Counting the Cost of Climate Change: Hits, Misses and the Future of Environmental Compensation After the ICJ’s Advisory Opinion

[Mohit Khubchandani is an International Disputes Resolution Attorney with over a decade of experience working at the government, UN, ICJ, ILC, ITLOS and at international arbitration firms in London and Paris.

Jason Rudall is Associate Professor of Public International Law at Leiden University.]

Introduction

The ICJ has handed down 3 of its 4 judgments in contentious cases on the question of compensation since 2012, and its last two on environmental compensation since 2018, i.e. Certain Activities Case (Costa Rica v. Nicaragua) (2018) and Armed Activities Case (DRC v. Uganda) (2022) (discussed here in relation to the ICJ’s climate advisory opinion). By contrast, no inter-state climate litigation has yet been instituted at the ICJ and so the Court has had no occasion to pronounce on compensation in a climate context. This absence is of course explained in part by jurisdictional and other obstacles. To begin with, the jurisdiction of the Court—under Article 36 of its Statute— is limited to states that have: (a) either accepted it through an instrument of declaration; (b) have signed a treaty that has a compromissory clause (i.e. conferring jurisdiction to the Court; such as the UNFCCC in our context); or (c) have accepted it through forum prorogatum (i.e. acceptance on a case-to-case basis; which is unlikely in climate change cases). If this was not all, it is also necessary for an Applicant state to establish standing before the Court. In this context, obligations erga omnes of States in the context of climate litigation are still yet to be tested.

For climate litigation utopians like us—notwithstanding the increasingly rich jurisprudence developed by international human rights bodies and domestic courts—the ICJ’s advisory opinion on climate change presented a unique opportunity for the Court to pave the way for contentious cases at international and other fora. Yet, as often happens, a gap has emerged between expectation and reality. The expectation being fully-formed tools to tackle, inter alia, questions of liability, causality and quantification in climate cases, and the reality that the Opinion is only a starting point before climate justice can in its true sense be rendered to the injured parties. That said, the Opinion also represents a ‘eureka’ moment that helps illuminate the distance left to travel from expectation to reality.

Against that backdrop, this post’s aim is to highlight: (a) the extent to which the Advisory Opinion assists future courts and tribunals deciding upon compensation in climate cases (Hits), and (b) what more the Court could have said—or what now should be done—to develop a framework around environmental compensation for cases involving climate change (Misses). As we will show, the Court did not sufficiently develop metrics for environmental compensation under the law of state responsibility. Nonetheless, its Opinion provides a foundation for developing such mechanisms at international, regional and domestic levels, not least given its recognition that compensation may indeed be owed for climate-related harms.

Hits: Key Contributions to the Law of Climate Compensation

The fact that both the ICJ and ITLOS have given deference to the science-based reports of the Intergovernmental Panel on Climate Change, and identified which treaty-based regimes (see, e.g., Tigre, Bönnemann and De Spiegeleir), customary norms (including affirming the erga omnes character of climate induced harms (see, e.g., Gehring and Pezzano)) and due diligence obligations apply (see, e.g., Voigt) is significant. More specifically, several other observations made by the Court assist in this endeavour but could have been elaborated upon:

The Diffuse Nature of Climate Change and Concurrent Causes Poses No Obstacle to Establishing State Responsibility and Making Reparation

The ICJ confirmed that causation of damage “plays a role in determining reparation” but is not required for establishing state responsibility as such (para. 433). To illustrate, this means that where Country A emits hazardous levels of transboundary air pollution into Country B and fails to take diligent measures to regulate or restrict polluters operating in its territory, this failure is sufficient to constitute a breach of international law and engage the international responsibility of Country A.

The Court bolstered its analysis by finding—despite concerns in the submissions of some participants that “causation is impossible to establish […] due to the diffuse nature of climate change” (para. 434)—that damage caused through concurrent causes does not preclude reparations, while relying in particular on the Armed Activities Case (para. 435). This is a helpful point of departure for future cases.

Elaborating on What a ‘Sufficiently Direct and Certain Causal Nexus’ is in the Context of Climate Change (But Does this Standard Need to Evolve Further?)

The Court emphasises the need for an ‘in concreto’ (or case-specific, fact-based) assessment to establish causality in the context of climate induced harms (para. 438; also discussed in posts by Paddeu and Jackson, Wewerinke-Singh and Viñuales and Reetz), which helps to clarify how a court is to apply and assess the ‘sufficiently direct and certain’ causal link when it comes to diffuse harm like climate change.

However, the million-dollar question is whether the test of ‘sufficiently direct and certain’ causality needs to be adapted for climate change? The Court said this test is “flexible enough” for complex climate cases (para. 436) and did not countenance an alternative. Some scholars have argued for other techniques that speak to normative causality in relation to climate change (see, e.g., Nollkaemper) while others have elaborated on the differences between pure simple causation (i.e. a direct link between cause A and effect B); transitive simple causation (A -> leading to B -> leading to C (very important in the context of understanding chained ripple effects in environmental ecosystems)); cumulative intervention (many small causes of As leading to one big B effect (important to understand the diffuse nature of small emissions collectively leading to big environmental harms)); parallel causation (causes A and B leading to effect C independently) and complementary causation (A and B collectively causing C (which is also the case of joint and several liability)) (see, e.g., Demaria). These differences are significant because different theories of causation may serve to expand the ‘direct causality’ threshold in climate-related litigation and associated compensation claims.

