Between the Devil and the Deep Blue Sea: Enhancing Flexibility in International Climate Change Law

An important way to adjust a treaty regime to exogenous developments over time is by providing for the review of the original treaties and including possibilities for amending them when deemed necessary.

Both climate treaties include various provisions for review. In a general review clause, the UNFCCC instructs the COP to ‘[p]eriodically examine the obligations of the Parties and the institutional arrangements under the Convention, in the light of the objective of the Convention, the experience gained in its implementation and the evolution of scientific and technological knowledge’.³³ It further provides for a review of the adequacy of some of its core commitments by its first session, drawing on ‘best available scientific information and assessment of climate change and its impacts, as well as relevant technical, social and economic information’.³⁴ This review led to the ‘Berlin Mandate’ at the first COP in 1995, which in turn launched the negotiations for the Kyoto Protocol.³⁵ While the provision could thus be said to underscore the dynamic character of the climate regime, a second review never followed. Finally, the Convention provides for a review of its Annexes by the end of 1998.³⁶ While this provision was hardly used,³⁷ its framing offers a potential avenue to reflect changes in socio-economic conditions over time.³⁸

The Kyoto Protocol also offers several opportunities for reviewing the adequacy of its commitments. It contains a general clause providing for a periodical review by the COP similar to that of the UNFCCC.³⁹ More specifically, for Annex B Parties (i.e., the developed countries that have signed up to specific emission limitation or reduction targets) Article 3(9) Kyoto Protocol suggests that the consideration of future commitments needs to start seven years before the end of the first commitment period (2008–2012).⁴⁰ The Protocol also provides for a separate periodical review in light of the latest scientific information. Unlike the one-off Article 3(9) Kyoto Protocol review, however, this review covers the commitments of all Parties (including developing countries), and is not limited to the mitigation-oriented commitments.⁴¹ Although two reviews were held in this context, these reviews were not to ‘lead to new commitments for any Party’.⁴² Moreover, the second review ended without a clear outcome or a plan for follow-up in 2008.⁴³

Amendments to the climate treaties may be adopted without any limitations as to their substance.⁴⁴ Both the UNFCCC and Kyoto Protocol indicate that amendments should be agreed upon by consensus or, if this turns out to be infeasible, a three-fourth majority.⁴⁵ While the majority voting procedure in theory enhances flexibility, as it could allow for overruling obstructionist states, in practice ‘international regimes are heavily dependent on voluntary action on the part of member states when it comes to matters of implementation’, meaning that ‘it is important to avoid alienating key members via the use of decision-making procedures that seem coercive’.⁴⁶ Nevertheless, as Farhana Yamin and Joanna Depledge point out, ‘[a]lthough voting rarely takes place in [multilateral environmental agreements], the possibility of voting ensures that no Party, or small group of Parties can veto evolution of the regime in the face of widespread agreement’.⁴⁷ Should amendments be adopted, they are binding only to those parties that have accepted them, creating a potentially complex network of legal relations between Parties.⁴⁸

Another way to adapt is to make changes in the Annexes to the climate agreements, and hence change the commitments only for certain Parties. For the UNFCCC, the adoption of changes to the Annexes follows the same procedure as an amendment.⁴⁹ However, as the entry into force requirements for changing the Annexes are more relaxed, such changes are ‘a more expedient way of incorporating material of an essentially factual nature into the regime than is possible through the use of amendments’.⁵⁰ Changing the Annexes to the Kyoto Protocol is more difficult, given the fact that they comprise the core commitments of individual Parties, and define the scope of the agreement by listing the greenhouse gases covered. More specifically, the Protocol states that ‘any amendment to Annex B shall be adopted only with the written consent of the Party concerned’.⁵¹

