Policy Brief - Harvard Project on Climate Agreements, Belfer Center

Whither the Kyoto Protocol? Durban and Beyond

  • Daniel Bodansky
| August 2011

W[h]ither the Kyoto Protocol? Durban and Beyond1

On December 31, 2012, the Kyoto Protocol's first commitment period will expire. Unless states agree to a second commitment period, requiring a further round of emissions cuts, the Protocol will no longer impose any quantitative limits on states' greenhouse gas emissions. Although, as a legal matter, the Protocol will continue in force, it will be a largely empty shell, doing little if anything to curb global warming.

Ever since the Kyoto Protocol's entry into force in 2005, the question of what to do after 2012, when Kyoto's first commitment runs out, has been a central focus of the U.N. climate change negotiations. Developing countries such as China and India want the Protocol to continue in its present form, imposing quantitative limits on developed country emissions but not their own. The European Union might be amenable to a new commitment period under the Protocol, but only as part of "a global and comprehensive framework engaging all major economies,"2 including the United States and China. Meanwhile, some Kyoto parties, such as Japan, Canada, and Russia, want to replace the Kyoto Protocol with a comprehensive new agreement with commitments by both developed and developing countries.

In 2005, the parties to the Kyoto Protocol established an ad hoc working group to negotiate further commitments under the Protocol for the post-2012 period. But since none of the countries with Kyoto targets are willing to proceed absent parallel action under the Convention to address the emissions of the other major economies, the working group has made only limited progress to date.3 In 2007, at the Bali conference, the parties to the U.N. Framework Convention on Climate Change established a parallel negotiating process, involving the other big emitters such as the United States and China, to consider long-term cooperative action under the Convention, with the goal of reaching a comprehensive outcome addressing mitigation, adaptation, finance, and technology.4

With little more than a year to go before the end of Kyoto's first commitment period, it appears likely that there will be a gap — of indefinite duration — before the establishment of any new legal commitments limiting greenhouse gas emissions. Even if the Kyoto Protocol parties agreed at the Durban conference this year to an amendment establishing a second commitment period — an extremely improbable outcome — there is virtually no chance that sufficient countries would ratify the amendment in time for it to enter into force before the end of 2012.

This discussion paper analyzes the options going forward for the Kyoto Protocol, including adoption of a legally-binding second commitment period, a "political" second commitment period, or no new commitment period. It also considers the legal implications of a gap between the end of Kyoto's first commitment period and the adoption of a new legal regime to limit emissions, the prospects for the Clean Development Mechanism in the absence of a second Kyoto commitment period, and the relationship between the Kyoto Protocol negotiations and the emerging regime under the Cancun Agreements.


The development of the U.N. climate change regime has followed a pattern familiar in international environmental lawmaking. First a framework convention is adopted, establishing the basic system of governance for a given issue area. Then, regulatory requirements are negotiated in a protocol to the convention.5 The ozone regime followed this pattern, starting with the adoption of the Vienna Convention for the Protection of the Ozone Layer in 1985, and continuing with the adoption of the Montreal Protocol on Substances that Deplete the Ozone Layer in 1987. Similarly, the climate change regime began with the negotiation of the 1992 U.N. Framework Convention on Climate Change, followed five years later by the Kyoto Protocol, which elaborates specific regulatory requirements to limit greenhouse gas emissions.

Three features of the Kyoto Protocol are noteworthy:

First, the Protocol sharply differentiates between Annex I and non–Annex I parties (roughly translatable as "developed" and "developing" countries respectively).6 The UNFCCC established the principle of "common but differentiated responsibilities and respective capabilities" (CBDR),7 but did not draw an absolute separation between developed and developing countries. It elaborated general obligations common to all parties, additional commitments relating to reporting and financial assistance for Annex I and Annex II8 parties respectively, a "degree of flexibility" for countries with economies in transition (i.e., the former Soviet bloc), special consideration for least developed countries, and a procedure by which the classification of countries could be reconsidered as circumstances change.9 In contrast, the negotiating mandate for the Kyoto Protocol categorically excluded any new commitments for non–Annex I countries, operationalizing a comparatively flexible principle in an extremely rigid way.10

