By Lidewij Nollkaemper[1]
Introduction
In December 2025, Shell initiated proceedings against the Kingdom of the Netherlands before the International Centre for Settlement of Investment Disputes (ICSID). Shell brought this case after the Netherlands decided to close the Groningen gas fields due to the gas extraction causing earthquakes, which put residents at risk. This legitimate policy response now faces a potentially billion-euro claim. While this case is not framed as a climate policy one, the same arbitration threat would hang over the Netherlands if it had phased out gas extraction for climate reasons instead. A government regulating fossil fuels for the climate faces identical legal jeopardy.
This legal jeopardy has become more imminent worldwide. As States enact and implement climate change regulations, arbitration cases brought by investors against States over climate policy regulations have surged. Such arbitral cases can concern claims of hundreds of millions, or even billions, of euros. By bringing such claims in response to climate-related regulations, investors hold considerable leverage over States. This dynamic is problematic, as it may have a chilling effect and cause States to hesitate to act in the public interest out of fear of being exposed to arbitral cases.
This blog post examines how investment arbitration tribunals should reconcile climate obligations with international investment law. It argues that, instead of redesigning or abandoning the investment law system, they should give effect to the systemic integration obligation of Article 31(3)(c) VCLT, and interpret investment law in the light of international climate law.
The tension between international climate change law and international investment law
The parallel development of climate regulation and investment arbitration is driven by legal obligations that, at face value, may collide. Under international climate law, especially since the ICJ Advisory Opinion on Climate Change, States are obliged to regulate the conduct of private actors (paras. 403, 427 of the Advisory Opinion). However, International Investment Agreements (IIAs) provide the conditions of lawful expropriation, prohibit States from engaging in unlawful expropriation, and, under the Fair and Equitable Treatment (FET) standard, as defined by the Tribunal in Tecmed v Mexico, oblige States ‘to act in a consistent manner, free from ambiguity and totally transparent in its relations with the foreign investor, so that it may know beforehand any and all rules and regulations that will govern its investments(…)’. These two sets of obligations can cause tensions, for example between phase-out regulation and expropriation.
Some scholars have concluded that this tension fundamentally threatens climate change regulation and can only be resolved by redesigning or abandoning the ISDS system as a whole. This blog post takes a different path. Though there are indeed strong arguments for redesigning the system, which has institutional shortcomings, opting for either fundamental reform or abandonment would create immediate practical problems that cannot wait for structural change. The volume of climate-related arbitration is increasing, and tribunals will be called upon to decide such disputes regardless of whether broader reform proceeds. International investment law works not only against climate change objectives but is also instrumental for achieving them, as it can protect critical investments in clean energy industries and promote decarbonisation. Wholesale system redesign will have uncertain outcomes and risks that this protection is lost.
Given that the investment arbitration system is here to stay and, moreover, has beneficial effects, a different solution to the tension between climate obligations and investment protection obligations is called for. Rather than overhauling the system, tribunals must deploy an interpretative tool already mandated by law: systemic integration. Tribunals should use interpretative tools available within the existing investment law framework to allow States to give effect to their climate change obligations without sacrificing investor protections for legitimate projects.
Systemic integration as a legal obligation
Systemic integration is not just a good idea. All judicial bodies interpreting treaties are subject to the rules of treaty interpretation reflected in Articles 31 to 33 of the Vienna Convention on the Law of Treaties (VCLT) that reflect customary international law, as the ICJ has repeatedly confirmed. Article 31(3)(c) VCLT requires that, when interpreting any provision of a treaty, ‘any relevant rules of international law applicable in the relations between the parties’ must be taken into account. Systemic integration is a mandatory step under customary international law.
There are two important elements of Article 31(3)(c) VCLT. First, ‘relevant rules’ need not share the treaty’s subject matter. Climate law counts even when interpreting investment treaties. Second, those rules need not bind all parties to be considered ‘applicable in the relations between the parties’, as clarified in the Advisory Opinion by the ICJ on the right to strike under ILO Convention No. 87 of May 21, 2026 (paras. 90-98). Rather, a rule may be ‘applicable in the relations between the parties’ where it expresses ‘their common understanding regarding certain provisions of the treaty under interpretation’. This may be presumed where the rules in question have been so widely adopted that all parties can be considered to have implicitly accepted them.
This is directly relevant to the climate context. The Paris Agreement has 194 parties, and the UNFCCC has 198, covering almost every State that has concluded an investment treaty. Under the standard articulated in the Right to Strike Advisory Opinion, both treaties express the common understanding of the parties regarding the legal framework that tribunals must apply. The customary duty to prevent significant transboundary environmental harm likewise satisfies the requirement of being widely adopted. The rules of international law contained in the climate regime are therefore ‘applicable in the relations between the parties’ to IIAs within the meaning of Article 31(3)(c) VCLT, and tribunals are required to take them into account when interpreting IIA provisions.
Implications of systemic integration
Systemic integration in relation to climate change operates through two key standards. First, under the police powers doctrine, non-discriminatory regulatory measures taken in the public interest do not amount to compensable expropriation, even where they substantially affect the value or use of an investment, as confirmed in Methanex v the United States. There is a strong argument that climate measures are relevant for demonstrating that the regulatory measure in question is a genuine exercise of regulatory authority for a public purpose. The ICJ Advisory Opinion confirms that the regulation of fossil fuel production, licensing, and subsidies is part of the customary duty to prevent significant transboundary environmental harm. If a State justified a regulatory measure on this basis, a tribunal that would reject that justification would contradict the Court’s reasoning. In Saluka Investments v Czech Republic, the Tribunal acknowledged that systemic integration is a part of treaty interpretation. Through systemic integration, the Tribunal integrated customary international law with the interpretation of the relevant treaty, specifically falling back on customary international law to assist with the interpretation of the police powers doctrine.
