should be attributed accordingly. The result is contested; it has not been refuted; and every packaging-EPR target conversation implicitly assumes it away.

Against the sceptics: the marine-leakage externality is enormous — where it applies. Nicola Beaumont and colleagues, in a 2019 article in Marine Pollution Bulletin, estimated a 1–5% reduction in marine ecosystem service delivery from the stock of plastic in the oceans as of 2011, equating to an annual loss of $500–2,500 billion — "each tonne of plastic in the ocean having an annual cost in terms of reduced marine natural capital of between $3,300 and $33,000." That per-tonne figure is routinely miscited in policy debate as if it were a marginal damage per tonne of packaging placed on the market. It is not: it is an annual ecosystem-service loss per tonne of the existing marine stock, and it cannot be transferred into fee-setting arithmetic. The policy-relevant implication is that the externality case for packaging policy runs through leakage — mismanaged waste reaching the environment — not through disposal per se. A tonne of packaging destined for a lined landfill and a tonne destined for a river carry externalities differing by three or four orders of magnitude, and a uniform fee prices neither correctly.

The synthesis, and the gap. For the working analyst: EPR fees are neither too high nor too low relative to measured externalities — they are uncalibrated to them. In jurisdictions where leakage is rare (most of Canada, the United States and Europe), fees probably exceed the marginal damage of well-managed disposal, while bearing no relation to the leakage externality that dominates globally. No study anywhere estimates marginal external cost per tonne of packaging by format and destination and compares it to an operating fee schedule. That is arguably the most fundamental unwritten paper in the field.

5. Where the argument stands¶

Five propositions summarise what the theory and evidence actually support.

The polluter-pays principle licenses cost allocation, not damage calibration. It began as a no-subsidy trade rule; its legal descendants permit assigning waste costs to producers and permit those costs to be rough. Invoking the principle settles far less than either side of the EPR debate pretends.
EPR's legal identity, where tested, is a service charge — not a tax, and not a Pigouvian instrument. Both the "hidden tax" framing used by opponents and the "pricing pollution" framing used by advocates are wrong, in opposite directions.
The efficiency canon ranks deposit-refund and two-part instruments above everything EPR schemes actually do, and its transaction-cost branch explains why the market gap EPR fills is nonetheless real.
The externality record supports pricing leakage heavily and disposal lightly — approximately the inverse of the attention distribution in current packaging policy.
The theory neither endorses nor rejects EPR. Its message is that the instrument's genuine achievements — stable financing and infrastructure formation — rest on institutional rather than externality-pricing foundations. Evaluating EPR as if it were a Pigouvian tax guarantees disappointment; evaluating it as a financing institution invites the governance questions taken up in Theme 2.

References¶

The polluter-pays principle

OECD (1972). Recommendation C(72)128 on Guiding Principles concerning International Economic Aspects of Environmental Policies; OECD (1974), Recommendation C(74)223 on the Implementation of the Polluter-Pays Principle; OECD (1992), The Polluter-Pays Principle: OECD Analyses and Recommendations.
Article 191(2) TFEU; Rio Declaration (1992), Principle 16.
CJEU, Case C-188/07 Commune de Mesquer v Total (Grand Chamber, 24 June 2008); Case C-254/08 Futura Immobiliare (16 July 2009).
OECD (1998). Extended and Shared Producer Responsibility (ENV/EPOC/PPC(97)19/REV2) — the producers-because-they-control-design reconciliation.
Pouikli, K. (2020). "Concretising the role of extended producer responsibility in European Union waste law and policy." ERA Forum 20, 491–508.

The efficiency canon

Dinan, T. (1993). "Economic Efficiency Effects of Alternative Policies for Reducing Waste Disposal." Journal of Environmental Economics and Management 25(3), 242–256.
Palmer, K. & Walls, M. (1997). "Optimal policies for solid waste disposal: taxes, subsidies, and standards." Journal of Public Economics 65(2), 193–205.
Palmer, K., Sigman, H. & Walls, M. (1997). "The Cost of Reducing Municipal Solid Waste." Journal of Environmental Economics and Management 33(2), 128–150.
Fullerton, D. & Wu, W. (1998). "Policies for Green Design." Journal of Environmental Economics and Management 36(2), 131–148.
Calcott, P. & Walls, M. (2000). American Economic Review Papers & Proceedings 90(2), 233–237; and (2005), "Waste, recycling, and 'Design for Environment.'" Resource and Energy Economics 27(4), 287–305.
Walls, M. (2006). Extended Producer Responsibility and Product Design. Resources for the Future Discussion Paper 06-08.

Externality magnitudes

Kinnaman, T. (2006). "Policy Watch: Examining the Justification for Residential Recycling." Journal of Economic Perspectives 20(4), 219–232. Characterised from the abstract and secondary literature; full text access-restricted.
Kinnaman, T., Shinkuma, T. & Yamamoto, M. (2014). "The socially optimal recycling rate: Evidence from Japan." Journal of Environmental Economics and Management 68(1), 54–70.
Beaumont, N.J. et al. (2019). "Global ecological, social and economic impacts of marine plastic." Marine Pollution Bulletin 142, 189–195.

Legal classification and cost coverage

Conseil d'État (France), 28 December 2017, n° 408425 (Fédération des industries nautiques).
Directive (EU) 2018/851, Art. 8a(4)(c).
BIO Intelligence Service & Deloitte (2014). Development of Guidance on Extended Producer Responsibility. Report for DG Environment, European Commission.

Verification note: citations in this article were checked against publisher and official records during this library's citation audit. Kinnaman (2006) is cited for the argument its abstract actually makes; the "~10% optimal rate" gloss is attributed to Kinnaman's restatement rather than to the 2014 article's text; and the Beaumont et al. per-tonne figure is presented with the stock-versus-marginal distinction the authors' own wording requires. See Sources and method.