Up to 2009, proposed amendments to the climate treaties were few and far between. Proposed amendments by Turkey and Kazakhstan to change the Annexes to the UNFCCC met with resistance.⁵² Belarus’ proposal to amend Annex B of the Kyoto Protocol to allow it to assume emission reduction obligations was adopted in 2006, but is unlikely to enter into force due to a lack of ratifications.⁵³ The number of proposals increased sharply in 2009. The reasons for this related to the heightened expectations that at the Copenhagen climate summit at the end of 2009 a new international agreement would be adopted, combined with the provision in the climate treaties that an amendment needs to be sent to Parties at least six months in advance of a meeting before it can be adopted.⁵⁴ Whereas several developing countries sought to act upon the mandate under Article 3(9) Kyoto Protocol by proposing amendments including specific new commitments for Annex B Parties,⁵⁵ some developed countries instead tabled amendments putting forward an entirely different legal architecture from that of the Protocol.⁵⁶ Parties in Copenhagen, as has been extensively discussed elsewhere,⁵⁷ did neither adopt any of these amendments nor any other legally binding agreement. Three years later, however, Parties in Doha ultimately adopted an amendment—yet to enter into force—in response to Article 3(9) Kyoto Protocol, inscribing new targets for a second commitment period (from 2013 to 2020) for the European Union, its Member States and a few other countries.⁵⁸ The long and winded inception period of this amendment can be considered a reflection of the challenges faced when Parties to the climate treaties need to make substantive changes that touch the core of the climate regime—i.e., who needs to do what by when.

Provisions on entry into force (of both treaties and amendments), as well as reservations and withdrawals by individual Parties offer another opportunity to enhance flexibility in the design of a legal regime. Such provisions set important parameters for Parties’ legal obligations by specifying when such obligations begin to apply, whether any unilateral opt-outs are allowed and whether and how a Party can cease to be bound.

The entry into force requirements of the UNFCCC were rather lenient,⁵⁹ ensuring a rapid entry into force of the framework treaty less than two years after its adoption. The Kyoto Protocol introduced more challenging hurdles. Not only did it require at least 55 Parties to ratify, it also stipulated that these Parties should include Annex I Parties responsible for at least 55 percent of the total carbon dioxide emissions in 1990. In terms of flexibility, this provision created a ‘handicap’.⁶⁰ While the provision sought to avoid a situation where the fate of the Protocol’s entry into force depended entirely on the United States,⁶¹ it led to several compromises on the part of the European Union to ensure the participation of other Annex I Parties such as Japan, Canada and, notably, the Russian Federation. These compromises are primarily reflected in rules on the use of carbon sinks (see further Sect. 10.2.6), and while they arguably enhanced the regime’s flexibility (in terms of methods for accounting for emissions as well as implementing commitments), they also jeopardized the environmental integrity of the Kyoto targets by creating new loopholes.⁶²

Like most other multilateral environmental agreements, the climate treaties state clearly that no reservations are allowed.⁶³ While this arguably limits the flexibility of the treaties, it also prevents the emergence of a complex web of bilateral legal relationships, and ensures that the fragile compromises reached in the treaties are not unravelled at a later stage.⁶⁴ Flexibility, in short, has to be built into the treaty itself.

The provisions on withdrawal in both climate treaties arguably enhance flexibility, as they allow for a Party to withdraw provided a prior notification has been submitted.⁶⁵ Withdrawal clauses can be seen as a form of ‘insurance’, where Parties seek to minimise the risks of potentially high compliance costs of a treaty by allowing for unilateral treaty exit.⁶⁶ Yet, the consequence of having a withdrawal provision (specifically, one that does not pose any high barriers) is that it may in fact be used, with potentially detrimental consequences for future international cooperation. While it was perhaps expected that the need to keep up a good international reputation would deter countries from exiting, Canada notoriously made use of the provision in 2011, when it submitted its notification of withdrawal.⁶⁷ Canada was well off track to meet its Kyoto targets, but rather than being found to be in non-compliance, it chose to exit. The risks of such a precedent are highlighted by Helfer, who points out ‘that states will invoke exit clauses opportunistically (or credibly threaten to do so) whenever economic, political, or other pressures make compliance costly or inconvenient’ and that ‘states that would prefer to cooperate but fear that their treaty partners may quit the agreement have less incentive to invest in treaty compliance’.⁶⁸ Although neither scenario described by Helfer has materialized yet in the climate regime, there is a possibility that they will in the future, particularly if states cannot avail of other flexibility mechanisms.

