Climate Change as a Security Interest: A Novel Defense in Fossil Fuel Investment Arbitration

I. Introduction

Fossil fuels are embedded in modern society. The public and private sectors annually invest billions of dollars in finding, extracting, refining, transporting, and combusting fossil fuels,1 creating an energy infrastructure the global economy relies on. Much of this investment happens across borders as fossil fuel reserves are dispersed around the world.2 Through treaties, strong protections are in place to ensure these foreign investments remain intact across the often decades-long lifetimes of the fossil fuel assets. These protections effectively prohibit state action that damages the fossil fuel investments—including measures to address climate change by targeting fossil fuel industry activity. Foreign investors are increasingly invoking treaty protections to counter state climate action, winning huge awards in arbitration while also stifling regulation targeting their assets. The current investment protection regime has become a major barrier to global climate change mitigation efforts.

This Comment analyzes the standard that has developed around the security exception, a common provision in investment agreements allowing a state to override its treaty obligations to foreign investors in order to protect its essential security interests, to conclude that this exception could be used to defend state action taken against fossil fuel investments for the purpose of reducing greenhouse gas (GHG) emissions, thereby mitigating climate change.

A. Investor-State Dispute Settlement (ISDS)

A critical mechanism protecting cross-border investment is Investor-State Dispute Settlement (ISDS).3 Enshrined in thousands of international investment agreements (IIAs), ISDS “allow[s] foreign investors to resort to international arbitration to seek monetary damages from the host State for violating . . . substantive rights” guaranteed within those IIAs.4 These agreements are intended to promote foreign direct investment (FDI) by offering strong protections for investors.5 Substantive protections often include national/most favored nation (MFN) treatment, prohibition of expropriation, and fair and equitable treatment (FET).6 If a government violates these commitments to foreign investors, it can be held liable for damages by an arbitral tribunal.

Though the first ISDS case was initiated in 1987, most of the cases have arisen more recently.7 73% of all ISDS cases, which total over 1,400, have been filed since 2010.8 These cases are typically decided by a tribunal of three arbitrators appointed by the parties.9 “Their decisions are not bound by precedent, and their legal reasoning is not subject to correction and harmonisation through appeals. Decisions are final and binding.”10 Two treaties in particular have solidified the ISDS mechanism as prevalent and binding: the ICSID Convention (the Convention on the Settlement of Investment Disputes between States and Nationals of Other States, entered into force in 1966)11 which is ratified by 158 Contracting States,12 and the New York Convention (the United Nations Convention on the Recognition and Enforcement of Foreign Arbitral Awards, entered into force in 1958)13 which has 172 Contracting States.14 Most countries and FDI are covered by these treaties that guarantee the finality of ISDS. 

Most ISDS cases are administered at ICSID, the International Centre for Settlement of Investment Disputes.15 It was established by the ICSID Convention as “an independent, depoliticized and effective dispute-settlement institution” under the World Bank Group in Washington, DC.16 The ICSID database holds the majority of concluded and ongoing ISDS cases.17 It should be noted, though, that ISDS allows for confidentiality. The ICSID Convention requires party consent to publish awards publicly; basic case information may be available in ICSID or UN databases, but the case-related documents are only available if the parties agree.18 The current rules assume party consent to publish awards unless a written objection is made within 60 days of award issuance.19 The practice of ISDS often happens behind closed doors.

B. Fossil Fuel Investors and ISDS

A significant portion of ISDS cases involve companies in the fossil fuel industry. A 2021 report found that “the fossil fuel industry is the most litigious industry in the ISDS system by number of cases, accounting for almost 20% of the total known ISDS cases across all sectors.”20 These cases primarily derive from the large quantities of FDI directed to fossil fuel exploration and extraction, or upstream activities.21 In 2024, energy and gas suppliers announced more greenfield projects, which are constructions on undeveloped land, by value than any other major industry worldwide, a total of $273 billion.22 International project finance deals for oil and gas (O&G) totaled $46 billion.23 Global investment in fossil fuels overall was more than $1 trillion in 2023.24 Most fossil fuel arbitrations involve the O&G industry, and most are related to upstream investments in exploration and extraction.25

Data from the publicly available ISDS cases indicate that claimant corporations in the fossil fuel industry are highly successful in using ISDS to win large awards. 26 Fossil fuel investors win 72% of their cases at the merits stage, and their average award amount is over $600 million, five times higher than the average award in non-fossil fuel cases.27

These fossil fuel ISDS cases are more frequently being initiated in response to “climate-related measures” undertaken by states to mitigate GHG emissions.28 The IIA protections investors enjoy, and the long-term nature of their fossil fuel investments, can conflict with national emission reduction goals: “to achieve the Paris Agreement goal of limiting the rise in global average temperatures to 1.5°C, states need to prevent the exploitation of many known fossil fuel reserves.”29 Given that combustion of fossil fuels is the largest contributor to climate change,30 “the potential for conflict between the climate regime on the one hand and existing investor protection on the other entails major consequences for the development of climate policies.”31 Fears of liability via ISDS have already deterred state climate action: France weakened its fossil fuel phase-out plan after threats of “a billion-dollar arbitration claim” from Canadian oil company Vermillion in 2017,32 and both Denmark and New Zealand redesigned their O&G phase-out plans in part “to reduce the likelihood of ISDS claims arising from existing projects.”33 This regulatory chill has been identified by the Intergovernmental Panel on Climate Change (IPCC) as a barrier to climate action.34

In response, some countries have attempted to exclude themselves from the jurisdiction of investment arbitration tribunals. The EU notably withdrew from the Energy Charter Treaty (ECT) in 2024 because the treaty was “outdated compared to the climate ambition at EU and international level.”35 The ECT was a multilateral agreement regulating energy investment, and its language protected long-term fossil fuel investments by “allow[ing] companies headquartered in any member state to sue the government of another member if it harms their existing energy investments.”36 The mismatch between the ECT and EU climate goals came to the fore in 2021 when German companies RWE and Uniper brought multi-billion-dollar claims under the ECT against the Netherlands for its post-2030 ban on coal-fired power plants.37 Canada also declined to sign the ISDS provision of the United States-Mexico-Canada Agreement (USMCA), having defended “at least 44 claims” under its predecessor the North American Free Trade Agreement (NAFTA).38

However, even if countries exit treaties that guarantee the ISDS mechanism to investors, or update agreements to remove ISDS provisions, the investment arbitration system will persist. This is because thousands of IIAs currently in force with ISDS provisions include sunset clauses in their text. Sunset clauses “are an almost universal feature in bilateral investment treaties” (BITs) and multilateral treaties.39 These clauses extend treaty protections for a defined period following the termination of the treaty.40 The ECT, for example, has a 20-year sunset period,41 meaning parties who made fossil fuel investments in certain EU countries before 2024,42 the year the EU left the ECT, can still bring claims in ISDS alleging ECT violations by the host country until 2044. Even if these thousands of IIAs are improbably rescinded or amended, their protections will nonetheless persist for decades. 

The success rate of these fossil fuel investment claims is a looming, long-term threat for governments attempting to enact climate policy by targeting fossil fuels. Managing this liability risk is a pertinent need for states. Though protections for investors are strong, states do have several tools available to defend their right to enact regulations aimed at limiting fossil fuel industry activity.

C. The Right to Regulate: Exceptions in Investment Treaties

States agree in IIAs to extend substantive protections to foreign investors, but they also retain some leeway to take action that violates these protections: this is the “right to regulate.”43 The right to regulate, which exists in customary international law as “the doctrine of necessity,”44 is defined as “the legal right exceptionally permitting the host state to regulate in derogation of international commitments it has undertaken by means of an investment agreement without incurring a duty to compensate.”45 Protections of this right of a state to pursue legitimate regulatory activity has become more commonplace in IIAs since the 1990s.46 The right rests on and seeks to solidify the fluid “police powers” of the state, which are “the ensemble of its sovereign powers relating to public policy, including the maintenance of public order, the protection of public health and the environment, and taxation.”47 The limits of these police powers, though, are unclear even after a century of domestic and international investment case law.48 One of the ways treaty drafters have attempted to preserve the right to regulate is to include exception clauses in the IIA.49

Exception clauses allow a state “to regulate subject to conditions designed specifically to limit the circumstances in which investors will lose their protection.”50 Under circumstances agreed upon beforehand, states can “deviate from their [IIA] obligations.”51 These clauses can be narrow in scope, such as carve-outs and reservations targeting specific sectors, or quite general, such as the security exception.52

The latter category, general exceptions, have become more popular in IIAs.53 They are often modeled on Article XX of the General Agreement on Tariffs and Trade (GATT 1994)54 and Article XIV of the General Agreement on Trade in Services (GATS).55 The exception affirms “the right of Members to pursue objectives identified in the paragraphs of these provisions even if, in doing so, Members act inconsistently with obligations set out in other provisions of the respective agreements, provided that all of the conditions set out therein are satisfied.”56 For a state action to qualify under such an exception, it must “fall[] within the scope of one of the subparagraphs”57 of the clause, and there must “be a sufficient nexus between the measure and the interest protected.”58

Successful invocation of a treaty-specific security exception has the same requirements as the GATS general exception: the state action must (1) fall within the relevant scope of the subparagraphs of the exception and (2) meet the measure-interest nexus.59 Article XXI of GATT serves as a model security exception,60 reading: 

Nothing in this Agreement shall be construed

(a) to require any contracting party to furnish any information the disclosure of which it considers

contrary to its essential security interests; or

(b) to prevent any contracting party from taking any action which it considers necessary for the protection of its essential security interests […]61

The key terms wrestled with in the case law, which ultimately determine whether the state action is within the bounds of the security exception, are “essential security interests” and “it considers necessary.” The state action must fall within the scope of protecting essential security interests, thereby meeting the relevance requirement, and that action must have been considered necessary to protect those interests, thereby bridging the measure-interest nexus. 

Invoking the security exception could help states win certain ISDS cases brought against them by fossil fuel investors claiming violations of treaty obligations. The current defenses states are presenting are failing to persuade tribunals in most cases.62 Utilization of the security exception in the context of climate change is a novel approach for states to avoid liabilities while pursuing emissions reduction goals. 

To demonstrate the viability of the security exception defense in the climate context, it is helpful to understand how states are currently defending their climate policies. Each dispute has an inherently unique fact pattern, but identifying common defenses and how tribunals have interpreted relevant clauses, obligations, exceptions, bodies of law, and state action contributes to the creation of a standard, or a suggestive precedent, that should inform a tribunal’s decision if the security exception is invoked in cases against fossil fuel investors. 

This Comment proceeds in three main sections: Part II, on Climate-Related ISDS Cases, Part III, on the Security Exception, and Part IV, on Climate Change as a Security Issue. Part II is an overview of the current landscape of climate-related ISDS cases involving fossil fuel assets. The analysis classifies the cases into Permitting cases and Explicit Phase-Out cases. Since the cases are few in number, significant detail is provided on both sides of the dispute, along with the tribunal’s method in determining the award. These cases reference one another frequently, indicating the influence past cases have on tribunals despite investment arbitration not being bound by precedent. This section contextualizes the nature of climate-related fossil fuel disputes to suggest a role for the security exception in state defenses. 