Second, because the Kyoto Protocol negotiations focused exclusively on developed country emissions reductions, the primary axis in the negotiations was between the two main developed country powers, the United States and the European Union, in the case of the United States with support from Japan, Australia, and other members of the so-called "Umbrella Group." Third, the Kyoto Protocol's regulatory approach was modeled on the Montreal Protocol on Substances that Deplete the Ozone Layer, adopted ten years earlier. Like the Montreal Protocol, the Kyoto Protocol establishes legally-binding commitments, consisting of quantitative national performance standards, defined through a process of "top-down" international negotiations.11 In contrast, the UNFCCC had elaborated a bottom-up process requiring countries to develop and report on nationally-defined policies and measures to mitigate climate change.12

Negotiations on a Post-2012 Climate Change Regime

Unlike the Kyoto Protocol negotiations, which focused exclusively on developed country emissions, the ongoing negotiations on a post-2012 climate change regime have also addressed developing country mitigation actions, without which a solution to the climate change problem is impossible. 13 This has made the current negotiations as much between developed and developing countries as between the U.S. and the European Union. Key issues include:

  • Legal Form: Will the post-2012 regime be established through a legally-binding agreement (or agreements), such as an amendment to the Kyoto Protocol, a new legal agreement defining mitigation commitments for states that do not have Kyoto targets (such as the United States and China), or a comprehensive agreement that embraces all states and replaces the Kyoto Protocol? Or will the post-2012 regime be defined through a political agreement or decisions of the parties?14 And will the post-2012 regime consist of separate outcomes under the Kyoto Protocol and the Convention, or will the two negotiating tracks merge into a single outcome?
  • Regulatory approach: Will the post-2012 regime continue the top-down approach of the Kyoto Protocol, in which internationally-defined commitments are adopted in order to drive national action? Or will the regime switch to a more bottom-up approach, in which countries unilaterally define their own national climate change approach, or adopt some kind of hybrid approach?
  • Differentiation: Will the post-2012 climate change regime continue to draw an absolute wall between developed and developing countries, as the Kyoto Protocol does? Or will it provide for greater parallelism or symmetry between developed and developing countries — for example, by imposing legally-binding commitments on both, or by adopting common rules on accounting, mechanisms, reporting, review and/or compliance?15

Positions vary widely on these issues. The European Union is open to considering a new round of legally-binding Kyoto targets, but only as part of a global and comprehensive framework that includes the United States and China. The United States would be willing to negotiate a legally-binding agreement, but only if the mandate provided that the agreement would apply with equal legal force to all of the major emitters (including China and India). Although it accepts that developing country commitments should be differentiated from those of developed countries as to content, it insists on symmetry of legal form, meaning that the provisions for major-emitting developing and developed countries should have the same legal character. Meanwhile, the big developing countries such as China and India would like developed countries to continue Kyoto's top-down, legally-binding approach, but are unwilling to accept this approach themselves. They insist on maintaining the Kyoto "firewall" between developed countries (which have emissions limitation commitments), and developing countries (which don't).

The negotiations on a post-2012 climate change regime were initially supposed to wrap up at the 2009 Copenhagen Conference. But the Copenhagen Accord — a political agreement establishing a bottom-up process based on national pledges — was not formally adopted by the conference.16 And although the following year, the Cancun conference adopted decisions that elaborate the Copenhagen framework and anchor the Copenhagen pledges in the Convention, it extended the negotiating process and left open the final legal form of the regime, including the possibility of a second commitment period under the Kyoto Protocol.17 So the battle over policy architectures will continue at this year's conference of the parties in Durban, South Africa, and most likely at the 2012 climate conference scheduled in Qatar.

Scenarios for Durban and Beyond

Although there are innumerable possible outcomes in Durban and beyond, this section focuses on three scenarios, to illustrate the range of options. One scenario represents the minimal outcome: no agreement in Durban on a Kyoto Protocol second commitment period, minimal progress in elaborating the Cancun Agreements, and no decisions about the longer-term direction of the regime. At the opposite extreme, a second scenario represents a politically ambitious outcome, in which the Kyoto Protocol parties agree to a second commitment period and the Convention parties agree to a mandate to negotiate a legally-binding agreement that addresses the emissions of countries without Kyoto targets. Finally, a third scenario represents an intermediate outcome, establishing a transitional regime aimed at the development of a legally-binding agreement (or agreements).

