A Bottom-Up Compliance Mechanism for the Paris Agreement

In: Chinese Journal of Environmental Law
View More View Less
  • 1 Luo Jia Distinguished Professor, Research Institute of Environmental Law, Wuhan
Full Access

The article discusses the development of the Paris Agreement’s provision on a compliance mechanism contained in Article 15 of the agreement. There is a risk that a compliance mechanism set up as a separate body within the new regime will be duplicative and dysfunctional. Keeping state parties to their Paris Agreement obligations can be achieved, instead, through the elaboration of pre-existing and already well-developed reporting-and-review processes under the United Nations Framework Convention on Climate Change (unfccc). In the past, these have generated sufficient pressure on states to maintain compliance with their unfccc obligations, and are likely to have the same effect for the new treaty. The fact that the Paris Agreement does not impose any onerous new obligations on states is a reason to continue to rely on existing processes. I show that the separate compliance body envisaged by the Paris Agreement has no obvious way to improve on the diffuse ‘compliance mechanism’ currently operating under the unfccc. The most efficient approach to ‘Paris Agreement compliance’, therefore, would be to delay implementation of the Article 15 mechanism and allow current practices, suitably modified, to continue for as long as their performance is satisfactory.


The article discusses the development of the Paris Agreement’s provision on a compliance mechanism contained in Article 15 of the agreement. There is a risk that a compliance mechanism set up as a separate body within the new regime will be duplicative and dysfunctional. Keeping state parties to their Paris Agreement obligations can be achieved, instead, through the elaboration of pre-existing and already well-developed reporting-and-review processes under the United Nations Framework Convention on Climate Change (unfccc). In the past, these have generated sufficient pressure on states to maintain compliance with their unfccc obligations, and are likely to have the same effect for the new treaty. The fact that the Paris Agreement does not impose any onerous new obligations on states is a reason to continue to rely on existing processes. I show that the separate compliance body envisaged by the Paris Agreement has no obvious way to improve on the diffuse ‘compliance mechanism’ currently operating under the unfccc. The most efficient approach to ‘Paris Agreement compliance’, therefore, would be to delay implementation of the Article 15 mechanism and allow current practices, suitably modified, to continue for as long as their performance is satisfactory.

1 A Placeholder for a Compliance Mechanism

Article 15 of the 2015 Paris Agreement makes provision for the creation of a compliance mechanism. According to the Decision of the Conference of the Parties to the unfccc adopting the Paris Agreement (‘cop Decision’), the mechanism is to consist of twelve persons ‘with recognized competence in relevant scientific, technical, socioeconomic or legal fields’.1 The purpose of the mechanism led by this group of twelve is ‘to facilitate implementation of and promote compliance with the provisions of this Agreement’.2 The mechanism is to be ‘facilitative in nature and function in a manner that is transparent, non-adversarial and non-punitive’.3 In addition, it ‘shall pay particular attention to the respective national capabilities and circumstances of Parties’.4

Neither the Paris Agreement nor the cop Decision develop the compliance mechanism of Article 15 beyond what I have just stated. The matter was taken up for discussion at the unfccc’s Conference of the Parties in November 2016, in Marrakech, but the mechanism was not further developed at that cop. A decision developing Article 15 is not expected before the end of 2017 or perhaps the end of 2018. There is still time to reflect on Article 15, before state parties commit to a particular approach.5

Even before the Paris Agreement materialized, scholars had begun to ponder the form of a compliance mechanism for the post-2020 regime. Commentary on this picked up after the Paris cop in December 2015. The presumption has been that a separate compliance mechanism is likely to be necessary and important for the success of the new treaty.6 This stance has not been unanimous, for there are some who have questioned the role of a compliance mechanism in the context of a ‘bottom up’ regime—which is the way the Paris Agreement tends to be characterized.7 Those in the doubting group emphasize the potential for state-on-state ‘peer pressure’ to act as a significant constraining influence on non-compliance. They suggest that if transparency of state action and an opportunity for communal questioning of states about their actions could be maintained, it may be enough to sustain an acceptable level of compliance.8

The doubting group is only half right, in my view. It is right in saying that peer pressure among states can be a significant force in the climate change regime.

However, little is ever said about how this pressure on states actually works. A lack of engagement with the mechanics of the exerted pressure is a weakness in the literature, which this paper attempts to correct. Of greater concern is the doubting group’s suggestion that a (top-down) compliance mechanism may be unsuited to the Paris Agreement because the treaty has a bottom-up design.9 This would be a superficial justification for an otherwise correct conclusion. Granted that there is a pleasant symmetry in having a bottom-up approach to mitigation kept in check by a bottom-up approach to compliance; however, the Paris Agreement is no more bottom-up in its design than the Kyoto Protocol, for whose second commitment period Annex I parties pledged essentially the same nationally determined mitigation targets which they had appended to the unfccc’s Copenhagen Accord. What could be more bottom-up than that? The relevant difference between the Paris Agreement and the Kyoto Protocol, which explains why a bottom-up approach to compliance is adequate for the former but not for the latter, is that the Paris Agreement lacks a feature which the Protocol enjoyed and which enabled the Protocol’s top-down compliance mechanism to at least run on one piston: namely, it had legally binding mitigation targets. Therein lies the critical difference, and it has nothing to do with any divergence between the two treaties’ approaches to the method of generation of mitigation targets.

I will advance two arguments in favour of delaying the operationalization of Article 15. However, before coming to those arguments, in a section that serves as a conceptual backdrop, I will outline why an ‘enforcement’ function for a compliance mechanism under Article 15 is completely excluded. Some readers may need no convincing of this—the article’s wording itself suggests that a compliance mechanism with an enforcement function has been rejected by the parties to the Paris Agreement. Here is the text of Article 15:

  1. A mechanism to facilitate implementation of and promote compliance with the provisions of this Agreement is hereby established.

  2. The mechanism referred to in paragraph 1 of this Article shall consist of a committee that shall be expert-based and facilitative in nature and function in a manner that is transparent, non-adversarial and non-punitive. The committee shall pay particular attention to the respective national capabilities and circumstances of Parties.

  3. The committee shall operate under the modalities and procedures adopted by the Conference of the Parties [to the Paris Agreement].10

We note that the word ‘facilitate’ appears twice in the article, ‘enforcement’ not at all, adversarialism and punitiveness are excluded, and relativism about party circumstances is required.11 So the orientation is quite clear. Nevertheless, my preliminary discussion on why an enforcement function is excluded from the Paris Agreement serves to introduce the Kyoto Protocol’s Enforcement Branch, which in my later argument provides the model for a top-down and centralized—as opposed to bottom-up and diffuse—compliance mechanism.

This article thus has three main parts. First, I develop my background point on enforcement. I note that the Paris Agreement’s binding requirements for states are modest. They are procedural, not substantive, and some of the procedural requirements are collective, not individual.12 It follows almost directly that an Article 15 compliance mechanism could not have an enforcement function; for, to have such a function, it should be possible, as I will explain, to sanction a state for non-compliance with a substantive obligation, or non-compliance with a procedural obligation related to a substantive one. In addition to having no substantive rules to supervise directly or indirectly, an enforcement body under the Paris Agreement would have no sanctions to apply that were not already in use by the regime’s other supervisory elements.

I then proceed to the two arguments against the immediate operationalization of Article 15. I argue that if Article 15 were to be endowed with the alternative to an enforcement function, namely with a facilitative function, which is what most commentators are betting on or urging, it risks becoming duplicative, wasteful, or impotent. The Paris Agreement’s provisions on state accountability, technical expert review (ter), and ‘facilitative, multilateral consideration of progress’ (fmcp), contained in Article 13 of the treaty, will almost certainly maintain and perpetuate the bottom-up, diffuse compliance dynamic that is currently found to operate in both the Kyoto Protocol regime and the Cancun arrangements for the ‘pledging’ period (2013–2020) under the unfccc. This dynamic may be sufficient, I suggest, to supervise the states’ procedural obligations under the Paris Agreement without setting up a separate compliance mechanism under Article 15. In developing this argument, I will present new information on the Kyoto Protocol’s Facilitative Branch and on developments in the Biennial Review process—the latter being the flagship supervisory process for the pledging period.

