Report 55 — Vienna–Montreal Ozone Regime: Substantial but Unfinished Successful Repair

v1.8 status note: This is a bounded retrospective comparison. Its adequacy gates were specified after the historical outcomes were known, not preregistered in advance. The case can refine or challenge the theory, but it does not by itself establish a general Repair-Lag mechanism.

Case question

What can the Vienna–Montreal ozone regime show about successful repair when institutions reverse a causal pressure before the affected substrate has fully recovered?

Short answer

The integrated Vienna–Montreal regime met a strong institutional-control criterion and reversed the causal pressure. Binding controls, differentiated schedules, finance, technology transfer, reporting, licensing, trade rules, monitoring, compliance procedures, domestic implementation, and repeated revision helped phase out more than 99% of controlled ozone-depleting-substance production and consumption. Atmospheric loading of major controlled substances and reactive halogen has declined.

Full return of total-column ozone to the 1980 benchmark has not yet been observed and remains conditional on continued policy and compliance. The safest classification is therefore substantial, adaptive, still-incomplete successful repair: strong control output and pressure reversal, emerging ozone response, and unfinished substrate recovery.

Bounded repair episode

The object is the Vienna–Montreal ozone regime, not the signature of the 1987 treaty treated as a single isolated cause.

Three analytic adequacy gates

These gates distinguish an intervention from its pressure and substrate outcomes. They are analytic operationalizations made with knowledge of the case history. They are not a preregistered causal test and do not identify the separate contribution of each institutional mechanism.

Empirical anchors by evidence type

Treaty and institutional design

• The official handbook records quantitative controls, reporting, licensing, trade restrictions, periodic assessment, non-compliance procedures, differentiated schedules, finance, and technology-transfer obligations. Source: S136.
• The Multilateral Fund supplies part of the differentiated finance and implementation-capacity architecture. Source: S140.
• As of 8 April 2026, the original Vienna Convention and Montreal Protocol had 198 parties; the Kigali Amendment had 173. This is participation status, not proof of compliance or legitimacy. Source: S142.

Administrative output

• WMO reported in 2025 that more than 99% of production and consumption of controlled ozone-depleting substances had been phased out. This is not a claim that all emissions, banks, feedstocks, or uncontrolled substances have disappeared. Source: S139.
• The Multilateral Fund reports cumulative finance and project outputs across 144 Article-5 countries. These are authoritative administrative self-reports, not independent causal or distributive evaluations. Sources: S140 and S141.

Observed pressure reversal

• The 2022 WMO/UNEP assessment found continuing declines in atmospheric chlorine and bromine from long-lived controlled substances. Source: S137.
• NOAA reported that nearly all abundant controlled ozone-depleting substances were declining in 2024. Its reactive-halogen index had moved 55% from peak toward the 1980 benchmark at mid-latitudes and 28% in the Antarctic stratosphere. Source: S138.

Observed atmospheric and ozone response

• Upper-stratospheric ozone recovery is progressing, and Antarctic total-column ozone continues to recover amid large year-to-year variability. Outside Antarctica, evidence of total-column recovery since 1996 was limited and assessed with low confidence in the 2022 assessment. Source: S137.

Conditional projection and model counterfactual

• Return to 1980 total-column ozone is projected around 2040 near-globally, 2045 in the Arctic, and 2066 in Antarctica if current policies remain in place. This is a projection, not an observed completed outcome. Sources: S137 and S139.
• Newman and colleagues model a severely damaged no-regulation world under continued ozone-depleting-substance growth. This supports the physical counterfactual; it does not establish what markets would certainly have done without the regime. Source: S143.

A nested causal sequence

The case is best read as a sequence rather than a treaty-to-outcome shortcut:

production and consumption controls
    → lower emissions and releases
    → declining atmospheric ODS stock
    → declining stratospheric reactive halogen
    → slow and variable ozone replenishment

Existing chemical banks continue to release some substances. Long atmospheric lifetimes delay decline in the tropospheric stock. Transport delays the stratospheric response. Chemistry, circulation, climate change, and natural variability affect ozone recovery. A long replenishment period can therefore coexist with effective institutional repair.

