The ICJ and the protection of foreign property under customary international law: quid novi?

Abstract

The protection of foreign property has been a key feature of CIL for at least the last century. The central protection under this regime is the MST. Still, the more specific investment protection regime reached its apex around 2010 with thousands of BITs in force. This article focuses on customary property protection, the scope of which exceeds that of BITs because it protects not only investors, but also any foreigner. To identify the relevant CIL rules, this article assesses the International Court of Justice Certain Iranian Assets case against the backdrop of previous cases. To this end, it first analyses elements characterized as CIL by the Court, eg police powers. Secondly, it deals with elements whose CIL character must be found between the lines of this case, such as the obligation to compensate indirect takings. Finally, it discusses legal questions the Court peremptorily set aside, for example whether assets’ blocking constitutes indirect expropriation.

INTRODUCTION

This article does not purport to reinvent the wheel. Instead, in times of uncertainty, it seeks, amid a plethora of articles on investment law and customary international law (CIL), to bring a modest contribution to the study of CIL norms relating to the protection of foreign property. It aims to identify those customary international norms that pertain to foreign property protection. As such, this article provides what some have called a ‘shopping list’¹ alongside what could be labelled as a ‘wish list’ of CIL norms applicable in the realm of property protection. To identify such norms and their content, this article analyses the recent Certain Iranian Assets judgment against the backdrop of previous decisions of the International Court of Justice (ICJ or ‘the Court’) in which the Court dealt with issues pertaining to the international protection of property. Rather than proceeding to identify CIL norms and their genealogy directly through an analysis of state practice and opinio juris,² this article relies on the case law of the ICJ and other international courts and tribunals, which, although not a source of law, constitutes a subsidiary means for the identification of CIL rules.³

Although this article takes the Certain Iranian Assets case as a starting point, it is not a case note and does not purport to cover all relevant legal aspects or arguments to be found in this case. Instead, the body of law examined in this article goes beyond this case and extends to the case law of the ICJ, alongside occasional references to other international courts and tribunals.

This CIL identification effort takes place within the context of the relative ice age experienced by bilateral investment treaties (BITs) and Investor–State Dispute Settlement (ISDS).⁴ Even though since the beginning of the ISDS era in the 1990s,⁵ BIT protections have overtaken CIL,⁶ this latter body of law has evolved and may now be the only protection available in a growing number of instances. Indeed, despite the importance of BIT networks, they only cover a small share of bilateral relationships between countries and remain the exception rather than the rule.⁷ In addition, BITs often incorporate CIL by reference,⁸ which makes CIL identification highly relevant, even within the context of international investment law.

Be that as it may, CIL protects a broader category of legal persons. Indeed, this body of law has the advantage of safeguarding foreigners’ interests in general rather than investors alone, as CIL does not distinguish between those two categories.⁹ Likewise, it protects investors even in the absence of any BIT.¹⁰ For instance, an individual who acquires real property abroad or who has a bank account in another country will benefit from CIL rules protecting the property of foreigners, regardless of whether they are considered an investor.

BITs and other related treaties, for example treaties of Friendship, Commerce, and Navigation (FCN), undeniably contributed to the evolution of CIL. But a question remains: to what extent? One issue with BITs is that most of the time they have been interpreted by ISDS tribunals which often offer too little support or evidence when characterizing a norm as CIL. Another concern at the time of assesing the contribution of BITs in shaping CIL is that the same body of investor-state awards has generally been criticized for its lack of coherence, which casts doubt on the extent to which ISDS arbitration contributes to the identification of CIL.¹¹ Yet, many legal notions, such as indirect expropriation or police powers, have been extensively developed in ISDS arbitral awards and have often been characterized as CIL. By way of illustration, the arbitrators in Saluka v Czech Republic declared the police powers doctrine, which shields states from the duty to compensate a foreigner impacted by general regulations when acting in the public interest, is CIL.¹² The arbitrators solely relied on other ISDS awards to support this conclusion and the paragraphs cited as authorities were often wrong. For instance, Saluka’s arbitrators cited Organisation for Economic Co-operation and Development (OECD) documentation on indirect expropriation while erroneously attributing it to the Methanex award.¹³ The arbitrators also relied on other awards’ paragraphs summarizing the parties’ arguments rather than the arbitral awards themselves. In light of the above, the varying quality of ISDS awards¹⁴ is yet another reason to examine general Public International Law (PIL), to determine which rules governing the protection of property, including in relation to foreign investment, are CIL.¹⁵

In any event, CIL identification beyond the realm of BITs and ISDS awards has particular merits. Indeed, as the Court said in the Diallo case, the existence of a myriad of BITs ‘is not sufficient to show that there has been a change in the customary rules […] it could equally show the contrary’.¹⁶ Moreover, as Bindschedler highlighted, the body of property or investment protection treaties is not necessarily indicative of opinio juris but rather of political opportunities. For instance, the former Soviet Union (USSR) was bound by treaties protecting the property of foreigners against expropriation without compensation within the USSR without considering private property to be a right.¹⁷

Given that this article focuses on CIL, it will mostly rely on the ICJ, which is the highest PIL authority as the principal judicial organ of the United Nations. So far, the attention given to ICJ cases in relation to property protection has been limited, in part due to the scarcity of relevant cases.¹⁸ Yet, one recent ICJ decision, Certain Iranian Assets, cleared out many debates within the field of property protection and, incidentally, customary investment law.¹⁹ Even though the core of the decision relates to the 1955 Treaty of Amity, Economic Relations, and Consular Rights Between the United States of America and Iran (1955 Treaty),²⁰ the Court had to interpret generic terms such as ‘expropriation’ or CIL concepts such as police powers. This was the first instance where the Court directly examined questions relating to property protection since the Elettronica Sicula S.p.A. (ELSI) was decided by the Chamber. Other ICJ cases mentioned property protection issues, but only in passing. Cases directly adressing the question of the protection of private property have also been dismissed at the preliminary objection stage on account of a lack of admissibility, for example the Interhandel case, or jurisdiction, such as the Anglo-Iranian Co. case.²¹ Still, in addition to elements to be found in Certain Iranian Assets and ELSI, which related to amity and FCN treaties, the Court also gave some indicia, but no clear answer, on its stance regarding property protection in the famous Barcelona Traction case, as well as in the Diallo and Armed Activities on the Territory of the Congo cases. For instance, before Certain Iranian Assets, it was debatable whether expropriation had to be a transfer of title or if there could indeed be such a thing as indirect expropriation under CIL as per the ICJ case law.²² Hence, Certain Iranian Assets has been the occasion of some jurisprudential evolution, nay reversal, by the ICJ on the very question of what constitutes expropriation. This evolution reflects the non-static nature of CIL²³ as well as the Court’s dawning reliance on arbitration awards to identify CIL and may also be indicative of BITs’ influence²⁴ on the ever-evolving body of custom, be it ‘sage’ or ‘savage’.²⁵

According to d’Aspremont, there are six potential ‘candidates’ that could be characterized as CIL within the field of foreign property protection: the Minimum Standard of Treatment (MST), Fair and Equitable Treatment (FET), protection against and conditions for expropriation, standards of compensation, denial of justice, and due process.²⁶ Nonetheless, the MST is an umbrella protection made out of many sub-standards which protects foreigners abroad. This umbrella protection includes, among other things, the prohibition of unlawful expropriation,²⁷ the prohibition of discrimination against foreigners in the realm of property,²⁸ the obligation to provide full protection and security,²⁹ and so forth. Indeed, the MST is characterized as a ‘series of interconnecting and overlapping elements’ which relate not only to the protection of property but also to the protection of the human person when visiting a foreign state.³⁰ In addition to property-related protections, the MST encompasses the prohibition of denial of justice and discrimination against individuals, as well as due process and due diligence requirements.³¹ For this reason, Newcombe and Paradell use the plural form, minimum standards of treatment, to discuss the MST. In fact, a landmark case relating to the MST, Neer,³² relates to the conduct of a criminal investigation rather than property protection. Some authors even wonder if it would be necessary to include a reference to expropriation in a treaty already containing a MST provision which would include expropriation by definition.³³

However, whether FET is an independent standard or a reflection of the MST remains a contentious doctrinal issue.³⁴ Therefore, this article mainly focuses on the MST and FET, which arguably encompass all the CIL candidates listed by d’Aspremont.³⁵

Accordingly, this article unfolds in three parts and critically addresses the Court’s findings—and omissions—alongside the parties’ allegations in relevant cases. The focus of this article is on CIL. However, it is important to recall that in Certain Iranian Assets as in most cases analysed below, the ICJ had to interpret and apply treaties and not CIL—except when treaties incorporate CIL by reference. The ICJ case law nonetheless sheds light on CIL in many ways. Indeed, in some instances, the Court directly ruled on CIL by stating that the content of some provisions differed from CIL or by clarifying how specific CIL norms had to be factored into the operation of a treaty. Still, the Court’s interpretation of treaty provisions provides indirect indications of the scope of CIL, as ‘undefined’ terms such as ‘expropriation’ have the same normative content in treaties and under CIL.³⁶ For instance, in interpreting the word ‘expropriation’ in provisions governing expropriations under different treaties, the Court gave indications of the meaning of ‘expropriation’ under CIL.