Having established that a separate Article 15 mechanism with a facilitative function would probably add no value, I discuss the underlying purpose of the Paris Agreement. Here, a second argument emerges against a hurried activation of Article 15. In brief, the new treaty is an agreement to the effect that participation in mitigation efforts is to be universal and is to proceed by ‘ratcheting up’ state mitigation ambition sufficiently to keep global warming below 2°C. This is to happen in five-year rounds, in what amounts to a kind of rolling negotiation. Given the centrality of ratcheting up, the new treaty is best conceptualized as a transitional agreement—an iterative, structured negotiation process—to create the conditions for a ‘final’ agreement on climate change. A final agreement would be characterized by state-level emission-reduction obligations consistent with the 2°C warming limit. Assuming that avoidance of that limit (and not some higher one) remains the climate regime’s objective, no further iteration of the regime would be necessary. The final agreement, which would look rather like the Kyoto Protocol except with mitigation commitments for all parties, would need a compliance mechanism to disincentivize deviation from what, at that point, would amount to substantive, legally binding emission restrictions on states; in other words, it would need a compliance body with an enforcement function. Even if one did not accept my postulation of a final agreement as a logically implied element of the current climate change regime, one might still accept my argument that, given that the Paris Agreement is just an agreement to stay engaged in rounds of mitigation-focused talks and action, a compliance mechanism is the wrong kind of instrument to use to deal with cases of state ‘engagement’ in which the engagement of the state in question seems to be less than wholehearted.

In the course of the article, I will be making a handful of axiomatic assumptions. I have already indicated the first of these, namely that there are two possible objectives or models for a compliance mechanism: either enforcement or facilitation.13 They may co-exist, one alongside the other, as they do in the Kyoto Protocol. A second assumption is that a compliance mechanism applies to individualized state obligations, not collective ones. (I do not seek to tackle here the question of collective obligations, whose existence has only recently begun to be recognized.14) Third, I regard it as a given that a separate compliance mechanism under the Paris Agreement should be effective in itself and not be duplicative of existing (diffuse, compliance-effecting) responsibilities and processes, as such duplication would lead to confusion and waste.15 Lastly, the experience to date with compliance mechanisms in the climate change regime should be taken into account in deciding how to approach the development of Article 15.16

I close this introductory section with a table. I will refer back to its elements in later sections. It compiles the Paris Agreement’s mitigation and related reporting-and-review requirements (the ‘shall’ provisions, mostly) and highlights several key concepts. The fact that the Agreement does not create substantive state obligations17 will be plainly apparent from the table. While I have created two categories of provisions—‘Mitigation’ and ‘Climate Finance’—I regard climate finance as a form of mitigation, for its main purpose is either to achieve mitigation (adaptation has always been, in practice if not in rhetoric, a secondary concern of the regime) or to support reporting on mitigation action.18

table 1

A compilation of the main ‘shall’ provisions on mitigation and finance for mitigation in the Paris Agreement. Three important non-‘shall’ provisions have been included (distinguished in bold font). An alphabetic label against each numbered provision is for ease of reference. The provisions are arranged in narrative order rather than numerical order.19

table 1
table 1
table 1

a The cop Decision, para 25, adds: ‘Parties shall submit to the secretariat their nationally determined contributions referred to in Article 4 of the Agreement at least 9 to 12 months in advance of the relevant session’ of the Paris Agreement cop, ‘with a view to facilitating the clarity, transparency and understanding of these contributions’ (emphasis added).

b Neither the Paris Agreement nor the cop Decision contain any further information on the fmcp process.

2 Enforcement Model: No Substance

Which of the breachable (‘shall’) provisions in Table 1 might be amenable to ‘enforcement’—and with the use of what sanction? One candidate is a state’s obligation to prepare and communicate a nationally determined contribution (ndc) every five years (items B and C in the Table). This is an important, foundational obligation in the context of the Paris Agreement. How might an enforcement body respond to a state’s failure to submit an ndc by the five-year deadline? Rules developed for Article 15 could enable a hypothetical enforcement body to resort to holding a hearing with the state to formally establish the breach and perhaps also its cause, followed by a declaration of non-compliance and an imposition of a new deadline for the submission of the ndc. This would constitute an appropriate, proportionate, response to a state’s non-compliance with a deadline. Because the Paris Agreement does not create any special privileges for states, such as the trading privileges created by the Kyoto Protocol (see below), which could be used as leverage against a state, any attempt at a stronger response by the enforcement body would have no intuitively plausible sanctions to back it up. Consequently, if the state were to miss the new ndc deadline set by the enforcement body, this body could do no more than set another deadline.20

What would be the point of creating a new body to undertake the kind of intervention I have just described? The original missed deadline would already be known to the Paris Agreement’s Secretariat; it would also be known—eg because of rules specifically adopted to ensure this—to the Technical Expert Review team (ter; item I in the Table) assigned to review the state’s progress, a task which presupposes the submission of an ndc. In due course, the breach would also be known to the body responsible for conducting the ‘facilitative, multilateral consideration of progress’ (fmcp; Table item J). With appropriate procedures in place, the public, as well as the various ngo pressure groups, could also quickly be informed of the lapse. A formal declaration and a new deadline would most naturally issue from a point in the fmcp process. A specialized enforcement body, therefore, is not needed for an outcome that can be achieved no less effectively in the course of the operation of already-prescribed processes overseen by already-named (or implied) bodies.

However, the irrelevance of the enforcement function in the context of the Paris Agreement is not only the upshot of the axiom against duplication. Let us recall the design of the Kyoto Protocol’s Enforcement Branch, which embodies the complete experience we have to date with the enforcement model applied to the climate change regime. The essential objective of the Kyoto Protocol is/was21 to keep states from exceeding the emission limits (assigned amounts) to which they were legally bound. Each assigned amount was important to meeting the Protocol’s reduction aim of ‘at least 5 per cent below 1990 levels in the commitment period 2008 to 2012’.22 The corresponding power in the Protocol’s compliance mechanism was the power of the Enforcement Branch to penalize a country for exceeding its emission limit. The mechanism enabled the Enforcement Branch to deduct a country’s excess commitment-period emissions, multiplied by 1.3, from its assigned amount for the subsequent commitment period.23 These three elements—objective of the Protocol, power of the compliance mechanism, and penalty for the breach—were tightly interlinked: a substantive breach was to be met with a substantive penalty that repaired the breach and ensured that the emission reductions sought by the Protocol were maintained.

A second penalty available to the Kyoto Protocol’s Enforcement Branch applied to certain procedural breaches. These were procedures that underpinned the Protocol’s substantive aims. To meet its primary aim of determining whether an Annex I party had (or had not) exceeded its emission limit, the Enforcement Branch needed to be sure that the country’s reporting of its emissions was complete and accurate. It needed to receive the state’s reports in a timely fashion (since the assigned amount was confined to a time period), and it needed to verify that any international trading of emission allowances engaged in by the Annex i party was properly accounted for. The Enforcement Branch was empowered to penalize a party in breach of one of these conditions. It could do so by imposing restrictions on its participation in emission trading.24 This penalty for a procedural breach potentially hurt the non-compliant state by increasing the cost of its compliance with its substantive commitment.25 Once again, the enforcement function’s design maintained a close link between the Kyoto Protocol’s ultimate objective, the power of its enforcement body, and the available penalty. Exclusion from trading was no less elegantly built into the system than the penalty of reducing the assigned amount.26

We should note that both penalties I have described amounted to economic sanctions against a state. They were not mere declarations, or requests to attend hearings, or requirements to produce reports or commit to new deadlines. The Enforcement Branch had these powers, as well, but its economic sanctions were in a class of their own.

An economic sanction requires delicate handling. Even where it does not cause any economic pain in fact, it is symbolically powerful. The availability of economic sanctions is itself a reason to have a dedicated enforcement body set up to supervise the penalties. An economic sanction must be judiciously imposed, of course; and when the party’s non-compliance has ceased, it must be efficiently withdrawn. Consider the mechanics of exclusion from, or re-inclusion in, one or more of the Protocol’s trading mechanisms. Status changes must be accurately and authoritatively communicated to all relevant departments of the climate regime’s administration, such as the International Transaction Log.27 An economic sanction is both politically too sensitive and practically too consequential to be left to the control of loose, generalist, collective procedures (such as those of the unfccc’s multilateral process for the assessment of Biennial Reviews28) or to large all-party bodies with fixed and inflexible meeting dates (such as the unfccc’s Subsidiary Body for Implementation).29

The Kyoto Protocol’s enforcement function not only had a lean and logical design, in practice it worked more or less as intended. I have argued elsewhere that the Expert Review Teams (erts) expanded their facilitative role and held cases from reaching the Enforcement Branch;30 even so, the Enforcement Branch has handled nine cases to date. (One case only, concerning the Ukraine, is still open.) Not only has the enforcement function worked, it is the only formal element of the Kyoto Protocol’s compliance mechanism to have worked, as I will further explain. To put it plainly, if value is given to experience and to systems that work as intended, then we should associate the words ‘compliance mechanism of the climate change regime’ with ‘Enforcement Branch of the Kyoto Protocol’. Yet, as we saw earlier, the text of Article 15 of the Paris Agreement wants us to believe that ‘compliance mechanism’ means not enforcement but facilitation.