Part III develops the security exception standard. The first portion uses case law to demonstrate the types of interests that have successfully qualified under the security exception. The second portion discusses how tribunals have interpreted the text of the exception itself, which can differ among IIAs. This section parses out the standard of review tribunals use to validate a security exception invocation under a self-judging clause versus non-self-judging clause, a distinction that arises from the specific exception’s phrasing. 

The final section of the Comment, Part IV, describes how a state can frame its climate action as an action acceptable under the security exception. Successfully doing so would disencumber the state from liability owed to foreign fossil fuel investors for treaty violations. The security exception is generous so long as the state can meet its generally low burden. Winning acceptance from a tribunal that mitigating climate change is an essential security interest, which this Comment argues is feasible, could protect states from owing billions as they pursue necessary climate action.

II. Climate-Related ISDS Cases: Fossil Fuel Assets

A. Classifying the Relevant Cases

Out of the bulk of existing ISDS cases, a handful can be categorized as “climate-related ISDS cases.”63  These cases “relate directly to the introduction, withdrawal or amendment of a policy measure explicitly developed to meet a country’s climate goals” though they “do not always contain explicit references to climate change.”64 The climate-related ISDS cases that are of relevance here are “stranded asset claims”65 in which investors’ fossil fuel assets “turn out to be worth less than expected as a result of changes associated with the energy transition.”66

Fossil fuel assets can become stranded due to various governmental actions. One way is through action taken against a specific asset, usually “involving claims of unfair treatment during individual, project-specific environmental processes.”67 These are Permitting cases. Another way is through actions of general application, “including bans, limits, or moratoria on high emitting activities.”68 These are Explicit Phase-Out cases. Though the former are currently more prevalent in ISDS, the latter are expected to proliferate “since energy supply side measures focused on fossil fuels are becoming more common.”69   

This Section will first examine Permitting cases, completed and ongoing, by discussing their factual background, legal issues, and party arguments. An analysis of these cases indicates how tribunals have approached the core issues underlying these disputes and the positions taken by investors and states, which is helpful for determining how a tribunal may react to the use of the security exception to defend state climate action. Explicit Phase-Out cases will be discussed afterwards—the case law there is scant, in part because few states have passed laws mandating fossil fuel phase-outs, but given the impending prevalence of such policies, understanding the landscape of existing disputes is useful. 

The documents and decisions from nonconfidential cases, plus public knowledge about high-profile confidential cases, indicate that the security exception has never been used to defend state action. This Comment argues that it could. 

B. Permitting Cases

1. Rockhopper v. Italy

Rockhopper v. Italy is one of two nonconfidential climate-related Permitting cases that an ICSID Tribunal has ruled on.70 It demonstrates the types of circumstances and rationales under which state action against foreign fossil fuel investment could be justified. Rockhopper, a British upstream O&G company, brought this case seeking compensation from Italy for violating the ECT in respect to its investment in the Ombrina Mare O&G field in the Adriatic Sea off Italy’s coast.71 This violation, Rockhopper alleged, occurred when Italy rejected its production permit application in January 2016.72 The rejection stemmed from a 2015 law banning O&G “research, prospection and exploitation in waters within a 12-mile limit of the Italian Peninsula.”73

The Italian government had granted Rockhopper a six-year exploration permit for the Ombrina Mare field in 2005.74 After conducting various exploration activities including drilling exploratory wells, which required an environmental impact assessment (EIA), Rockhopper found oil and applied for a production concession in 2008. Italy informed Rockhopper it would “begin the procedure to grant the concession” in 2009.75

In 2010, though, Italy passed Decree 128/2010 banning new offshore drilling projects within twelve miles of the coastline.76 This law was “for the purposes of protecting the environment and the ecosystem, within the perimeter of marine and coastal areas.”77 The details were unclear as to how this law would affect pending production concession applications, so in 2012, an exception was added “exclud[ing] from its scope those who already had started a production concession application at the time of Decree 128/2010.”78 This meant that Rockhopper’s application could still have been approved. Then, in 2015, “the exception for pending applications, which had been made in 2012…was removed.”79

Rockhopper claimed that Italy’s denial of the production permit constituted a violation of “impairment, fair & equitable treatment, and unlawful expropriation,” the latter of which would breach Article 13 of the ECT.80 That provision of the ECT81 provides four exceptions under which expropriation can happen. Rockhopper claimed violation because Italy’s expropriation of its investment violated three of the exceptions: it was “(a) not undertaken for a public purpose; (b) discriminatory; and (c) without payment of prompt compensation.”82

Italy countered by arguing that Rockhopper’s extraction business had “never started and had never been authorised,” thus could not have been expropriated, and that “reasonable regulatory measures that a State passes to pursue societal policies without discriminating among its addresses constitute the legitimate exercise of police powers and, therefore, any economic impact that they might cause on investors is not compensable.”83 Because the goal of the offshore drilling ban was “the protection of the environment,” Italy argued it could “justify its expropriation on the basis that it was acting in the public interest.”84

The Tribunal sided with Rockhopper. It found that Rockhopper “had a right to be granted a production concession”85 because Italy approved Rockhopper’s EIA report.86 Under Italian law, a production concession is granted within fifteen days of the Ministry of Economic Development’s receipt of the environmental compatibility decree, which in this case had occurred on August 14, 2015.87 Once this approval happened, “the Rubicon was crossed insofar as the right to the subsequent grant of the production concession was concerned.”88      

The production concession could have been lawfully expropriated if Italy had met the four requirements of Article 13 of the ECT; however, the Tribunal found Italy’s police powers/public interest environmental protection argument, the first requirement under Article 13, to be pretextual.89 The environmental concerns around extracting O&G from Ombrina Mare likely existed from the beginning, and were not flagged in the EIA, so the Tribunal concluded “[t]he more likely reason for the position taken by the Respondent…is the political and civic engagements as discussed earlier in this Award.”90 Local protest in response to the approval of Rockhopper’s EIA was “enormous,”91 and evidence indicated that the 2015 law removing the exception for pending production permits was politically motivated to resolve a “power struggle” between the federal and regional governments.92 It was therefore concluded that the denial of Rockhopper’s permit did not stem from the public interest. This meant Italy unlawfully expropriated Rockhopper’s investment.

Italy failed to defend itself from liability because it had already approved Rockhopper’s concession. The last-ditch attempt to deny the approval was driven by political motives, the Tribunal found, not the “public interest.” The Tribunal exercised significant discretion in concluding for itself why the 2015 Italian law was passed. 

This ruling presents a few key points. One is the challenge presented by the long-term nature of fossil fuel investment. Because these assets have long lifespans, what one government administration approves can bind future administrations. Italy owed Rockhopper over €184 million in 202293 because of what its government from 2005–2015 had done. 

Next is the Tribunal’s rejection of Italy’s public interest argument. This was decided because of perceived political motives underlying the passage of the 2015 total ban on new offshore drilling projects. Essentially, the Tribunal found that the measures Italy took were not truly taken to protect its environment or public interests. But what if there was no political motive? Would a Tribunal rule differently if the state truly acted, in good faith, against a fossil fuel asset for environmental protection purposes in the public interest? This situation arises in the other nonconfidential climate-related permitting case, Lone Pine Resources v. Canada. 

2. Lone Pine Resources v. Canada

Lone Pine, an American O&G company, sought compensation for Canada’s alleged breach of NAFTA Article 1110(1) and 1105(1),94 a breach that arose from “the Government of Québec’s revocation of nine exploration permits [for fracking for oil and gas] held under the St. Lawrence River.”95 Forest Oil, Lone Pine’s parent company at the time, acquired exploration permits from Quebec in 2006 on land adjacent to the St. Lawrence River and areas under the river itself.96

In 2009, Quebec began conducting EIAs on O&G development in the Gulf of St. Lawrence and its estuary basin.97 Its findings indicated these areas were “ill-suited to oil and gas development,”98 and coupled with local opposition to shale gas operations, resulted in Quebec prohibiting “all oil and gas exploration and exploitation activities in the maritime Estuary and northwestern Gulf of St. Lawrence” in 2010.99 This moratorium was expanded in 2011 to cover the St. Lawrence River as well through Bill 18, An Act to Limit Oil and Gas Activities.100 Under the new law, Quebec revoked nine of Lone Pine’s exploration permits that “were completely situated under the St. Lawrence River,” and reduced the areas of twenty permits that “were partially situated in the St. Lawrence River”.101

Akin to Rockhopper, Lone Pine alleged that its investment was unlawfully expropriated without compensation, and that Bill 18 “was not passed for a public purpose, and neither was it a valid exercise of Respondent’s police powers.”102 They estimated the appropriate relief for damages incurred was CAN$250 million.103

Canada asserted that “there is no expropriation because the passage of Bill 18…was a valid exercise of Québec’s police powers.”104 For policies passed for the public’s welfare, Canada asserted that “a tribunal’s only task is to decide whether a claimant has proven that the State’s actions ‘exhibit a gross denial of justice, manifest arbitrariness, blatant unfairness, a complete lack of due process, evident discrimination, or a manifest lack of reasons.’”105 It argued that Lone Pine failed to prove an absence of good faith in the passage of Bill 18 because “it has not elaborated on what those political reasons were” that it claimed were the true motive behind the passage of Bill 18.106

Lone Pine argued that the revocation of its exploration permits under Bill 18 violated the FET standard to which Canada was bound under NAFTA Article 1105.107 It claimed Quebec revoked “valid exploration permits without a valid basis,” and it challenged “Québec’s attempt to justify Bill 18 with environmental and scientific bases that did not exist.”108

The Tribunal disagreed with Lone Pine. It found no breach of the minimum standard of FET,109 “acts or omissions that are manifestly or demonstrably arbitrary, grossly unfair, inherently unjust, or idiosyncratic.”110 Bill 18 was passed with “proper democratic procedures,”111 and the Tribunal afforded “a high measure of deference…to the right of the host State to make regulatory changes in light of the public interest.”112 It found no “disconnect between the impugned Act and the objective of protecting the fluvial environment under the St. Lawrence River,” for the FET minimum standard is a high one.113

3. Comparing Rockhopper and Lone Pine

In both Rockhopper and Lone Pine, the state passed a law barring upstream fossil fuel activities for the explicit purpose of environmental protection. Both claimant fossil fuel investors alleged the regulations breached treaty obligations because they were not actually enacted to protect the environment. The Tribunal in Rockhopper found the stated purpose of the new offshore drilling projects ban to be pretext for political motives, and ruled that Italy owed compensation for expropriating Rockhopper’s assets. The Tribunal in Lone Pine found no evidence that Quebec’s ban on O&G operations in the St. Lawrence River was for any reason other than environmental protection, and ruled that Canada had not breached its treaty obligations. There were other issues and specifics in each case that contributed to their opposite results, but the key takeaway is that the Tribunals looked for good faith intention in determining whether the regulations were within the state’s right to regulate. 