Scenario 1: No Agreement on a Second Commitment Period

The business-as-usual scenario — the most likely scenario for Durban and beyond — is that the current negotiating dynamic will continue and nothing will be agreed about a second commitment period by the end of 2012, when Kyoto's first commitment period expires. The result will be a period of uncertain duration during which the U.N. climate regime will not impose any legally-binding quantitative limits on states’ greenhouse gas emissions. Instead, the only limits that would continue to apply would be the political commitments that states made in their Copenhagen/Cancun pledges.

Why is this scenario likely? The biggest single reason is that even those Kyoto Protocol parties that would, in principle, be willing to accept a second commitment period are reluctant (and possibly unwilling) to do so on their own, without reciprocal commitments by other states. As discussed earlier, the European Union is willing to consider a second commitment period only as part of a comprehensive framework including the United States and China. The United States is unwilling to accept a new legal agreement unless it includes new commitments of the same legal character by all of the world's major economies (although these could be very different in terms of stringency and content). And China seems unwilling to accept any legal commitment to limit its emissions, no matter how differentiated. The gridlock can be relieved only if one or more parties back out of their current positions.

What would be the consequence of the first commitment period expiring with no successor regime in place? Although Kyoto's emissions targets are time-limited, the Protocol as a whole is not, so the agreement would continue in force. But what would this mean if the Protocol did not impose any limits on states' greenhouse gas emissions?18

Obviously, the absence of emissions targets would deprive many Protocol provisions of any effect. Most importantly, without emissions targets, there would be no emission allowances or "assigned amount units" (apart from those units carried over from the first commitment period), so all of the many provisions relating to assigned amount would be largely without effect — for example, the system of national registries to track each state's assigned amount, or the rules for crediting of land use change and forestry activities.

But other provisions of the Protocol are not dependent on emissions targets, and these would continue to operate. For example, Article 5 requires developed countries to have "national systems" for the estimation of their greenhouse gas emissions.19 Similarly, Article 7 requires developed countries to include in their national communications whatever supplementary information is necessary to demonstrate compliance with the Protocol. 20 Although these requirements were included to promote compliance with the Protocol's emissions targets, they are not dependent on the existence of targets and would hence continue in effect even after Kyoto's first commitment period targets expire.21 The same is true of the general obligations set forth in Article 10 and the financial obligations for Annex II parties set forth in Article 11.

Similarly, the Protocol's institutions — the Meeting of the Parties (Article 13), the secretariat (Article 14), the Subsidiary Body for Scientific and Technological Advice, the Subsidiary Body for Implementation (Article 15), the Adaptation Fund22, and the Compliance Committee (Article 18)23 — are not tied to the existence of commitment periods under the Kyoto Protocol and do not depend on national emissions targets. The same is true of the expert reviews of national inventories provided for by Article 8, as well as the periodic reviews of the Protocol by the parties pursuant to Article 9. Although one of the main purposes of expert reviews is to determine compliance with a party's Annex B target, they serve an important function in verifying a country's emissions inventory, even in the absence of targets.

Perhaps most importantly, the CDM would continue to operate after the expiration of the first commitment period, since it too is not dependent on emissions targets. In establishing the CDM, Article 12 identifies as one of its purposes to "assist [developed] countries in achieving compliance with their quantified emission limitation and reduction commitments under article 3."24 But Article 12 does not define this as the CDM's only purpose. Indeed, Article 12 begins not with the CDM's role in complying with emissions targets, but with its role in "assist[ing] developing countries in achieving sustainable development and contributing to the ultimate objective of the Convention."25 Whether or not states agree to a second commitment period, there is nothing in either the Protocol or the CMP decisions implementing Article 12 that would prevent the CDM's infrastructure, modalities, and procedures from continuing.26 The Executive Board could still register projects. Operating entities could still verify and certify emission reductions. The resulting certified emission reduction units (CERs) could still be deposited in the country's registry. And a "share of the proceeds" could still be used to finance the Adaptation Fund.27