Lag diagnosis

1. Scientific sensing and interpretation narrowed an initial epistemic lag.
2. Treaty schedules, reporting, trade rules, and repeated meetings reduced coordination lag.
3. Differentiated finance and technology transfer addressed part of the capacity lag.
4. National implementation reduced the damaging production and consumption flow.
5. Chemical banks, atmospheric lifetimes, transport, chemistry, and natural variability slow the later substrate response.

The remaining atmospheric response time is not, by itself, proof of ongoing institutional failure. Strategic lag is not established by this source set; claims about purposeful industry obstruction or support require actor-level evidence.

Power and distribution

The regime combined legal obligation, market access, trade restrictions, scientific expertise, finance, and domestic implementation. Article-5 countries received differentiated schedules and financial and technical support, and the Multilateral Fund's Executive Committee has equal Article-5 and non-Article-5 membership. Those design features matter, but they do not prove equal bargaining power, sufficient finance in every setting, or complete distributive legitimacy.

Benefits are global, diffuse, and intergenerational. Conversion costs were concentrated in producing and using sectors and in countries with less technical and financial capacity. Substitute chemicals created climate and safety burdens that required later correction. Evidence about workers, consumers, patent access, local effects, and the distribution of gains and losses remains incomplete.

Serious rivals and scope conditions

• Technology and market substitution: alternatives and commercial advantage may have driven a substantial share of the phase-out.
• Domestic regulation: national bans and implementation may explain timing more directly than the international agreement alone.
• Trade leverage and authority: market access and binding rules may matter more than shared evidentiary ground.
• Unusual tractability: a measurable group of chemicals, available substitutes, concentrated sectors, and trade leverage may make Montreal unlike economy-wide decarbonization.
• Natural variability and climate change: these materially influence annual and regional ozone outcomes, though they do not explain the collapse in controlled production and consumption.
• Formal compliance versus actual emissions: unexpected emissions, banks, feedstocks, by-products, exemptions, and monitoring gaps remain live challenges.
• Problem displacement: HCFC and HFC substitutes transferred some harm into climate and safety domains; later adaptation is part of the repair story.
• Policy-package attribution: the treaty, assessments, trade rules, Fund, national regulation, technical change, and market response are entangled. Calling them one system does not identify their separate causal shares.

What would lower confidence

Lower confidence or downgrade the case if:

• controlled atmospheric chlorine or bromine reverses persistently outside anticipated bank release;
• reported phase-out and top-down emissions remain persistently inconsistent and the regime fails to correct the discrepancy;
• expected ozone recovery fails over an appropriate multi-decadal window after climate and natural variability are accounted for;
• Article-5 implementation fails systematically because finance or technology transfer is inadequate;
• successor harms outweigh the bounded ozone and climate benefits after later amendments and life-cycle effects are included; or
• a credible counterfactual shows comparable global decline without the integrated regime's controls, finance, reporting, trade rules, and adaptive tightening.

Current theory disposition

C1 and C7 remain developing.

Next test

Compare one specified carbon-dioxide-emitting sector and jurisdiction against the Vienna–Montreal regime. Declare the observation period, lag stage, outcome, political-economy rival, and the conditions under which the institutional features should or should not travel. North American acid deposition supplies a general environmental boundary for C1, not the missing carbon-dioxide-sector boundary for C7.

Misuse warning

Do not present this case as proof that every planetary crisis can be repaired by elite technocracy, one universal treaty, information alone, or a simple ban. Do not erase developing-country claims, industry power, national implementation, transition costs, substitute harms, exemptions, enforcement gaps, or the unusual tractability of a measurable family of chemicals with available alternatives. The Vienna–Montreal regime is a comparison to learn from, not a template to copy mechanically.

Canonical sources used

• S136 — Montreal Protocol Handbook
• S137 — Scientific Assessment of Ozone Depletion 2022: Executive Summary
• S138 — NOAA Ozone Depleting Gas Index
• S139 — WMO 2025 bounded recovery update
• S140 — Multilateral Fund: About
• S141 — Multilateral Fund: Our Impact
• S142 — Ozone Secretariat ratification status
• S143 — Newman et al., *The World Avoided by the Montreal Protocol*
• S144 — U.S. EPA ozone-protection retrospective

Official administrative sources are strong records of rules, reported outputs, and monitoring. Several are produced by institutions that designed or administer the regime. They do not substitute for independent causal attribution or distributional review.