If a state attempts to use its permitting discretion to block fossil fuel industry activity, and the fossil fuel investors demand compensation, the state’s defense will be much stronger if it acted truly for the public interest rather than for political reasons. Tribunals have exercised broad discretion in making this determination, for there is no clear demarcation defining when a law was passed for political reasons rather than for public interest reasons. Questions remain as to what “political reasons” actually are, and how influential those reasons must be to the passage of the law for the tribunal to rule that the purpose of its passage was political. This is a tricky decision that two tribunals could rule differently on. For a state to argue its action was undertaken to mitigate climate change, which is ostensibly not only in the public interest but also is an essential security interest, it would need to make concerted efforts to remove all doubt that the action was taken for political rather than public purposes. 

4. Ongoing cases: Ascent Resources v. Slovenia, Ruby River Capital v. Canada

Ongoing climate-related Permitting cases show that the same disputes and arguments found in Rockhopper and Lone Pine continue to arise. In Ascent Resources v. Slovenia, the London-based upstream O&G company alleges its investment in the Petišovci O&G field was unlawfully expropriated by Slovenia in violation of the ECT and UK-Slovenia BIT.114 Other alleged breaches include the FET standard and discriminatory treatment.115 Slovenia required Ascent to provide an acceptable EIA before it could frack, which Ascent says was neither a legal requirement nor a decision supported by Slovenia’s expert bodies.116 Then, in May 2022, Slovenia completely banned low-volume fracking. Ascent alleges this “specifically targeted” it and was the result of “a populist campaign carried out by Slovenia against the Investors and their Investment.”117 The damages claim is for €656.5 million.118

Climate change concerns explicitly arise in Ruby River Capital v. Canada, an ongoing dispute regarding a natural gas pipeline and liquefaction complex in Quebec.119 The American investors claim Quebec and the Canadian federal government backtracked on supporting the project by changing the environmental review process at the last minute to require “(i) that the Project made a ‘positive net contribution’ to global GHG emissions reductions; (ii) that the Project ‘promoted energy transition’; and (iii) that the Project achieved ‘social acceptability’.”120 This standard, it claimed, was “manifestly arbitrary and discriminatory.”121 They claim the government’s concerns over GHG emissions and beluga whales were “pretextual and discriminatory”122 because (1) other industrial projects in the region were approved that would disturb the whales,123 (2) Canada and Quebec were supporting liquified natural gas projects before,124 and (3) other high-emitting projects like a vanadium processing facility and oil terminals had been approved by Quebec in 2019.125 Ruby River is seeking over $588 million in damages for breaches of NAFTA (national treatment, MFN treatment, minimum standard treatment, expropriation) and USMCA/CUSMA.126

In all these Permitting cases, countries are faced with the challenge of rescinding approvals given in the past for fossil fuel infrastructure without incurring liability for violating treaty obligations. Investors are consistently accusing governments of using environmental concerns as a pretext to discriminate against them. The U.S. faced this accusation in TC Energy and TransCanada Pipelines v. USA, a high-profile dispute about the Keystone XL oil pipeline.127 That case was dismissed in July 2024 because of a jurisdictional issue regarding NAFTA, but the U.S. was facing liabilities exceeding $15 billion because of flip-flopping by successive presidential administrations as to approval of the pipeline.128 The fact that many governments are in the process of reversing decades of policy support for fossil fuel investment means there is a need for strong arguments to defend their right to do so.  

C. Explicit Fossil Fuel Phase-Out Cases

In Explicit Phase-Out cases, there is less contention about the underlying motives driving state action against fossil fuel assets. There is no discrimination specific to certain assets. These are policies of “general application” applying to all targeted fossil fuel assets.129 Examples include the Netherlands’ 2019 law requiring a coal phase-out by 2030 and Germany’s 2038 coal phase-out law.130 Both policies were enacted to help meet domestic climate targets,131 and both resulted in high-profile ISDS cases brought by fossil fuel investors.

Two German energy companies, RWE and Uniper, brought cases against the Netherlands in 2021 for its coal phase-out plan.132 Much of both disputes is confidential, but extensive press coverage and their spillover into Dutch and German courts allowed the public to learn details of the claims, like that RWE was seeking over $1.4 billion in compensation.133 Similar to TC Energy, the case involving the Keystone XL Pipeline, both of these arbitrations were dismissed on jurisdictional issues rather than the substance of the policy itself—the German Supreme Court ruled an award could not be enforced because the Tribunal lacked jurisdiction over intra-EU disputes,134 and Dutch courts rejected petitions for compensation from the investors.135 Swiss company AET brought a case against Germany in 2023 for its coal phase-out.136 AET is alleging an ECT violation,137 but further details are confidential and the case is currently pending. 

Westmoreland v. Canada is a nonconfidential case arising from a coal phase-out plan. Westmoreland, an American coal company, brought claims against Canada three times regarding Alberta’s 2030 coal phase-out policy; the first two were interrupted by bankruptcy and dismissed for jurisdictional issues.138 The substance of the third claim was the same as those before, that Alberta and Canada violated their NAFTA obligations of national treatment, minimum standard of treatment, and expropriation.139 Alberta’s coal phase-out was explicitly for the purpose of mitigating climate change, as it was made law in 2015’s Climate Leadership Plan.140

Westmoreland’s main complaint was that Alberta compensated the coal-fired power plants affected by the climate legislation but not Westmoreland,141 which owns coal mines “adjacent to and in conjunction with a power plant so that the coal can be delivered to the power plant economically.”142 Canada responded that the Tribunal lacks jurisdiction, but on the substance, asserted that paying coal mine owners “would not have advanced that legitimate public policy objective” of reducing emissions from electricity generation while maintaining grid reliability.143 It reiterated that the coal phase-out is a “non-discriminatory regulatory measure[] designed and applied to protect legitimate public welfare objectives: protection of the environment and of human health.”144

The Tribunal ruled in favor of the state.145 It found the statute of limitations had run for Westmoreland’s claims, placing them outside the Tribunal’s jurisdiction.146 The Tribunal made no commentary on the substance of the matter, so it is unclear whether the Tribunal would have accepted Canada’s defense that its “general application” action was non-discriminatory and for the public welfare. The effectiveness of that argument remains ambiguous. 

D. Lessons from the Case Law

There are several commonalities between the various climate-related Permitting and Explicit Phase-Out cases. The fossil fuel investors consistently allege treaty violations, specifically unlawful expropriation and discriminatory treatment of their investment, and they attack the state climate action as pretext for other motives that are primarily political in nature. States defend themselves by claiming their action was within their police powers as a sovereign nation, that they have the right to act in the public interest to protect the environment by blocking or rescinding upstream O&G permits or by phasing out certain fossil fuel activities within their jurisdiction. Sprinkled into these disputes are various case-specific issues, but from a birds-eye perspective, they resemble one another. 

The broad similarities within this subset of ISDS cases involving fossil fuel investors present an opportunity for states to build an established portfolio of defenses. The security exception could be one of those defenses to invoke. Though it has not been used yet in a climate-related ISDS case, the analysis in this Comment indicates that, based on the security exception standard that has developed over the past decades, the exception could cover state action taken to mitigate climate change by reducing GHG emissions. 

Substantively, securitizing climate change would provide another rationale justifying the state action—that protecting its security interests, one of which is to mitigate climate change, is well within a state’s police powers and the public interest. A state could add the security rationale to its arguments involving environmental protection and human health, providing more reason why it, as a sovereign nation, is allowed to act against fossil fuel assets. The inclusion of climate change as a relevant issue to consider, and framing it as a security issue among other priorities, may help tip the scales in favor of the defendant state. 

As will be explained in the next Section, the threshold to meet the security exception standard is generally low. There are a few key requirements for successful invocation, including good faith, plausibility, and protection of a legitimate essential security interest. Tribunals tend to grant states wide discretion in invoking the security exception, especially when the exception is a self-judging clause. As it is a provision commonly embedded in IIAs that is relatively easy to successfully invoke, the security exception could be a defense that states utilize to defend climate action. The following Section explains why this should be so. 

III. The Security Exception

Determining whether climate action fits within the security exception requires analyzing prior applications of the exception, which, while not binding, are often cited by tribunals in their rulings as suggestive.147 Case law indicates that tribunals have accepted a broad variety of interests as falling within the scope of the security exception, ranging from economic to environmental concerns. A pertinent question, though, is who decides what the state’s essential security interests are: the tribunal or the invoking state? Whether the power of self-judgment has been conferred on the signatory state hinges on the text of the provision at hand. 

A. Expanding the Exception’s Boundaries

1. The Argentina cases

The security exception in IIAs protects a state’s right to regulate.148 The language of this clause usually includes the term “‘essential security interests’ in a way that is not further defined, leaving it for Tribunals to determine the nature of security concerns that can be the basis for invoking the exception.”149 One of the overarching questions in determining the applicability of the security exception is what qualifies as an essential security interest—what is the scope of this provision? The Argentina cases played a large role in defining this unknown. 

Traditionally, protecting security interests meant “self-defence and war if necessary,” because security issues were primarily conceived as when “a state’s territory was threatened by other states’ military activities.”150 Since the 1980s, though, the conception of security interests has expanded to include “the paradigm of human security . . . economic welfare, environmental concerns, cultural identity, political rights, health, social problems, and other internal sources of instability.”151 With this enlarged understanding of security grew the range of the security exception’s applicability. 

Argentina tested the limits of this expanded concept of security by invoking the security exception defense in multiple cases in the early 2000s against fossil fuel investors from the U.S.152 The provision had never been invoked as a defense before, and the security exception standard was undefined, so the Argentina cases have been cited often since by tribunals in ISDS cases involving the security exception.153 These early invocations were critical to fleshing out the exception’s standard. The cases were brought by American investors alleging that the Argentinian government had violated the Argentina-U.S. BIT by passing “emergency measures enacted as a result of the financial crisis of the early twenty-first century.”154 Energy companies including CMS, LG&E, Enron, Sempra, Continental, El Paso, and Mobil brought cases.155

Article XI of the Argentina-U.S. BIT, the security exception clause which Argentina invoked in its defense,156 reads: “This Treaty shall not preclude the application by either Party of measures necessary for the maintenance of public order, the fulfillment of its obligations with respect to the maintenance or restoration of international peace or security, or the Protection of its own essential security interests.”157

In these cases, “Argentina argued that security interests included economic security, political and economic interests, and political stability.”158 While the claimants argued for a more traditional understanding of security as “safety from external threats,” Argentina argued that “measures aimed at ensuring internal security” are in its essential security interests, and should trigger the liability shield of the security exception.159 The Tribunals agreed with Argentina: “a severe economic crisis may thus qualify under Art. XI as affecting an essential security interest . . . ‘there is nothing in the context of customary international law or the object and purpose of the treaty that could on its own exclude major economic crisis from the scope of Art. XI.’”160 Ruling that economic matters can fall within the scope of essential security interests expanded the exception’s reach, granting states “a large margin of discretion in defining its security interests.”161

When the security exception applies, “the substantive obligations under the Treaty do not apply.”162 Tribunals have ruled that the exception applies beyond a traditional understanding of “measures taken in times of war,”163 accepting economic crisis as a valid scenario in which a state can override investor protections. Though the Argentina cases do not directly deal with actions taken to mitigate climate change via reducing GHG emissions, their expansionary precedent creates an opening for the security exception to cover nonmilitary, nontraditional “security” actions too. 