Of course, one of the main reasons that states undertake CDM projects is to generate CERs, which they can use to comply with their first commitment period emission targets. So it is unclear how many CDM projects would be funded if Annex I countries no longer had any targets. But to the extent that a state still wanted to pursue a CDM project — for example, as a means of achieving its Copenhagen/Cancun pledge to limit emissions, or because CDM credits were recognized under a national or regional trading system — there is nothing in the Protocol that would preclude it from doing so.28

Some adjustments of the CDM's modalities and procedures would obviously be necessary if there were no second commitment period. Since the existing rules for afforestation and reforestation projects explicitly apply only to the first commitment period, 29 new rules would be needed. New rules would also be needed if CERs were used for compliance with regional and national emissions reduction targets (including those listed under the Copenhagen Accord), to ensure that the credits were not re-used. Under the current system, the Kyoto registry tracks CERs and ensures that they cannot be used more than once for compliance purposes. When a country uses a CER to meet its Kyoto target, the CER is cancelled in the state's registry and cannot be used again.30 To the extent that CERs were used to comply with a regional or national emissions target, they would similarly need to be cancelled, presumably at the time that the CERs were transferred out of the Kyoto system and entered a regional or national trading system.

If there were an extended period without Kyoto Protocol emissions targets, this would give added importance to progress under the Copenhagen/Cancun framework, which establishes a more incremental, bottom-up process, involving political commitments to reduce emissions, significant new financial assistance for developing countries, and a process for international analysis and/or review of national actions. A crucial question is whether failure to reach agreement on a Kyoto second commitment period would provoke a backlash by developing countries, stymying efforts to further elaborate the Copenhagen/Cancun framework. Assuming this framework continued to evolve, however, the infrastructure established by the Protocol might be adapted to play a role under it — for example, by supplying emission reduction credits to countries that have made quantitative emissions reduction commitments under the Cancun Agreements.

Scenario 2: Adoption of a Kyoto Protocol Amendment Establishing a Second Commitment Period

At the other end of the spectrum from the no-agreement scenario, the Kyoto Protocol parties could agree to an amendment establishing a second commitment period prior to the expiration of the current period. Given Japan, Russia, and Canada's stated opposition to a second commitment period, such an agreement would presumably include only a rump of the original Kyoto Protocol parties. Within this small group, the European Union would play the decisive role. Agreement on a second commitment period would be possible only if the EU were determined to demonstrate that the Kyoto Protocol is not only alive but well.

Adoption of a second commitment period amendment faces huge political obstacles. Even the EU would find it politically difficult to accept a second commitment period amendment unless there were parallel progress in the Convention track of the negotiations — for example, agreement to a mandate to negotiate a new legal agreement that included the United States, China, and the other big emitters without Kyoto targets. For the reasons discussed earlier, agreement to such a mandate is highly unlikely.

The countries assuming new Kyoto targets might also insist that the Kyoto amendment include a linked entry-into-force requirement, providing that it not enter into force until the simultaneous entry into force of the parallel agreement to be negotiated under the Convention. This would help ensure mutuality of legal obligation between the countries with Kyoto targets and the other major emitters. But it would be fiercely resisted by developing countries, as they have already made clear.

Even assuming a Kyoto amendment could be adopted without a linked entry-into-force requirement, a second commitment period amendment almost certainly would not enter into force quickly enough to prevent a gap between the first and second commitment periods. Amendments to the Kyoto Protocol require acceptance by three-quarter of the Protocol parties, and enter into force 90 days later.31 So to prevent a gap between the first and second commitment periods, an amendment would need to be accepted by 143 countries32 by October 3, 2012 — a virtual impossibility, given the often lengthy time required for domestic ratification.

