How does EPR differ from product stewardship, deposit-return systems, and packaging taxes?¶

"EPR" shares a crowded field with several instruments that also make producers pay for packaging: product stewardship programs, deposit-return systems, advance disposal fees, packaging taxes, and negotiated covenants. The labels are used loosely — sometimes interchangeably, sometimes strategically — and the confusion is not harmless, because the instruments differ in who pays, who controls the system, and what behaviour each can actually change. This article sets out the distinctions that matter, the official tests that exist for drawing them, and what the comparative evidence says about how the instruments perform alone and in combination.

1. Three questions that classify any instrument¶

The most reliable way to classify a packaging policy is to ignore its title and ask three questions: who pays, who decides, and who can be sanctioned? Every instrument in this family is a different configuration of those three answers, and — as this article shows — the configuration predicts behaviour far better than the name does.

Instrument	Who pays	Who decides system design	Behavioural target
EPR (full form)	Producers (passing costs into prices)	Producers, within regulatory constraints	System financing; design change (aspirational)
Product stewardship (Canadian sense)	Consumers via visible fees, or government	Government or a delegated agency	Financing only
Deposit-return	Consumers who fail to redeem; producers for system costs	Statute (deposit level, scope); operator for logistics	Consumer return behaviour
Advance disposal fee / eco-fee	Consumers, visibly at the till	Government	Financing only
Packaging tax	Producers/importers (passing into prices)	Legislature	Material choice, via the tax base
Covenant	Negotiated	Negotiated between government and industry	Whatever the covenant specifies

The remainder of this article works through the comparisons a reader is most likely to need.

2. EPR versus product stewardship: the control test¶

The sharpest formal distinction comes from Canada. The Canadian Council of Ministers of the Environment (CCME) — the intergovernmental body of federal, provincial and territorial environment ministers — defined the two terms precisely in its Canada-wide Action Plan for Extended Producer Responsibility (2009). EPR programs are those "in which manufacturers and importers are fully and directly responsible for both the funding and operation of programs"; product stewardship programs are those in which producers are "neither directly responsible for program funding nor for program operations." The action plan then supplies the operational sentence:

"To a consumer an EPR and a product stewardship program can look similar but it is the lack of ability of producers to directly influence program funding, cost, design and operations that distinguishes the two approaches."

This is a control test: if the parties paying for a system cannot influence its cost, design and operations, it is stewardship, whatever the statute calls it. The test matters because control is what connects payment to incentive. A producer who funds a system it cannot manage has no lever with which to economise or innovate; a producer who controls operations concentrates both the incentive to reduce cost and the political leverage to seek weaker obligations. Programs on either side of the line behave differently in documented ways, which is why this library treats the distinction as substantive rather than terminological.

The United States uses the vocabulary differently, for reasons rooted in policy history rather than analysis. The President's Council on Sustainable Development, in its 1996 report Sustainable America: A New Consensus, adopted the phrase "extended product responsibility" — the same acronym with the word producer removed — a substitution that contemporaneous work commissioned by the US Environmental Protection Agency attributes to a preference for language implying "shared responsibilities in the product chain." American statutes have used "stewardship" and "EPR" interchangeably ever since; California's packaging law equates the terms explicitly. The practical consequence: a US bill titled "product stewardship" may impose more on producers than a Canadian program titled "EPR," or far less. The word is not information; the CCME's three questions are.

3. EPR versus deposit-return: siblings with opposite strengths¶

Deposit-return systems (DRS) are both EPR's ancestor and its most important present-day companion instrument. Producer-run recovery of refillable beverage containers is roughly a century old — the Container Recycling Institute, a US research organisation focused on beverage container recovery, describes deposit systems as "perhaps the earliest form of EPR system" — and the legislated versions in British Columbia (1970) and Oregon (1971) predate the EPR concept by two decades.

Analytically, however, the two instruments work on different margins, and the difference explains nearly everything about how they perform.

A deposit prices an individual consumer's behaviour. The consumer pays the deposit at purchase and recovers it personally upon return. The incentive is individually appropriable, salient, and large relative to the decision it targets. An EPR fee prices a producer's membership in a collective system. It is invisible to the consumer, small relative to product price, and — under the market-share cost allocation used everywhere — only weakly connected to any individual firm's choices (the individual-versus-collective article).