In light of the above considerations, a compliance mechanism with an enforcement function is not relevant to the Paris Agreement, not simply because it would create duplication, and not at all because the Agreement is supposedly different from its predecessors for having a bottom-up design.31 An enforcement function is irrelevant because even a breach of the Agreement’s core rule on the submission of ndcs by the prescribed deadline is not a breach of a substantive state commitment, whether directly or indirectly. The Agreement has no such state-level commitments. Non-submission of an ndc by the five-year deadline is of course not desirable in the framework of the Paris Agreement, as it would impact on the work of the general ‘stocktaking’ session which comes at the end of the ndc-ter-fmcp process, causing the stocktake to be less than fully informed about the state of ‘progress’ under the treaty.32 However, while this would be undesirable, it would not undermine the regime’s aims as listed in Article 2, because meeting those aims does not depend on any given country meeting any given target every five years, nor even does it depend on the parties collectively meeting any given target over the same period—in contrast to the Kyoto Protocol’s aforementioned 5% reduction target, for example. The new treaty is legally so weakly formulated that it does not even forbid countries from reducing their mitigation ambition from one ndc offering to the next. The treaty is merely aspirationally opposed to such reductions.33 Thus an enforcement model for the further development of Article 15 is ruled out. Peer pressure, not enforcement, would seem to be the tool most suited to keeping up collective aspirations.

3 Facilitation Model: Duplication and Waste

In this section I discuss the prospect of a facilitative function for the Paris Agreement’s compliance mechanism. If scholarly commentary is anything to go by, this is the model that will be implemented.

3.1 A Facilitation Model for Article 15 and What it Would Do

Christina Voigt has had the most to say about the operationalization of Article 15 in the form of a facilitative committee. Voigt also goes further than anyone else in her interpretation of Article 15, reading into the article’s opening words (‘A mechanism to facilitate implementation of and promote compliance with the provisions of this Agreement …’) two different mandates, namely, as she puts it, ‘the implementation of provisions of the Agreement that are not legally binding, as well as promoting compliance of parties with their legally binding obligations’.34 It is a rather strained reading. A more natural one, it seems to me, is that the mechanism is intended to promote compliance, inter alia by facilitating implementation of the Paris Agreement’s legally binding provisions. Nevertheless, Voigt postulates that ‘facilitate implementation’ is intended to cover one set of provisions, whereas ‘promote compliance’ is intended to cover another set. She thus proceeds to propose a bifurcated Article 15 committee, consisting of two different ‘chambers or branches’, viz. ‘a compliance [in the sense of facilitation] branch and an implementation branch’.35

What would this two-chamber committee do? Voigt suggests that ‘the compliance branch should seek to determine the cause of non-compliance’, without seeking to punish the party; it should also ‘work with the party concerned to rectify the problem’.36 After giving the party a chance to come into compliance, the branch would, if the problem persists, ‘issue a declaration of noncompliance’, as part of an extended procedure involving written submissions and hearings.37 As for the ‘implementation branch’ of the Article 15 mechanism, its purpose would be to give ‘assistance’ in the form of ‘technical or capacity-building initiatives’, to which Voigt adds the qualification: ‘It would not be advisable, however, [for it] to provide financial assistance’.38

In another example of support for the establishment of a committee with a facilitative function, Harro van Asselt and colleagues write that the Article 15 committee ‘may identify and verify cases of non-compliance, and could issue declarations and recommendations to the Party concerned’ or to the Paris Agreement cop.39 This is similar to what Voigt’s first chamber would do. However, the van Asselt group also indicates support for the use of ‘consequences’, which is a euphemism for measures against a state. Two ‘consequences’ are mentioned: ‘requiring a plan to return to compliance or finding a Party in non-compliance (which [counts as a consequence because it] may entail reputational costs)’.40 Van Asselt et al. envisage that the facilitative committee, having prescribed a so-called consequence, would proceed to keep track of the state’s progress with the implementation ‘of such facilitative measures’.41 On the supportive side, the proposed committee would ‘help parties identify the causes of non-compliance and overcome barriers to compliance’, as well as point to examples of ‘successful cases of implementation’, which would serve as models for all parties.42

Neither Voigt nor van Asselt et al. pause to consider whether the tasks they assign to their proposed committees have not already been assigned, explicitly or by implication, to other components of the Paris Agreement architecture.

3.2 Kyoto Protocol Experience With the Facilitative Branch

At this advanced stage in the evolution of the climate change regime, a preference on how to operationalize Article 15 should not be advocated without reference to relevant past experience. We are 25 years into the regime’s history. Quite a lot of compliance-relevant experience has accumulated. Advocates of a facilitative compliance mechanism under Article 15 must contend with the fact that the Kyoto Protocol’s Facilitative Branch has been a failure. This is no exaggeration. It has been moribund throughout its existence. We could not have known this would be the fate of the Facilitative Branch when it was established. There is no theory (yet) to tell us how such designs will pan out. But whereas there was no relevant practice prior to 2005—the year when the Facilitative Branch was established—there is some now. It is incumbent upon supporters of an Article 15 facilitative committee to distinguish the Kyoto Protocol’s facilitative compliance body from their own proposed committee and explain why theirs will enjoy a useful and productive life even though the Kyoto Protocol’s did not.

I have described the Facilitative Branch’s problems in another publication,43 so here, after a brief background paragraph, I will confine myself to an update drawing on information from the last two years (mid-2014 onward).

Early in its life, the Facilitative Branch was assigned a case it could not figure out how to handle and promptly abandoned;44 since then it has never been assigned another case. Worse than that—for it is at least conceivable that no state actually needed any facilitative assistance throughout these years—the Facilitative Branch itself still has no clear idea about what its role is within the Kyoto Protocol’s compliance mechanism. On paper, the measures at the disposal of the Facilitative Branch are, in Sebastian Oberthür’s helpful summary, ‘permutations on the provision of advice and the facilitation of assistance, including as the strongest measure the formulation of recommendations’.45 They are not so different, we note, from those that Voigt et al. have proposed for the Article 15 committee.

The ‘Monaco Incident’ from late 2014 is an illustration of the Facilitative Branch’s problems. In that year, Monaco was, not for the first time, late in submitting its National Communication. (Annex i parties to the Kyoto Protocol submit a single, combined, National Communication for both the unfccc and the Protocol.) The Facilitative Branch took the initiative to write to Monaco, ‘asking whether it could provide any advice and facilitation’.46 Monaco replied, not by letter, but by email, giving reasons for the delay in its submission, and informing the Branch that it did not need its involvement. Upon receiving the email, the Branch ‘agreed to conclude its consideration of the matter, as Monaco had indicated that it did not require any support’.47 The Branch gave no indication of what ‘support’ it could have provided Monaco had its offer of support been accepted. Monaco’s dismissal of the Facilitative Branch’s gesture was not quite as biting as Canada’s in the ‘Canada Incident’ from 2012, which I have described elsewhere.48

These two have been the Facilitative Branch’s only attempts at facilitative outreach to states. Each time, the Branch has been rebuffed. These incidents add to the evidence that states do not take a facilitative compliance body seriously.