The article’s first part relates to CIL elements that appear in plain sight in the Certain Iranian Assets case. The Court has spoken expressly on (i) whether police powers apply under CIL, (ii) whether the standard of the most constant protection and security extends to legal protection, (iii) whether FET is subsumed in MST and lastly, (iv) whether the corporate veil must be upheld and under what conditions it might be pierced. The second part addresses two issues that are only indirectly dealt with by the Court in Certain Iranian Assets, which are read between the lines of the judgment, namely (i) whether the MST forms part of CIL and (ii) whether rules governing indirect expropriation are customary. The third and last part deals with issues that the Court either partly addressed or refused to address and which it should nonetheless have tackled: (i) whether the freezing of funds can constitute an indirect expropriation and (ii) whether expropriatory measures can be challenged on the basis of their prospective application or if such measures can only be challenged once they have been implemented.

WHAT APPEARS IN PLAIN SIGHT

The Court’s decision in Certain Iranian Assets sheds light on topical questions. The first subsection discusses how the Court settled the debate on the CIL nature of the doctrine of police powers in international law, confirming that the invocation of such powers to regulate can be a valid defence on the merits while clarifying the conditions for successfully invoking it. The second subsection outlines that the standard of the most constant protection and security does not extend to legal protection, that is, the stability of the legal regime, but is instead confined to protection from ‘physical’ threats. The third subsection clarifies that, according to the Court, absent any express reference to CIL, the FET is a self-standing standard not to be equated with the MST. The last subsection shows that the Court brought additional certainty vis-à-vis the ‘strength’ of the corporate veil in international law, that is whether the corporate veil can be pierced and under what conditions. While the first three subsections deal with topics relatively new for the Court, it already had the occasion to pronounce on the last one.

Police powers as an affirmative defence

In the Certain Iranian Assets case, the key measures analysed by the Court related to US Executive Orders, which enabled domestic courts in the USA to use foreign state-owned assets to satisfy judgment creditors when a given state had been found guilty of terrorism. In this case, the property of affected companies, which possessed a separate legal personality albeit being ultimately owned by Iran, had been attached and used to execute judgments rendered by US courts against Iran, the companies’ ultimate owner.³⁷ In addition, the measures led to the blocking of additional assets held by states condemned for terrorism by US domestic courts, including when assets were held indirectly, for example when a state, in this case, Iran, ultimately owned a corporation.³⁸

In their pleadings, the two parties, Iran and the USA, put forward a different understanding of the applicability and scope of the police powers doctrine. This doctrine has been traditionally portrayed as emanating from general international law and being applicable regardless of it being explicitly referred to in a treaty. It provides for the right of states to enact general regulations in the public interest, in good faith, and in a non-discriminatory manner, without owing compensation to those affected.³⁹ The term ‘police’ has to be understood as a reference to public ‘policy’, for example taxes or environmental regulations, rather than ‘police forces’.⁴⁰ In ISDS case law, there is little doubt as to the applicability of the police powers under CIL.⁴¹ Some tribunals, such as the one established in the case Tecmed v Mexico, even affirmed that their applicability was ‘indisputable’.⁴²

In Certain Iranian Assets, however, Iran challenged the applicability of the police powers doctrine, stating that this doctrine was not mentioned directly in the text of the 1955 Treaty⁴³ and had not been considered during treaty negotiations—according to Iran, it was therefore not applicable.⁴⁴ In the alternative, Iran alleged that if the doctrine were applicable, the exercise of police powers would have to be non-discriminatory, designed and applied for a legitimate public purpose, proportionate, and not otherwise contrary to international law.⁴⁵ According to Iran, these conditions were, in any event, not satisfied, among other reasons, because the US measures, the attachment and execution of property and interests in property of Iranian companies without compensation, violated the 1955 Treaty. The USA, on the other hand, affirmed that the police powers doctrine was, in fact, a valid justification under international law regardless of the treaty’s text;⁴⁶ that the measures were taken in good faith and in a non-discriminatory manner;⁴⁷ and that, in any event, proportionality was not relevant to the police powers doctrine.⁴⁸ The USA also added that the judicial decisions at stake did not ‘give rise to a claim for expropriation’.⁴⁹

The Court did not spend much time on this issue. It plainly affirmed that the police powers doctrine was applicable under CIL and that a provision on the prohibition of uncompensated expropriation, such as Article IV:2 of the 1955 Treaty,⁵⁰ even in the absence of an express reference to police powers, did not undermine the rights of states to regulate. Indeed, as the Court put it, ‘[i]t has long been recognized in international law that the bona fide non-discriminatory exercise of certain regulatory powers by the government aimed at the protection of legitimate public welfare is not deemed expropriatory or compensable’.⁵¹ The Court, relying on the Saluka ISDS award, however, added that the powers of governments are not unlimited and that the reasonable character of a measure is ‘one of the considerations that limit the exercise of the governmental powers’.⁵² Although the Court did not clarify whether proportionality between the invoked public interest and the impacted nationals is a criterion, as alleged by Iran, there seems to be a certain overlap between the notions of reasonableness and proportionality. In this specific instance, the Court judged that US measures were contrary to the treaty and thus ‘not a lawful exercise of regulatory powers’.⁵³ The core of the Court’s finding seems to take its source in the fact that the US measures were already in breach of Article IV:1—a provision embodying reasonableness considerations⁵⁴—which could indeed indicate, as upheld by Iran, that measures taken in the exercise of police powers must be consistent with international law to be a valid exercise of police powers.⁵⁵

The standard of the most constant protection and security

A second issue that the Court swiftly settled relates to the standard of the most constant protection and security (sometimes referred to as full protection and security), which often appears in BITs and FCNs. This standard has traditionally been conceived as protecting foreigners against physical threats from either the host-state or third parties.⁵⁶ This standard appears in Article IV:2 of the 1955 Treaty, which provides that the level of protection can ‘in no case [be] less than [that] required by international law’, thus incorporating CIL by reference. Within the ISDS framework, this standard of protection appears to be subsumed in the MST⁵⁷ and has been understood as including standards of ‘due diligence or vigilance with respect to the physical protection’ of foreigners.⁵⁸ However, some ISDS tribunals, such as CME v Czech Republic, decided that the most constant protection and security included protection against legislative or administrative changes that could endanger an investment.⁵⁹

In ELSI, which is the only other ICJ case that has dealt with the standard of constant protection and security, the ICJ Chamber held that the standard, invoked in relation to the ‘physical’ protection of workers occupying a factory, was not an absolute standard because the most constant protection and security protection did not provide any ‘warranty that property shall never in any circumstances be occupied or disturbed’.⁶⁰ The Chamber noted that the obligation to provide full protection and security in the applicable treaty ‘must conform to the minimum international standard’.⁶¹

In Certain Iranian Assets, Iran argued that this standard also covered legal protection (as opposed to mere physical protection) and that the USA had failed to provide such protection by ‘removing generally applicable legal defences otherwise available to Iranian companies and by making those companies liable for purportedly wrongful acts of Iran in proceedings to which they were not parties’.⁶² The USA argued that this standard was ‘limited to protection from physical harm, and concerns the level of police protection required under CIL against acts of physical harm to a person or property’.⁶³

The Court sided with the USA on this issue and found that this standard only covers ‘physical’ protection and not legal security. Given the direct reference to CIL constituting a legal protection ‘floor’ in the 1955 Treaty, the Court’s position is, in fact, that CIL does not provide for legal security protection but rather for ‘physical’ protection only.⁶⁴ The Court also specified in an obiter dictum that this would be the case in most BITs and FCNs displaying similar drafting and that such protection extended to the actions of third non-state parties.⁶⁵ Although the Court did not directly address the CIL status of the full or most constant protection and security, it neither contradicted the US statements regarding its alleged customary nature nor the past decision by the Chamber in the ELSI case, which shows the Court acknowledges that this standard forms part of CIL.

FET

The third issue that the Court settled relates to FET, which is a standard commonly found in FCNs and BITs, that ensures the protection of citizens conducting business abroad. Scholars and ISDS tribunals are split between two approaches, with one side considering that a reference to the ‘fair and equitable treatment’ in a BIT constitutes an implicit reference to CIL, and another claiming that there must be an express reference to general international law for this standard to ‘be reflecting’ the MST.⁶⁶ On the one hand, the tribunals in Siemens v Argentina or Azurix v Argentina viewed FET as an independent standard from CIL, which would create favourable conditions and foster the investment climate.⁶⁷ On the other hand, BITs include FET under MST provisions, such as the US-Uruguay 2005 BIT and the US and Canadian BIT models,⁶⁸ which led many tribunals to apply the CIL-FET standard instead.⁶⁹ In general, the practice of many Western states, including the USA, Canada, Spain, and Switzerland, shows that they consider FET to be part of the MST.⁷⁰ This CIL-FET is not static concept. Many tribunals applying the CIL-FET have deemed that this standard, while setting a high threshold, has evolved and is not confined to the ‘outrageous’ or ‘shocking’ situations described in the Neer award and which have been considered in some instances to be the applicable thresholds under CIL.⁷¹ However, the majority of ISDS cases—when applying treaties that do not expressly frame FET as part of MST—seem to consider it to be a separate and independent standard offering heightened protection when compared to the CIL-FET standard.⁷²

In addition, other tribunals have taken a different approach and reconciled the two standards by equating them. For instance, in Gabriel Resources v Canada, the tribunal had to apply two BITs: one containing an autonomous FET provision and the other one referring to CIL-FET. In the end, it was decided that the standard was the same in both treaties, given the constant evolution of the customary FET standard.⁷³ Likewise, in Ruiz v Spain, the tribunal asserted in an obiter dictum that even if it had applied the autonomous FET standard labelled as ‘a more generous FET standard’, the conduct of Spain would not have reached the level of a breach.⁷⁴