The Court has previously endorsed flexible approaches to the standard of proof requirement, not least as evidence is sometimes “inaccessible” (para. 124, Armed Activities Case). Even back in 2010, Judge Greenwood categorically noted in Pulp Mills that “the nature of environmental disputes is such that the application of the higher standard of proof would have the effect of making it all but impossible for a State to discharge the burden of proof” (Separate Opinion, para. 26). This begs the question: if we can have flexible standards of evidence for environmental harm, then why not—as a corollary—causation?

A lowered flexible ‘preponderance of probability’ category of causation threshold could aid in the calculation of climate harm. And if a court or tribunal encounters insufficient or partial evidence before it, other techniques might be used to address this. For instance, to appreciate the ‘proportionate probative value’ of that evidence, recourse to ‘evidentiary discount factors’ (which although advanced in the Armed Activities Case (para. 248) by the experts, were ultimately neither rejected nor endorsed by the Court for reasons specific to that case) could facilitate the valorisation of climate harm. By way of illustration, if the evidence that a catastrophic event was caused by a climate-related x-factor is 50+ percent sufficient, then a reduced amount of compensation should be awarded (using appropriate calculation methodologies) rather than no amount being awarded if a more stringent standard is not met vis-à-vis that x-factor being the cause of the catastrophic event. There are, of course, many other techniques used in awards of environmental damage in international arbitration that could be taken as inspiration in such situations as well (see, e.g., Burlington v. Ecuador, Decision on Counterclaims (2017)).

Apportionment, Shared Responsibility and Composite Acts

The Court’s observation that reparation may be owed by multiple States, and that each State’s share of cumulative global greenhouse gas emissions—past and present—can be scientifically assessed, is welcome (paras. 429 and 452). However, the Court might have gone further by elaborating on the principles and considerations relevant to the apportionment of responsibility.

The law of State responsibility provides different mechanisms for addressing situations involving multiple contributing States but how these apply in the context of climate change remains unclear in many respects (see, e.g., de Arcos Tejerizo), particularly where a State or several States may have repeatedly emitted harmful greenhouse gasses at both lawful and unlawful levels over a period of time. Judge Nolte expressed scepticism at the role that composite (or continuing) acts could play (Declaration, para. 28) in the appraisal of individual States. While the Guiding Principles on Shared Responsibility shed some light with the development of concepts like supplementary contributions, the law of reparations needs additional clarity on equitable apportionment in the context of climate change involving multiple acts and actors to ensure courts and tribunals have the tools they need to hold those who emit harmful levels of GHGs properly to account.

Misses: Questions Left Unaddressed but not Unimportant

The International Liability Regime

The Court’s Advisory Opinion omits any discussion of the regime of international liability for lawful acts, despite its potential to address situations where wrongfulness is hard to prove but the hazardous nature of an activity is clear. Judges Yusuf and Nolte flagged this omission, with Judge Yusuf noting that such a regime could complement State responsibility by filling remedial gaps (Separate Opinion of Judge Yusuf, paras 46-7, Declaration of Judge Nolte, paras 16-7).

The International Law Commission’s Draft Articles on Prevention of Transboundary Harm from Hazardous Activities (2001) and Draft Principles on the Allocation of Loss (2006) provide a helpful starting point. These instruments promote equitable loss allocation, cooperation and accessible remedies. The framework for international liability may also be an approach that moves beyond ARSIWA’s fault-based model in a way that is more responsive to transboundary environmental harm given that such a framework would focus on harm itself and have strict liability at its heart. Indeed, the international liability regimes for oil spills and nuclear accidents typically impose strict, sometimes joint-and-several, liability on operators regardless of fault, providing a potential model for designing an emissions regime that ensures accountability and reparation for damage. The Institut de Droit International’s 1997 Resolution similarly integrates responsibility and liability, advocating strict liability for hazardous activities with limited exceptions. It also calls for presumptions of causality where harm is cumulative or long-term and linked to a sector rather than a single actor—an idea directly relevant to climate change. While the Request arguably did not require the Court to consider this regime, Judge Yusuf observed that the two regimes—one concerned with wrongful conduct, the other with actual injury—are complementary rather than mutually exclusive (Separate Opinion, para. 42).

Environmental Compensation Approaches

Perhaps the most important element in the context of environmental compensation that is conspicuously missing from the ICJ’s Advisory Opinion is any guidance on compensation approaches or methodologies. Judge Yusuf emphasised that, over 50 years ago, the Stockholm Convention called on States to cooperate in developing international rules on liability and compensation for transboundary environmental harm, and this was reiterated by the Rio Declaration in 1992 (Separate Opinion, para. 43). While this has as yet not been done satisfactorily, scholarly efforts are afoot (see Khubchandani).