There was a second element to the Monaco Incident. The Expert Review Team which had conducted the review of Monaco’s 2014 annual National Inventory Report (nir) did not raise a ‘question of implementation’ in respect of Monaco’s failure to submit the nir by the deadline. (That year Monaco was late both with its National Communication and its nir.) In the framework of the Kyoto Protocol, nir submission is a procedure belonging to that special class of procedure I described earlier, which is necessary to determine whether a party has met its substantive obligation on the assigned amount. A question of implementation raised by the ert would have brought the issue before the Kyoto Protocol’s Compliance Committee. The Facilitative Branch decided to request the unfccc Secretariat to ask the ert’s Lead Reviewers to appear before the Facilitative Branch ‘with a view to understanding the reasons for the team’s decision’ not to raise a question of implementation.49 The meeting with the Lead Reviewers was duly held, but a record of it was not made public, and no further action came of it. We can only assume that the Lead Reviewers told the Facilitative Branch that they saw no point in raising a question of implementation after they had received the nir, albeit late. If this is what happened, it is another case of an ert exercising unscripted control over the Kyoto Protocol’s compliance mechanism while the Facilitative Branch sits powerlessly on the sidelines. I have argued this point before using pre-2014 evidence.50 It is a feature of international regimes (not only of the climate change regime, but also, in my experience, of international criminal law51) that ad hoc bodies tend to take unto themselves powers that were never intended for them, whereas other bodies are unable to exercise the very powers that were formally bestowed on them. If this is indeed a general feature, it is all the more reason for a conservative approach to Article 15.

A last word on deadlines: It is a noteworthy fact that Multilateral Environmental Agreements (meas) in general have not been able to solve the problem of how to get states to submit their periodic reports in a timely fashion. Only the Kyoto Protocol has solved it, and then only in relation to nirs, and then not without exceptions, as the Monaco Incident shows. Under the Protocol, non-receipt of an nir may lead the Enforcement Branch to limit a country’s participation in international emission trading (hence nirs are rarely submitted late); by contrast, National Communications are submitted late with impunity, and neither the Kyoto Protocol’s Facilitative Branch nor the facilitation mechanisms of other meas have any sanctions to use against such breaches of procedure unrelated to substance.52 This general weakness with the facilitation model led the Basel Convention’s Secretariat to convene a cross-mea workshop in search of solutions to the problem of missed deadlines.53 No solutions emerged from the workshop—more evidence that facilitative compliance mechanisms struggle to add value to international environmental regimes.

The Kyoto Protocol’s Facilitative Branch was set up with an ‘early warning’ function.54 The Branch has never been able to figure out how to operationalize it. The original idea was that the Branch would be able to foresee, say, that a state will not meet its assigned amount. Where it had such foresight, it would be able to deal with the risk of non-compliance before it was actualized. Late in its life, the Facilitative Branch developed certain ‘indicative working arrangements’ for the early-warning function. It claims to be ‘continuing to refine’ them.55 However, it has applied them only once, unsuccessfully, in the Canada Incident.56 In its ongoing effort to work out what role it could play in early warning, the Facilitative Branch has relied on the advice of the Expert Review Teams, with whose practitioners it periodically holds joint meetings.57 The Branch has acknowledged that it sits at the end of a process, at which little information arrives, because most of it is filtered out at earlier stages. What little information does arrive, arrives too late.58 The Facilitative Branch is set up as a facilitation body in name only; it has no money to hand out, and no allocation of time for carrying out investigations. It meets just four days a year in total. Almost each time it meets, it reflects on ‘the need for further work in order to advance its understanding on how to exercise its mandate’;59 or it ‘agree[s] to continue its general discussion on [how to] provid[e] advice and facilitation’.60 It has become inward-looking.

The Facilitative Branch has always been a step removed from the actual compliance challenges a state faces. From that distance, it is difficult to diagnose their true causes or to know what advice to give the state to fix them. The erts, by contrast, have an ear to the ground and operate in well-balanced teams with expertise covering all emission sectors. ert members collectively spend thousands of hours each year picking through state reports, while remaining in direct communication with the governmental authorities of the countries under review. Together with staff at the unfccc Secretariat, they dominate the field of advice and facilitation. The ten members of the Facilitative Branch (and ten alternates) cannot hope to match an ert’s fine-grained understanding or expertise in issues arising from state reporting. Nevertheless, the Facilitative Branch continues to pretend that it has useful powers as a facilitation body. It refuses to acknowledge that it has been sidelined by the wider ert system.

3.3 Why the Paris Agreement Should Not Rush to Establish a Facilitative Compliance Body

The Paris Agreement keeps in place, in essence, the unfccc’s reporting-and-review processes, including the Cancun arrangements for the period to 2020,61 only now more states are to participate. States are to submit National Communications, greenhouse gas inventories (nirs), and Biennial Reports. Under the Cancun arrangements, developed countries had ‘nationally appropriate mitigation commitments or actions’, which included ‘quantified economy-wide emission reduction targets’.62 The extent of pro-active state participation in mitigation under the Paris Agreement is much broader, and the Agreement’s ndcs also cover more topics than the Cancun pledges did. Yet, despite differences in detail, there is no paradigmatic shift between the reporting-and-review elements currently in place for the unfccc and those envisaged for the Paris Agreement. The latter can easily be implemented through piecemeal amendments to existing rules. It is important to keep in mind that the Paris Agreement does not call for any overhaul of the current system of ‘transparency’.

I will proceed to make three arguments about why the Ad-Hoc Working Group on the Paris Agreement should think twice about setting up a separate facilitative compliance body. First, the erts (which will be called ters under the Agreement) will continue their current pro-active facilitative role, rendering impotent any body with overlapping functions. Second, the Paris Agreement demands a ‘flexible’ approach to transparency, which makes ‘compliance’ with procedural requirements all the more difficult to operationalize. Third, the Paris Agreement increases the number of facilitative services offered to parties, making it ever more unlikely that parties will turn to a ‘facilitative’ compliance committee for assistance.

I turn to my first argument. Although there is no designated compliance body set up by the unfccc (as opposed to the Kyoto Protocol), it does not follow that the treaty fails to generate any compliance pressure on states, or even that it lacks a kind of compliance system. Oberthür has correctly observed that the unfccc’s current reporting-and-review system for the period through to 2020 incorporates compliance features. The processes of International Assessment and Review (iar) and International Consultation and Analysis (ica) go beyond simple reporting and review and ‘may . . . be considered to move into the field of an assessment of compliance and a response to implementation issues’.63 As Oberthür further explains, iar and ica ‘form part of the grey area between review and compliance [as] they provide the opportunity to raise questions regarding implementation’64 and ‘to publicly identify, discuss, and address shortcomings’.65

Yet, Oberthür’s insight is only half the story. Below the iar/ica level, Expert Review Teams along with Secretariat staff push compliance incessantly. They pressure states to fill information gaps in order to achieve greater transparency, even if states are not literally (legally) required by the rules to provide that information. I will give an example from my own experience as an ert member taking part in the unfccc’s Biennial Reviews.

The erts control the Biennial Review process in a manner that a facilitation committee could never hope to attain. ert Lead Reviewers, in cooperation with the unfccc’s Secretariat, have produced a manual called Review Practice Guidance (rpg). It covers Biennial Reviews among other types of review.66 The rpg gives ert members guidance on how to deploy two weapons they have available to use against not-fully-conforming states in Biennial Reviews. They are called ‘recommendations’ and (ungrammatically) ‘encouragements’. Both are euphemisms. They roughly correspond to the labels ‘unacceptable error’ and ‘avoidable error’. The threat of use of such language in an ert’s report on a state’s Biennial Report is, as a matter of fact, sufficiently motivating for states to persuade them to make most changes requested by an ert. States fight hard during the Biennial Reviews to avoid having recommendations or ‘encouragements’ written into ert assessments of their biennial submissions.