When discussing the FET standard, found within Article IV:1 of the 1955 Treaty, the Court in Certain Iranian Assets decided that it would not examine ‘the content of the customary minimum standard of treatment’ due to the absence of an express reference to the MST. The Court hence rejected the American argument that the use of the wording ‘fair and equitable treatment’ within this provision ‘reflect[ed] one of the components of the customary minimum standard of treatment of aliens’.⁷⁵ It instead sided with Iran, which relied on ISDS cases and argued that the protection set forth in Article IV:1 entailed ‘freestanding obligations’ that were not ‘tied to the minimum standard of treatment under CIL’.⁷⁶

Although the Court decided that FET in the 1955 Treaty was not linked to the MST, it should have justified further how FET constitutes an independent standard, for example what does the free-standing FET offer in comparison with an MST-aligned FET? Despite the lack of further development on FET, the Court’s decision in Certain Iranian Assets settles a larger debate: absent any express reference to CIL, FET has to be construed as an independent standard offering protection going beyond the MST. This is congruent with the ISDS majority case law which opted for the same solution.⁷⁷

The strength of the corporate veil in international law

A fourth issue straightforwardly addressed by the Court relates on the one hand to the extent to which the ultimate (or even intermediate) owners of companies are shielded or hidden behind a corporate veil. On the other hand, the Court adressed whether there can be a ‘piercing’ or lifting of the corporate veil under some circumstances, that is identifying the owner or shareholders of a corporation and substituting their personality to that of the company, in spite of the independent legal status of a corporation. The treatment of these issues is relatively straightforward as it did not constitute new ground for the Court, which had ruled on the corporate veil in both Barcelona Traction and Diallo. In those cases, the fact that a corporation was validly established under the law of one state sufficed to confer the nationality of that state for diplomatic protection purposes, provided that a corporation possessed a legal personality. In contrast, the nationality of shareholders, the ultimate owners, was deemed irrelevant for such purposes, given that the corporations in both Barcelona Traction and Diallo had an independent legal personality. While is true that in the ELSI case, the USA could act in the name of the shareholders of an Italian corporation, this could only be done as the treaty at stake provided for protection extending to ‘interests held directly or indirectly’ in a given corporation.⁷⁸ ISDS tribunals frequently found that a corporation’s nationality depends on the law of the country of incorporation.⁷⁹ For instance, in the Tokios v Ukraine case, the tribunal held that the corporation was Lithuanian, as incorporated in this country, albeit being almost entirely owned by Ukrainians.⁸⁰ ISDS cases where tribunals accepted to pierce the corporate veil of corporations exist but remain marginal, for example Venkolim v Venezuela.⁸¹

However, those ISDS cases did relate directly to the question of the investors’ nationality. In Certain Iranian Assets, there was no doubt regarding the corporations’ nationality, which were undoubtedly Iranian. The relevant question was whether the corporate veil could be lifted to hold Iran responsible as the ultimate owner of the corporations. Iran argued that the affected corporations—despite all being ultimately owned by Iran—should be considered separate and independent entities, whose conduct could not be attributed to the State, because they were constituted so under domestic law.⁸² For Iran, the upholding of the ‘independent legal personalities’ of corporate entities constituted a ‘fundamental rule’ grounded in ICJ case law and notably in the Barcelona Traction and Diallo cases.⁸³ The USA argued that it was ‘both reasonable and justified’ to pierce the companies’ corporate veil, thus tying them to Iran, so as to enable American victims ‘holding terrorism-related judgments against Iran to attach assets of Iran’s agencies and instrumentalities’.⁸⁴ The USA argued that there was a principle of law, grounded in both common law and civil law, ‘that it may be appropriate to pierce the corporate veil or otherwise disregard the distinction between a corporation and its shareholders in the interests of justice’.⁸⁵

The Court sided with Iran as it found that upholding the independent legal personality of a corporation is indeed the basic rule under PIL, confirming its past jurisprudence.⁸⁶ However, the Court added, mostly relying on the Barcelona Traction case, that the corporate veil could be lifted when ‘justified and equitable in certain circumstances or for certain purposes’.⁸⁷ It decided that the conditions for lifting the corporate veil were not met in this case because the American measures ‘caused an impairment of the Iranian companies’ rights that was manifestly excessive when measured against the protection afforded to the purpose invoked’. The Court found the companies were targeted by liability judgments in which they could not participate in and ‘in relation to facts in which those companies do not appear to have been involved’.⁸⁸ Here, in addition to confirming its own jurisprudence, the Court confirmed what the majority of ISDS tribunals have held, that is, that circumstances must truly be extraordinary for the corporate veil to be pierced. The court also confirmed that to determine whether those circumstances are reunited, a balance must be struck between a corporation’s interests and the ‘purpose invoked’ by the party which wishes to pierce said corporation’s corporate veil.

WHAT MUST BE FOUND BETWEEN THE LINES

In addition to what appears in plain sight, one may also find indicia in the Certain Iranian Assets case on the state of international law vis-à-vis (i) the scope of the customary MST and (ii) the existence of legal safeguards for property owners facing situations of indirect expropriation (as opposed to a valid transfer of title).⁸⁹

The customary international MST

As held above, the MST is the core of the individual and property customary protection in international law. The MST virtually encompasses all the minimum standards under CIL, such as the prohibition of expropriation without compensation.⁹⁰ The MST counter-doctrine, the Calvo doctrine,⁹¹ which provides that foreigners only benefit from states’ national treatment obligations appears to have fallen into desuetude, although some countries, such as Venezuela, still refer to it episodically.⁹²

As held above, in Certain Iranian Assets, the Court found that FET treaty provisions are not to be conflated with the MST but are rather to be considered as a self-standing protection in the absence of any express reference to CIL. For this reason, one can discern what is not part of the MST, that is, the protection afforded by the autonomous FET. However, the Court did not provide much explanation as to what might be part of the MST, except for the standard of full protection and security. Indeed, although there now seems to be a general recognition of the MST as a reflection of CIL,⁹³ it still remains the object of many debates as to what it does include and the scope of included rights.

Yet, clarifying the scope of the MST is paramount for at least two reasons: first, the MST applies even when no treaty law is applicable. Second, as noted above, many BITs do incorporate by reference or mention the MST,⁹⁴ which gives rise to various and sometimes conflicting interpretations of the MST’s scope by ISDS tribunals.⁹⁵

In fact, the MST has often been equated with CIL. For instance, the Permanent Court of International Justice already recognized the existence of the MST under the generic denomination of ‘international law’ in Certain German Interests in Polish Upper Silesia when it affirmed that the exceptions on expropriation in the convention at stake were derogations ‘from the rules generally applied in regard to the treatment of foreigners and the principle of respect for vested rights’ and referred to measures prohibited under ‘international law […] in respect of foreigners’ such as uncompensated expropriation ‘and similar measures’.⁹⁶ In French, the judgment’s authoritative text, the Court even refers to the ‘droit international commun’.⁹⁷ A few years earlier, in the Norwegian Shipowners’ case, the arbitral tribunal had held that the ‘inviolability’ or respect for the ‘private property of foreigners’, at least from friendly states, was protected and ‘in complete accord with the international public law of all civilized countries’.⁹⁸

The ICJ also indirectly referred to the MST in a few cases. In Barcelona Traction, the Court indicated that ‘[w]hen a State admits into its territory foreign investments or foreign nationals, whether natural or juristic persons, it is bound to extend to them the protection of the law and assumes obligations concerning the treatment to be afforded them’ (emphasis added).⁹⁹ It also indirectly touched upon the MST when ordering provisional measures in the case of the United States Diplomatic and Consular Staff in Tehran. In this case, the Court clarified that the purpose of the consular functions (protecting, assisting, and safeguarding the interests of nationals) under the 1963 Vienna Convention on Consular Relations is ‘precisely to enable the sending State, through its consulates, to ensure that its nationals are accorded the treatment due to them under the general rules of international law as aliens within the territory of the foreign State’ (emphasis added).¹⁰⁰

Then, in the ELSI case, the ICJ Chamber had no choice but to examine the content of the MST since Article V of the 1948 FCN between the USA and Italy provided for ‘the full protection and security required by international law’.¹⁰¹ In this case, the USA affirmed, as it did in Certain Iranian Assets, that most treaty provisions did not set higher standards than the MST,¹⁰² even in relation to provisions not making an express reference to international law, such as the treaty’s provision on due process.¹⁰³ Italy did not object, possibly because it had no interest in doing so under the particular circumstances of the case. The Chamber did not directly address the issue, yet it stated that it had to interpret provisions that set standards ‘which may go further in protecting nationals of the High Contracting Parties than general international law requires’ (emphasis added).¹⁰⁴ Without further explanation, the Chamber decided that, in any event, the delays at stake were not ‘falling below that standard’.¹⁰⁵ Hence, the Chamber seems to have mostly judged the Italian measures against the backdrop of CIL as incorporated by the treaty.

In the Diallo case, Guinea alleged that the arrest, detention, and expulsion of Mr Diallo from the Democratic Republic of Congo amounted to violations of the ‘minimum standard of civilization’, which included, amongst other things, the obligation to ‘respect the freedom and property of aliens’.¹⁰⁶ The Court did not pronounce on expropriation claims given that the companies were Congolese rather than Guinean (their claims could, therefore, not be espoused by Guinea despite their ultimate ownership by a Guinean national). Nonetheless, the Court quoted the aforementioned Guinean legal understanding of the MST without challenging it. The Court also added in an obiter dictum that although diplomatic protection was an institution ‘originally limited to alleged violations of the minimum standard of treatment’, it had expanded to cover human rights violations.¹⁰⁷ This might be considered an additional, yet timorous, recognition of the MST and its contours by the Court, as well as its status within the realm of diplomatic protection.