The performance evidence tracks this difference. The empirical record for deposits is the strongest in the packaging field: Germany's system returns 98% of containers; Finland's 97%; Norway's and Denmark's 92%. The field's cleanest natural experiment is Connecticut, which doubled its deposit from five to ten cents in January 2024 and expanded redemption infrastructure — and saw its return rate rise from 43% to 92% within two years, demonstrating that deposit level and convenience, not culture or demographics, drive performance. The efficiency literature reaches the matching conclusion from theory: the reference calculation by Karen Palmer, Hilary Sigman and Margaret Walls (1997, in the Journal of Environmental Economics and Management) found a deposit-refund achieves waste reduction at roughly half the cost per ton of single-margin alternatives, because it works on both the consumption and recovery margins simultaneously (the economic-foundations article).

Two cautions complete the picture. First, a badly designed DRS underperforms a good kerbside system: Massachusetts, with an unchanged five-cent deposit, returns 36% of containers — less than well-run collection captures — while still imposing the system's costs. Deposit systems are powerful, not automatically so. The Netherlands supplies the governance version of the same caution: with return rates well below the statutory 90% target (roughly 68% for plastic bottles and 65% for cans), the Dutch environmental inspectorate has issued repeated enforcement orders against the system operator, Verpact — which had meanwhile accumulated some €374 million in unredeemed deposits over three years while producer fees were being reduced. Where the operator keeps unredeemed deposits and faces no enforcement pressure, under-performance is self-financing. Norway's design inverts the incentive — its environmental tax on producers falls as the return rate rises, so performance itself is what reduces producer cost — and Norway returns 92% of containers. The comparison is a compact lesson in how the same nominal instrument produces opposite results depending on who captures the money that failure generates. Second, when the two instruments run together — as almost every jurisdiction in Canada, the US and Europe is now committed to doing — the DRS strips the highest-value materials (PET bottles and aluminium cans) out of the kerbside stream that packaging EPR finances, leaving the EPR system with heavier, dirtier, lower-revenue material. This interaction, known as cannibalisation, is the most consequential instrument-interaction problem in the field; modelling commissioned by both the waste industry and the can-manufacturing industry (which sit on opposite sides of the debate) agrees it is real and bounded, disagreeing mainly over compensation. Theme 7 of this library treats the interaction in full; what belongs in a foundations article is the design lesson that the two instruments are complements with a contested boundary, not substitutes.

4. EPR versus the advance disposal fee: the visible-fee trap¶

The advance disposal fee (ADF) — a per-unit charge levied at sale to fund end-of-life management — was the market-based instrument the United States debated through the 1980s as its alternative to bottle bills; Florida briefly enacted one (1988, effective 1992, later repealed). The instrument survives today inside EPR vocabulary as the "eco-fee" displayed at the till in some programs.

The distinction from EPR is precise and frequently missed. An ADF is levied on the consumer, visibly, at a rate set by government; the producer is a collection agent. An EPR fee is levied on the producer, at a rate emerging from the system's costs; whether it reaches the consumer depends on market pass-through. The consequence for incentives is decisive: an ADF gives the producer no design incentive whatsoever — the fee is not the producer's cost, so no redesign can reduce it. In the CCME's classification, a visible eco-fee is a stewardship device, not an EPR one. The efficiency literature adds the second strike: as a single-margin instrument, the ADF achieves waste reduction at roughly twice the cost per ton of a deposit-refund (Palmer, Sigman and Walls's $85 against $45).

The practical reading rule follows: when a program's funding appears as a separate line on consumer receipts at a government-set rate, its producer-responsibility label should be discounted accordingly.

5. EPR versus packaging taxes: instrument against institution¶

A packaging tax — a legislated levy on packaging placed on the market, typically varying by material or recycled content — is the closest fiscal relative of EPR, and the comparison illuminates both.

The similarities are real: both charge producers per unit of packaging, both pass substantially into prices, and both can differentiate by material. The differences are structural. A tax's rate is set by a legislature on whatever base it chooses — which means it can be set above system cost, calibrated (at least in intent) to environmental harm, and pointed at a specific margin such as virgin-plastic content. Its revenue goes to the treasury. An EPR fee, by contrast, is legally anchored to system cost — in the EU, capped at the cost of cost-efficient service provision; in France, classified by the Conseil d'État as consideration for a service (the fee-or-tax article) — and its revenue stays inside the waste-management system, funding collection, sorting and processing.

That last difference is the important one, and it cuts both ways. The tax is the sharper incentive instrument: it can be made as large as the legislature dares, and it requires no producer organisation, no cost allocation, and no governance apparatus. But it builds nothing: a tax finances a treasury, not a collection network, and no packaging tax anywhere has created the physical infrastructure that EPR systems finance as their core function. The theoretical literature's verdict on single-margin instruments applies to taxes with full force — and the institutional literature's verdict on EPR applies equally: it is best understood not as a price instrument but as an institutional arrangement for financing a system (the economic-foundations article). A jurisdiction choosing between them is not choosing between two versions of the same thing; it is choosing between an incentive without an institution and an institution whose incentives are weak. Several European jurisdictions now operate both simultaneously — a combination the evidence base has barely begun to evaluate.