The rpg manual explains that a ‘shall’ requirement in the unfccc-approved rules on Biennial Reviews is a mandatory requirement. When breached, it is to be addressed with a recommendation. On the other hand, a ‘should’ requirement is not a mandatory one and any deviation should issue only in an encouragement.67 This might seem like a tidy rule that limits an ert’s discretion. Yet it does not. The cop’s rules on Biennial Reviews are remarkably vaguely formulated. They leave plenty of scope for gap-filling. This allows erts, under the guidance of the Lead Reviewers and the sharp eye of Secretariat staff, to develop their own ideas about what information should be provided under a ‘shall’ or ‘should’ requirement. Nothing sinister is going on here. All ert members are state-nominated to the international Roster of Experts, and the vast majority of them work for their national governments when they are not performing ert duties. Nevertheless, once they step into their formal reviewing role, ert members become infused with a desire to raise the quality of reporting. Outside the Biennial Reviews, reviewers and those reviewed are peers, yet the review system lends the former a short-term power to pressure the latter to do better. Combined with the determination of states to avoid aspersions about their willingness to uphold treaty requirements, this arrangement generates enough pressure on parties to ensure that a general conformity among states is achieved at the ert-dominated review level of the process and is not left to any later stage, such as a higher-up facilitation committee. These are facts which only practice reveals.

ert Lead Reviewers hold their own regular meetings in Bonn. Consider the mindset behind the following statement emerging from a Lead Reviewers’ meeting:

the lrs [Lead Reviewers] recommend that lrs promote the following procedures during the 2016 review: … Interacting with the Parties is fundamental in assessing whether sufficient progress has been made in the implementation of previous recommendations for the purposes of including the issue in a prominent paragraph.68

There is a clear awareness in this passage that the threat of reputational consequences is a powerful lever for erts to use with states. Here, the Lead Reviewers can be seen to be considering ways of maximizing the threat of loss of face to states so as to achieve quick transparency gains with them. The Lead Reviewers have a better-developed sense of their own role in a de facto compliance system than the Facilitative Branch of the Kyoto Protocol has ever had in its de jure role. The Lead Reviewers also have the advantage of, as they put it, ‘interacting with the Parties’ outside of the formalities and floodlit public forum that a de jure compliance procedure introduces.

Expert Review Teams insist on giving non-mandatory as well as mandatory language its maximum effect.69 But they also push interpretations that go beyond the plain language of the applicable cop rules. I will give an example. Lead Reviewers take the view that certain reporting tables that states must use in connection with the Biennial Reviews, known collectively as the Common Tabular Format (ctf) tables, are to have all of their cells completed by the reporting states. Thus:

[where] the reported information included in the tables of the ctf, in particular information on the effects of individual policies and measures (PaMs) [is missing,] the ert [must] clarify … why a Party did not report some information in the ctf … including on the effects of its individual PaMs and summarise the Party’s explanation in the review report. Gaps identified … might not necessarily mean incomplete reporting, provided that they are adequately explained by the Party as being due to national circumstances. The ert should recommend [i.e. use its strongest sanction, not just give an ‘encouragement’] that the explanation for gaps be included by the Party in its next submission of the br [Biennial Report].70

This passage exemplifies a ‘rule’ originating entirely at the level of the ert system. It is plucked out of thin air by the Lead Reviewers at their ad hoc forum and written into the rpg manual. There is no mandate for it, so it is technically ultra vires. But no state has complained about regulatory innovation at the ert level (yet). And so the pressure on states to improve their reporting practices gradually builds up.

Here is what the relevant cop-approved Biennial Review rules actually say, in relevant part:

Each Annex i Party shall provide information on its mitigation actions, including on the policies and measures it has implemented or plans to implement since its last national communication or biennial report.71

[The cop] Adopts the common tabular format … noting that developed country Parties … shall use the reporting guidelines for the preparation of their biennial reports and will provide information on all the elements of the guidelines in these reports.72

[The cop] Decides that developed country Parties shall use the electronic reporting application [for the ctf] when preparing and submitting their biennial reports.73

Nowhere in the formal rules does it say or imply that quantification of the effects of individual PaMs shall be provided in the ctf, or that all cells of the ctf are mandatorily to be filled in, or that if they are not filled in a completeness or transparency issue will arise and is to be reflected in the ert’s review report. Yet, this is exactly how it works in practice, to the credit of the ert system. My point is that erts not only have the power to enforce rules; they have taken it upon themselves to gap-fill the rules and even to make up their own. They are an incarnation of ‘bottom-up’ ideology. This has not been recognized in the literature supporting the operationalization of Article 15 of the Paris Agreement, nor is it discussed in compliance literature generally.

My second argument on why a facilitation committee under the Paris Agreement may end up being dysfunctional is that the Agreement requires ‘flexibility’ in the implementation of its transparency framework: the ‘enhanced transparency framework for action and support, [has] built-in flexibility which takes into account Parties’ different capacities and builds upon collective experience’.74 The framework is to ‘avoid placing undue burden on Parties’.75 This is quite unlike the Kyoto Protocol’s generally inflexible rules for Annex i parties.

How is a compliance mechanism, even a merely facilitative one, to be grafted onto a system with flexible rules? How could subordinate rules attempt to limit state conduct when an unqualified notion of flexibility is written into the treaty itself? Flexibility suggests broad discretion, and broad discretion suggests political, not legalistic, judgement. Implementation of a ‘flexible’ transparency framework is thus best left to a political forum, such as the Subsidiary Body for Implementation, which is likely to be the actual forum to supervise the Paris Agreement’s process of ‘multilateral, facilitative consideration of progress’.76

My third and final argument under this section is that, even without Article 15, the new treaty significantly enhances the facilitative support options available to parties. Several scholars have claimed that non-compliance results from a lack of capacity.77 Whether or not this is true,78 the Paris Agreement provides for new support for capacity-building. In particular, Article 11 promises capacity-building in all areas, including in connection with the transparency framework. Institutional arrangements for capacity-building are to be adopted at the Paris Agreement cop’s first session.79 In parallel, the cop Decision, without even waiting for the cop’s first session, ‘establish[es] the Paris Committee on Capacity-building’, devoting eleven paragraphs to laying out the scheme.80 It also, separately, establishes ‘a Capacity-building Initiative for Transparency in order to build institutional and technical capacity [to] support developing country Parties … in meeting enhanced transparency requirements as defined in Article 13’.81 The Global Environment Facility is instructed to make arrangements to financially support the new Initiative.82 Financial support is also given generously for such purposes in several other provisions of the treaty.83 Article 15 could be launched onto the crest of this wave of facilitation, but it could also be kept in reserve until a need for further facilitative support emerges.

4 Nature of the Paris Agreement

A final argument against the activation of an Article 15 ‘compliance’ mechanism in the short- or even medium-term finds a foothold in the very nature of the Paris Agreement.

It might be said that the purpose of the Paris Agreement is to keep global warming ‘well below’ 2°C, as Article 2(a) of the treaty suggests. Yet this interpretation exaggerates the Paris Agreement’s purpose. Countries ratifying the Agreement have not agreed to emission reductions consistent with a future below 2°C. The Agreement does not lay out any such agreement. State parties have only agreed to work towards an agreement to make the emission reductions implied by the 2°C limit. To be successful, the Paris Agreement will need to induce all countries to agree to take on individual mitigation targets (whether supported by climate finance or not) adding up to a total reduction in emissions consistent with the 2°C warming limit. Unless the 2°C limit is abandoned, by definition all countries must, sooner rather than later,84 agree to a set of reductions consistent with the stated warming limit and stay true to their promised mitigation targets. This implication of the Paris Agreement is politically too unpalatable to be broadly recognized at the present time, but necessity will make the notion palatable again before long. States cannot keep working through ndc iterations indefinitely.85 On a logical reading, the purpose of the Paris Agreement is to assist countries to quickly arrive at a point of a final agreement on permissible emissions. The latter agreement would operate to keep warming below 2°C by keeping states from departing from their individualized, rationally apportioned, targets.

The mechanism intended to help states reach the point where their emission pledges under the Paris Agreement are collectively consistent with a global emission budget that avoids 2°C of warming is the Agreement’s ratcheting-up mechanism. The mechanism (states pledge—states implement and report—states participate in multilateral review—the cop stocktakes—states pledge—states implement and report—states participate in multilateral review—the cop stocktakes—and so on and so forth) rolls around on a five-year cycle, until, in theory, the point is reached where no more ratcheting up is necessary because the collective pledges do not exceed the remaining global emission budget. I say ‘in theory’ because there is nothing about the ratcheting-up mechanism itself that guarantees that states will eventually agree to individual emission pledges consistent with the warming limit. There is no scientific/behavioural theory according to which states bound into such an arrangement will eventually reach an agreement. On the other hand, there is more than enough scientific evidence that the Paris Agreement has only a small number of ndc cycles available to it before it must settle on a final agreement—maybe three cycles, almost certainly no more than six, with each round making strong gains over the previous one.86

States have impliedly already bid for a share of the global emission budget through their ‘Intended’ ndcs. Where a country claims that its indc, or, later, its ndc, represents not only the best it can do but also its fairest contribution to the global effort, it is making a bid for recognition that it owns that share of the global emission budget corresponding to the emissions implied by its ndc’s embodied emission trajectory. A state cannot consistently accept the 2°C limit imposed by the Paris Agreement, as well as the global budget implied by that limit, and at the same time deny that its ‘contribution’ defines the size of the allocation it wants for itself from that budget. The bidding war for a burden-sharing formula has already begun, notwithstanding the euphemistic language of cooperative rounds of progressively higher ambition being used to describe what will happen under the Paris Agreement. The Agreement has launched a process to negotiate a budget-setting agreement.

