How should a jurisdiction manage the transition to EPR?¶

Moving a jurisdiction from municipally run recycling to a producer-run or producer-funded system is the highest-risk period in any packaging EPR program's life. Five things have to move — contracts, assets, data, money and blame — and the documented record from the completed transitions shows that the recurring failures occur not within any of the five transfers but in the gaps between them, where functions too minor for the negotiating table simply fall between the parties. This article reviews what is actually being transferred; the three sequencing models jurisdictions have used; the failure modes the record documents, drawn principally from Ontario and Quebec; and the design principles a jurisdiction planning a transition should adopt before setting a date.

1. What is actually being transferred¶

Contracts. Collection and processing agreements must expire, novate or transfer, and this — not policy ambition — sets the feasible calendar. Assignment usually requires counterparty consent; early termination has a price. Because the answer differs contract by contract, the transition calendar becomes a function of hundreds of separate expiry dates. The Canadian transition schedules were substantially reverse-engineered from municipal contract expiries — communities with long-dated contracts transitioned late — which is why those schedules look arbitrary from the outside and are anything but.

Assets. Municipalities own carts, bins, trucks and sometimes sorting facilities, bought with capital budgets and depreciated over long horizons. The options are sale, lease, transfer at book value, or stranding; each has fiscal consequences for parties who did not choose the transition, and no jurisdiction has published a general approach to stranded-cost settlement.

Data. The incoming system needs supply data (who placed what on the market), cost data (what the service actually costs), and performance data (what was collected and recycled). The first comes from producers; the second and third from municipalities and their operators — and the second is precisely what disappears if the transition terminates the instruments that collected it, as Section 3 documents.

Money. Fees must be assessed and collected before service obligations begin, or the producer organisation is capitalising a system with no revenue history. Mistiming in either direction creates either producer resentment (paying for nothing) or a cash crisis (obligations without revenue).

Blame. Resident-facing accountability stays with visible local governments long after operational control departs. Residents call their council about a service the council no longer runs, and councils cannot simply refuse. This asymmetry is documented in every completed transition, and no transition agreement has solved it — because it is not allocable by agreement.

2. Three sequencing models¶

Geographic staging. Ontario moved community by community over two and a half years, from July 2023 to December 2025, with full operational transfer complete on 1 January 2026. Alberta split by household type instead — single-family homes from April 2025, multi-family from October 2026. Staging spreads operational risk and lets the new operator learn; its cost is a long period in which two systems run in parallel and public communication must be locality-specific.

Functional staging. Quebec moved by function rather than geography, and the sequence is the design: regulation in force July 2022; the producer organisation designated October 2022; contract negotiations with municipal bodies obligatorily opened by March 2023; the decade-old municipal compensation regime terminated 31 December 2024; full producer operation from 1 January 2025. Residents saw no seam — at the price of compressing hundreds of contract negotiations against a fixed operational start date, which the system's higher-than-expected first-year costs may partly reflect.

Big-bang with soft-start provisions. The United Kingdom brought packaging fees in nationally at once, but with the first year unmodulated, enforcement deliberately patient, and the payment methodology explicitly provisional. The soft-start elements are the risk management: the scope is national from day one, but the parameters tighten over time rather than arriving at full force.

No model dominates. Geographic staging manages operational risk and extends the seam period; functional staging eliminates the resident-facing seam and concentrates negotiation risk; big-bang eliminates the parallel-running period and bets everything on the soft-start calibration.

The four most instructive transitions, compared:

Jurisdiction	Model	Timeline	Documented outcome
Ontario	Geographic staging	July 2023 – December 2025; full transfer 1 January 2026	Completed; mid-transition rollback of targets and scope (2025); measurement infrastructure terminated at handover; bin-responsibility gap at the seam
Quebec	Functional staging	Regulation July 2022 → full operation 1 January 2025	Seamless for residents; first operational year materially over cost expectations, with staggered producer payments as mitigation
Alberta	Staging by household type	Single-family April 2025; multi-family October 2026	In progress at the time of writing
United Kingdom	National big-bang, soft-start	First fee year unmodulated; methodology provisional	Roughly £1.46 billion collected in the first year; payment-model dispute with local authorities live

The money column of this comparison deserves emphasis, because the first-year sums are large enough to make sequencing errors expensive: approximately £1.46 billion in the UK's first fee year, Can$457 million in producer contributions in Quebec's first modernised year, US$167.9 million in Oregon's first program year. Fee collection at that scale, arriving before the operating system has a performance history, is precisely why the optimism-then-escalation cycle described below has political consequences — the sums are visible to producers from the first invoice, while the promised system benefits mature years later.

