Csongor István Nagy*
Editor’s Note: This piece is part of a special collaboration with the Harvard International Law Students Association on international arbitration.
I. Introduction
In February 2022, Russia launched a full-scale war against Ukraine. The overwhelming majority of the international community condemned this as a gross violation of international law. Although the U.N. Security Council did not adopt sanctions, several countries introduced unilateral measures freezing Russian assets. It has been argued that countries should go beyond asset freezes and use these assets for the indemnification of Ukrainian war damages. Confiscation would, however, be unprecedented and raise serious international legal concerns. While states have, with good reason, been reluctant to react to one wrongful act with another,[1] this question has given rise to intensive debate. Recently, the EU set up a working group to inquire if and how Russian assets could be used to reconstruct Ukraine.
In this paper, I argue that instead of frontal approaches involving direct seizure, “maneuver lawfare” and international investment law provide a solution. Under international law, sovereign immunity rules out confiscation both as a countermeasure and a compensatory measure responding to acta jure imperii, such as military operations. Nonetheless, sovereign immunity does not extend to commercial matters, where judgments and awards can be enforced against state assets. International investment law, notably the Russia-Ukraine BIT (“RUBIT”), “commercializes” acta jure imperii. It converts public law violations into quasi-commercial claims “immune from sovereign immunity.” Although not the norm, mass claims are not unknown in investment arbitration. This implies that a substantial part of Ukrainian war damages can be submitted to arbitration and that their incorporation into an arbitral award offers a solid legal basis for enforcement against Russian assets. Although the proposed approach promises no one-click solution and has not yet undergone baptism by fire, it is legally capable of producing substantial practical results.
II. Confiscating Foreign State Assets Under International Law
The wholesale freezing and confiscation of foreign state assets is generally prohibited under customary international law (Art 9 of the Declaration on the Human Rights of Individuals Who Are Not Nationals of the Countries in Which They Live). There are, however, exceptions, most notably countermeasures. These are illegal acts whose wrongfulness is precluded by the fact that they react to a pre-existing wrongful act (see e.g. Case Concerning the Gabčíkovo-Nagymaros Project (Hungary v. Slovakia), ¶¶ 82-87; Air Services Agreement Case (France v. United States), ¶ 83), provided they are reversible and aim to compel compliance with international law (Article 22, 49(1) and 49(3) of the 2001 Draft UN Treaty on the Responsibility of States for Internationally Wrongful Acts (“TRSIWA”). Although generally countermeasures can be imposed only by injured countries, this limitation does not apply to the violation of erga omnes obligations. “Any State (. . .) is entitled to invoke the responsibility of another State (…) if (…) the obligation breached is owed to the international community as a whole” (Article 48(1) of TRSIWA). Russian aggression in Ukraine clearly violates the erga omnes prohibition on use-of-force in international relations (Article 2(4) of the U.N. Charter) and, hence, authorizes countermeasures. However, countermeasures are expected to preserve assets so they can be returned once their legal basis ceases (the perpetrator terminates the wrongful act and provides reparations (Article 48(3) of TRSIWA) or the countermeasure is revoked. This requirement implies that countermeasures may not go beyond asset-freezing and that the assets of the foreign state cannot be confiscated via countermeasures (Bederman, p. 824; Elegab, p. 196; Zoller, p. 15).
The confiscation of frozen Russian assets may also be described as a garnishment assisting the enforcement of Ukrainian international law-based claims to compensation. Undoubtedly, Russia is “under an obligation to make full reparation for the injury caused by the internationally wrongful act” (Article 31 of TRSIWA). Simultaneously, notwithstanding this substantive obligation, Russia has sovereign immunity in these matters. “A State enjoys immunity, in respect of itself and its property, from the jurisdiction of the courts of another State” (Article 5 of 2004 UN Convention on Jurisdictional Immunities of States and Their Property (“CJISP”)). Although exceptions apply, the war is an actum jure imperii and all war-related claims may come under the general rule of sovereign immunity (Jurisdictional Immunities of the State (Germany v. Italy), ¶¶ 60, 77, 92-97, 134-35, 140-42).
While the confiscation of frozen Russian assets, both as countermeasure and garnishment, raises serious international law concerns, commercial matters (Article 10 of CJISP), as well as state consent (Article 7(1) of CJISP), are exceptions to sovereign immunity. This applies a fortiori to arbitration, which is, by definition, less intrusive in state sovereignty than proceedings by the courts of another sovereign. Section II demonstrates how investor-state arbitration embedded in BITs enables the use of this exception to claim compensation for war damages.