2. Enviro-social stability

Since the Argentina cases, states have attempted to securitize various interests: Caroline Henckels groups these attempts into “politico-economic contingency, infrastructure sovereignty, safeguarding law and order, and enviro-social stability.”164 This last category is most relevant here as it is most fitting for climate action. The only instance of a state using the security exception for the “enviro-social stability” paradigm is in Discovery Global LLC v. Slovak Republic. 

In Discovery Global, the American O&G company claimed that Slovakia expropriated its investment and breached various treaty obligations owed under the 1991 Slovakia-U.S. BIT by blocking it from drilling for O&G.165 This blockage allegedly occurred when Slovakia allowed protestors to obstruct the drilling sites, then, after already having granted exploration licenses to Discovery, ordered the company to produce an EIA before drilling.166 In response to the initiation of arbitration, Slovakia invoked the Article X(1) security exception,167 arguing that its “environment and drinking water, including respecting the right of affected residents to advocate for such protection, clearly qualify as part of the Slovak Republic’s ‘essential security interests.’”168

Discovery countered that (1) the measure-interest nexus is not met because Slovakia’s measures (requiring an EIA) were not necessary to protect its essential security interests (protecting its environment and drinking water), (2) the retrospective EIA requirement was pretextual, and (3) Slovakia is stretching the concept of essential security interests past its “natural and ordinary meaning.”169 Among other arguments, Discovery also claimed that “fostering domestic oil and gas exploration and production (to improve Slovakia’s energy security) was an essential security interest of Slovakia.”170 Discovery sought damages of $2.11 billion.171

The Tribunal ruled in favor of Slovakia.172 Though Discovery was deemed unable to prove its claims on their merits, resulting in Slovakia’s win, Slovakia’s security exception defense was rejected. The Tribunal stated that Slovakia’s measures were “not necessary to maintain public order or protect an essential security interest,”173 and because they were not necessary, Article X(1) could not be used as a defense to liability. The Tribunal clarified that “if it can be shown that any measure adopted by the Slovak Republic was necessary to maintain public order or to protect an essential security interest, the consequence would be that the Respondent could avoid liability for such measure since the substantive protections in the BIT would not apply.”174 The Tribunal did not comment on whether environmental interests can be understood as essential security interests. Such a classification was not outright rejected either, though; the analysis stopped at Slovakia’s failure to meet the measure-interest nexus. Climate action is an even more compelling “essential security interest” than the interests presented by Slovakia: this will be demonstrated in Part IV. 

The Tribunal in the recent ICSID case Seda v. Colombia listed some examples of essential security interests: “‘political or economic survival, the maintenance of conditions in which its essential services can function, the keeping of its internal peace, the survival of part of its population[…]’, protection of the environment, and economic security and stability.”175 Furthermore, “the essential security interests of a State are an expression of its sovereignty,”176 and “a State’s essential security interests are no longer understood to be limited to the sphere of military threats and territorial integrity.”177 The two qualities these interests must have, in the view of this Tribunal, are that “it must be of higher importance than just any interest—vital and going to the core of the State’s main functions, and it must be related to the matters of security, i.e., protection from threats.”178

Along with lingering uncertainty around the substantive scope of the security exception is the question of who decides what a state’s essential security interests are. In Discovery, the claimant argued that the Slovakia-U.S. BIT security exception clause (Article X(1)) is non-self-judging and that “the Tribunal must conduct an objective and substantive assessment of whether Slovakia can rely on Article X(1).”179 Does a tribunal defer to the state to decide its essential security interests for itself (self-judging), or does a tribunal decide what qualifies as a state’s essential security interests?

B. Is the Security Exception Self-Judging?

Whether a tribunal interprets the security exception to be self-judging depends in part on the wording of the specific provision in question. For example, Article 18 of the 2004 U.S. Model BIT specifies that the treaty does not preclude the state from “applying measures that it considers necessary for the fulfillment of its obligations with respect to the maintenance or restoration of international peace or security, or the protection of its own essential security interests.”180 In particular, the phrase italicized above “confers a large degree of ‘self-judgment’ on the party invoking the exception.”181

On the other hand, provisions like Article XI of the Argentina-U.S. BIT do not include language specifying that the measures taken to protect essential security interests should be what the state itself considers necessary.182 This implores the tribunal to determine whether an essential security interest was being protected by an action, and whether the action taken was necessary for this protection. Without the “explicit self-judging language, the host State’s actions will be subject to greater scrutiny.”183 The Tribunal in Enron v. Argentina confirmed Article XI is not self-judging, saying “clauses such as a self-judging provision normally must be expressly drafted to reflect that intent.”184

1. Non-self-judging: customary international law

If a security exception provision is deemed non-self-judging, like Article XI of the Argentina-U.S. BIT, then the tribunal determines whether the state action falls within the purview of the exception. Some tribunals have decided this involves “determining whether the measure is necessary under the customary international law and/or the terms of the governing IIA.”185 In Sempra v. Argentina, for example, the Tribunal accepted that managing an economic emergency was in Argentina’s essential security interests, but said “in view of the fact that the Treaty does not define what is to be understood by an ‘essential security interest,’ the requirements for a state of necessity under customary international law . . . become relevant to the matter of establishing whether the necessary conditions have been met for its invocation.”186 The Tribunal used Article 25 of the International Law Commission’s (ILC) Draft Articles as reflective of the customary international law doctrine of necessity: Article 25 requires that the state action “is the only way for the State to safeguard an essential interest against a grave and imminent peril.”187 The Tribunal found that Argentina’s actions were not the only way to safeguard its essential interest of managing an economic crisis; therefore, it could not invoke the security exception in its defense.188

The connection of customary international law’s doctrine of necessity with the Article XI security exception resulted in a murky interpretive standard. In Continental Casualty v. Argentina, the Tribunal “separated the customary defense of necessity from Article XI and chose to apply only Article XI in the majority of the claims.”189 That Tribunal found that Article XI covered Argentina’s actions, as “there was no reason to analyze the defense of necessity, which required meeting a higher standard.”190 This contrasted with the analytical approach Tribunals took in Sempra, Enron, and LG&E.191 Similar fact patterns were interpreted using distinct standards, leading to contradictory rulings. ISDS decisions are not binding precedent, after all192 —rather, they are “instructive” as “illustrations of custom.”193

The customary international law standard is a stringent one, for it requires states to demonstrate “that their measures were the only available means of safeguarding their essential interests against grave and imminent harm.”194 This is hard to prove, and “states . . . have faced significant difficulties in making those required showings.”195 If a security exception is deemed non-self-judging, and the tribunal chooses to utilize the doctrine of necessity to determine whether the exception applies, it appears unlikely that a state could utilize the exception as a defense for climate action against fossil fuel assets. They would need to prove that GHG emissions and/or climate change are a grave and imminent peril, that mitigating those perils is within their essential security interests, and that the only way for the state to safeguard these interests is to act against the specific fossil fuel investment at hand. The tribunal would use a high standard. This route appears to be an uphill battle. 

If the tribunal sidelines the customary international law analysis, as the Tribunal in Continental decided and the ad hoc committees on the application for annulment for CMS, Enron, and Sempra decided was ultimately appropriate in those cases, the state would just need to meet the lower threshold requirements of the exception itself.196 This approach is more likely given the controversy that emerged from cases like Sempra regarding the Tribunal’s use of customary international law.197 Without the higher standard required by customary international law, tribunals interpret the exception to cover “a wide variety of interests,”198 and its invocation “does not require that ‘total collapse’ of the country or that a ‘catastrophic situation’ has already occurred before responsible national authorities may have recourse to its protection…[t]here is no point in having such protection if there is nothing left to protect.”199

2. Self-judging: GATT Article XXI

Many security exceptions are expressly drafted with explicit self-judging language like “it considers necessary” that reflects a self-judging intent.200 This helps avoid confusion around the use of customary international law standards. Tribunals treat such provisions differently than those without self-judging language. One of these provisions is GATT Article XXI, specifically section (b),201 which acts as a model for many IIA security exception provisions.

It should be noted, though, that GATT is an agreement in international trade law, not international investment law, so the contexts for their application are not identical.202 However, given the near-identical language of the GATT security exception to security exceptions in IIAs, “the considerable body of case law interpreting and applying Article XXI of the GATT [i]s a helpful, albeit inconclusive, supplementary means of interpretation.”203

The first time a tribunal ruled on the GATT national security exception was in 2019 in the case of Russia - Measures Concerning Traffic in Transit.204 The complaint here was brought by Ukraine regarding Russia’s restriction on international transit cargo by road and rail from Ukraine destined for Kazakhstan or Kyrgyzstan.205 Russia argued that this restriction was taken due to an “emergency in international relations,” thereby permitting it under Article XXI(b)(iii), and that “a [WTO] Member’s essential security interests and the determination of whether any action is necessary for the protection of a Member’s essential security interests are at the sole discretion of the Member invoking the provision.”206 Russia asserted that the WTO Panel and Dispute Settlement Body (DSB) “lacks jurisdiction to evaluate measures taken pursuant to Article XXI of the GATT 1994”207 because the provision is “totally ‘self-judging.’”208 In their view, this invocation “cannot be ‘doubted or re-evaluated by any other party’,” as that would “result in interference in [the] internal and external affairs of a sovereign state.”209

The Panel disagreed with Russia: “Article XXI(b)(iii) of the GATT 1994 is not totally ‘self-judging’ in the manner asserted by Russia.”210 A state action “must objectively be found to meet the requirements in one of the enumerated subparagraphs of that provision.”211 The objective scrutiny the Panel concludes is required in its review of a security exception invocation is based on “its textual and contextual interpretation of Article XXI(b)(iii),”212 including the structure of the provision, the purpose of GATT 1994, and the negotiating history of the agreement. That the action was taken in a time of an emergency in international relations, qualifying it as acceptable under Article XXI(b)(iii), “is that of an objective fact, subject to objective determination.”213

Though the Panel rejected Russia’s interpretation of the security exception as totally self-judging, it did agree that Russia’s action was taken in time of an emergency in international relations.214 Once this hurdle was passed, the Panel required that Russia “interpret and apply” the security exception “in good faith.”215 This good faith requirement “applies not only to the Member’s definition of the essential security interests said to arise from the particular emergency in international relations, but also, and most importantly, to their connection with the measures at issue.”216 The interests protected must legitimately be essential security interests, and the action taken must plausibly relate to those essential security interests.217 The standard here is low: A Member must articulate its essential security interests “sufficiently enough to demonstrate their veracity,”218 and the action taken must “meet a minimum requirement of plausibility in relation to the proferred [sic] essential security interests.”219

Russia was able to meet this low standard because its actions were deemed at least plausibly implemented to protect its essential security interests in a time of emergency in international relations: the military conflict in Crimea beginning in 2014.220 The Panel ruled that it had jurisdiction to objectively review an invocation of the security exception, but it left “the ‘necessity’ of the measures for the protection of its essential security interests” to Russia’s sole discretion.221 Ultimately, Russia met the requirements to invoke Article XXI(b)(iii), allowing its actions to be covered by the security exception.222

The Russia case made clear that even if a security exception has self-judging language, it is not entirely self-judging; a state’s invocation of the provision will prompt some level of objective review. The action taken must fall within the appropriate purview of the provision—in GATT, the purview is the subparagraphs of Article XXI(b). The interest being classified as an essential security interest must be objectively so, and it must be classified as such in good faith by the invoking country. Finally, the action in question must plausibly have been taken to protect this essential security interest. Whether the action was necessary to defend the interest is for the invoking state to decide.