3. The failure modes, from the record¶

The residual-responsibility gap. When Ontario's municipalities exited bin provision in January 2026, no public communication established who pays for blue-box repair and replacement. Toronto's own notices directed residents to the producer organisation without stating coverage, and popular consumer explainers were still describing the pre-2025 program scope months later. The dollars are small; the example is the cleanest documented instance of the failure mode that matters most publicly — a function too minor for the negotiating table falling between the parties, in full view of every affected household.

Mid-transition renegotiation. Ontario's 2025 regulatory amendment deferred scheduled target increases and dropped planned service expansions — multi-residential buildings, schools, long-term care facilities and public spaces — under producer cost pressure, while the transition was still in progress. The general lesson: a transition schedule and obligations sitting in regulation are renegotiable throughout, because a regulation can be amended by the government of the day without legislative process. Jurisdictions that want end-state obligations to survive the first cost overrun need them in statute.

Loss of evaluability. Ontario terminated its Continuous Improvement Fund and municipal data call — the province's long-running cost and performance data infrastructure — in July 2025, at the moment of handover. No independent post-transition evaluation exists, and the before/after comparison that the transition's claimed savings invite is now, strictly, unmeasurable. A measurement instrument was switched off before its successor reported.

Deferred data assurance. The instinct to soft-start verification while a system beds in is understandable and compounds. Ontario's packaging verification requirement was waived for two consecutive submission years (the verification article), against a backdrop in which the province's Auditor General found audit requirements unenforced across nearly every other material stream (the capacity article). The foundational years — when fees are first set and baselines established — are precisely the years left unchecked.

Beyond these, two patterns recur across all documented cases. The optimism-then-escalation cycle: cost projections made at design are consistently exceeded in operation — Quebec's first operational year ran materially over expectations; Ontario's producers found costs high enough to secure the rollback. A jurisdiction should assume the first two years cost more than modelled and design its political protections accordingly. Communication lag: public understanding trails the design by a year or more, which becomes a fairness problem when households are being asked to sort correctly under rules they are misinformed about.

4. Design principles the record supports¶

Anchor the calendar to contract expiries, or budget for early termination. The contracts are the schedule; pretending otherwise produces either delay or termination payments.

Put the transition schedule in regulation but the end-state obligations in statute. Regulations move at the speed transitions need; statutes resist the renegotiation that cost pressure will predictably demand. Ontario's rollback was structurally easy because everything sat in regulation. A jurisdiction that accepts both in regulation should do so knowing both are renegotiable throughout.

Assign every residual function explicitly, however trivial. Bins, litter, public-space collection, resident communication, complaint routing. The seam failures are always in the unassigned remainder, and the assignment costs nothing at the drafting stage.

Never terminate a measurement instrument before its successor reports. Evaluation-data continuity is a design requirement, not an afterthought. Ontario's experience shows the alternative: the largest recently completed EPR transition in North America cannot now be evaluated against its baseline.

Treat the first operational year's costs as structurally unrepresentative in both directions — under-collection from soft enforcement, over-cost from compressed procurement — and say so publicly in advance, so that the first-year numbers do not become ammunition for premature renegotiation in either direction.

Verify during the transition, not after it. The transition years produce the baseline data that anchors everything after; they are the worst years to leave unaudited, however sympathetic the bedding-in argument sounds.

5. The next test: the US states¶

The transition lessons above were learned in jurisdictions moving an existing municipal recycling system to producer responsibility. The US states now approaching their operational dates face a variant of the problem with one harder feature and one easier one.