As an elaboration of the principle of ‘common but differentiated responsibilities and respective capabilities’,⁶⁹ the UNFCCC divides its Parties into two major blocks, emerging from the initial classification of Parties into Annex I (including a richer Annex II) and the default category of non-Annex I countries. The same approach is reinforced in the Kyoto Protocol, which focuses on Annex B and non-Annex B countries. Both treaties contain commitments applying to all countries, but add specific commitments for Annex I/B countries. Notably, non-Annex B countries do not have any emission reduction commitments under the Kyoto Protocol.

While the division into two Annexes arguably simplifies the negotiation process, it can be criticized for not taking into account changing circumstances in countries. Developed countries, in particular, have argued for abandoning the distinctions introduced by the Annexes. Such calls have met with fierce resistance from several developing countries, particularly large emerging economies such as China and India, which have insisted that the ‘firewall’ of the Annexes needs to be maintained. Therefore, since ‘the addition and deletion of names of countries from Annex I is closely related to the broader issues raised by the evolution of mitigation commitments of non-Annex I Parties’,⁷⁰ any proposed change to the Annexes has been a sensitive issue.

The diverging views between developed and developing countries came sharply into focus during the negotiations on a follow-up agreement to the Kyoto Protocol in the mid-2000s. A first departure from the ‘firewall’ approach to differentiation was agreed in the 2007 Bali Action Plan, which called for ‘nationally appropriate mitigation commitments or actions’ by developed country Parties (rather than referring to Annex I/B countries) and ‘nationally appropriate mitigation actions’ for developing country Parties (rather than referring to non-Annex I/B countries).⁷¹ This trend continued in subsequent COPs in Copenhagen, Cancún and Durban. Notably, the decision launching the Durban Platform on a future climate agreement specifies that such an agreement will be ‘applicable to all Parties’.⁷²

More importantly, however, further differentiation is already quite common in the climate regime.⁷³ The UNFCCC refers, for instance, to the ‘special situations of the least developed countries’;⁷⁴ to ‘economies that are vulnerable to the adverse effects of the implementation of measures to respond to climate change’;⁷⁵ to countries ‘undergoing the process of transition to a market economy’;⁷⁶ and to the special needs and circumstances of a range of different countries, including ‘small island countries’, ‘countries with low-lying coastal areas’, fossil-fuel producing countries, etc.⁷⁷ Differentiation can also be observed in specific issue areas. With respect to the CDM, initiatives have sought to promote the ‘equitable distribution’ of CDM projects, for instance through capacity-building initiatives in sub-Saharan Africa⁷⁸ and a loan scheme for countries with less than 10 projects.⁷⁹ Likewise, over the years a differentiated approach to climate change finance has been followed. The Cancún Agreements, for example, stipulate that ‘funding for adaptation will be prioritized for the most vulnerable developing countries, such as the least developed countries, small island developing States and Africa.’⁸⁰

The principle of common but differentiated responsibilities and respective capabilities provides a legal argument in favour of differentiation between Parties with different contributions to the climate problem and different capacities to reduce emissions. Essentially, the principle offers a basis for treating countries differently if national circumstances vary, and thereby allows for enhancing flexibility in the climate regime.

The extent to which a treaty prescribes a certain conduct is another important determinant for its flexibility. Given its nature as a framework treaty, it is unsurprising that the conduct prescribed in the UNFCCC is rather broadly formulated, with the important exception of provisions on monitoring and reporting. For instance, the most concrete elaboration of an emissions goal is for all Parties, individually and/or jointly, to return to their 1990 emissions level.⁸¹ Moreover, the treaty stipulates that Annex I Parties are to ‘adopt national policies and take corresponding measures on the mitigation of climate change’,⁸² without specifying what types of policies and measures would be desirable.