To address this issue, a Kyoto Protocol amendment establishing a second commitment period could provide for the amendment's "provisional application" pending entry into force. Provisional application is a recognized technique in treaty law by which states undertake to apply a treaty pending its entry into force.33 Provisional application has been used most frequently in arms control agreements, but the technique has been used in other areas as well, most notably in the trade arena, where it was used to bring the original General Agreement on Tariffs and Trade (GATT) into effect. In the environmental arena, a number of treaties have been provisionally applied pending entry into force, including the 1964 European Fisheries Convention, the 1979 Long-Range Transboundary Air Pollution Convention, the 1991 Protocol on Environmental Protection to the Antarctic Treaty, and the 1998 Agreement on the International Dolphin Conservation Program.34 One of the accepted purposes of provisional application is to prevent legal gaps between successive treaty regimes. For example, the 1994 agreement modifying Part XI of the 1982 U.N. Convention on the Law of the Sea (UNCLOS) was provisionally applied so that it would be in effect when UNCLOS entered into force in 1994.35

Although provisional application raises conceptual puzzles (how can a treaty provision providing for provisional application have legal effects if the treaty as a whole is not yet in force?), the prevailing view is that an agreement by states to provisionally apply a treaty creates legal obligations that are largely the same as if the treaty entered into force. For example, the International Law Commission concluded in its commentary on the Vienna Convention on the Law of Treaties that "there can be no doubt that ... clauses [providing for provisional application] have legal effect and bring the treaty into force on a provisional basis."36

Important issues that would need to be resolved in adopting an amendment to the Kyoto Protocol establishing a second commitment period would include:

  • Which countries would agree to new Kyoto targets, and which would prefer to proceed under the Copenhagen/Cancun framework?
  • What modifications, if any, would be made to the Kyoto rules — for example, concerning accounting, land-use change and forestry credits, the Clean Development Mechanism, reporting, review and compliance?
  • What kinds of linkages, if any, would be developed between the Kyoto Protocol and the new legal agreement that would be negotiated under the Convention? For example, could countries with emissions reduction targets trade across the two agreements? Similarly, could countries in the Convention track make use of CDM credits?

Scenario 3: Political Agreement on a Second Commitment Period

An intermediate outcome between no agreement on the one hand and adoption of a Kyoto Protocol amendment on the other would be a transitional regime, establishing a "political" second commitment period — that is, a commitment period in which the emission targets were political commitments rather than legally-binding obligations. 37 This scenario has received considerable attention recently in the run-up to the Durban conference as a more politically realistic option than a legally-binding second commitment period. Because the second commitment period targets would be political rather than legal commitments, they could be adopted through a decision of the parties or a political declaration, rather than requiring a protocol amendment. And although agreement even to a "political" second commitment period would depend on parallel progress in the Convention track of the negotiations — for example, agreement in the longer term to negotiate a comprehensive legal agreement — less progress would be needed to make Scenario 3 politically viable than Scenario 2.

What would be the difference between a "political" second commitment period set forth in a decision of the parties and a "legal" second commitment period adopted in a Protocol amendment? In certain respects, not much. As critics of international law are fond of noting, even legal obligations are generally not "enforceable," given the absence of sanctions in international law.38 Despite this shortcoming, treaties are usually seen as more credible and effective than non-legal agreements because they require greater domestic buy-in (particularly in countries with special requirements for treaty ratification), signal a greater intensity and seriousness of intent, create a stronger internal sense of obligation, and involve higher reputational costs for violation.39 "For these reasons, states typically take the negotiation of a legally-binding agreement more seriously than a non-binding instrument, and the outcomes therefore better reflect what states are, in fact, prepared to do."40

The relative effectiveness of "political" versus "legal" commitments, however, varies depending on the circumstances. Sometimes, states take political agreements seriously, or legal agreements lightly. On the one hand, the 1975 Helsinki Accords and the U.N. General Assembly resolution on high seas driftnet fishing were extremely influential, even though not legally-binding. On the other, the Kyoto Protocol has had little if any influence in curbing the emissions of some parties, such as Canada. Potentially, an agreement by the Kyoto parties to a political second commitment period would have comparable gravity and visibility as a treaty amendment. And, whatever differences might exist between a political and legal commitment period, they would be even less if a legal second commitment period were being applied only provisionally, as is likely to be true for an extended period of time. Finally, to the extent that states do see political commitments as leaving them greater flexibility (and hence less risky) than legal obligations, they may be willing to accept more environmentally-ambitious commitments under a political than a legal second commitment period.