While a stepwise, gradual approximation to a plausible emission budget could lead to a final agreement on emissions, the same result could be achieved in one great leap (say, after two or three disappointing iterations of the ndc ratcheting-up mechanism), by states agreeing to cut their losses and accept a burden-sharing formula consistent with a <2°C limit. Under either scenario (ratcheting up steadily to the end point, or ratcheting up with a great leap at the end), states will need a new agreement: either an agreement to stick with the pathways they have achieved through iteration (which might be possible under the umbrella of the Paris Agreement, suitably amended), or an agreement to speed up mitigation through the application of a final burden-sharing formula, which would require a new international agreement. Either way, the Paris Agreement’s lauded ratcheting-up mechanism is really no more than a practically minded, deadlock-breaking, albeit highly structured, framework for further negotiations.

The commitment of states to stay engaged with the Paris Agreement’s structured negotiation process (their agreement to work towards an agreement) is not the kind of commitment appropriately managed through a compliance mechanism. No Paris Agreement mechanism, whether under Article 15 or another clause, could meaningfully sanction a state for not wholeheartedly engaging in the pursuit of a final-stage agreement. The standard of ‘wholehearted engagement’ is too ambiguous to be assessed with any precision. It is a political concept for a political agreement to be weighed in a political forum. The Paris Agreement is too political for a top-down compliance mechanism.

5 Conclusion

Does the Paris Agreement need an Article 15 compliance mechanism to be activated right now? I have argued that states already come under considerable pressure to comply with reporting-and-review rules through the operation of the unfccc’s ert/iar system, a diffuse collection of actors which nevertheless operates as a de facto compliance system. Under the Kyoto Protocol, the ert system, especially, functioned so effectively that it paralyzed the Facilitative Branch. Given our experience with how technical review of state reports actually works, the most efficient approach to the implementation of the Paris Agreement would be to suspend the operation of Article 15 until such time as practice attests that some kind of Article 15 mechanism would actually add value.

The Kyoto Protocol’s Enforcement Branch is the only compliance mechanism to have worked in the quarter-century of the climate change regime. There is thus a basis for saying that only this kind of formal body will work in the future. A compliance mechanism less like the facilitative body of Article 15 and more like the Enforcement Branch of the Kyoto Protocol will become relevant again once the Paris Agreement leads to another, final, agreement, not unlike the Kyoto Protocol, or is amended to create its own substantive obligations to the same effect.

1 unfccc, Decision 1/CP.21, Adoption of the Paris Agreement, fccc/cp/2015/10/Add.1 (2015), para 102. The author is grateful to the journal’s two referees for their detailed comments on earlier versions of this article.

2 Paris Agreement, Article 15.1.

3 Ibid. Article 15.2.

4 Ibid.

5 The parties to the Paris Agreement were until recently still at the stage of compiling party views on how to develop Article 15; see Ad-Hoc Working Group on the Paris Agreement, Informal Note by the Co-Facilitators: Agenda item 7: Modalities and Procedures for the Effective Operation of the Committee to Facilitate Implementation and Promote Compliance Referred to in Article 15.2 of the Paris Agreement (14 November 2016), available at <>.

6 See, for example, Michael MEHLING, ‘Enforcing Compliance in an Evolving Climate Regime’, in Jutta BRUNNÉE, Meinhard DOELLE, and Lavanya RAJAMANI (eds), Promoting Compliance in an Evolving Climate Regime (Cambridge University Press 2012) 214; Lavanya RAJAMANI, ‘Developing Countries and Compliance in the Climate Regime’, ibid. 393–394; Lavanya RAJAMANI, Jutta BRUNNÉE, and Meinhard DOELLE, ‘Introduction: The Role of Compliance in an Evolving Climate Regime’, ibid. 11; and Harro VAN ASSELT et al., Maximizing the Potential of the Paris Agreement: Effective Review in a Hybrid Regime (Stockholm Environment Institute 2016) 1 and 6.

7 Eg Laurence BOISSON DE CHAZOURNES, ‘Editorial on Paris Agreement’ (2016) 27(2) European Journal of International Law 253, 254 (‘[The ndcs of the Paris Agreement are a] bottom-up approach … in contrast to an approach whereby targets are imposed from “above” in a treaty. The Kyoto Protocol has become a symbol of this failed top-down approach, and, hence, the idea of a new approach has since prevailed, creating an opportunity for another normative model’); and Meinhard DOELLE, ‘The Paris Agreement: Historic Breakthrough or High Stakes Experiment?’ (2016) 6(1–2) Climate Law 1, 20 (‘The Paris Agreement is an experiment in a bottom-up, managerial, transparency-building, and norm-building approach to global cooperation. The shift in approach is a reasonable gamble in light of the failed efforts over the past two decades to implement the top-down approach under the Kyoto Protocol’).

8 See Daniel BODANSKY, ‘The Legal Character of the Paris Agreement’ (2016) 25(2) Review of European Community and International Environmental Law 142, 149; idem, ‘The Paris Climate Change Agreement: A New Hope?’ (2016) 110(2) American Journal of International Law 288, 291 (‘peer and public pressure’); Harro VAN ASSELT and Thomas HALE, Reviewing Implementation and Compliance under the Paris Agreement: Arizona State University Workshop Background Note (2016), available at <>, 3, Table 1 (where the claim is made that ‘An expert review process flagging problems with implementation can help facilitate compliance without the intervention of a compliance mechanism’; however, the authors give no argument in support of this claim); and VAN ASSELT et al. (n 6) 2 (‘opportunities to apply political pressure’).

9 See, for example, the works cited in note 7, above.

10 Emphasis added.

11 BODANSKY, ‘Legal Character’ (n 8) 149, makes the point that legal compulsion was excluded from the Paris Agreement in order to encourage a high mitigation ambition.

12 The only formally substantive ‘shall’ (and thus mandatory) requirements in the Paris Agreement relate to the supply of climate finance; however, while formally substantive, these are both unquantified and collective, which means that in practice, if not in international law, they are not substantive, since they lack specificity in both of those dimensions and no state-level accountability can be derived from them.

13 The same assumption is found, for example, in Sebastian OBERTHÜR, ‘Options for a Compliance Mechanism in a 2015 Climate Agreement’ (2014) 4(1–2) Climate Law 30, 34 and 42.

14 See BODANSKY, ‘Legal Character’ (n 8) 145 (recognizing the possibility of ‘general institutional obligations for the regime as a whole, but not obligations for individual parties’); 146 (‘the obligation to pursue domestic mitigation measures [in Article 4.2 of the Paris Agreement] is arguably a collective rather than an individual obligation’); and 147 (‘these [several listed provisions] represent collective rather than individual obligations’); see also Lavanya RAJAMANI, ‘The 2015 Paris Agreement: Interplay between Hard, Soft and Non-Obligations’ (2016) 28 Journal of Environmental Law 337, 343.

15 Note that avoidance of duplication of effort is a principle forming part of the Paris Agreement itself; see cop Decision, paras 73(a) and 92(d).

16 This too may be an implication of the new treaty; see Articles 13(1), 13(4), and 13(13), as well as cop Decision, paras 37(f), 73(e), and 93, which however are not directly linked to Article 15.

17 I reiterate my exclusion of the unquantified, non-individualized, obligations on climate finance, which I alluded to in note 12.

18 The position is fully developed in Alexander ZAHAR, Climate Change Finance and International Law (Routledge 2017).

19 For a more elaborate tabular presentation of the Paris Agreement’s key provisions, see RAJAMANI (n 14) 344–351, as well as BODANSKY, ‘Legal Character’ (n 8) 146–47.

20 I would like to offer a challenge to those who disagree with this article’s skeptical conclusions to try to devise other, tougher, sanctions and explain how they would work. On this point, I am grateful to one of the journal’s referees for suggesting that Paris Agreement negotiators can maintain a space for creative thinking on sanctions, and that the case for new sanctions is not closed. See also note 29, below.

