¹ The Implementation and Effectiveness of International Environmental Commitments (David G. Victor, Kai Raustiala & Eugene B. Skolnikoff eds., forthcoming 1998).

² Among the issues we examine and some of our findings: (1) We examine how patterns of participation, especially by nonstate actors, have influenced outcomes. Among our findings is that regulated enterprises, such as industry associations, typically provide technological and economic information that is crucial for effective implementation. We explore ways to involve such enterprises while avoiding “regulatory capture”; (2) Several studies have focused on the dismal situation in Russia, as well as ways that the West can help; and (3) At the international level, we have examined how implementation is monitored and implementation problems are handled. The institutions formally charged with the job are typically weak (or nonexistent), but we show that a wider array of actors and institutions perform these tasks. Moreover, contrary to fears of duplication and poor coordination, this decentralized approach—what we term systems for implementation review (SIRs)—actually functions as a well-coordinated system. For more on these topics, and the fourteen empirical case studies, see id.

³ Schachter, Oscar, The Twilight Existence of Nonbinding International Agreements, 71 Am. J. Int’l L. 296 (1977)CrossRefGoogle Scholar; Handl, Günther F. et. al., A Hard Look at Soft Law, 82 ASIL Proc. 371 (1988)Google Scholar; Donoghue, Joan E. et. al., Environmental Law. When Does it Make Sense to Negotiate International Agreements?, 87 Proc. Am. Soc’y Int’l L. 377 (1993)Google Scholar.

⁴ However, Russian compliance with many other regulatory commitments has been high because most pollution has declined with contraction of the economy; compliance has been achieved without implementation. When international commitments require costly actions—such as the programmatic activities to monitor radioactive wastes dumped at sea, mandated under the London Dumping Convention—Russian compliance has been low. For more detail, see also Vladimir Kotov, Elena Nikitina, Alexei Roginko, Olav S. Stokke, David G. Victor & Ronnie Hjorth, The Implementation of International Environmental Commitments in Countries (mainly Russia) Undergoing Economic and Social Transition (MOCT-MOST; Economic Policy in Transition Countries 7, no. 2, forthcoming 1997).

⁵ Protocol to the 1979 Convention on Long-Range Transboundary Air Pollution on the Reduction of Sulphur Emissions or Their Transboundary Fluxes by at least 30 Per Cent, July 8, 1985, 27 I.L.M. 1442.

⁶ 1988 Protocol to the 1979 Convention on Long-Range Transboundary Air Pollution Concerning the Control of Emissions of Nitrogen Oxides or their Transboundary Fluxes, Oct. 21, 1988, 28 I.L.M. 212 (entered into force Feb. 14, 1991).

⁷ Among the exceptions is the second Sulphur Protocol (1994 Protocol to the 1979 Convention on Long-Range Transboundary Air Pollution on Further Reduction of Sulphur Emissions, June 14, 1994, 33 I.L.M. 1540), in which targets were set mainly on the basis of ecological sensitivity (see next note). That case illustrates the crucial role that data reporting and review can play in improving environmental governance. Targets in the second Sulphur Protocol were computed using models that were developed over two decades as part of an extensive system for gathering, assessing and using data on emissions, transport and deposition of acidifying substances.

⁸ Effectiveness and compliance are the same only when goals are clear and commitments are identical with the changes in behavior that are needed to meet those goals. Examples mainly include agreements that require a ban on some activity, such as the Montreal Protocol’s ban on the consumption of ozone-depleting substances and the London Dumping Convention’s ban on dumping radioactive waste into the oceans. (Another possible example is the “critical loads” concept which has been employed to define the pollution reductions needed to protect European ecosystems from acid rain. However, as yet, that concept has not been fully implemented. The 1994 Sulphur Protocol (supra note 7) uses the method, but emissions targets are set so that only approximately half the reduction needed to achieve sustainable load is achieved.) But those special cases are not the problems of interest here, where goals and commitments are uncertain and commitments must be adjusted as well as behavior. When both behavior and the yardstick move, compliance indicates little. Moreover, I do not suggest that compliance is irrelevant. In some cases, institutions that influence behavior are triggered by compliance problems. The Montreal Protocol’s Non-compliance Procedure is a prominent example. A similar procedure is being established under the 1994 Sulphur Protocol and may have a similar compliance-related impact on behavior.