In Armed Activities on the Territory of the Congo (Democratic Republic of the Congo v. Uganda), Uganda alleged that its public property was expropriated.¹⁰⁸ The Court, when analysing this issue (see infra for further explanations), held that this part of Uganda’s second counterclaim was based ‘on the international minimum standard relating to the treatment of foreign nationals who are present on a State’s territory’.¹⁰⁹

In Certain Iranian Assets, the Court omitted to elaborate on the MST. Yet, between the lines, it provided some useful information on the MST’s scope. As stated above, it decided not to refer to it within the context of FET as it deemed it was a separate standard given that Article IV:1 did not include an express reference to CIL.¹¹⁰ Even when analysing Article IV:2 of the 1955 Treaty regarding the most constant protection and security, which ‘in no case’ can be ‘less than that required by international law’, the Court refrained from referring to the MST. Nonetheless, the Court held that ‘formulations’ such as ‘required by international law’ are ‘sometimes associated with’ the MST whose existence it recognized in passing by saying it did not ‘need to examine the content of the customary minimum standard of treatment’.¹¹¹ As seen above, the Court recognized that the standard of the most constant security and protection, understood as ‘physical’ protection, is part of the MST. Likewise, the Court minimally clarified that formulations such as ‘required by international law’ were to be understood as references to the MST¹¹² and that the MST is indeed part of CIL.¹¹³

All in all, the jurisprudence of the Court and its predecessor already acknowledged—albeit succinctly—the applicability and importance of the MST under CIL. However, in Certain Iranian Assets, the Court missed an occasion to clarify the scope of the MST. It would have been of interest that the Court indeed examine further the content of the MST and the links between it and other standards for three reasons. First, given the generic wording of the 1955 Treaty's provisions, it is likely that at least part of them are reflective of the MST and CIL. Secondly, the Court could have engaged with it because of the direct reference to CIL in Article IV:2 concerning the most constant protection and security. Third, the Court, as the principal judicial organ of the United Nations, ought to have taken a position on this standard, which is fundamental and, reflects a universal sense of justice, despite conflicting interpretations in the case law.¹¹⁴  In casu, explaining further the scope of the MST was not strictly necessary to settle the dispute. However, the Court often resorts to obiter dicta, to serve as guidance and shed light on topical international law questions.¹¹⁵ Indeed, the task of the Court is not only to settle disputes but also to be the guardian of the international legal order and of international law’s coherence by stating the law more largely, that is, by exercising its juris dictio.¹¹⁶ Thus, even when some pronouncements are not strictly needed, the Court can resort to obiter dicta,¹¹⁷ and should do so to ensure a minimum of coherence at a time when international courts and tribunals are multiplying.¹¹⁸

Indirect expropriation under CIL

Another issue the Court indirectly clarified in Certain Iranian Assets relates to whether there is a prohibition on indirect expropriation or taking (without compensation) under CIL, that is whether the prohibition on uncompensated expropriation extends to situations other than those involving a formal transfer of property title following an expropriatory measure by a state. As a reminder, in this case, the property of affected companies had been attached and used to execute judgments rendered by US courts against Iran, the ultimate owner of those companies.¹¹⁹ Although the property indeed was forcibly transferred, it had been transferred in the course of domestic legal proceedings, which determined that the companies’ ultimate owner, Iran, had to make good debts it incurred on account of US domestic judgments.¹²⁰ The property at stake was thus not openly expropriated.

The scope of the protection of foreign private property in international law remains somehow unclear, both under CIL and in relation to BITs.¹²¹ As a matter of fact, the range of what can be considered indirect expropriation is now a ‘central issue in international investment arbitration’,¹²² although there seems to be little doubt that indirect expropriation is now included in the plain meaning of ‘expropriation’ under BITs and related treaties.¹²³ Indeed, a significant share of BITs, such as BITs made as per the American and British models¹²⁴ and other multilateral agreements, such as the North American Free Trade Agreement (NAFTA), the Energy Charter (ECT) or the Association of Southeast Asian Nations Treaty,¹²⁵ and more recently the Canada-United States-Mexico Agreement (CUSMA), refer explicitly to indirect expropriation. Those treaties use formulas such as ‘measure tantamount to nationalization or expropriation’,¹²⁶ ‘having an effect equivalent to nationalization or expropriation’,¹²⁷ or ‘measure equivalent to expropriation or nationalisation’.¹²⁸ Therefore, most of the body of investment arbitration awards is of little help to determine whether indirect expropriation is covered within the CIL prohibition against expropriation for the very simple reason that most investment treaties contain explicit references to indirect expropriation. This also extends to the case law of the Iran-US Claims Tribunal, which is competent to assess ‘other measures affecting property rights’. Therefore, the arbitral case law should thus be considered with great care at the time of assessing CIL.¹²⁹

If one follows the body of ISDS awards, the mere notion of ‘expropriation’ without any qualifier indeed includes indirect expropriation.¹³⁰ For instance, in Pope & Talbot v Canada, the tribunal held that measures ‘tantamount to nationalization or expropriation’ were in line with the ‘ordinary concept of expropriation under international law' (emphasis added),¹³¹ which was later confirmed in SD Myers v Canada.¹³² Direct expropriation has been understood as a ‘forcible taking by the government […] by means of administrative or legislative action to that effect’¹³³ or ‘open, deliberate and acknowledged takings of property’.¹³⁴ Indirect expropriation, rather, takes place when governmental measures taken by any organ¹³⁵ have a similar effect to a taking which can be labelled as de facto or disguised expropriation and relates to the ‘deprivation’ of a property’s value or ‘prevention of enjoyment’ of property rights such as the ‘ability to manage, use or control’ property.¹³⁶ Some ISDS awards further recognize that indirect expropriation can take the form of orders by domestic courts.¹³⁷ For example, in Saipem SpA v Bangladesh, it was found that the refusal by a domestic court to enforce a valid award was tantamount to expropriation.¹³⁸ In Swisslion Doo Skopje v Macedonia, the tribunal reiterated the findings of the tribunal in Saipem and held that domestic court’s decisions could be expropriatory when illegal.¹³⁹ Indeed, in principle, a transfer of property ordered by a court will not be considered expropriation if it serves to make good a debt, but it will constitute expropriation as soon as its nature is confiscatory.¹⁴⁰ Accordingly, other ISDS tribunals considered that a judicial decision could only be expropriatory when there was a denial of justice and that the ‘executive or legislative branches interfered with court rulings so as to effectively cause an expropriation’.¹⁴¹

In addition to BITs and similar instruments, the prohibition of indirect expropriation is also present in the US Third Restatement of the Foreign Relations Law of the US,¹⁴² the OECD 1967 draft Convention on the Protection of Foreign Property,¹⁴³ and the 1961 Harvard Draft Convention on the International Responsibility of States for Injuries to Aliens.¹⁴⁴ It is interesting to note that all these restatements or drafts aimed to represent ‘recognized rules of international law’ in relation to expropriation rather than to provide for heightened protection.¹⁴⁵

Nonetheless, since the aforementioned awards concerned BITs that explicitly refer to indirect expropriation, whether indirect expropriation is prohibited under CIL remains unclear. In particular, the ICJ qualms on indirect expropriation entertain doubts as to the CIL character of rules governing indirect interpretation. Indeed, the ICJ did distinguish between what it labelled as expropriation ‘in the technical sense’, that is, a forcible transfer of legal title, which it considered illegal, and indirect expropriation, which it only dealt with in a hypothetical fashion until Certain Iranian Assets.

In ELSI, the Chamber engaged hypothetically with the question of whether the word ‘taking’ did encompass ‘unreasonable interference with [property’s] use, enjoyment or disposal’ in addition to ‘outright expropriation’.¹⁴⁶ It held, ‘without deciding’, that the word expropriation ‘might be wide enough to include not only formal and open expropriation, but also a disguised expropriation’ of ‘an Italian corporation’ (emphasis added).¹⁴⁷ Yet, it affirmed that ‘the possibilities of disguised expropriation or of a “taking” amounting to expropriation [did not] have to be resolved’ in ELSI since the most important factor to consider was the poor financial situation of the company, which would have likely gone bankrupt regardless of the Italian measures.¹⁴⁸ The Chamber nonetheless indicated in an obiter dictum that had ELSI remained solvent and had the treaty covered indirect taking, the requisition could have constituted a ‘significant deprivation’, that is, a potential expropriation.¹⁴⁹

In light of this case, the ICJ, in 1989, was not ready to affirm the possibility of indirect expropriation being subsumed in the notion of expropriation tout court. Indeed, the Chamber appears to have applied CIL standards in general, given the fact that the parties did not dispute that general international law standards were applicable (see supra). This interpretation is relevant when analysing the Certain Iranian Assets case, as the provisions at stake were drafted in a similar fashion in the 1948 USA–Italy and 1955 USA–Iran treaties and simply referred to ‘taking’ without including further detail. In this case, the Chamber adopted a rather restrictive approach vis-à-vis the customary (and even conventional in this case) nature of the rules governing expropriation under international law.