Second, systemic integration is also relevant for interpreting the FET standard. In Philip Morris v Uruguay, the Tribunal interpreted the FET standard in light of the WHO Framework Convention on Tobacco Control and the public health obligations of Uruguay, concluding that the regulatory measures at issue were a bona fide exercise of regulatory authority. In relation to climate change, particularly the legitimate expectations dimension of the FET standard is relevant. The expectations that investors can have regarding regulatory stability changes when international climate law obligations become stronger. Judge Nolte, in his Declaration to the Advisory Opinion, argued that the customary duty to prevent significant environmental harm became applicable to GHG emissions when their risks were generally understood and recognised, which he locates in the late 1980s.
Certainly, for investments made after the Advisory Opinion in July 2025, investors cannot credibly claim they legitimately expected States to refrain from adopting new regulatory measures. Even for investments made between the Paris Agreement and the Advisory Opinion, investors cannot be deemed to have expected no regulation, as, in that period, nationally determined contributions (NDCs) had already been made and were known and foreseeable. For investments made before the Paris Agreement, investors could make a stronger case for the legitimate expectations that no climate-driven regulation might be forthcoming.
Systemic integration cannot resolve every conflict. Where treaty text is genuinely clear, for example the 2021 Canada Model BIT, where a mandatory three step test regarding expropriation was laid down, tribunals cannot read in climate exceptions that the parties never agreed to. In such cases, other techniques (fundamental norms such as jus cogens or lex posterior) may be required. If these principles did not suffice to protect climate change regulation, States may have to renegotiate investment treaties.
Systemic integration in practice
Tribunals have begun applying systemic integration, though inconsistently. In Urbaser v Argentina, the Tribunal interpreted the relevant BIT in light of the human right to water, considering whether such rights could ground obligations binding on the investor as well as the State. While the Tribunal ultimately did not find a corresponding investor obligation, its reasoning, and the framework it used, affirmed that human rights law is to be applied following Article 31(3)(c) VCLT.
Further, Philippe Sands’ dissent in Odyssey Marine Exploration v Mexico illustrated how Article 31(3)(c) VCLT can operate in an investment dispute concerning environmental measures. In assessing whether Mexico’s denial of an environmental permit for seabed phosphate mining constituted arbitrary conduct under Article 1105 NAFTA, Sands argued that the standard of arbitrariness ‘falls to be interpreted and applied by reference to the relevant provisions of NAFTA, and the relevant or applicable rules of international law’. Invoking Article 31(3)(c) VCLT, he stated that the assessment of arbitrariness must take account of customary international law obligations concerning the protection of biodiversity and the marine environment, including the obligation under Article 208 UNCLOS to prevent pollution arising from seabed activities and the precautionary principle.
However, though the principle of systemic integration under Article 31(3)(c) VCLT is an obligation by which tribunals are bound, the limited practice so far suggests that tribunals often ignore it or apply it in a superficial way.
Rockhopper v Italy is illustrative. When Italy banned offshore fossil fuel exploration, the Tribunal found unlawful expropriation. The Tribunal based its finding of unlawful expropriation on Italy’s decision-making process. The Tribunal thus essentially classified the dispute as one concerning the process behind the measure, and not one concerning climate, which resulted in Article 31(3)(c) VCLT not being applied to include climate obligations.
What explains the gap between law and practise?
Three factors may explain this gap between law and practice. First, tribunals resist stepping beyond investment law and prefer to stay close to the subject matter of the dispute. This was voiced by the Tribunal in Grand River v United States, where the Tribunal stated that, though it understood that the obligation to ‘take into account’ other rules of international law required it to respect the VCLT’s rules governing treaty interpretation, it ‘does not understand this obligation to provide a license to import […] legal elements from other treaties.’ Such reluctance to interact with other regimes may be fuelled by a fear of annulment, or of exceeding their mandate.
Second, tribunals have discretion regarding arbitral proceedings. Although a tribunal must address every question submitted by the parties, it is not required to engage with every argument in detail, or to explain the entire reasoning in depth.
Third, the willingness and ability of tribunals to engage in systemic integration depend largely on State argumentation. Tribunals are unlikely to use climate change obligations unless the respondent State raises such obligations as part of its defence. If the respondent State downplays climate claims, treating the measure as domestic policy to avoid political backlash at home, tribunals will not bring up the argument. Even where States do raise climate arguments, the quality and specificity of the argumentation matter. What is required is the identification of specific obligations under specific instruments, for instance, the customary duty to prevent significant transboundary environmental harm.
None of these barriers is insurmountable. The ongoing development of international law, notably through the Advisory Opinions on climate change and the right to strike, the fact that States are starting to rely on climate change argumentation, the recognition of systemic integration in the dissenting opinions, and the role of experts and amicus curiae submissions can all assist in eroding these barriers. Systemic integration does not require a complete overhaul; it deploys tools already embedded in the law.
Conclusion
In conclusion, tribunals are obliged under international law to integrate climate change law in their interpretation of IIAs. Through the expropriation and FET standards, such integration protects both climate regulation and legitimate investment. The barriers to a fuller application of systemic integration are real, but this blog post has demonstrated that these are surmountable. Parties and tribunals need not wait for systemic reform. They should do what the law already requires, and thus allow for regulation of private sectors, and thereby help to reduce GHG emissions. Doing so would go a long way to reducing the threat posed by private parties who aim to challenge legitimate and necessary climate change regulations.
[1] Lidewij Nollkaemper is an LL.M. graduate in Public International Law from the University of Amsterdam, and an LL.M. candidate in Arbitration and Dispute Settlement at the University of Hong Kong. This blog post is inspired by her Master thesis: Climate Regulation and Investment Protection: Towards a Harmonious Interpretation