3. Reinforced self-judging: Seda v. Colombia

Seda v. Colombia marked “the first publicly available award where an investment tribunal accepted the ‘self-judging’ character of a provision within an investment agreement.”223 Despite this recognition, the Tribunal still assessed the invocation of the security exception. The IIA allegedly breached here was the U.S.-Colombia Trade Promotion Agreement of 2012 (TPA), and Colombia invoked Article 22.2(b),224 the security exception, to defend its actions.225 The final award for this case was dispatched to the parties in June 2024.226

In this case, Colombian authorities seized a luxury real estate property owned by American investors “because of the historic connection of the property to organized crime.”227 The investors claimed that Colombia breached “fundamental obligations under the TPA” including unlawful expropriation of their investment, the FET standard, the national treatment standard, and the full protection and security standard.228 In response, Colombia invoked Article 22.2(b) of the TPA and argued that when read with the interpretive footnote 2, the treaty allows a state to invoke the exception at its discretion—essentially, that the Tribunal lacks the jurisdiction to review its invocation.229 The footnote is notable because it “extends beyond a classic self-judgment provision.”230 It explicitly says that tribunals must apply the exception if it is invoked: it is a “reinforced” self-judging security exception.231 This is the first time a “reinforced” provision was invoked in ISDS.232

The Tribunal agreed with Colombia that the security exception provision is self-judging, for it contains the phrase “it considers necessary,” which has been held to indicate self-judging intent by tribunals before.233 In examining whether Colombia’s interests were truly essential security interests, the Tribunal states that “it must be of higher importance than just any interest – vital and going to the core of the State’s main functions, and it must be related to the matters of security, i.e., protection from threats.”234 This includes interests beyond “the sphere of military threats and territorial integrity” including “protection of the environment . . . it is the State itself which can best identify the scope of its own essential security interests.”235 Because the TPA does not have qualifying clauses for what counts as essential security interests like GATT 1994 does, the Tribunal affords “a broad margin of appreciation” to the state “in identifying its essential security interest.”236 Though the Tribunal acknowledges that a state decides its essential security interests for itself, it aligns with the Panel in Russia in saying “such determination may be restrained by the good faith obligations.”237

The Tribunal also gives the state leeway in the measures-interest nexus analysis, the determination of whether the action taken was for the purpose of protecting the essential security interest. “The Tribunal finds the plausibility standard an appropriate benchmark . . . it carries an implication of a ‘light-touch’ good faith review – not too restrictive as to infringe on the explicit self-judging language.”238 However, the Tribunal rejects the notion that the security exception is totally non-justiciable because neither Article 22.2(b) nor footnote 2 contain explicit non-justiciable language depriving the Tribunal of the ability to conduct any legal assessment regarding the invocation of the provision.239 This would also contradict the ordinary meaning of the terms of the provision and would act as an escape clause, “an omnipotent tool at a State’s disposal.”240 To respect the wording of the provision while still holding the state accountable to an extent, “the Tribunal considers a good faith review—a standard supported by jurisprudence and legal scholars—sufficiently balanced to ensure proper application of Article 22.2(b) of the TPA without infringing on its self-judging nature.”241

Using this cursory standard of review, the Tribunal agreed that fighting organized crime and drug trafficking “directly relate[s] to public safety, national security, and socio-economic stability . . . and therefore constitute[s] essential security interests.”242 It found that classifying these efforts as essential security interests was done in good faith,243 and that “the nexus between the measures enacted against Claimants and the essential security interest invoked by Respondent in the present case satisfies the plausibility threshold.”244 This plausibility standard is low—“It suffices that the measures could serve such purpose on their face.”245 In conclusion, the Tribunal found “no indications in the case record that the ESI (essential security interest) Provision was not invoked by Respondent in good faith . . . this means that the measures taken by Respondent are excluded from the scope of the TPA coverage and Tribunal’s inquiry must stop here.”246

Seda confirmed that tribunals will exercise some level of review, albeit minimal, for self-judging clauses that attempt to reinforce their non-justiciability. There are some agreements with even more explicit self-judging language that tribunals have yet to rule on,247 but given concerns about the exception acting as “an omnipotent tool,”248 it appears unlikely that a tribunal would permit a security exception invocation without any scrutiny, at the very least a “light-touch good faith review.”249

State climate action could ostensibly pass this good faith and plausibility standard of review. The state would need to show that its action was undertaken to protect an essential security interest; that the interest of mitigating climate change by reducing GHG emissions is being classified as an essential security interest in good faith; and that the action taken against fossil fuel assets is plausibly related to protecting that essential security interest of mitigating climate change. This is a much clearer path than if the exception is deemed non-self-judging, under which the state may need to meet the stringent doctrine of necessity standard, or at least a higher level of scrutiny from the tribunal than good faith and plausibility. Given that self-judging language in IIAs is increasingly in vogue,250 the ability of a state to use the security exception to defend its climate action looks increasingly possible.

The wide discretion tribunals have given states in invoking the exception, especially with self-judging exceptions, makes it appear possible that the security exception could be a tool for states to defend climate policy against fossil fuel investors, as long as the invocation meets the relatively low standards for good faith and plausibility, and mitigating climate change is accepted as an essential security interest. This last point is critical to the successful invocation of the security exception to defend state climate action in any future ISDS case. Part IV will discuss how climate change can be viewed through a security interest lens.

IV. Climate Change as a Security Issue

A. Climate Action Under Self-Judging and Non-Self-Judging Provisions

Climate change has been understood as a global security issue for some time. The United Nations Security Council publicly discussed the connection between climate change and security in 2007,251 the U.S. Department of Defense (DoD) has been studying climate change’s impact on national security since at least 2003,252 and the North Atlantic Treaty Organization’s (NATO) 2010 Strategic Concept stated climate change “will further shape the future security environment.”253 A Naval War College report from 1990 was titled “Global Climate Change Implications for the United States Navy.”254 Alongside the environmental issues climate change will exacerbate, ranging from changes in precipitation patterns to sea level rise to more extreme weather events,255 climate change is viewed as a “threat multiplier” that has “the potential to exacerbate pre-existing threats and other drivers of instability to contribute to security risks.”256

Climate change is an even more compelling essential security interest than any acute environmental issue, for it threatens vital core state functions by exacerbating security issues: this is what is meant by the term “threat multiplier.” For example, higher levels of interpersonal and intergroup conflict occur as temperature rises,257 increasing the risk of war, insurgency, and political instability. The National Intelligence Council found that climate change is poised to intensify geopolitical flashpoints, specifically regarding migration and competition over resources like water, as well as impact country-level instability by straining energy and food systems, fomenting internal insecurity and conflict and straining military readiness.258 These stresses threaten socioeconomic and political systems.259 Protecting these aspects of a state’s security is undoubtedly within its essential interests, as threats to such fundamental operations threaten the state’s functionality as a sovereign entity. Examples of essential security interests, according to the Tribunal in Seda, include political and economic stability, maintaining conditions for the functioning of essential services, survival of the population, and protection of the environment:260 these are all threatened by climate change. Whether a tribunal would accept climate change mitigation as an essential security interest for a country may depend on how the tribunal categorizes the specific security exception being invoked—as a self-judging or non-self-judging provision.

For self-judging provisions, which are those that include phrasing like “it considers necessary,” tribunals tend to allot substantial discretion to the state in determining its own essential security interests.261 In Russia, the WTO Panel required good faith regarding what a country classifies as its essential security interests,262 and in Seda, the Tribunal acknowledged these interests could include non-military matters like environmental protection as long as they are vital to core state functions and are related to security. The state should define for itself, in good faith, what its essential security interests are.263 Slovakia already pushed the boundaries of the exception in this direction by arguing protection of its environment and drinking water is an essential security interest.264 This lays the groundwork for arguing that climate change mitigation should be classified the same way. New investment agreements now almost universally include self-judging language in security exception drafting.

If the security exception is non-self-judging, the pathway for climate action to qualify is more tenuous. The tribunal may use the necessity standard from customary international law, which requires the state action to be “the only way for the State to safeguard an essential interest against a grave and imminent peril.”265 This is a high standard that climate action would be unlikely to meet—the state would “have to establish that their measures were the only available means of safeguarding their essential interests against grave and imminent harm.”266 Climate change may be grave, but it is not exactly imminent, and state action phasing out certain emitting activities or blocking/rescinding permits for specific projects is likely not the “only” way to safeguard against climate change. The exception under the standard established in customary international law “can only be accepted on an exceptional basis.”267 Luckily for potential future invocators, tribunals have mostly moved away from using the customary international law standard in analyzing security exception invocations.

In the circumstance where a tribunal rejects the customary international law standard in favor of the standard of the specific non-self-judging security exception provision in the IIA at hand, which was done in Continental, the Tribunal grants “a significant margin of appreciation for the State applying the particular measure.”268 There is still an objective review to avoid “allowing a party unilaterally to escape from its treaty obligations in the absence of clear textual or contextual indications,”269 and regarding necessity, an analysis of “whether the Measures were apt to and did make such a material or a decisive contribution to” the protection of the essential security interests,270 but the argument that climate action falls under the security exception has a higher chance of success with this standard than with the customary international law standard.

B. Satisfying the Measure-Interest Nexus

Climate change threatens all parts of the planet;271 it is commonly framed as a collective action problem that “cannot be solved by any one agent acting unilaterally.”272 This is because it is caused by GHG emissions from human activities happening globally.273 No nation is large enough to totally mitigate climate change by reducing just its own emissions, not even the largest emitter, China, which accounted for 29% of global emissions in 2024.274 The international nature of the problem means that a country reducing its own emissions will not necessarily protect it from the effects of climate change. This presents the question: can the measures-interest nexus be bridged if the measures taken to supposedly protect the security interest will not actually do so? In other words, if a country attempts to reduce its GHG emissions (measure) to mitigate climate change (interest), but the emissions reduction will not effectively mitigate climate change, can the measure qualify under the security exception?