Later, in the Armed Activities on the Territory of the Congo (Democratic Republic of the Congo v Uganda) case, 16 years after ELSI and 18 before Certain Iranian Assets, the stance of the Court appeared unchanged as it seemed to entertain further doubts. In that case, Uganda’s first counterclaim related to the mistreatment of its national, while the second counterclaim was that Ugandan diplomatic premises in the DRC had been ‘forcibly seized’ in Kinshasa and that Ugandan property had been stolen.¹⁵⁰ The premises were occupied for at least four years, and the parties agreed that all the ‘movable property’ of Uganda was missing and that the premises were ‘in a state of total disrepair’.¹⁵¹ At the time of the ICJ judgment, the premises were still occupied—they were so until April 2005¹⁵²—and Uganda claimed it had to rent new diplomatic premises on account of this situation.¹⁵³ Accordingly, in addition to its allegations of violations of the 1961 Vienna Convention on Diplomatic Relations (VCDR), Uganda claimed that the ‘seizure of the Embassy of Uganda, the official residence of the Ambassador and official cars of the mission, these actions constitute an unlawful expropriation of the public property of Uganda’ (emphasis added).¹⁵⁴ As seen above, although Uganda did not provide the Court with additional information on the source of the prohibition of unlawful expropriation, the Court understood that it was based ‘on the international minimum standard […]’.¹⁵⁵ However, the Court held that it would not examine this claim of expropriation in relation to Uganda’s property as the property at stake, the embassy, was already protected under the VCDR.¹⁵⁶ In any event, the Court stated a few paragraphs later that ‘there [was] nothing to suggest that in this case any confiscation or expropriation took place in the technical sense. The Court, therefore, finds neither term suitable in the present context. Uganda appears rather to be referring to an illegal appropriation in the general sense of the term’ (emphasis added).¹⁵⁷ In the end, the Court declared in the Armed Activities case that the seizures constituted unlawful uses of the property but did not amount to an expropriation since ‘no valid transfer of the title to the property ha[d] occurred and the DRC has not become, at any point in time, the lawful owner of such a property’ (emphasis added).¹⁵⁸

This shows that for the Court in 2005, there could only be direct expropriation involving a transfer of title under the MST. Indeed, the Court emphasized that expropriation ‘in the technical sense’ includes a valid transfer of title by contrast with indirect expropriation. Thus, had the DRC transferred the title, there might have been a violation of the MST. However, the theft of movable property—by the Congolese army, a state organ—was not analysed by the Court despite the fact that there was hardly any possibility that Uganda would recover its property. In the end, Uganda, in spite of its financial losses, renounced most of its claims (including the loss of movable property), as it deemed that satisfaction made good the loss.¹⁵⁹ It finally relinquished all its claims during oral pleadings.¹⁶⁰

However, despite the hesitations of the Court on indirect expropriation in the Armed Activities and ELSI cases, several separate opinions recognize that there could be such a thing as ‘disguised expropriation’ in other ICJ cases. For instance, the opinions of Judges Fitzmaurice and Gros in the Barcelona Traction case include references to indirect expropriation.¹⁶¹ In addition, Switzerland in the Interhandel case¹⁶² (see explanations infra on the question of the freezing or blocking of assets) or France in the case of the Compagnie du Port, des Quais et des Entrepôts de Beyrouth and Société Radio-Orient¹⁶³ both alleged there was already customary protection against uncompensated indirect expropriation in the 1960s. For instance, France deemed that imposing various taxes on a company supposed to be exempted from them under a treaty could be assimilated to an expropriation of the company’s rights, even without a transfer of title.¹⁶⁴

Although the Certain Iranian Assets case did not settle the question of indirect expropriation directly, it impliedly answered the question. Indeed, although the Court had to interpret a treaty rather than CIL, its assessment is particularly helpful to grasp the scope of expropriation under CIL given the treaty only refers to ‘taking’ or ‘expropriation’ without further clarification. Indeed, the 1955 Treaty is not a BIT in the modern sense. The fact that a treaty of amity is at stake is relevant when assessing the scope of CIL. First, the protection of the 1955 Treaty is directly afforded to all citizens—rather than to the narrower category of investors under BITs. This is in line with the MST, which protects the citizens of any country wherever and whenever they travel or conduct business abroad. While this article does not purport to engage in this debate, it can be said that regardless of their effective contribution, investors, in principle, receive heightened protection by BITs on account of their economic contribution (the investment), while the MST is not concerned with such considerations.¹⁶⁵ Second, in this case, the stipulations of the treaty are rather plain and simple when compared to modern-day BITs and, therefore, shed more light on the protections afforded under CIL.

In Certain Iranian Assets, the Court assessed whether the ‘attachment and execution’ of the property and interests in property constituted ‘takings’ contrary to Article IV:2 of the 1955 Treaty.¹⁶⁶ This provision provides that a ‘taking’ is only authorized when a measure is taken for a public purpose and is accompanied by the prompt payment of a just compensation.

According to Iran, the American measures had the effect of ‘stripping’ their banks and financial enterprises of their assets.¹⁶⁷ Iran, by claiming indirect or creeping expropriation, equated the word ‘taking’ with ‘expropriation’.¹⁶⁸ Indeed, according to Iran, the verb to ‘take’ included both direct and indirect acts of expropriation and covered the actions of the legislative, executive, and judiciary organs. Iran added that domestic courts’ decisions could constitute expropriation when the legislative or executive acts applied by the courts were ‘expropriatory in nature’.¹⁶⁹ Iran thus affirmed that the property had been ‘taken’ without compensation since the affected companies received no payment.¹⁷⁰

The USA first argued that the measures could not constitute expropriation because they had been ordered by US courts (although these courts merely applied the law adopted by US executive and legislative organs) and that a decision by a court could only amount to expropriation if there were ‘an additional element of illegality’ in the court’s conduct or in the ‘chain of events leading to the decision’.¹⁷¹ In a way, the USA was trying to revive the argument that states could use their internal law to justify the non-performance of a treaty. The USA affirmed that its courts only applied the law.¹⁷² However, in the history of international expropriations, and especially in the wake of WWII, there has been a plethora of confiscations ordered by domestic courts based on domestic law, including penal law. Although respecting the provisions of their national orders, such confiscation orders were often only a ‘pretext’ for spoliation as the impacted private individuals themselves had done no wrong.¹⁷³ The second unsuccessful line of argument of the USA was related to the doctrine of police powers, as previously examined in this article.

The Court sided with Iran and equated the term ‘taking’ with ‘expropriation’. It first clarified that judicial decisions ordering the attachment and the execution of property did not ‘per se constitute a taking or expropriation of that property’, but could be considered ‘compensable expropriation’ if an element of illegality was present.¹⁷⁴ It then explained that such an element of illegality could be present when ‘a deprivation of property results from a denial of justice’ or when a judicial organ applies ‘legislative or executive measures that infringe international law’.¹⁷⁵ Then, it concluded that the application of sections 201 (a) of the Terrorism Risk Insurance Act and 1610 (g) of the Foreign Sovereign Immunities Act by the USA ‘amounted to takings without compensation of the property’ (emphasis added).¹⁷⁶ Hence, the Court recognized that indirect expropriation is subsumed in expropriation by recognizing that the measures were not as such takings or expropriations but that those measures amounted to expropriation. This conclusion is welcome as it seems to be indicative of a change of attitude of the ICJ when considered against the backdrop of past Court decisions.

Indeed, the conclusion of the Court in Certain Iranian Assets appears in contradiction with the doubts expressed by the Chamber in the ELSI or Armed activities cases. This is a radical change considering that the provisions of the 1955 Treaty of Amity between Iran and the USA, on the one hand, and the 1948 Treaty of Friendship, Commerce and Navigation Between the United States of America and the Italian Republic,¹⁷⁷ on the other hand, are quite similar. For instance, Article IV:2 of the 1955 treaty provides that ‘such property shall not be taken’, while Article V:2 of the 1948 treaty between Italy and the USA provides that ‘the property […] shall not be taken’. Neither treaty specifies whether the ‘taking’ can be indirect or not.

Therefore, the Certain Iranian Assets case settled the issue of indirect expropriation as the Court now considers that the verb to ‘take’ is indeed ‘wide enough’ to include more than ‘formal and open expropriation’ as the Chamber put it in ELSI.¹⁷⁸ This change of attitude between the Court’s judgments in 1989 and 2023 is significant since the Chamber and the Court had to interpret two treaty provisions that are identical in essence and yet arrived at different conclusions, against the backdrop of CIL. Despite a significant body of investment awards having interpreted ‘expropriation’ and ‘takings’ in a broad fashion, this was not a given. Indeed, the Court was reluctant to recognize indirect expropriation in past cases, despite many BITs explicitly specifying that indirect and creeping expropriation are prohibited in the same way as direct expropriation. In Certain Iranian Assets, the Court might have done what it has done in the past and proceeded to a contrario interpretation of similar provisions in other treaties.¹⁷⁹ Indeed, many treaties do explicitly mention indirect expropriation, which could have led the Court to find that the provisions of the 1955 Treaty must be strictly interpreted to include direct expropriation only.¹⁸⁰ For instance, as discussed above, the Court decided that FET was not to be equated to the MST, given that Article IV:1 of the 1955 Treaty did not include an express reference thereof while other provisions and treaties did. Yet it did not do so in relation to expropriation, and a comparative analysis of the Court’s and the Chamber’s findings in Certain Iranian Assets, Armed Activities on the Territory of the Congo, and ELSI thereby confirms that CIL¹⁸¹ does in fact protect individuals and corporations’ property against both direct and indirect taking or expropriation. This evolution might well be an indication of the impact of investment law, through BITs and ISDS, on the development of CIL between the early 2000s and the 2020s.