In its Advisory Opinion, the ICJ endorsed its ‘Global Sum’ approach to calculate environmental compensation (para. 454), which has been the subject of academic debate. The global sum or overall valuation approach assesses the total harm caused by a wrongful act and determines reparations as a single aggregate amount, rather than calculating damages separately for each component of the harm. Experience tells us that in both of the ICJ’s environmental compensation cases (Certain Activities Case and Armed Activities Case), less than 5% of the amounts claimed by the parties (notwithstanding the weaknesses in the evidence adduced by the parties) was awarded. One can only imagine the added complexity with climate change harms.

To fully and fairly capture and compute environmental harm, an analysis on the basis of types of harm, actors, degree of fault, forum, and relevant international treaty involved, combined with other economic, social and cultural considerations (and others) is needed to design environmental compensation approaches suitable for climate and other extensive environmental damage.

To provide an example, the appraisal of climate change-induced harm and amount of compensation awarded should be different for a small island developing state on the verge of submersion due to sea-level rise (with additional considerations where cultural rights of Indigenous Peoples are at stake (as in the Torres Strait Islanders Case)) compared to that for a developed state experiencing the same degree of sea-level rise but not having the former’s special considerations. In the latter’s case, the risk of the loss of territory (or even the entire State) is likely much lower. Thus, here, the higher likelihood of loss of territory for the former State should be reflected in the environmental compensation awarded. Adding to the complexity could be the contribution of that developed State to the injury (Art. 39, ARSIWA), which may offset or mitigate the amount owed. This is not an abstract possibility because contentious climate litigation will likely target the world’s largest emitters, whose actions or omissions—whether in breach of the letter or spirit of the Paris Agreement—substantially contribute to the climate harm they experience (see, e.g., Khubchandani).

While the Court may not have expressed a view in its Advisory Opinion, it is argued that a manifestation of the ‘ecosystem services approach’ advanced by Costa Rica in the Certain Activities Case could be embraced given that it best accounts for the interconnected character of natural world. It takes into account not just the replacement value of environmental goods and services, but also intergenerational, cultural and other considerations (such as transitive simple causation (above) that deals with ripple effects and damages to other forms of the environment as a result of harm to one part of it). Therefore, rightly, we should heed Judge ad hoc Dugard’s emphasis on the hidden costs and context of environmental damage in that case (Dissenting Opinion, paras 18 and 20). One may also consider Judge Bhandari’s call for an evolution of the ARSIWA and international law in general to include punitive or exemplary damages (in line with certain municipal practices) for environmental harm as a supplemental lump-sum (Separate Opinion, paras. 16 to 21). And there are many more dimensions of and approaches to environmental damage still being discovered that courts and tribunals should be alive to (see, e.g., Leucci). Indeed, such metrics for compensation approaches may not only assist in contentious inter-state cases, but are also likely to help in investor-state and international commercial arbitration, human rights and municipal litigation and, as Judge Bhandari has suggested, any climate change-related claims commissions established under the auspices of the UN General Assembly (Separate Opinion, para. 8; see also, Tigre, Martini, Cohen and Rocha).

In short, we suggest that the ICJ’s Global Sum approach is not reasoned enough and risks significant under-compensation, particularly for climate change. It is also unlikely to be of any real assistance to future adjudicative bodies dealing with compensation for environmental or climate damage. In the short term, we need more inter-disciplinary studies to carve out bespoke approaches for particularly climate-induced harms, which can—in the longer term—be taken up by courts and tribunals.

Conclusion

The Court has articulated several key principles that will guide future climate litigation—the ‘hits’ as we have outlined above. Notably, the Court rejected the position held by many states that the diffuse nature of climate change absolves emitting states of responsibility, affirming that compensation for climate-related damage is indeed possible. Equally significant is the Court’s recognition that current scientific evidence is sufficient to establish direct causality for climate harms. The Court also endorsed standards of due diligence and environmental impact assessment, and affirmed the applicability of obligations erga omnes in addressing climate-related harms.

That said, there were also some notable ‘misses’ in the Court’s Opinion. While it is understandable that the scope of the Request was limited and did not directly address quantification methods, significant questions remain regarding approaches to compensation for climate-related harm. Greater clarity on how to calculate and allocate damages is still needed, and it is hoped that this will be addressed in future cases or through the International Law Commission’s forthcoming work on ‘Compensation for the damage caused by internationally wrongful acts’.

To tackle the dark scourge of climate harm, calculating reparations precisely may be the only practical deterrent to unfettered polluters, and we need to be creative in designing climate compensation. As the 140-page Opinion acknowledges in its concluding paragraph, climate change is a “daunting, and self-inflicted, problem” that will require illumination from “all fields of human knowledge, whether law, science, economics or any other” (para. 456). A sobering prospect, but as Judge Kittichaisaree said in the ITLOS Advisory Opinion on Climate Change, “[e]ven in the darkest places, we can find light if we only search for it” (Declaration, para. 36).