As discussed earlier, the first step a tribunal takes in its analysis of whether a general exception like the security exception applies to a state action is whether the action is within the scope of the exception. This means proving that mitigating climate change is an essential security interest. Given the geopolitical, economic, social, and environmental consequences of climate change, and the discretion tribunals give states in most cases in defining their own interests, this step should not be exceedingly difficult as long as the classification is done in good faith.

The second step is the measure-interest nexus: whether the state action was taken in good faith to protect the essential security interest. This could face a plausibility test, which is easy to pass, or a customary international law necessity test, which is nearly impossible to pass.275 Given that self-judging security exception provisions are increasingly common in IIAs, states in most cases will now have to pass the plausibility test.

Is action to reduce domestic GHG emissions plausibly related to the essential security interest of mitigating climate change? It must be. The Paris Agreement, an international treaty on climate change adopted by 196 parties in 2015,276 is predicated on Nationally Determined Contributions (NDCs) that “embody efforts by each country to reduce national emissions.”277 All party states “are to undertake and communicate ambitious efforts . . . with the view to achieving the purpose of this Agreement”278 which is primarily to “reduce the risks and impacts of climate change.”279 These efforts include undertaking “rapid reductions” in GHG emissions,280 and each party “shall pursue domestic mitigation measures, with the aim of achieving the objectives of such contributions.”281

The best way to mitigate climate change, which can be argued is in the essential security interests of all nations, is to lower global GHG emissions. The best way for a nation to do that is to reduce its own emissions. By complying with its Paris Agreement NDCs and demonstrating the priority it gives to emissions reduction, countries can use their domestic action to spur collective action. This is the most effective way for any government to achieve its interest of mitigating climate change. At the very least, the measure-interest nexus is plausibly achieved. That being met, a tribunal can apply the security exception to the state’s climate action taken against fossil fuel assets while complying with the standard set by past tribunals.

IV. Conclusion

The ISDS mechanism is used by foreign investors to hold host nations liable for damaging their investments. Much of this legal activity is brought by fossil fuel investors. The record indicates that these investors win in the majority of disputes, and the awards they reap dwarf those given to investors in other industries.282 As a result, states are increasingly wary of acting against fossil fuel assets for fear of crippling liability, a fear that is smothering state action intending to mitigate climate change by reducing domestic GHG emissions.

This Comment argues that states could use the security exception, a provision common in IIAs that protects a state’s right to regulate to protect its essential security interests, to defend certain claims brought by fossil fuel investors. Fossil fuel investor claims tend to be either Permitting cases, wherein the state blocks or rescinds upstream activity licenses for a specific project, or Explicit Phase-Out cases, in which the state passes a policy banning an emitting activity (like combustion of coal for electricity generation) by a set date. If a state successfully invokes the security exception found in the relevant IIA, it will not be liable to the investors for the regulatory action targeting those assets.

The standard that has emerged through case law around invocation of the security exception is favorable for state climate action. For exceptions containing self-judging language, the state is given wide discretion in classifying its own interests, and in past cases these interests have included economic security, infrastructure sovereignty, and combating organized crime—states are attempting to securitize environmental protection as well.283 Once the interest is accepted as an essential security interest, the tribunal requires the state to have classified and acted in good faith, and for the action in question to plausibly relate to protection of the essential security interest. This is a low standard that presents an opportune method of defense for states.

For exceptions without that self-judging language, which tend to be in older IIAs, tribunals have used the customary international law doctrine of necessity to determine applicability, but the stringency of that standard has led tribunals to apply a more lenient standard of review based on the threshold of the exception itself. In all scenarios, the tribunal will review the state’s invocation, and in most cases, the state is afforded ample respect to define its own essential security interests.

The security exception can act as a liability shield for states to defend their climate action that impacts fossil fuel assets. At present, fossil fuel investors are winning most cases, and states are chilled from enacting policies targeting fossil fuels. This hinders not only a state’s ability to achieve its NDCs under the Paris Agreement, but also humanity’s efforts to mitigate climate change. The securitization of climate action could help states acting in good faith to reduce their emissions avoid liabilities. It is a promising defense that is much needed if climate goals are to be achieved.

• 1See Overview and Key Findings, International Energy Agency (iea), https://perma.cc/UT58-PUGS (last visited Jan. 3, 2025); Simon Black et al., Fossil Fuel Subsidies Surged to Record $7 Trillion, IMF Blog(Aug. 24, 2023), https://perma.cc/S62K-CT4Z.
• 2See Introduction to Fossil Fuels, Stanford Doerr School of Sustainability: Stanford University, https://perma.cc/M9CR-SSDS (last visited Jan. 3, 2025).
• 3See generally Primer on International Investment Treaties and Investor-State Dispute Settlement, Columbia Center on Sustainable Investment(Jan. 2022), https://perma.cc/9VHG-6PYK (explaining the basic framework of investment treaties and ISDS).
• 4Martin Dietrich Brauch, Reforming International Investment Law for Climate Change Goals, in Research Handbook on Climate Finance and Investment Law (Michael Mehling & Harro van Asselt eds., 2020).
• 5See Primer on ISDS, supra note 3.  
• 6See Brauch, supra note 4.
• 7See Primer on ISDS, supra note 3.
• 8See U.N. Conference on Trade and Development (UNCTAD), World Investment Report 2025, at 80 (2025).   
• 9See Primer on ISDS, supra note 3.
• 10Brauch, supra note 4 at 8–9. 
• 11See International Centre for Settlement of Investment Disputes (ICSID) Convention, Oct. 14, 1966, https://perma.cc/LS5T-WKCD (last visited Nov. 17, 2024).
• 12See ICSID Convention, Oct. 14, 1966, https://perma.cc/LS5T-WKCD (last visited Nov. 17, 2024).
• 13See United Nations Convention on the Recognition and Enforcement of Foreign Arbitral Awards (New York Convention), Jun. 7, 1959, New York Convention Text, https://perma.cc/P6N9-94JM (last visited Nov. 17, 2024). 
• 14See New York Convention, Jun. 7, 1959, Contracting States, New York Convention, https://perma.cc/X569-A7F3 (last visited Nov. 17, 2024).
• 15See About ICSID, ICSID, https://perma.cc/A369-4MFK (last visited Nov. 17, 2024).
• 16Id.
• 17See Search Cases, ICSID, https://perma.cc/WK67-PVFD (last visited Nov. 17, 2024).
• 18See Spotlight on Transparency at ICSID, ICSID (May 15, 2023), https://perma.cc/XR4S-LALD.
• 19See id.
• 20Lea Di Salvatore, Investor-State Disputes in the Fossil Fuel Industry iii (2021). 
• 21See id.
• 22See UNCTAD, supra note 8, at 17.
• 23Id. at 61.
• 24See IEA, supra note 1. 
• 25See Di Salvatore, supra note 20.
• 26See id.
• 27See id. at iv. This $600 million calculation excludes the outlier $40 billion award in Hulley Enterprises v. Russia, the largest investment arbitration award ever.
• 28Joshua Paine & Elizabeth Sheargold, A Climate Change Carve-Out for Investment Treaties, 26 J. of Int’l Econ. L. 285, 286 (2023).
• 29Id.
• 30See Causes and Effects of Climate Change, U.N., https://perma.cc/ZP2Z-WQXN (last visited Nov. 17, 2024).
• 31M. Fermeglia et al., Mapping Climate-Related Investment Arbitrations, 21 Transnat’l Disp. Mgmt. 1, 2 (2024).
• 32See David R. Boyd, Paying polluters: the catastrophic consequences of investor-State dispute settlement for climate and environment action and human rights 16 (2023); Kyla Tienhaara et al., Submission to the Special Rapporteur on human rights and the environment call for inputs: “Should the interests of foreign investors trump the human right to a clean, healthy and sustainable environment?” 3 (2023).
• 33Paine & Sheargold, supra note 28, at 286; Tienhaara et al., supra note 32, at 3 ; Boyd, supra note 32, at 16.
• 34See Boyd, supra note 32, at 16 (citing Intergovernmental Panel on Climate Change, Climate Change 2022: Mitigation of Climate Change).
• 35Press Release, European Commission, Energy Charter, https://perma.cc/249V-JJLR (last visited Nov. 22, 2025).
• 36Monika Dulian, EU Withdrawal from the Energy Charter Treaty 3 (2023).
• 37See id.
• 38Temitope Badejo, Canada’s Withdrawal from Investor-State Arbitration in the USMCA: Implications and Alternative Dispute Resolution Mechanisms for Investors, 48 Can.-U.S. L. J. 187, 189 (2024).
• 39Antonios Kouroutakis, Sunset Clauses in International Law and their Consequences for EU Law 22 (2022).
• 40See Boyd, supra note 32, at 13.
• 41Paine & Sheargold, supra note 28, at 289 n.32.
• 42Some EU members withdrew from the ECT before the entire bloc did in 2024. For example, Italy withdrew in in 2016 and France and Germany withdrew in 2023. See Dulian, supra note 36, at 2.
• 43See Kilian Wagner, Regulation by Exception – The Emergence of (General) Exception Clauses in International Investment Law?, 26 Austrian Rev. of Int’l and Eur. L. 77, 79–81 (2023).
• 44Eric David Kasenetz, Desperate Times Call for Desperate Measures: The Aftermath of Argentina’s State of Necessity and the Current Fight in the ICSID, 41 Geo. Wash. Int’l L. Rev. 709, 720 (2010) (citing the International Law Commission’s (ILC) Draft Articles on Responsibility of States for Internationally Wrongful Acts (Draft Articles)). Article 25 of the Draft Articles reads:

  Article 25. Necessity

  1. Necessity may not be invoked by a State as a ground for precluding the wrongfulness of an act not in conformity with an international obligation of that State unless the act:

  (a) is the only way for the State to safeguard an essential interest against a grave and imminent peril; and

  (b) does not seriously impair an essential interest of the State or States towards which the obligation exists, or of the international community as a whole.

  2. In any case, necessity may not be invoked by a State as a ground for precluding wrongfulness if:

  (a) the international obligation in question excludes the possibility of invoking necessity; or

  (b) the State has contributed to the situation of necessity.