The Kyoto Protocol is more concrete when it comes to emission reduction targets, which are specified for all Annex B Parties. It also adds detail on the means of implementation by including a list of possible policies and measures, but the text of the provision—‘such as’—makes clear that the list is meant to be non-exhaustive.⁸³ The non-mandatory and non-exhaustive nature of the list of policies and measures means that states are in principle free to decide how to reduce their greenhouse gas emissions, leaving them with much discretion in achieving their targets. For instance, countries can opt to reduce emissions by putting in place renewable energy or energy efficiency policies or by adopting policies to enhance the carbon uptake of sinks (e.g., by increasing forested land). Moreover, Parties are given ‘what’ flexibility with respect to the sources of emissions: the Protocol covers a ‘basket’ of greenhouse gases,⁸⁴ with Parties free to choose which emissions to reduce. This is of significance, as it allows Parties to target emission sources that are not deeply entangled with their economies. The flipside of this approach is that it will inevitably pose challenges in measuring progress, as the equivalence of different types of emission reductions will need to be established before they can be aggregated.

Flexibility in the implementation stage is further enhanced by the introduction of three market-based mechanisms in the Kyoto Protocol. Although some form of flexibility mechanism was already included in the UNFCCC—in the guise of Activities Implemented Jointly⁸⁵—the idea of market mechanisms in international climate policy was only fully embraced through the Kyoto Protocol. Importantly, the Kyoto mechanisms offer ‘where’ flexibility to achieve cost-effective emission reductions by allowing Parties to reduce emissions outside of their jurisdiction.⁸⁶ First, through international emissions trading, Annex I Parties can buy or sell emission units representing 1 metric tonne carbon dioxide equivalent from or to other Annex I Parties.⁸⁷ Second, through the CDM, Annex I Parties may invest in emission reduction projects in non-Annex I countries. The emission reduction credits subsequently generated can be used up to a certain extent to meet the Kyoto targets.⁸⁸ The third flexibility mechanism, Joint Implementation, follows a similar logic, but is aimed specifically at countries undergoing the transition to a market-based economy (i.e., countries in Central and Eastern Europe).⁸⁹ The only limitation to this ‘where’ flexibility comes in the form of a vaguely formulated recital that the use of the Kyoto mechanisms ‘shall be supplemental to domestic action and that domestic action shall thus constitute a significant element of the effort made by each Party’.⁹⁰

Following Kyoto, and with a view to ensuring sufficient participation to allow the Protocol to enter into force (see Sect. 10.2.4), new rules were agreed by Kyoto Parties that provided for even more flexibility, particularly by relaxing rules in the area of carbon sinks.⁹¹ In short, it should be clear that the climate treaties leave Parties considerable flexibility regarding the means of implementation.

The design of a compliance mechanism can directly affect a Party’s costs of compliance, and therefore provides another avenue for either enhancing or limiting flexibility. Moreover, non-compliance itself can be seen as the ultimate unilateral flexibility mechanism:⁹² a Party can allow itself room for manoeuvring by not complying with its commitments (although such a strategy may backfire in cases where non-compliance involves sanctions).

Although Parties to the UNFCCC were invited to consider a ‘multilateral consultative process’,⁹³ no compliance mechanism has been established under the Convention. By contrast, the Kyoto Protocol put in place a compliance mechanism that combined the virtues of the two main schools of thought in compliance theory: through the Enforcement Branch of its Compliance Committee, it allows for penalties in case Kyoto Parties do not meet key obligations (reflecting the ‘enforcement’ school in compliance theory); and through its Facilitative Branch, it assists countries to return to compliance (corresponding to the ‘managerial’ school).⁹⁴ The managerial school, in particular, emphasises how compliance should be viewed as a continuous process requiring ‘compliance management’ rather than one-off findings of non-compliance combined with sanctions,⁹⁵ highlighting the role of compliance mechanisms in promoting flexibility.

Although the flexibility built into the functioning of the Enforcement Branch is rather limited (it is to ‘determine’ non-compliance and ‘apply’ consequences, such as penalties⁹⁶), the room for manoeuvring for the Facilitative Branch is greater.⁹⁷ The mandate of the latter includes

providing advice and facilitation to Parties …, promoting compliance by Parties with their commitments under the Protocol, taking into account the principle of common but differentiated responsibilities and respective capabilities …. It shall also take into account the circumstances pertaining to the questions before it. ⁹⁸