21 Some readers may consider that the Kyoto Protocol has already run its course.

22 Kyoto Protocol, Article 3.1.

23 Kyoto Protocol, Decision 27/CMP.1, Procedures and Mechanisms Relating to Compliance under the Kyoto Protocol, fccc/kp/cmp/2005/8/Add.3 (2005), Annex, para xv.5.

24 Ibid. para xv.4.

25 At the same time, exclusion of the country from trading would not have had any negative environmental impact, as the country remained bound to its assigned amount.

26 Throughout the text of the Kyoto Protocol, strong language supports the operationalization of the strong enforcement function. For example, Article 5.1 requires a ‘national system’ to be established and the use of methodologies developed by the Intergovernmental Panel on Climate Change (ipcc); where the latter are not used, unfavourable ‘adjustments’ may be applied to state emission reports, under the authority of the Conference of the Parties to the Kyoto Protocol. Article 7.1 makes compliance with the Protocol’s key Article 3 provision (on meeting assigned amounts) dependent on submitting reports and information listed in Article 7. Also, the trading mechanisms, outlined in Articles 6, 12, and 17, all require full accountability as a condition of state participation in them.

28 The process is called International Assessment and Review (iar).

29 Later in the article I discuss the important influence of ‘reputational costs’ in a certain context. Could reputational costs be substituted for economic costs to construct a functional Article 15 compliance mechanism? While this idea (which was suggested to me by one of the journal’s referees) might be worth exploring, a reputational consequence is not subject to the same degree of control as an economic one, for it might be realized long before a matter reaches the compliance committee (eg through the publication of an expert review report) and it might persist long after the compliance committee has declared that the state has returned to compliance. This type of penalty not being fully controllable by a committee, the establishment of a compliance committee with ‘reputational cost’ as its main penalty instrument would create uncertainty and unpredictability and would not actually be just a simple variant of the successful Enforcement Branch model.

30 Alexander ZAHAR, International Climate Change Law and State Compliance (Routledge 2015) 75–76.

31 To reiterate a point I made above, an ndc is no more bottom-up than an assigned amount under the Kyoto Protocol. Each represents a state’s domestically determined ambition. Neither is determined by a higher power or a global (as opposed to the far less meaningful collective Annex i) mitigation target. The difference between the two is merely one of legal status: whereas states are not legally committed to ndc targets, they are/were bound by their Kyoto Protocol assigned amounts.

32 For the aim of the global stocktake, see Paris Agreement, Articles 13.5 and 14.1.

33 See item D in Table 1, which employs ‘will’ instead of ‘shall’.

34 Christina VOIGT, ‘The Compliance and Implementation Mechanism of the Paris Agreement’ (2016) 25(2) Review of European Community and International Environmental Law 161, 166.

35 Ibid. 166.

36 Ibid. 167.

37 Ibid.

38 Ibid.

39 VAN ASSELT et al. (n 6) 5.

40 Harro VAN ASSELT, ‘International Climate Change Law in a Bottom-up World’ (2016) 26 Questions of International Law 5, 9–10.

41 VAN ASSELT et al. (n 6) 5.

42 Ibid.

43 ZAHAR (n 30) Chapter 3.

44 Ibid. 79.

45 OBERTHÜR (n 13) 40. For more detail on the Facilitative Branch’s responsibilities and powers, see Decision 27/cmp.1 (n 23) Annex, Parts iv.4–iv.6 and xiv.

46 Compliance Committee of the Kyoto Protocol, Annual Report, fccc/kp/cmp/2014/2 (2014), para 37. For the letter from the Facilitative Branch to Monaco, and Monaco’s reply, see the appendix to Compliance Committee, Facilitative Branch, Report on the Sixteenth Meeting, cc/fb/16/2014/2 (2014).

47 Compliance Committee, Annual Report (n 46) para 37.

48 ZAHAR (n 30) 79.

49 Compliance Committee, Enforcement Branch, Report on the Twenty-Sixth Meeting, cc/eb/26/2015/2 (2015), para 7.

50 ZAHAR (n 30) Chapter 3.

51 Eg Alexander ZAHAR, ‘Civilizing Civil War: Writing Morality as Law at the ICTY’ in Bert SWART, Alexander ZAHAR, and Göran SLUITER (eds), The Legacy of the International Criminal Tribunal for the Former Yugoslavia (Oxford University Press 2011), 469–504.

52 Compliance Committee of the Kyoto Protocol, Report by the Chairperson of the Enforcement Branch and the Secretary to the Compliance Committee of the Kyoto Protocol on the First Session of the Dialogue with Other Compliance Bodies Held in Paris, France on 6 December 2013, cc/14/2014/2 (2014), para 9(d) (‘Challenges in reporting [across meas] have often been rooted in the lack of clear consequences for late or non-submission of the relevant reports, or the absence of national legislation that supports the preparation of these’). René LEFEBER, who is the author of this report, proceeds to draw out a contrast with the Kyoto Protocol’s control over nirs: ‘In contrast to the lack of guidance concerning consequences for the late submission of national communications, it is clear that failure by a Party to submit its annual inventory within six weeks of the submission date established by the [Conference of the Parties to the Kyoto Protocol] can result in the suspension by the enforcement branch of a Party’s eligibility to participate in the market-based mechanisms’; ibid. 8.

53 See the report referred to in note 52.

54 Decision 27/cmp.1 (n 23) Annex, Part iv.6.

55 Compliance Committee, Dialogue on Compliance (n 52) 8.

56 See ZAHAR (n 30) 79.

57 Compliance Committee, Facilitative Branch, Report on the Fifteenth Meeting, cc/fb/15/2014/2 (2014), para 11.

58 Facilitative Branch, 16th Meeting (n 46) paras 10–11.

59 Compliance Committee, Facilitative Branch, Report on the Seventeenth Meeting, cc/fb/17/2015/2 (2015), paras 5–6; and Compliance Committee, Facilitative Branch, Report on the Eighteenth Meeting, cc/fb/18/2015/2 (2015), para 5.

60 Compliance Committee of the Kyoto Protocol, Annual Report, fccc/kp/cmp/2015/3 (2015), paras 31–32.

61 The cop Decision, at para 98, provides that ‘the modalities, procedures and guidelines of this transparency framework shall build upon and eventually supersede the measurement, reporting and verification system established by decision 1/cp.16 [The Cancun Agreements], paragraphs 40–47 and 60–64, and decision 2/cp.17 [Outcome of the Work of the Ad Hoc Working Group on Long-term Cooperative Action under the Convention], paragraphs 12–62’.

62 unfccc, Decision 1/CP.16, The Cancun Agreements: Outcome of the Work of the Ad Hoc Working Group on Long-Term Cooperative Action under the Convention, fccc/cp/2010/7/Add.1 (2010), Part iii.

63 OBERTHÜR (n 13) 32.

64 Ibid. 42.

65 Ibid. 43.

66 unfccc Secretariat, Conclusions and Recommendations: Third Meeting of Lead Reviewers for the Review of Biennial Reports and National Communications (3–4 March 2016), para 22.

67 unfccc Secretariat, Review Practice Guidance 2016 for National Communications and Biennial Reports of Annex I Parties (2016), 4; and unfccc Secretariat (n 66) para 23.

68 unfccc Secretariat, Conclusions and Recommendations: Thirteenth Meeting of Greenhouse Gas Inventory Lead Reviewers (29 February–2 March 2016), para 32 (emphasis added).

69 Ibid. para 31 (‘The lrs agreed that findings related to non-mandatory language are important to meeting the objectives of the review process to improve national ghg inventories and therefore agreed that they should continue to be reflected in the annual review report’).

70 unfccc Secretariat, Conclusions and Recommendations: First Meeting of Lead Reviewers for the Review of Biennial Reports and National Communications (6–7 March 2014), para 5; cf. unfccc Secretariat, rpg 2016 (n 67) 6.

71 unfccc, Decision 2/CP.17, Outcome of the Work of the Ad Hoc Working Group on Long-Term Cooperative Action under the Convention, fccc/cp/2011/9/Add.1 (2011), Annex i, para 6.

72 unfccc, Decision 19/CP.18, Common Tabular Format for ‘UNFCCC Biennial Reporting Guidelines for Developed Country Parties’, fccc/cp/2012/8/Add.3 (2012), para 1.