⁹ In this short essay, I have not addressed the reasons why states adopt voluntary commitments. Rather, my main purpose is to explore what types of commitments are most effective when states want to make commitments but are uncertain what and how to commit. I touch on the role of domestic pressure, which is probably the most important factor in most cases, at the end. In the example used here, Norway adopted the NO* declaration because it was under pressure from other Nordic nations (which were especially vulnerable to acid rain).

¹⁰ The United Kingdom has also been supportive of the North Sea regime because, in practice, its commitments are less onerous. The costly commitments to regulate nutrient pollution apply only to pollution that flows into “sensitive ecological areas.” Britain has declared that its North Sea coastline includes no such areas. In contrast, the Norwegian parliament declared that Norway’s entire coastline met that classification. For more on the case, see J.B. Skjserseth, The Making and Implementation of North Sea Commitments: The Politics of Environmental Participation, in The Implementation and Effectiveness of International Environmental Commitments, supra note 1.

¹¹ These cases also illustrate that even when international commitments appear to be uniform, much differentiation is possible. In the case of the North Sea and the designation of sensitive ecological areas, national actions and decisions can result in differentiation of commitments. In the North Sea and NO, cases, the nonbinding instruments also resulted in differentiation of commitments because they allowed leaders to adopt more stringent commitments. Such differentiation is not exclusive to nonbinding instruments—reservation clauses in binding instruments similarly allow nations to select the stringency of commitments that they will implement. Moreover, all legal instruments allow differentiation by allowing different parties to over- or undercomply. Nonbinding instruments may allow more such unilateral differentiation because they compel all parties, perhaps especially leaders, to do more; however, that hypothesis needs further investigation.

¹² I am mindful that a firm determination of the effectiveness of nonbinding PIC is severely hampered by the lack of relevant data, which is a major failing of the nonbinding PIC system. (The binding system currently under negotiation is poised to make the same mistake.) It underscores the need to build data systems and self-evaluation for nonbinding and binding regimes alike. Moreover, as in all research on complex phenomena, a definitive judgment on the counterfactual—what would have happened if PIC had been binding—is impossible to make. The discussion here focuses on suggestive and likely conclusions, based on detailed analysis of the history and judgment of possible outcomes if one factor (the legal status of PIC) were different.

¹³ Both of these “ifs” are widely assumed but in need of empirical investigation. In the case of enforcement, perhaps binding agreements can be incorporated more readily into domestic law. But comparative studies have not yet been conducted to explore whether nonbinding instruments are less readily incorporated. Moreover, some modes of enforcement—such as by nongovernmental organizations (NGOs)—seem to be as valid for nonbinding agreements as for binding ones. Indeed, many “enforcement” actions—such as by Greenpeace against commercial whaling or against dumping in the North Sea—do not concern formal violations of a binding treaty but rather an informal (nonbinding) norm.

¹⁴ Some weaken review by not providing data (or by providing inaccurate data), as in the case of the Soviet bloc regimes in many instances; some weaken mechanisms by weakening their design, as in the case of developing-country participation in the design of the Montreal Protocol Non-Compliance Procedure.

¹⁵ Victor, David G. & Salt, Julian, Keeping the Climate Treaty Relevant, 373 Nature 280-82 (1995)CrossRefGoogle Scholar.

¹⁶ Some means of regulating greenhouse gases may require highly quantified, binding targets. For example, a system of tradeable permits—envisioned, for example, in the current United States proposal for strengthening the FCCC—probably requires binding targets because, absent such hard targets, the value of permits will be uncertain; experience at the domestic level with permit systems shows that high certainty and security of permits is vital to ensuring that an active market emerges. Without an active market the benefits of such a market-based approach— namely, cost reduction and flexibility—will not be forthcoming. A less ambitious form of trading—known as “joint implementation” (JI)—probably could work without hard targets. At this early stage in cooperation, with high uncertainty about goals and means, a full-blown tradeable-permits scheme strikes me as premature, especially if it perpetually narrows the range of legal instruments that might be used to address global warming in the future.

¹⁷ An example is the Montreal Protocol, in regard to which there has hardly been zero uncertainty, but compared with the problem of global warming, clarity in goals and means has been much higher. Because ozone-depleting substances could be substituted in most cases, most governments were certain that they could implement the international commitments. (That was not true at the outset, but commitments were also more modest in the early stages.) Applications where substitution is impossible are exempted under the Montreal Protocol’s “essential uses” provision. An illustration of the crucial role of implementation review regardless of the legal status of an agreement, the essential uses exemption is extensively and regularly reviewed on a case-by-case basis.