Since the 1972 Stockholm Conference launched the modern era of environmental diplomacy, states have signed more than 500 multilateral environmental agreements. Emissions keep rising. Biodiversity keeps declining. Fisheries keep collapsing.

This pattern is not simply a failure of political will. Environmental problems have structural features that make them uniquely resistant to the tools of international cooperation. The same institutional architecture that manages trade disputes or coordinates civil aviation struggles when applied to the atmosphere or the oceans.

Understanding why requires looking past the rhetoric of summits and communiqués to the underlying incentive structures. When benefits are diffuse and costs are concentrated, when verification is technically difficult, and when free-riding is rational, agreements bend toward the lowest common denominator. Yet a few notable exceptions demonstrate that better design is possible. Examining what separates the Montreal Protocol's success from the Kyoto Protocol's disappointment offers a practical template for reformers working on the next generation of environmental governance.

Environmental goods like a stable climate or intact biosphere are what economists call global public goods. They are non-excludable, meaning no country can be prevented from benefiting, and non-rival, meaning one country's benefit does not diminish another's. This creates a textbook free-rider problem: every state has an incentive to enjoy the benefits of others' restraint while continuing to pollute.

The asymmetry between costs and benefits deepens the challenge. Reducing emissions imposes concentrated costs on specific domestic industries and workers, whose losses are visible, measurable, and politically organized. The benefits, by contrast, are diffuse across all nations and future generations who cannot lobby, vote, or negotiate.

This dynamic explains why environmental agreements tend to codify what states were already willing to do rather than push them toward genuinely costly commitments. The Paris Agreement's nationally determined contributions represent this pattern in institutional form: countries set their own targets, adjust them as convenient, and face no real penalty for missing them.

The problem compounds when we consider the temporal mismatch. Political leaders operate on electoral cycles of two to six years, while climate benefits unfold over decades. Sacrificing GDP today for atmospheric stability in 2075 is a trade few political systems are structured to make.

Takeaway

When benefits are shared globally but costs are borne locally, the natural gravitational pull of any agreement is toward commitments states would have made anyway. Institutional design must actively counteract this pull, not simply document it.

Trade agreements work in part because violations are visible. A tariff appears in customs data; a subsidy shows up in government budgets. Environmental compliance is far harder to observe. Emissions from thousands of facilities, fishing catches at sea, or logging in remote forests can be underreported, mislabeled, or hidden entirely.

This information asymmetry weakens agreements in two ways. First, it makes cheating rational: if you cannot be caught, restraint becomes a competitive disadvantage. Second, it makes cooperation fragile. Even honest compliers grow suspicious that others are gaming the system, which erodes the political constituency for continued participation.

The problem is not merely technical. Even where remote sensing and independent audits are technically feasible, states resist intrusive monitoring on sovereignty grounds. The Kyoto Protocol's verification mechanisms were substantially weaker than those governing nuclear materials or chemical weapons, despite the arguably higher stakes.

Newer approaches attempt to route around this. Satellite-based methane monitoring, blockchain-verified carbon credits, and civil society reporting networks bypass state gatekeepers. These transnational verification systems, championed by scholars like Anne-Marie Slaughter, suggest that non-state actors may become as important to environmental compliance as governments themselves.

Takeaway

Agreements without credible verification are essentially promises. The future of environmental governance may depend less on what states commit to and more on who is watching, from satellites to sensors to citizens.

The 1987 Montreal Protocol on ozone-depleting substances stands as the most successful environmental treaty in history. It achieved near-universal ratification, effectively phased out CFCs and related chemicals, and has begun to heal the ozone layer. Understanding what made it work reveals the design principles that other agreements have lacked.

First, the problem had a manageable scope. A small number of chemicals produced by a limited number of companies caused the damage. This concentration made monitoring feasible and negotiation tractable, unlike carbon emissions which permeate every sector of every economy.

Second, viable substitutes existed. DuPont and other chemical manufacturers had already begun developing alternatives to CFCs, meaning the transition imposed transformation costs rather than absolute economic sacrifice. Industry became an ally rather than an opponent, a stark contrast to the fossil fuel sector's posture on climate.

Third, the treaty included robust mechanisms for adjustment, trade sanctions against non-parties, and a multilateral fund to help developing countries comply. These features created both carrots and sticks, aligning incentives across the North-South divide. The lesson is not that Montreal was easy, but that it was well-designed for the specific problem it addressed. Replicating that success requires matching institutional architecture to problem structure, not simply copying the treaty template.

Takeaway

Success in international environmental cooperation is not a matter of political willpower alone. It emerges when the problem is bounded, alternatives exist, and the agreement builds in both financial support and enforcement teeth.

The disappointment surrounding international environmental agreements is not evidence that global cooperation is impossible. It is evidence that the standard treaty toolkit was built for problems with different structural properties than the ones we now face.

The path forward likely involves smaller coalitions of committed actors, sector-specific agreements with clearer verification, and hybrid governance that draws in cities, companies, and civil society alongside states. Slaughter's networked model of global governance may prove more resilient than the treaty conferences we have inherited.

The Montreal Protocol shows that international environmental cooperation can work. The question is whether we are willing to design future agreements for the problems we actually have, rather than the ones we wish we had.