73 Ibid. para 4.

74 Paris Agreement, Article 13.1, emphasis added.

75 Ibid. Article 13.3.

76 See Table 1, items J and U.

77 E.g. Jutta BRUNNÉE, ‘Promoting Compliance with Multilateral Environmental Agreements’, in Promoting Compliance in an Evolving Climate Regime (n 6) 48 (‘compliance problems result not from lack of respect for international standards, but from gaps in economic, regulatory, and technical capacity’); VAN ASSELT and HALE (n 8) 4 (‘much of the time non-compliance … stems from a lack of capacity to implement commitments or unforeseen shocks that undermine countries’ efforts to comply’); and VOIGT (n 34) 167 (‘Most cases of non-compliance are related to a lack of capacity’).

78 It is far from clear that non-compliance is always, or even usually, a consequence of lack of capacity. Scholars (such as those cited in the previous footnote) have not backed up their claims on this point. None of the Kyoto Protocol Enforcement Branch’s cases arose from a lack of capacity. (For summaries of some of those cases, see Meinhard DOELLE, ‘Experience with the Facilitative and Enforcement Branches of the Kyoto Compliance System’, in Promoting Compliance in an Evolving Climate Regime (n 6) 106–120.) In the Enforcement Branch’s latest case, on the Ukraine, capacity is again not the issue. (See Compliance Committee Enforcement Branch, Preliminary Finding on Ukraine, cc-2016-1-4/Ukraine/eb (2016), paras 2, 19, and 21(a); and Compliance Committee, Written Submission from Ukraine, cc-2016-1-5/Ukraine/eb (2016), paras 12, 15, 18, 35, 39, and 45.) In my reading of the case material, states were called to appear before the Enforcement Branch because they ignored or overlooked technical rules. This is what lies at the bottom of all these cases. Similarly, the Monaco Incident discussed earlier did not involve any lack of capacity by Monaco to have its National Communication submitted on time. It should be obvious even without reading the documentation that a country like Monaco has no relevant ‘capacity’ shortcomings. Actual experience with the Kyoto Protocol’s compliance mechanism attests that non-compliance is caused by self-serving assumptions, or not taking obligations seriously, or assigning them low priority, or being less than frank about what a country is doing, or not doing, about its emissions. Even in the case of developing countries, the most advanced among them (e.g. those in the brics group), which host Olympic Games and have space-exploration programs, etc., surely do not lack capacity to submit their ndcs on time or report their emissions in accordance with the less demanding of the ipcc’s methodologies. Of course, some developing countries will genuinely lack capacity to respond to the Paris Agreement’s demands, and therefore should be assisted. But no non-compliant Annex i party has failed to comply because of a lack of capacity.

79 Paris Agreement, Article 11.5.

80 cop Decision, paras 71–81.

81 Ibid. para 84.

82 Ibid. para 86.

83 Eg Paris Agreement, Articles 13.14 and 13.15. This is in contrast to Voigt’s proposed implementation branch, which is to engage in capacity-building without providing any financial assistance.

84 See n 86, below.

85 BODANSKY, ‘New Hope’ (n 8) 28: ‘The ratchet mechanism will … continue on a five-year cycle indefinitely’. No, not indefinitely!

86 Some experts, focusing on the state of the ice caps, believe that it is already too late to avoid irreversible warming of 2°C or more: see, eg, Elizabeth KOLBERT, ‘A Song of Ice’ (24 October 2016) The New Yorker 50–61 (Greenland’s two ‘ice streams’ are melting quickly, possibly out of control); Carl-Friedrich SCHLEUSSNER et al., ‘Science and Policy Characteristics of the Paris Agreement Temperature Goal’ (2016) 6 Nature Climate Change 827, 833 (‘Recent findings from West Antarctica suggest that irreversible marine ice-sheet instability might have already been triggered for several basins’); and Warren CORNWALL, ‘Sea Ice Shrinks in Step with Carbon Emissions’ (2016) 354 (6312) Science 533, 533–34 (‘If ... current emission trends [ie, around 35 Gt CO2 eq./year] hold ... the Arctic will be ice free by 2045—far sooner than some climate models predict’). On the other hand, if we allow that there is still time to act to stay within the Paris Agreement’s Article 2 temperature limit, an argument can be mounted to show that the Agreement’s shelf-life is no greater than 3–6 ndc rounds (ie, it extends to 2035, at most to 2050). The limited shelf-life is implied by the Agreement itself. Article 4.1 states: ‘In order to achieve the long-term temperature goal set out in Article 2, Parties aim to reach global peaking of greenhouse gas emissions as soon as possible, and to undertake rapid reductions thereafter ... so as to achieve a balance between anthropogenic emissions by sources and removals by sinks of greenhouse gases in the second half of this century’ (emphasis added). The cop Decision, para 17, acknowledges that ‘the estimated aggregate greenhouse gas emission levels in 2025 and 2030 resulting from the [indcs] do not fall within least-cost 2°C scenarios but rather lead to a projected level of 55 gigatonnes in 2030 [20 Gt CO2 eq./year more than the present], and [therefore] much greater emission reduction efforts will be required than those associated with the [indcs] in order to hold the increase in the global average temperature to below 2°C’. In Article 4.1 of the Paris Agreement, quoted above, the term balance ‘essentially means that globally aggregated anthropogenic ghg emissions are required to reach zero’ at some point around 2070, with no let-up in reductions after that (SCHLEUSSNER et al., above, 830–31). Given that aggregate emissions are projected to continue to rise through to 2020 and probably through to 2030, the Paris temperature goal necessitates either that states urgently revise their projections downward and accept a ‘much more severe near-term mitigation’, or that future ‘negative emissions’ are made in even greater amounts than what current models assume to be necessary: Kevin ANDERSON and Glen PETERS, ‘The Trouble with Negative Emissions’ (2016) 354 (6309) Science 182. The concept of negative emissions is thus critical to the analysis. Even if aggregate emissions were to start falling in 2016, enormous amounts of negative emissions (‘comparable in size with the remaining carbon budget’) must be achieved, starting already in 2030, for a ‘likely’ chance of not exceeding 2°C: see, ibid., Figure 1. Anderson and Peters argue that negative-emission technologies are currently essentially non-functioning; because of this, but for reason of other drawbacks with them as well, it is unlikely that they will be able to deliver the reductions needed for a net-zero state by around 2070 from a peaking in 2016, let alone from a peaking in 2030 (ibid. 183). (On this, see also SCHLEUSSNER et al., above, 831.) Thus, Anderson and Peters warn, ‘the mitigation agenda should proceed on the premise that [negative emissions] will not work at scale’ (ibid.). If so, ndcs must be very ambitious already in their first round: emission levels of approximately 37–39 Gt CO2 eq. must be set for 2030: SCHLEUSSNER et al., above, 832. Moreover, each ndc round must be much more ambitious than the previous one, and the progression must proceed without interruption and must conclude quickly. RAUPACH et al. calculate the necessary aggregate mitigation rates (for a 2°C warming limit) as being greater than 5% per year (‘Sharing a Quota on Cumulative Carbon Emissions’ (2014) 4 Nature Climate Change 873). As they put it, the ‘biophysical realities of climate change demand solutions within decades’, involving ‘very high national mitigation targets’ combined with emission pricing and trading (ibid. 878). (The same conclusion is found in Joeri ROGELJ et al., ‘Differences between Carbon Budget Estimates Unravelled’ (2016) 6 Nature Climate Change 245, at 245 and 251.) The latest science, therefore, supports the view that the Paris Agreement must deliver significant reductions already in its first 5-year cycle and conclude the bulk of its work well before 2050, if it is not to render itself pointless. This means that, very early on, we will be in a position to know whether or not the Paris Agreement is working. It also means that, in the most optimistic rational scenario, the Paris Agreement will deliver a maximum of six ndc iterations.


E.g. Jutta BRUNNÉE, ‘Promoting Compliance with Multilateral Environmental Agreements’, in Promoting Compliance in an Evolving Climate Regime (n 6) 48 (‘compliance problems result not from lack of respect for international standards, but from gaps in economic, regulatory, and technical capacity’); VAN ASSELT and HALE (n 8) 4 (‘much of the time non-compliance … stems from a lack of capacity to implement commitments or unforeseen shocks that undermine countries’ efforts to comply’); and VOIGT (n 34) 167 (‘Most cases of non-compliance are related to a lack of capacity’).

  • Search Google Scholar
  • Export Citation

Content Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 87 87 21
PDF Views & Downloads 174 174 40