WHAT HAS BEEN LEFT OUT… SOME MISSED OPPORTUNITIES

In addition to issues that appear in plain sight in Certain Iranian Assets and others that can be found between the lines of the judgment, there are issues that the Court either omitted or simply refused to address. Yet, the Court should have settled these issues as their omission compromises the stability of economic relations between states. This section will first deal with (i) whether the freezing of funds can constitute an indirect taking under CIL and (ii) whether states can challenge the prospective application of expropriatory measures adopted by other states.

Frozen assets and indirect expropriation

The first issue relates to whether the freezing or blocking—freezing being a subcategory of blocking—of assets can constitute indirect expropriation under CIL. When funds are blocked or frozen, their owners cannot use (usus), transfer (abusus), or access them or their revenues (fructus), although there is no formal transfer of ownership.¹⁸² The legality of freezing measures is a pressing legal question that the Court ought to have answered because Iran alleged the blocking constituted a violation of the 1955 Treaty and because blocking measures are increasingly being used by states.¹⁸³ Indeed, in addition to the measures affecting Iran and Iranian corporations and citizens, one could also think of other countries facing all-encompassing sanctions from the Western Powers (and vice versa), which include assets or funds freezing (eg North Korea, Russia, Sri Lanka, or Zimbabwe). Indeed, even though in a not-so-distant past, the direct transfer of titles was the ‘distinctive feature of traditional expropriation under international law’,¹⁸⁴ measures of indirect expropriation now appear to be the norm,¹⁸⁵ while measures freezing funds or assets are multiplying.

By definition, a freezing measure is temporary. Under the existing ICJ and ISDS case law, a taking must, in principle, be permanent to be characterized as expropriation.¹⁸⁶ However, the freezing of assets could well constitute expropriation under CIL.¹⁸⁷ Indeed, many investment arbitral tribunals have insisted that the main criterion to find expropriation is the ‘lasting removal of the ability of an owner to make use of its economic rights’, which can constitute an indirect expropriation even if ‘partial or temporary’.¹⁸⁸ However, the practice of arbitral tribunals is inconsistent on this question. For instance, in S.D. Meyers v Canada, the tribunal did not find a breach of provisions on expropriation on account of an 18-month interference,¹⁸⁹ while, in contrast, the tribunal in Wena Hotels v Egypt found that a 1-year interference was not ‘ephemeral’ and constituted indirect expropriation.¹⁹⁰ Likewise, the tribunal in Señor Tza Yap v Peru deemed that a period of three years could not be considered ‘indirect, remote or temporary’.¹⁹¹ Even more relevant here, the tribunal in Belokon v Kyrgyz Republic found that a sequestration of bank funds which lasted more than four years ‘with no end in sight’ constituted indirect expropriation.¹⁹² In any event, at the ISDS level, despite some tribunals insisting on the permanency of a measure as a criterion to find expropriation,¹⁹³ there seems to be consensus, or jurisprudence constante on the fact that beyond a certain period of time, a temporary measure can be considered expropriatory.¹⁹⁴ It has also been highlighted in the legal scholarship that temporary measures are often only labelled as temporary by governments to avoid international responsibility with no intention to give back control to the affected owners.¹⁹⁵

However, this consensus on the fact that a lasting albeit temporary measure may be expropriatory seems relatively recent as, for instance, the Iran–US Claims Tribunal in the Eastman Kodak Company case deemed in 1987 that the fact of freezing funds ‘did not rise to the level of an expropriation or of a deprivation of ownership of rights’.¹⁹⁶ Nonetheless, the tribunal considered that it constituted a measure covered under Article II of the Algiers Accords, that is, ‘other measures affecting property rights’¹⁹⁷ because the funds at stake had been frozen for about two years, between 1979 and 1981, until the affected company went bankrupt.¹⁹⁸

The obligation to compensate even temporary deprivations of property is also included in several international instruments, notably as part of indicative lists.¹⁹⁹ For example, in the 2013 BIT between Kenya and Kuwait, or the 1991 Protocol between France and Kuwait, the indicative list of measures having ‘de facto expropriatory effects’ includes ‘the freezing or blocking of the investment’.²⁰⁰ Express references to temporary measures within a BIT are also far from being a recent phenomenon. The 1927 Treaty of Establishment and Legal Protection concluded between Greece and Switzerland already provided that ‘nationals of each of the contracting parties (…) cannot be expropriated or deprived, even temporarily, of the enjoyment of their property, safe for public utility purposes and general interest’ (emphasis added).²⁰¹

One could also think of the arguments of Switzerland in the ICJ Interhandel case at the end of the 1950s. In this case, Switzerland argued that general international law prohibited the blocking of foreign assets in times of peace (and even in times of war as regards neutral countries).²⁰² Switzerland tried to exercise diplomatic protection for the Interhandel conglomerate whose assets had been ‘blocked’ by the USA.²⁰³ Switzerland not only invoked Article IV of the Washington Agreement that provided that the USA would ‘unblock Swiss assets in the United States’,²⁰⁴ but it also relied on CIL. According to Switzerland, under jus gentium (‘droit des gens’), foreign states have to afford certain protection to the property of the nationals of neutral countries in times of war as well as in times of peace.²⁰⁵ In this case, Switzerland argued that the blocking or freezing of assets fell within the broader category of ‘confiscation’ and ‘expropriation’.²⁰⁶ Nonetheless, the Court eventually found Switzerland’s claims inadmissible since the Interhandel conglomerate had not exhausted domestic remedies.

There is thus uncertainty whether the freezing of funds or assets constitutes indirect expropriation under CIL. Nevertheless, as seen above, the Court in Certain Iranian Assets helped answer a question that was put in the following words by Rosalyn Higgins in the course on expropriation she gave at the Hague Academy in 1982: ‘[d]o interventions by the State that leave title untouched in the hands of the plaintiff, but nonetheless occasion him loss, give rise to a right of compensation?’.²⁰⁷ Indeed, there seems to be a consensus, which includes the Court following the case Certain Iranian Assets, on the question of indirect expropriation. However, the question is now, ‘[h]ow far does [indirect expropriation] extend’?²⁰⁸ In fact, although the state of CIL might have evolved to cover the freezing of assets over an extended period of time as expropriatory, there is no clear answer as to what amount of time is necessary.²⁰⁹

In the Certain Iranian Assets case, Iran claimed that the ‘blocking’ of its corporations’ assets pursuant to the US Executive Order 13599 was a further ‘taking’, violating Article IV:2 of the 1955 Treaty. According to Iran, which referred abundantly to investment arbitral awards, the said blocking ‘radically deprived the owners of the economic use and enjoyment of their property as if the rights had ceased to exist’.²¹⁰ Iran considered that the freezing amounted to indirect or creeping expropriation given the ‘combined effect of the series of legislative, executive and judicial acts’ at stake. It also argued that the ‘blocking’ measures were ‘themselves expropriatory in nature’ and deprived the impacted owners of the economic use and enjoyment of their property.²¹¹ Iran insisted that the blocking orders were ‘temporary only in the most technical sense that the United States could repeal them if it chose to do so’.²¹²

Iran sought to rely on the Tecmed v Mexico arbitral award, a case which has also been relied upon by the USA, to identify what constitutes a taking, that is, a measure ‘radically depriv[ing] [the owner] of the economic use and enjoyment of its investment, as if the rights related there […] had ceased to exist’.²¹³ Both parties seemed to—at least partly—acknowledge the ‘sole effect doctrine’, which holds that the severity of the taking’s impact on the owner’s use and enjoyment of its property constitutes the main element to find indirect expropriation.²¹⁴ However, the USA also argued that measures had to be permanent to constitute a taking since ‘blocking orders […] are not expropriatory, including because they are by nature temporary and do not themselves alter ownership of the blocked assets’.²¹⁵

Yet, the Court expeditiously rejected the Iranian claim without examining its merits. The Court peremptorily affirmed that Iran had ‘failed to identify the property or interests in property of Iranian companies that were specifically affected by Executive Order 13599’ other than one bank, Bank Markazi (Iran’s Central Bank, whose operations are not covered by the 1955 Treaty).²¹⁶ Thence, the Court deemed that Iran had ‘not substantiated its allegations in relation to takings under Article IV, paragraph 2, of the Treaty’.²¹⁷ However, the Court regrettably omitted to consider Iran’s allegations and clarifications offered during oral proceedings. Indeed, one of the Iranian legal advisers clearly established that besides Bank Markazi, all the goods and interests of identifiable private Iranian Banks had been frozen.²¹⁸ Iran argued that although the USA affirmed that Order 13599 only affected Markazi, several private banks’ assets (Melli, Saderat, Mellat, Sepah, and EDBI banks) had been frozen after the entry into force of the Executive Order.²¹⁹

Still, if the Court disagreed with Iran’s factual assessment, it should have said so expressly rather than avoid the question. Indeed, the decision of the Court not to entertain this claim on the basis that Iran had ‘not substantiated its allegations’²²⁰ is surprising, given that it did the exact opposite regarding Article X of the Treaty, which relates to the freedom of commerce and navigation. Indeed, in its findings on Article X, the Court found that ‘[b]y its own terms, this Executive Order constitutes an actual impediment to any financial transaction or operation to be conducted by Iran or Iranian financial institutions in the territory of the United States’ and thus a treaty violation (emphasis added).²²¹ The Court also seemed to explicitly recognize that the Executive Order ‘comprehensively blocked the property and interests in the property of Iran and Iranian financial institutions’²²² when discussing its findings under Article X. Nonetheless, the Court came to the opposite conclusion when examining Article IV of the 1955 Treaty.