• 45Wagner, supra note 43 (citing Catherine Titi, The Right to Regulate in International Investment Law (2014)).
• 46See id.
• 47Catharine Titi, Police Powers Doctrine and International Investment Law, in Principles of Law and International Investment Arbitration 323–43 (2018).
• 48See id.
• 49See Wagner, supra note 43, at 80–81.
• 50Robert Brew, Exception Clauses in International Investment Agreements As a Tool for Appropriately Balancing the Right to Regulate with Investment Protection, 25 Canterbury L. Rev. 205, 217 (2019).
• 51Lise Johnson, International Investment Agreements and Climate Change: The Potential for Investor-State Conflicts and Possible Strategies for Minimizing It, 39 Env’t L. Rep. News & Analysis 11147, 11158 (2009).
• 52Caroline Henckels, Permission to Act: The Legal Character of General and Security Exceptions in International Trade and Investment Law, 69 Int’l & Compar. L. Q. 557, 559 (2020).
• 53Wagner, supra note 43, at 83.
• 54The General Agreement on Tariffs and Trade 1994 art. XX, Apr. 15, 1994, 1867 U.N.T.S. 187.

  Article XX reads: 

  ‘Subject to the requirement that such measures are not applied in a manner which would constitute a means of arbitrary or unjustifiable discrimination between countries where the same conditions prevail, or a disguised restriction on international trade, nothing in this Agreement shall be construed to prevent the adoption or enforcement by any contracting party of measures:

  (a) necessary to protect public morals;

  (b) necessary to protect human, animal or plant life or health; […]’

• 55Wagner, supra note 43, at 84.
• 56WTO, WTO Analytical Index: GATS, Article XIV (DS reports), 2 (2024), https://perma.cc/4URF-KR4R.
• 57Id. at 3.
• 58Id.
• 59Wagner, supra note 43, at 87.
• 60Id.
• 61WTO, Analytical Index of the GATT 1994, 599 (2024) (emphasis added),https://perma.cc/C2XX-Y9NR.
• 62Boyd, supra note 32, at 4 (citing Di Salvatore, supra note 20, at 17–19) (“At the merits stage, fossil fuel investors win 72 per cent of cases, forcing Governments to pay more than $77 billion in compensation to date.”).
• 63Fermeglia et al., supra note 31.
• 64Catherine Higham & Joana Setzer, ‘Investor-State Dispute Settlement’ as a new avenue for climate change litigation, Grantham Research Institute on Climate Change and the Environment: The London School of Economics and Political Science (June 2, 2021), https://perma.cc/9ALQ-88FJ.
• 65Fermeglia et al., supra note 31, at 8.
• 66Stranded Assets, Carbon Tracker (Aug. 23, 2017), https://perma.cc/32JR-Q28G.
• 67Fermeglia et al., supra note 31, at 9.
• 68Id.
• 69Id. at 9 n.35.
• 70Id. at 13.
• 71Rockhopper Expl. PLC v. It,, ICSID Case No. ARB/17/14, Final Award, ¶ 5 (Aug. 23, 2022).
• 72Id. ¶ 92.
• 73Id. ¶ 97(2).
• 74Id. ¶ 97(4).
• 75Id.
• 76Id. ¶ 105.
• 77Id. ¶ 103.
• 78Id. ¶ 101.
• 79Id. ¶ 105(3).
• 80Id. ¶ 183.
• 81Energy Charter Treaty (ECT), art. 13, Apr. 16, 1998, 2080 U.N.T.S. 95.

  Investments of Investors of a Contracting Party in the Area of any other Contracting Party shall not be nationalised, expropriated or subjected to a measure or measures having [effect] equivalent to nationalisation or expropriation (hereinafter referred to as “Expropriation”) except where such Expropriation is:

  (a) for a purpose which is in the public interest;

  (b) not discriminatory;

  (c) carried out under due process of law; and

  (d) accompanied by the payment of prompt, adequate and effective compensation.

• 82Rockhopper Expl. PLC, Final Award, 2022 ICSID ¶ 188.
• 83Id. ¶ 195.
• 84Id. ¶ 275.
• 85Id. ¶ 191.
• 86Id. ¶ 124.
• 87Id. ¶ 129.
• 88Id. ¶ 130.
• 89Id. ¶ 198.
• 90Id.
• 91Id. ¶ 127.
• 92Id. ¶ 112.
• 93Id. ¶ 319.
• 94Lone Pine Res. Inc. v. Can., ICSID Case No. UNCT/15/2, Final Award, ¶ 239 (Nov. 21, 2022).
• 95Id. ¶ 144.
• 96Id. ¶ 177.
• 97Id. ¶ 205.
• 98Id.¶ 206.
• 99Id. ¶ 212.
• 100Id. ¶ 226.
• 101Id. ¶ 231.
• 102Id. ¶ 422.
• 103Lone Pine Res. Inc. v. Can., ICSID Case No. UNCT/15/2, Notice of Arbitration, ¶ 58(a) (Sept. 6, 2013).
• 104Lone Pine Res. Inc., Final Award, 2022 ICSID ¶ 457.
• 105Id. ¶ 458.
• 106Id. ¶ 460.
• 107Id. ¶ 538.
• 108Id. ¶¶ 538–39.
• 109Id. ¶ 619.
• 110Id. ¶ 612.
• 111Id. ¶ 620.
• 112Id. ¶ 624.
• 113Id. ¶ 626.
• 114See Ascent Res. Plc. & Ascent Slovn. Ltd. v. Slovn., ICSID Case No. ARB/22/21, Arbitration Initiation and Revised Damages Estimate (Aug. 15, 2022).
• 115Ascent Res. Plc. & Ascent Slovn. Ltd. v. Slovn., ICSID Case No. ARB/22/21, Treaty Claims against the Republic of Slovenia by Ascent Resources Plc. & Ascent Slovenia Ltd, ¶ 1.3 (May 5, 2022).
• 116Ascent Res. Plc. & Ascent Slovn. Ltd., Arbitration Initiation and Revised Damages Estimate, 2022 ICSID.
• 117Id.
• 118Press Release, London Stock Exchange, Update on ECT Claim, RNS No. 1859E (Feb. 23, 2024),  https://perma.cc/SZ8M-YMLS.
• 119See Ruby River Capital LLC v. Can., ICSID Case No. ARB/23/5, Request for Arbitration (Feb. 17, 2023).
• 120Id. ¶ 93.
• 121Id. ¶ 148.
• 122Id. ¶ 106.
• 123Id. ¶ 107.
• 124Id. ¶ 125.
• 125Id. ¶¶ 133-34.
• 126See Ruby River Capital LLC, Request for Arbitration, 2023 ICSID.
• 127TC Energy Corp. & TransCan. Pipelines Ltd. v. United States, ICSID Case No. ARB/21/63, Award (July 12, 2024).
• 128Id.
• 129Fermeglia et al., supra note 31, at 9.
• 130Ending coal-generated power, The Federal Government, https://perma.cc/8FCE-9VX2 (last visited Nov. 17, 2024).
• 131Stan Putter, The Netherlands Coal Phase-Out and the Resulting (RWE and Uniper) ICSID Arbitrations, Kluwer Arbitration Blog (Aug. 24, 2021), https://perma.cc/F8W3-CPQM.
• 132Id.
• 133RWE v. Neth., Request for Arbitration, ICSID Case No. ARB/21/4 (Jan. 20, 2021).
• 134High-profile coal phase-out intra-EU arbitration discontinued, International Institute for Sustainable Development (Jan. 13, 2024), https://perma.cc/9JLW-EC2Z.
• 135Dutch court denies compensation to RWE and Uniper, International Institute for Sustainable Development (Dec. 26, 2022), https://perma.cc/P4JR-82GY.
• 136Azienda Elettrica Ticinese v. Ger., ICSID Case No. ARB/23/47, UN Trade & Development Investment Policy Hub, https://perma.cc/E8CT-QETQ (last visited Nov. 17, 2024).
• 137Luigi Jorio, Why Switzerland is diverging from Europe on a key energy treaty, Swissinfo.ch (July 18, 2024), https://perma.cc/KLS6-SP3E.
• 138Westmoreland v. Can. (III), ICSID Case No. UNCT/23/2, Claimant’s Notice of Arbitration ¶ 63-72 (Oct. 11, 2022).
• 139Id. ¶ 75.
• 140NAFTA tribunal in Westmoreland v. Canada declines jurisdiction, finding that the claimant did not own or control the investment at the time of the alleged breach, International Institute for Sustainable Development (July 4, 2022), https://perma.cc/ZJ8X-6678.
• 141Westmoreland, Claimant’s Notice of Arbitration, 2022 ICSID ¶ 8.
• 142Id. ¶ 5.
• 143Westmoreland v. Can. (III), ICSID Case No. UNCT/23/2, Memorial on Jurisdiction and Response to Notice of Arbitration ¶ 150 (June 28, 2023).
• 144Id. ¶ 155.
• 145Westmoreland v. Can. (III), ICSID Case No. UNCT/23/2, Award, (December 17, 2024).
• 146Id. ¶ 134. 
• 147See Henckels, supra note 52.
• 148See Wagner, supra note 43.
• 149Caroline Henckels, Whither Security? The Concept of ‘Essential Security Interests’ Investment Treaties’ Security Exceptions, 27 J. of Int’l Econ. L. 114, 114–15 (2024).
• 150Id. at 116.
• 151Id.
• 152Kasenetz, supra note 44.
• 153Henckels, supra note 52, at 571.
• 154Kasenetz, supra note 44, at 709–10.
• 155Henckels, supra note 149, at 117.
• 156Kasenetz, supra note 44, at 726–27 (“Argentina argued its position with identical defenses . . . Argentina’s strongest defenses rested on the customary doctrine of necessity and Article XI of the U.S.-Argentina BIT.”).
• 157Treaty Between United States of America and the Argentine Republic Concerning the Reciprocal Encouragement and Protection of Investment, U.S.-Arg., art. XI, Nov. 14, 1991, 94 T.I.A.S 1020.
• 158Henckels, supra note 149, at 118.
• 159Continental Casualty Company v. Arg., ICSID Case No. ARB/03/9, Award ¶¶ 170–72 (Sep. 5, 2008).
• 160Id. ¶ 178.
• 161Wagner, supra note 43, at 87.
• 162Deutsche Telekom AG v. India, PCA Case No. 2014-10, Interim Award ¶ 227 (Dec. 13, 2017).
• 163Wagner, supra note 43, at 87.
• 164Henckels, supra note 149, at 118.
• 165Discovery Global LLC v. Slovk., ICSID Case No. ARB/21/51, Request for Arbitration (Sep. 30, 2021).
• 166Id.
• 167Treaty Between the Czech and Slovak Federal Republic and the United States of America Concerning the Reciprocal Encouragement and Protection of Investment, Slovk.-U.S., art. X(1), Oct. 22, 1991, S. Treaty Doc. No. 102-31. Article X(1) of the BIT reads:

  This Treaty shall not preclude the application by either Party of measures necessary for the maintenance of public order, the fulfillment of its obligations with respect to the maintenance or restoration of international peace or security, or the protection of its own essential security interests.