In line with the notion of compliance management, the Facilitative Branch can enter into a dialogue with Parties that are at risk of non-compliance. However, mainly due to the later-than-expected entry into force of the Protocol, the facilitative function of the Compliance Committee has been under-utilized.⁹⁹ Nevertheless, this facilitative role has in part been assumed by the expert review teams, which were established by the Kyoto Protocol to review Parties’ reports and submit ‘questions of implementation’.¹⁰⁰ Evaluations of the functioning of the compliance system underline that Parties’ national circumstances have been taken into account, but point out that this has been done in an inconsistent fashion.¹⁰¹

Adaptive flexibility can be observed, for instance, in the withdrawal provision that allowed Canada to exit from Kyoto in 2011, or in the Protocol’s compliance mechanism which allows for a measured response to different cases of non-compliance. Transformative flexibility is clear from the provisions on review and amendment, as well as the system of delegated law-making through the COP, which allows for a response to changing insights and circumstances without seeking recourse to formal re-negotiation processes. Means flexibility is present in the climate treaties through the provisions allowing significant discretion with respect to mitigation policies and measures, as well as through the Kyoto Protocol’s flexibility mechanisms.

These examples show that the designers of the Convention were well aware of the temporal nature of the subject matter. While the UNFCCC establishes an overarching objective and guiding principles for the climate regime, and the Kyoto Protocol sets clear rules for developed countries to reduce their emissions (thus enhancing legal certainty and predictability), the flexibility mechanisms offer Parties (individually and collectively) significant leeway.

Yet notwithstanding the flexibility built into the design of both climate treaties, the discussion has also clearly shown that there are problems with the existing regime when it comes to adapting to changing circumstances in practice. Despite institutional arrangements for the provision of scientific input into the policy-making process and a reference to precaution in the UNFCCC,¹⁰² it remains unclear whether the UNFCCC process is able to properly reflect changing scientific insights that stress the importance of rapid and deep emission cuts. An interesting test case in the immediate future in this regard will be to examine whether Parties will acknowledge and embrace scientific findings that to stay below 2 °C with over 75 % certainty, it is necessary to limit cumulative carbon dioxide emissions between 2000 and 2050 to 1000 gigatonnes.¹⁰³ Although Parties have established a review process from 2013 to 2015 ‘to periodically review the adequacy of the long-term global goal …, in the light of the ultimate objective of the Convention, and overall progress towards achieving it’,¹⁰⁴ it remains to be seen whether the longer-term objectives will influence short- to mid-term goals and actions. The experience of the climate regime in terms of setting ambitious and concrete goals does not bode well, and it is perhaps too much to expect a translation of scientific insights into specific emission reduction commitments. As a consequence, it will remain difficult to measure progress toward any long-term objective.

Related to this, the current climate regime is arguably not well-aligned with the structure of the climate change problem. As Oran Young explains,

[t]his misfit is, first and foremost, a function of the fact that the relevant biophysical systems are dynamic and prone to changes that are nonlinear, irreversible, and surprising, whereas the regime is sluggish and unable to respond in a nimble fashion to such changes in the climate system.¹⁰⁵

The climate regime is marked by an incremental approach, with each COP making only minimal progress. Variations of the criticism ‘the process has been saved, but not the climate’ underline this incremental approach.¹⁰⁶ This step-by-step approach is related to the challenge of achieving consensus between more than 190 countries,¹⁰⁷ which forms a significant limitation to the flexibility with which the UNFCCC’s decision-making bodies can respond to changing circumstances.

Another limitation to the flexibility of the climate regime is its lack of a dynamic system of commitments. In the face of continuing resistance to change the current system of Annexes (i.e., the ‘firewall’), it will be difficult to adapt the regime to changing circumstances (e.g., a change in greenhouse gas emissions; economic growth or stagnation; increased vulnerability; etc.) in different countries. Although further differentiation among countries is slowly creeping into the regime (as discussed in Sect. 10.2.5), it remains doubtful that abandoning the Annexes system entirely will be possible—at least in the short term.

The discussion so far has highlighted that the adaptability of the climate regime is limited in at least two important respects: the evolution of commitments in accordance with changing national circumstances; and changing the overall ambition in light of new scientific evidence. Critics of the regime have not only pointed to these weaknesses but, in response, have also proposed alternative models that could fare better in terms of flexibility and adaptability. This section will examine one such model in particular, namely a pluralist approach to international climate change law and governance.