On the one hand, the Court should have settled the matter by deciding whether the freezing of the private bank’s assets and interests in property constituted indirect expropriation by examining the claims presented by Iran’s legal team. On the other, as will be detailed in the next subsection, the Court should have examined the Executive Order’s prospective application or its general impact as it did under Article X of the 1955 Treaty.

Although Iran might not have succeeded in establishing a violation on this basis, its claim was certainly not meritless as, indeed, one of the corollaries of property is effective possession as opposed to a mere possession on paper without the possibility to dispose of it. In the case of frozen assets or funds, the interests in property, usus, fructus, and abusus are likewise frozen.²²³ The animus, or intent to possess, of the owner is all that is left out of the control of the authority that ordered the blocking or freezing, rendering the on-paper title rather useless. As the arbitral tribunal said in El Paso Energy International Co v Argentina, which was relied upon by Iran and thus could have inspired the Court, there is an expropriation when ‘at least one of the essential components of the property rights […] have disappeared’.²²⁴ Had the Court entertained Iran’s claim, it would have had reasonable chances of success since the indefinite duration of the Executive Order strip affected Iranian corporations of nearly all the attributes of property. Indeed, it is likely that MST includes a duty to compensate for the seizing or requisition of foreign goods, because under most domestic systems, including in the USA, even temporary takings²²⁵ or requisitions are to be justly compensated.²²⁶

Property, state responsibility and the existence of an expropriatory measure ‘as such’ vs ‘as applied’

Lastly, as seen in the previous subsection, the Court refused to judge whether the freezing or blocking of assets amounted to a violation of Article IV:2 of the 1955 Treaty because it affirmed that Iran did not adduce concrete examples of impacted banks.²²⁷ However, regardless of the evidence adduced by Iran’s agents, the measure as such, or as the Court put it in the context of Article X of the 1955 Treaty, the measure ‘by its own terms’, should have been examined by the Court. The refusal of the Court to entertain this claim thus raises the question of whether states can complain about measures as such, for instance, the text of a law or of an Executive Order, or only complain about a measure as applied, that is the way a legal text has been applied.

This question is particularly relevant within the field of property protection as government measures that provide for the freezing or the expropriation of assets or goods can have a severe impact on their value regardless of whether such a measure is indeed applied. On the one hand, it will be harder to dispose of a good that may be expropriated and the value of such a good would eventually depreciate. On the other hand, if a government act (executive, legislative, or judicial) provides for a freezing or an expropriation, it constitutes a sword of Damocles that affects the capacity of individuals and businesses to plan ahead. It might push them to rush the selling of their assets, at an undervalued price, to avoid such a measure. Having to wait for their property to be effectively expropriated in the technical sense (which might never happen) before seeking redress would strip of the affected individuals or companies from the possibility to avoid suffering further damage.

Under general international law, there is no doubt that a measure ‘as such’ can constitute a violation because it is still an ‘action’ attributable to the state within the meaning of Article 2 of the International Law Commission (ILC) 2001 Articles on State Responsibility.²²⁸ For instance, in the advisory opinion on the Applicability of the Obligation to Arbitrate under Section 21 of the United Nations Headquarters Agreement of 26 June 1947, the ICJ made clear that although claims do arise ‘out of the behavior of or a decision by one of the parties, it in no way requires that any contested decision must already have been carried into effect’.²²⁹ This does not mean that under all circumstances, a domestic law conflicting prima facie with international law will constitute a breach, but it may result in a breach of international law without having already been implemented.²³⁰ Alternatively, in some instances, a domestic law may not be contrary to international law, but responsibility could rather arise from the application of said law by domestic authorities.²³¹

Fitzmaurice deemed that states can directly challenge ‘legislation contrary’ to international law without ‘await[ing] the occurrence of a concrete case’.²³² On the issue of property protection, he specifically added that ‘[i]n some cases the legislation itself undoubtedly constitutes the breach—as for instance if a law is passed expropriating foreign property without provision for adequate compensation, or if there were legislation denying to foreigners the right to appear in the Courts’.²³³ These examples particularly resonate with the measures at stake in Certain Iranian Assets case.

The refusal of the Court to engage with a measure as such or ‘by its own terms’ is reminiscent of the jurisprudential distinction in World Trade Organization (WTO) law between measures ‘as such’ and ‘as applied’. Without entering into the specifics, when it comes to legislation, there can be two types of violations in WTO law. On the one hand, a measure can, under certain circumstances, violate a legal norm on the basis of its very existence, which is a violation by the measure, for example a law, ‘as such’. On the other hand, a measure can constitute a violation on account of its application, in which it is considered a measure ‘as applied’. Indeed, the WTO Appellate Body (AB) ruled that laws or norms ‘intended to have a general and prospective application’ could also violate the WTO Covered Agreements.²³⁴ In US—Continued Zeroing, the AB indicated that the distinction between ‘as such’ and ‘as applied’ measures was developed ‘as an analytical tool to facilitate the understanding of the nature of a measure at issue’.²³⁵ As regards proving the existence of violation ‘as such’, the AB stated that ‘the starting point for an analysis must be the measure on its face. If the meaning and content of the measure are clear on its face, then the consistency of the measure as such can be assessed on that basis alone’.²³⁶ One of the criteria retained by the AB to assess if a measure, which would be illegal if applied to the letter by national authorities, is legal or illegal is whether a measure is ‘mandatory’, i.e., national authorities must apply the measure to the letter regardless of international law, or ‘discretionary’, i.e., national authorities have sufficient leeway to apply the measure in a way which is consistent with international law.²³⁷ However, the AB later clarified that, in principle, even non-mandatory measures could be challenged ‘as such’.²³⁸ The ILC itself took inspiration from this distinction between ‘mandatory’ and ‘discretionary’ measures in WTO law as the commentaries refer to the jurisprudence of the AB. Indeed, on the basis of AB jurisprudence, the ILC affirmed that the ‘enactment of legislation may not in and of itself amount to a breach, especially if it is open to the State concerned to give effect to the legislation in a way which would not violate the international obligation in question’.²³⁹

In the Certain Iranian Assets case, the Court seems to have adopted a methodology akin, albeit more severe, to that developed by the WTO AB. In this case, an Executive Order—indeed issued by a state’s executive organ—qualifies as state conduct.²⁴⁰ For instance, in the Military and Paramilitary Activities in and Against Nicaragua case, Nicaragua could challenge an Executive Order issued by the President of the USA, which imposed a trade embargo on Nicaragua (which was also an applied measure).²⁴¹ Here, Executive Order 13599, which was adopted under the Obama administration in 2012, is imperative as it prescribes that ‘[a]ll property and interests in property of any Iranian financial institution, including the Central Bank of Iran, that are in the United States, that hereafter come within the United States, or that are or hereafter come within the possession or control of any United States person, including any foreign branch, are blocked and may not be transferred, paid, exported, withdrawn, or otherwise dealt in’ (emphasis added).²⁴²

Here, the authorities having to apply this Executive Order did not have any leeway and did not seem to have been endowed with any discretion. In fact, on account of Executive Order 13599, ‘all property and interests of property’ of ‘any Iranian financial institutions’ are subjected to the blocking provided for in the Executive Order. Unless the Court was only referring to Bank Markazi (which was in any event excluded from the Court’s jurisdiction), it seems surprising that the Court found that Executive Order 13599 violated Article X:1 of the 1955 Treaty on the freedom of commerce and navigation, but not Article IV:2 on the protection of property. As a reminder, according to the Court, when discussing the claims raised under Article X, the Executive Order ‘by its own terms’ constituted ‘an impediment to any financial transaction or operation to be conducted by Iran or Iranian financial institutions in the territory of the United States’ (emphasis added).²⁴³ Therefore, even if the Court were to apply a WTO-like test, Executive Order 13599 could have been challenged by Iran because ‘on its face’ the measure clearly mandates the blocking and is mandatory. It would not be ‘open to’ US officials to apply the Executive Orders otherwise than by freezing the listed assets.

In any event, in the interest of legal certainty, the Court should have clarified what is needed for the Court to entertain a claim relating to a measure ‘as such’ and whether the Court’s position has evolved since the above-mentioned advisory opinion on the Applicability of the Obligation to Arbitrate under Section 21 of the United Nations Headquarters Agreement of 26 June 1947. Indeed, the Court’s stance on the question of measures ‘as such’ seems even more stringent than within the WTO context as in WTO law mandatory measures would a minima have been examined. This is of particular importance given that most treaties, including the 1955 Treaty, do not define the term ‘measures’.²⁴⁴

In addition, pursuant to another US Executive Order (13902) adopted after the beginning of the Court proceedings in Certain Iranian Assets, many Iranian firms operating in the construction, mining, manufacturing, and textiles sectors have been impacted by blocking measures, which shows how important it would have been for the Court to settle this question.²⁴⁵

Finally, it is hard to accept that laws that would be inconsistent with international law if they were to be applied—laws are usually adopted with a view to being applied—could not be challenged before the Court. This would have the effect of indefinitely maintaining a sword of Damocles over the heads of individuals or corporations targeted by measures pending application (which might indeed be applied at any moment). The Certain Iranian Assets case concerned financial institutions, yet excluding measures of prospective application from the scope of the Court’s analysis could have dire consequences in other fields of international law, such as international human rights law. On this issue, it is important to recall that human rights bodies such as the Human Rights Committee have steadily affirmed that laws ‘as such’ must be compatible with human rights obligations undertaken by states.²⁴⁶

CONCLUSION

In conclusion, this article sought to contribute to the understanding of the substance and scope of CIL on the protection of the private property of foreigners. It did so by recalling the genealogy of norms that form part of the corpus of CIL by analysing previous ICJ and PCIJ cases, first and foremost the recent Certain Iranian Assets case, alongside arbitral ISDS awards.