The harder feature is that several are building programs and regulatory frameworks simultaneously. The first consolidated US reporting round (June 2026) documented what that looks like from the inside: producers filed up to eight separate reports across six states; several states lacked final regulations at the reporting deadline; one required data categories some producers had never tracked; and submissions could not be edited once filed. A reporting obligation that lands before the rules defining it are final is a transition sequencing error of exactly the kind the Canadian record warns against — money and data obligations arriving ahead of the framework that gives them meaning.

The easier feature is that most US programs are not transferring operations at all. Oregon's producers fund enumerated system elements while ratepayers continue paying for collection; the seam between municipal operation and producer funding is narrower than in a full operational transfer, and the residual-responsibility gaps correspondingly smaller. The US transitions are, in the terms of the pay-or-run article, mostly financial-only — which trades the seam problem for the payment-model problem, and relocates the predictable disputes into fee methodology and cost boundaries rather than bin ownership (where those disputes have already begun).

What the US states inherit unambiguously from the Canadian record is the data lesson. Their foundational reporting years are now — and the verification arrangements attached to those years are, on the available analysis, the weakest of any operating jurisdiction (the verification article). Ontario's experience of waived verification during transition, followed by fee-setting on unaudited baselines, is the specific precedent the US programs are currently repeating at larger scale.

6. A transition checklist¶

Ten questions a jurisdiction should be able to answer before setting a date:

When does every affected municipal contract expire, and what does early termination cost?
Who owns the bins, and who replaces them after transfer?
What happens to municipally owned sorting infrastructure?
Which measurement instruments continue through the transition, and who funds them?
What is verified in the first two years, and by whom?
When do fees start relative to service obligations?
Who answers resident complaints, and how are they routed?
What is the plan for the first cost overrun — and which obligations are protected from renegotiation?
Which functions are explicitly not transferring, stated publicly?
Who will evaluate the transition, with what data, and by when?

Question 8 is the one the Ontario record makes unavoidable: cost pressure will arrive, and the time to decide which obligations can bend is before it does. Question 10 is the one no completed transition has answered adequately — and answering it is nearly free at the design stage and impossible afterwards.

7. Where the argument stands¶

The completed transitions support a synthesis that is neither alarmist nor complacent. The transfers themselves — contracts, assets, data, money — are manageable engineering, and the Canadian record shows systems of millions of households moving to producer operation without service collapse. What the record does not show is a transition that protected its own evaluability, verified its foundational data, assigned its residual functions completely, or held its obligations stable through the first cost overrun. Every documented failure is a governance failure at a seam, not an operational failure at a transfer.

The uncomfortable implication for the field is that the transitions' headline claims — smooth handover, municipal relief, maintained service — are largely true, while the questions that would establish whether the new systems are better — cost against baseline, performance against baseline, data integrity from year one — were rendered unanswerable by choices made during the transitions themselves. The next jurisdictions to transition — the US states now approaching their operational dates — have the complete checklist available and the failure modes documented in advance. Whether they use them is the live test.

References¶

Lindsay, D. (2019). Renewing the Blue Box: Final report on the blue box mediation process. Ontario — the design document for Ontario's transition.
Ontario O. Reg. 391/21 and the 2025 amendments (ERO 025-0009) — the transition schedule, target deferral and scope changes.
RPRA, transition-completion notice (January 2026); RPRA notice on the termination of the Continuous Improvement Fund and municipal data call (July 2025).
City of Toronto, resident transition notices (2026) — the bin-responsibility communication gap.
Québec MELCCFP, modernisation deployment schedule and regulation; Éco Entreprises Québec, first-operational-year reporting (2025) — the functional-staging sequence and first-year cost experience.
Alberta EPR program launch documentation — the household-type staging model.
PackUK documentation — the UK soft-start design (unmodulated first year, provisional payment methodology).
Auditor General of Ontario (December 2025), value-for-money audit of RPRA — the audit-enforcement backdrop.

Verification note: claims about Ontario's transition draw on primary regulatory notices and the Auditor General's audit rather than on program self-reporting; the evaluability finding — that no independent post-transition evaluation exists and the baseline comparison is now unmeasurable — is an absence finding from targeted searches. See Sources and method.