• 168Discovery Global LLC v. Slovk., ICSID Case No. ARB/21/51, Respondent’s Counter-Memorial ¶ 243 (Mar. 31, 2023) (emphasis in original).
• 169Discovery Global LLC v. Slovk., ICSID Case No. ARB/21/51, Claimant’s Reply ¶ 223 (Sep. 18, 2023).
• 170Id. ¶ 226(6).
• 171Discovery Global LLC., Request for Arbitration, 2021 ICSID ¶ 109.4.
• 172Discovery Global LLC v. Slovk., ICSID Case No. ARB/21/51, Award (Jan. 17, 2025).
• 173Id. ¶ 359.
• 174Id. ¶ 362.
• 175Angel Samuel Seda et al. v. Colom., ICSID Case No. ARB/19/6, Award ¶ 643 (June 27, 2024).
• 176Id. ¶ 644.
• 177Id. ¶ 643.
• 178Id. ¶ 642.
• 179Discovery Global LLC, Claimant’s Reply, 2023 ICSID ¶ 220.
• 1802004 Model BIT, U.S.-Government of [Country], art. 18 (emphasis added), https://perma.cc/M5VV-7HE9 (last visited Nov. 22, 2025).
• 181Johnson, supra note 51, at 11156 (citing Organization for Economic Cooperation & Development, Negotiating Group on the Multilateral Agreement on Investment (MAI): National Security Measures: Note by the Chairman 4 (1995)).
• 182Treaty Between United States of America and the Argentine Republic, supra note 157.

  Article XI reads: This Treaty shall not preclude the application by either Party of measures necessary for the maintenance of public order, the fulfillment of its obligations with respect to the maintenance or restoration of international peace or security, or the protection of its own essential security interests.

• 183Johnson, supra note 51, at 11157.
• 184Enron Corp. & Ponderosa Assets, L.P. v. Arg., ICSID Case No. ARB/01/3, Award ¶ 335 (May 22, 2007).
• 185Johnson, supra note 51, at 11157.
• 186Sempra Energy International v. Arg., ICSID Case No. ARB/02/16, Award ¶ 375 (Sep. 28, 2007).
• 187Responsibility of States for Internationally Wrongful Acts, United Nations Art. 25(1)(a) (2005).
• 188Sempra Energy International, Award, 2007 ICSID ¶ 351 (“It is instead the Tribunal’s duty only to determine whether the choice made was the only one available, and this does not appear to have been the case.”).
• 189Kasenetz, supra note 44, at 730–31.
• 190Id. at 731.
• 191Id. at 729–30.
• 192Martin Jarrett, ISDS 2.0: time for a doctrine of precedent?, 27 J. Int’l Econ. L. 41 (2023).
• 193Lone Pine Res. Inc., Final Award, 2022 ICSID ¶ 600.
• 194Johnson, supra note 51, at 11158.
• 195Id.
• 196Angel Samuel Seda et al., Award, 2024 ICSID ¶¶ 677–82.
• 197Joanne Greenaway, Sempra v Argentina: award annulled for manifest excess of powers in failure to apply BIT defence of necessity, Thomson Reuters, Jan. 3, 2021.
• 198Continental Casualty Company, Award,  2008 ICSID ¶ 175.
• 199Id. ¶ 180.
• 200Johnson, supra note 51, at 11156.
• 201General Agreement on Tariffs and Trade (GATT) art. XXI, Oct. 30, 1947, 61 Stat. A-11, 55 U.N.T.S. 194.

  Article XXI Security Exceptions

  Nothing in this Agreement shall be construed

  (b) to prevent any contracting party from taking any action which it considers necessary for the protection of its essential security interests

  (i) relating to fissionable materials or the materials from which they are derived;

  (ii) relating to the traffic in arms, ammunition and implements of war and to such traffic in other goods and materials as is carried on directly or indirectly for the purpose of supplying a military establishment;

  (iii) taken in time of war or other emergency in international relations

• 202Angel Samuel Seda et al., Award, 2024 ICSID ¶ 687.
• 203Id. ¶ 688.
• 204Lode Van Den Hende et al., Landmark ruling on the WTO national security exception, Herbert Smith Freehills (June 7, 2019), https://perma.cc/8ZXD-J3H7.
• 205GATT Secretariat, One-page summary of key findings of Russia - Measures Concerning Traffic in Transit, DS512, https://perma.cc/KD48-EFPK (last visited Nov. 23, 2025).
• 206Panel Report, Russia – Measures Concerning Traffic in Transit¶ 7.27, WTO Doc. WT/DS512/R (adopted Apr. 5, 2019).
• 207Id. ¶ 7.28.
• 208Id. ¶ 7.26.
• 209Id. ¶ 7.29.
• 210Id. ¶ 7.102.
• 211Id. ¶ 7.101.
• 212Id. ¶ 7.83.
• 213Id. ¶ 7.77.
• 214Id. ¶ 7.126.
• 215Id.¶ 7.132.
• 216Id. ¶ 7.138.
• 217Id.
• 218Id. ¶ 7.134.
• 219Id. ¶ 7.138.
• 220Id. ¶ 7.145.
• 221Id. ¶ 7.146.
• 222Id. ¶ 7.149.
• 223Fabian Eichberger, When Reinforced Self-Judgment Meets Judicial Review: Insights from Seda v. Colombia, EJIL: Talk! (Sep. 10, 2024), https://perma.cc/PD3M-T53S.
• 224United States-Colombia Trade Promotion Agreement (TPA), U.S.-Colom., art. 22.2, May 15, 2012, U.S. Trade Representative, 125 Stat. 462.

  Article 22.2: Essential Security

  Nothing in this Agreement shall be construed:

  (a) to require a Party to furnish or allow access to any information the disclosure of which it determines to be contrary to its essential security interests; or

  (b) to preclude a Party from applying measures that it considers necessary for the fulfillment of its obligations with respect to the maintenance or restoration of international peace or security, or the protection of its own essential security interests.2

  2 For greater certainty, if a Party invokes Article 22.2 in an arbitral proceeding initiated under Chapter Ten (Investment) or Chapter Twenty-One (Dispute Settlement), the tribunal or panel hearing the matter shall find that the exception applies.

• 225Angel Samuel Seda et al., Award, 2024 ICSID.
• 226Id.
• 227Eichberger, supra note 223.
• 228Angel Samuel Seda et al., Award, 2024 ICSID ¶¶ 97–121.
• 229Id. ¶¶ 422–23.
• 230Eichberger, supra note 223.
• 231Id.
• 232Id.
• 233Angel Samuel Seda et al., Award, 2024 ICSID ¶ 638.
• 234Id. ¶ 642.
• 235Id. ¶ 643.
• 236Id. ¶ 646.
• 237Id. ¶ 650.
• 238Id. ¶ 655.
• 239Id. ¶¶ 712–13.
• 240Id. ¶¶ 719–22.
• 241Id. ¶ 748.
• 242Id. ¶ 765.
• 243Id. ¶ 769.
• 244Id. ¶ 786.
• 245Id. ¶ 787.
• 246Id. ¶ 795.
• 247India–Singapore Comprehensive Economic Cooperation Agreement, India-Sing., art. 6.12(4), Aug. 1, 2005, https://perma.cc/L89R-ZNA2 (“This Article shall be interpreted in accordance with the understanding of the Parties on non-justiciability of security exceptions as set out in their exchange of letters, which shall form an integral part of this Agreement.” (emphasis added)).
• 248Angel Samuel Seda et al., Award, 2024 ICSID ¶ 720.
• 249Id. ¶ 655.
• 250See id. ¶¶ 700–05. The newer generation of U.S. IIAs include the “it considers necessary” clause to avoid doubt around self-judging. See also Eichberger, supra note 223, noting that India also wants to establish self-judgment as a tool.
• 251Press Release, Security Council, Security Council Holds First-Ever Debate on Impact of Climate Change on Peace, Security, Hearing Over 50 Speakers, U.N. Press Release SC/9000 (Apr. 17, 2007).
• 252Peter Schwartz & Doug Randall, An Abrupt Climate Change Scenario and Its Implications for United States National Security 1 (2003).
• 253NATO, Strategic Concept for the Defence and Security of the Members of the North Atlantic Treaty Organization 13 (2010).
• 254Terry P. Kelley, Global Climate Change Implications for the United States Navy, U.S. Naval War College Archives (May 15, 1990), https://perma.cc/PM3L-3DWZ.
• 255Environment, Climate Change and Security, NATO (July 18, 2024), https://perma.cc/T8K6-ECHG.
• 256Sherri Goodman & Pauline Baudu, Climate Change as a “Threat Multiplier”: History, Uses and Future of the Concept 1 (Erin Sikorsky & Francesco Femia eds., 2023).
• 257Marshall Burke, Solomon M. Hsiang & Edward Miguel, Climate and Conflict, 7 Ann. Rev. of Econ. 577, 577 (2015).
• 258National Intelligence Council, Climate Change and International Responses Increasing Challenges to US National Security Through 2040 (2021).
• 259Renée Cho, Why Climate Change Is a National Security Risk, Columbia Climate School (Oct. 11, 2023), https://perma.cc/5YTH-FMAS.
• 260Angel Samuel Seda et al., Award, 2024 ICSID ¶ 643.
• 261Id.; Russia, supra note 206.
• 262See Russia, supra note 206 ¶ 7.138. The good faith requirement “applies not only to the Member’s definition of the essential security interests said to arise from the particular emergency in international relations, but also, and most importantly, to their connection with the measures at issue.”
• 263Angel Samuel Seda et al., Award, 2024 ICSID ¶¶ 650, 722.
• 264See Discovery Global, Respondent’s Counter-Memorial, 2023 ICSID ¶ 243.
• 265See Kasenetz, supra note 44.
• 266Johnson, supra note 51, at 11158.
• 267Gabčíkovo-Nagymaros Project (Hung./Slovk.), Judgment, 1997 I.C.J. 7, ¶ 51 (Sep. 25).
• 268Continental Casualty Company, Award, 2008 ICSID ¶ 181.
• 269Id. ¶ 187.
• 270Id. ¶ 196.
• 271See generally IPCC, Climate Change 2023 Synthesis Report (2023), https://perma.cc/LY92-H4W7.
• 272Säde Hormio, Collective Responsibility for Climate Change, 14 WIREs Climate Change 1, 2 (2023).
• 273See generally IPCC, supra note 271.
• 274Monica Crippa, et al., GHG Emissions of All World Countries 2025, at 21, 2025, https://perma.cc/GZ5K-B5FJ (last visited Nov. 23, 2025).
• 275See Kasenetz, supra note 44.
• 276Paris Agreement to the United Nations Framework Convention on Climate Change, Dec. 12, 2015, T.I.A.S. No. 16-1104.
• 277Nationally Determined Contributions (NDCs), UNFCC, https://perma.cc/27SK-5VU9 (last visited Nov. 23, 2025)
• 278Paris Agreement to the United Nations Framework Convention on Climate Change, Dec. 12, 2015, T.I.A.S. No. 16-1104.
• 279Id. art. 2(1)(a).
• 280Id. art. 4(1).
• 281Id. art. 4(2).
• 282See Di Salvatore, supra note 20.
• 283See generally Henckels,supra note 149.