The first part of this article dealt with issues that the Court did indeed settle in Certain Iranian Assets and which brought further clarity to the customary status of certain norms. This part first offered an analysis of the customary status of the police powers doctrine, which the Court indeed recognized as a justification for the adoption of measures of general application that are undertaken in good faith, for a public purpose, and in a non-discriminatory manner. Second, this part contained an examination of the most constant protection and security standard. The Court clarified that under CIL, this standard relates to ‘physical’ protection only, for example from physical threat, and does not extend to legal protection or security, that is, a guarantee that the legislative framework will stay equal. Third, the first part of the article dealt with FET, which, according to the Court, must not be equated with the customary MST in the absence of an express reference to the MST. Indeed, absent any CIL reference, the FET standard must be considered a freestanding obligation which goes beyond CIL. Finally, this part of the article also addressed the issue of the corporate veil in PIL. The Court indeed recalled that when corporations possess an independent legal personality as provided for in domestic legislation, such a personality must be upheld regardless of a corporation’s ultimate owners. The Court nevertheless held that, in theory, the corporate veil could be pierced when warranted by the circumstances. Yet the Court did not indicate which circumstances might warrant this outcome. Thus, the article's first part discussed the findings of the Court, which coincide with the majority of the ISDS awards acquis, with regard to the police powers and the corporate veil while it settled some enduring investment law debates regarding the nature of FET and the extent of the most constant protection and security standard.

The article’s second part adressed issues for which the Court provided additional certainty albeit without fully engaging with them. The second part began with an assessment of the MST, which embodies virtually all the rules related to the protection of individuals as well as property abroad. In Certain Iranian Assets, although the 1955 Treaty contains express references to the MST, the Court decided not to elaborate on this standard. However, an analysis of this decision alongside the Court’s jurisprudence shows that it has acknowledged that the MST is the applicable CIL standard for the protection of individuals and property abroad and that this standard might now include human rights protection. In addition, as seen above, the Court minimally clarified that FET is not to be read in the MST. The second issue analysed within this part of the article was whether indirect expropriation or takings are covered by the term ‘expropriation’. Despite hesitations from the Court in the past, the Certain Iranian Assets judgment appears to confirm that the notion of expropriation in CIL now covers both direct and indirect expropriation. In turn, this signals a change in the jurisprudence of the Court, which until recently was reluctant to consider indirect expropriation as part of the broader concept of expropriation. The Court’s decision thus shows that the large body of BITs and investment awards likely had an impact in shaping CIL.

Finally, the last part of this article engaged with issues that the Court omitted or refused to settle despite their importance and despite the claims raised by Iran. This part first discussed whether the freezing or blocking of funds or assets could be considered an indirect taking despite the—sometimes hypothetical—temporary nature of such measures. This article posits that such measures can be expropriatory on account of the existing international practice. Given the growing concerns related to all-encompassing freezing or blocking measures adopted by states and the relative inconsistency of ISDS awards on this question, the Court should have elaborated on this issue and brought more clarity thereupon. Indeed, while most ISDS awards recognize that after a certain period of time, even a temporary measure could be considered expropriatory, some tribunals insist that expropriation only covers permanent measures. This article’s last section also appraised an issue that is of particular importance in relation to property protection. This issue relates to whether the adoption of a measure ‘as such’, for instance, the adoption of a law or decree, can trigger state responsibility, or if a measure needs to be ‘applied’ to trigger a state’s international responsibility. Indeed, the Court adopted a method leading to rejecting some violation claims in the absence of evidence of a measure’s application. This is reminiscent of the WTO-developed distinction between measures ‘as such’, for example an Executive Order, and ‘as applied’, for example the same Executive Order’s application; the former being harder to challenge. However, the Court’s decision in Certain Iranian Assets contrasts with its previous jurisprudence and now appears to have become more stringent than the WTO AB at the time of deciding whether it will entertain a legal claim or not. Indeed, within the WTO system, if a measure is mandatory and leaves no doubt as to its prospective application, WTO panels will entertain relevant claims. On the contrary, in Certain Iranian Assets, the Court decided not to examine two Executive Orders that were mandatory on their face. Despite the Court’s refusal to engage with measures of prospective applications in this case, the possibility for states to challenge such measures appears crucial for them to protect both their rights and the rights of their nationals proactively.

In relation to these last two issues, the article advocates for a legal framework that would truly be protective of foreigners, including their property, when abroad. The Court should clarify whether temporary measures can be expropriatory. Indeed, there appears to be no reason why, against the backdrop of existing state practice, temporary property deprivation should not be covered by the MST. In turn, the Court should engage with measures of prospective application or at least explain under what set of circumstances it will entertain claims concerning ‘as such’ measures. This is especially important within the field of property protection, given the fact that once a measure has been adopted, the mere possibility of it being applied already has an economic impact on individuals and corporations. Further, beyond the freezing of assets, it is crucial to clarify the types of measures that can be challenged before the Court—especially with regard to measures that may impact human rights.

In a nutshell, on the one hand, the CIL shopping list mentioned at the beginning of this article appears to cover the MST, including the standard of the most constant ‘physical’ protection and security; the prohibition of uncompensated indirect expropriation; the possibility for states to rely on police powers; and the respect for corporations’ separate legal personality in accordance with domestic legislation. Nevertheless, the Court made it clear that absent any express reference to CIL, a FET provision is an independent treaty standard not to be derived from the MST, which is allegedly less protective than the independent FET standard. On the other hand, the author’s wish list now includes clarifications on whether freezing or blocking assets can constitute indirect expropriation under CIL and whether measures of prospective application can trigger state responsibility.

What is more, the importance of the issues addressed in this article is likely to grow given the increasing use of sanctions and counter-sanctions around the world. All in all, this article constitutes a call for consistency within the realm of foreign property protection. Indeed, for more than a century, the CIL rules on property protection have almost been sanctified by the Western Powers and imposed upon the rest of the world, with attempts to limit the scope of property protection norms by Global South countries, for example the Calvo doctrine, remaining infructuous. Yet, some Western countries now appear to be trying to escape from the rules they themselves contributed to establishing, such as those relating to the existence of the corporate veil and the prohibition on non-compensated expropriation. It seems rather odd that the historical flag bearers of private property protection, such as the USA, bypass property protection norms when it seems convenient to do so. At all events, legal rules must be applied consistently.

One topical example pertains to those countries blocking the assets of Russian oligarchs, individuals, and corporations following the war of aggression launched on Ukraine by Russia in February 2022.²⁴⁷ The blocking of Russian assets by countries not at war with Russia will most certainly pose the question of the violation of the MST, especially when corporations possess a separate and non-Russian legal personality. So far, states have taken different approaches. By way of example, even though Canada and Switzerland froze certain Russian assets, they have adopted two relatively opposite approaches within the bloc of countries that took measures aiming to halt the hostilities between Russia and Ukraine. For instance, Canada announced its decision to sell Russian or Russian-controlled assets to use their value for the reconstruction efforts in Ukraine,²⁴⁸ while Switzerland considers that confiscating or expropriating frozen assets that were legally acquired would violate Swiss law and Switzerland’s ‘international commitments’.²⁴⁹

Therefore, given the current situation and the ever-growing body of sanctions (which at least have the merit of replacing gunboat diplomacy or politique de la canonnière), many rules of international law, especially within the field of property protection, are yet to be clarified. This factual background is also creating a need for clearer rules in other fundamental fields of international law, such as those relating to state immunities and human rights when it comes to the seizing of foreign states central bank’s assets and property (eg Markazi)²⁵⁰ or the humanitarian impact of sanctions. The Court will likely rule on those two issues soon in other disputes involving Iran and other states, such as Canada and the USA.²⁵¹

Finally, this article could have extended to other contentious issues in scholarly work pertaining to investment arbitration, such as the issues of compounded interest or the nature of compensation (viz, the mere value of the expropriated goods or full prospective profits) which also remain on the author’s CIL-identification wish list. Nevertheless, the Court in Certain Iranian Assets gave the parties a chance to settle those questions themselves, and they may not be further elaborated on by the Court, at least in the short run. If the parties to the case do not succeed in settling an agreed amount of compensation, the Court will likely have to pronounce on the nature of applicable compensation standards and might add to the above-mentioned CIL shopping list.

While this article criticized part of the Court’s reasoning and its refusal to address some key issues of the dispute, the criticism is not intended to be blunt. Rather, it embodies a call to settle legal questions and legal disputes, as far as possible, in order to foster the stability of the international system. In general, it constitutes an invitation to the Court to keep clarifying important aspects of international law whenever the opportunity arises.

ETHICS APPROVAL

No need for approval.

The author wants to thank Professor Fuad Zarbiyev, Pascal Blickle, and Rebecca Walker for their kind help and comments.

Footnotes