In establishing a political second commitment period, a crucial issue would be the extent to which it would continue along the same lines as the first commitment period. At one extreme, a political second commitment period could extend the Kyoto Protocol essentially unchanged, with the same types of targets and the same rules for accounting, mechanisms, reporting, review and compliance. The targets would not be legally-binding, but everything else would look essentially the same. Of course, the Kyoto Protocol rules wouldn’t apply directly, because the targets would not be adopted through an amendment to Annex B. But the parties could adopt decisions that apply, mutatis mutandis, the Kyoto/Marrakesh rules to the new commitment period. The targets would be specified as a single, fixed number and would apply on an economy-wide basis; they would generate assigned amount units (AAUs); the AAUs would be deposited in a state's registry and could be traded; sinks would be accounted for as they are under the Marrakesh rules; states would be able to undertake CDM projects that generate credits that states could use to satisfy their targets; states would submit inventories, which would be reviewed by expert review teams; and questions about compliance would be addressed by the Compliance Committee. Since the targets would be political, failure to meet them would not represent a legal violation and would not have legal consequences. But, then again, under the existing Kyoto regime, states have never adopted an amendment to make the decisions of the Compliance Committee legally-binding,41 so the difference might not be so great.

Alternatively, the Kyoto Protocol parties could decide to establish a less ambitious second commitment period, which incorporated elements of the Copenhagen/Cancun approach. For example:

  • The second commitment period targets could be defined through unilateral pledges, as in the Copenhagen/ Cancun process, rather than through top-down international negotiations.
  • The targets could be conditional or be specified as a range, as are many of the Copenhagen pledges, rather than reduced to a single, fixed number, like the existing Kyoto Protocol targets.
  • As a result, the targets would not be able to generate assigned amount units, which could be deposited in a registry and traded. Instead, trading would have to be done on an ad hoc basis, through bilateral arrangements between countries that mutually recognize each other's allowances.
  • The rules on accounting, sinks, and MRV could be changed to reflect the political rather than legal character of the regime. For example, states could apply their own rules on accounting, sinks and project-based credits, rather than apply internationally-defined rules.

As in Scenario 2, Scenario 3 would raise the issue: Which countries would assume second commitment period targets and which would prefer to proceed under Copenhagen/Cancun track? This scenario would also raise questions about linkages between the KP and Convention tracks during the transitional period before the development of a new legal agreement (or agreements). For example, could states with targets under Copenhagen/ Cancun be able to buy CDM credits? More generally, could states with targets be able to trade across the tracks?

Finally, an important issue would concern the longer-term. Scenario 3 is intended as a transitional arrangement, in anticipation of the development of a legally-binding regime. But it leaves open the question: In the longer-term, would the regime consist of two legal agreements or a single comprehensive agreement?


The Kyoto Protocol establishes a very complex and ambitious regime, in architecture if not stringency. The problem is that relatively few states, representing only about a quarter of the world's emissions, have been willing to assume emission targets under Kyoto. And even some of these seem unwilling to continue down the same path, certainly not if others do not join the effort as well. The future of the Protocol thus seems doubtful at best. Even in the most optimistic scenario, a new round of emissions targets couldn't be agreed in time to prevent a legal gap between the first and second commitment periods.

A possible middle ground would be to establish a transitional regime that would be political in nature, but that could evolve over time into a legally-binding regime.42 Under the Convention, the Copenhagen/Cancun process has already begun down this road, starting with a bottom-up process of national pledges, coupled with significant financial assistance and an embryonic process of international "consultation and analysis." A political second commitment period would establish a parallel process under the Kyoto Protocol, thereby keeping it alive so that it can (potentially) fight again another day.

For more information on this publication: Please contact Harvard Project on Climate Agreements
For Academic Citation: Bodansky, Daniel. “Whither the Kyoto Protocol? Durban and Beyond.” Policy Brief, Harvard Project on Climate Agreements, Belfer Center, August 2011.

The Author