6. The covenant route: the negotiated alternative¶

One further alternative deserves mention because it defined an era. The Netherlands' 1991 Packaging Covenant was a negotiated agreement between government and industry — targets and obligations accepted contractually rather than imposed by regulation. The covenant model offered speed and flexibility, and its weaknesses were the mirror image: enforceability against non-signatories, free-riding, and renegotiation under pressure. The Netherlands itself subsequently migrated to a regulated EPR system, and the covenant survives mainly as a transitional or supplementary device. Its lesson for the classification exercise is that the binding/voluntary axis is independent of the who-pays axis — a fact the "shared responsibility" vocabulary tends to obscure.

7. Reading any program: a decoding guide¶

For the reader confronting an unfamiliar program, the terms most often encountered, decoded:

"Product stewardship" — in Canada: producers fund but do not control, often via a visible eco-fee. In the US: usually a synonym for EPR. Apply the control test.
"Shared responsibility" — a signal that producers will cover part of the cost. The operative questions are what part, and who covers the rest.
"Eco-fee" / "advance disposal fee" — a visible charge to consumers. A stewardship device: no producer design incentive.
"Producer responsibility organisation" — the collective body producers fund. Its existence says nothing about who controls it; its board composition and the regulator's powers do (Theme 2).
"Deposit" — the one charge in this family the consumer can personally recover. Behaviourally the strongest instrument per dollar; institutionally the narrowest in scope.

And the general rule, which the CCME's 2009 formulation captures best: ignore the title; ask who pays, who decides, and who gets fined. No combination of those answers is reliably signalled by any of the labels.

8. Where the argument stands¶

Three conclusions organise the comparative evidence.

First, the instruments are complements more than competitors. The strongest-performing packaging systems in the world combine a deposit system for beverage containers (the behavioural instrument), EPR for the remaining stream (the financing institution), and, increasingly, regulatory standards for design (the mandate instrument) — a division of labour that matches each instrument to the margin it actually moves. The design questions this combination raises — above all the DRS–EPR cost boundary — are among the most active disputes in the field.

Second, classification errors have consequences. Jurisdictions that adopt a visible eco-fee and expect design change, or a packaging tax and expect infrastructure, or a "shared responsibility" program and expect full municipal relief, are asking instruments for what they cannot deliver. Most documented disappointments in this policy area trace to a mismatch between the instrument chosen and the outcome expected.

Third, the labels are strategic, and have been since the 1990s. "Stewardship" was adopted in the United States precisely to soften producer obligation; "hidden tax" is deployed against EPR precisely because the charge is legally not one. The analyst's defence is the three-question test, applied without regard to what anyone calls the program.

References¶

Canadian Council of Ministers of the Environment (2009). Canada-wide Action Plan for Extended Producer Responsibility, p. 4. Quoted directly in Section 2.
President's Council on Sustainable Development (1996). Sustainable America: A New Consensus; Davis, G., Wilt, C., Dillon, P. & Fishbein, B. (1997), report for the US EPA — the documented origin of the US terminology shift.
Container Recycling Institute — deposit systems as proto-EPR; US return-rate data.
Palmer, K., Sigman, H. & Walls, M. (1997). "The Cost of Reducing Municipal Solid Waste." Journal of Environmental Economics and Management 33(2), 128–150 — the $45/$85/$98 instrument comparison.
Reloop (2024). Global Deposit Book and subsequent updates — international return-rate data cited in Section 3 (Germany 98%; Finland 97%; Norway and Denmark 92%; Connecticut 43%→92% following the January 2024 deposit increase; Massachusetts 36%).
NWRA/RRS (2022) and CMI/Circular Matters (2022–23) — opposing-industry modelling of deposit-system effects on materials-recovery-facility economics, cited for their agreement that cannibalisation is real and bounded. Both are interested-party studies and are identified as such.
Directive (EU) 2018/851, Art. 8a(4)(c); Conseil d'État, 28 December 2017, n° 408425 — the cost-anchoring of EPR fees discussed in Section 5.
Florida advance disposal fee: 1988 act, fee effective 1992, subsequently repealed.
On the Netherlands Packaging Covenant (1991) and its successors: standard accounts in the comparative EPR literature, including the OECD guidance documents.

Verification note: return-rate figures mix reporting years and national methodologies and should be normalised before quantitative reuse; the qualitative rankings cited here are robust to those differences. Interested-party modelling is identified as such wherever cited. See Sources and method.