III. Investment Awards Create a Legal Title for Enforcement Against Frozen Russian Assets
International investment law, and specifically the RUBIT, converts claims emerging from acta jure imperii into private law, providing a basis of arbitral proceedings that rest on the consent of Russia. This makes the resulting arbitral award immune from sovereign immunity.
Investment arbitration is the only mechanism that authorizes compensatory claims by individuals for breaches of public international law.[2] While claims emerging from acta jure imperii are suppressed by sovereign immunity, investment arbitral awards are not. BITs have a dual nature and, by blending public international law with private law, they vest claims emerging from public law violations with a commercial law character. They convert public law disputes into private law controversies with a quasi-commercial character, where states lack immunity.
This commercial character manifests itself in both scholarship and practice (Moses, pp. 243-45; Sweet & Grisel, pp. 72-73). Some even argue that investment arbitration is “international commercial arbitration.” (Gaillard & Savage, pp. 42-43). This thinking is also reflected in a wealth of case law that applies the New York Convention’s rules on recognition and enforcement to investment awards (Van Harten, p. 378). U.S. courts take a similar approach (Argentina v. BG Group, pp. 117-19; Chevron v. Ecuador, pp. 207-08; Gold Reserve v. Venezuela; Crystallex v. Venezuela).
International law limitations apply to “non-commercial” assets, such as military or central bank holdings, which are immune from enforcement even if the award or judgement was rendered in a commercial matter (Article 21 of CJISP). Nonetheless, sovereign direct investments, airplanes, ships and the assets of persons attributable to the state can, however, be used to satisfy enforcement creditors.
IV. Do War Damages Fall Under RUBIT’s Scope?
The case law on BITs’ application to illegally occupied territories is scant. It is limited to a few cases concerning Crimea, where arbitral tribunals held that effective control is sufficient for the RUBIT to apply ratione loci (Ukrnafta; Stabil; Everest; Belbek; Privatbank and Finilon; Naftogaz; Oschadbank; Lugzor). The issue has attracted more attention and produced some scholarship on BITs’ application in times of war (Zrilic; Schatz; Fach Gómez; Ryk-Lakhman; Ackermann & Wuschka; Ackermann; Schreuer) and civil disorder (Greenman), but this has predominantly focused on interpretation of substantive provisions, such as rules on expropriation and treatment. Interpretive questions concerning BITs’ scope and their application to illegally occupied territories have remained largely unsettled.[3]
BITs apply to investments by citizens of one Contracting Party situated on the territory of another. They provide for protection against direct and indirect expropriation without compensation and set out various treatment standards for foreign investors. Translating this to the current question, the RUBIT applies to investments of Ukrainian nationals and legal entities within Russian territory. Article 12 of the RUBIT provides for its application to “investments carried out (…) as of January 1, 1992.” Accordingly, the RUBIT may apply to war damages caused by Russia in Ukraine, if the damages are done to an investment by a Ukrainian national made after January 1, 1992, within territory that could be considered Russian under the RUBIT.
Although the application of a BIT to seek compensation for war damages is uncharted territory, the real question of first impression is clearly the construction of territorial scope under the RUBIT and whether the territories of Ukraine illegally occupied or claimed by Russia can be regarded as Russian for purposes of the RUBIT. The other elements of scope have relatively settled meaning in international arbitral and state practice (Brown) and interpretation of these elements in the context of war damages raises no unprecedented issues.
Article 1(4) of the RUBIT defines the treaty’s territorial scope as “the territory of the Russian Federation or the territory of the Ukraine and also their respective exclusive economic zone and the continental shelf as defined in conformity with the international law.” Areas that legally belong to and are effectively controlled by a state are unquestionably that state’s territory. Nonetheless, what if the area legally belongs to but is not effectively controlled by a state or is effectively controlled by but does not legally belong to the state? Does it matter if a state frivolously claims the area it controls? What if a state makes continuous efforts to occupy an area it frivolously claims but momentarily does not control?
The possible scenarios can be presented in a three-dimensional coordinate system, whose axes are “legal title,” “effective control” (occupation) and “unilateral claim.” An additional question is if an area can, under the RUBIT, simultaneously belong to both Russia and Ukraine. Can Crimea qualify as Ukrainian territory via legal title and as the territory of Russia via effective control?
Based on the above factors, the following matrix emerges:
- Ukrainian territory controlled by Ukraine and neither claimed, nor controlled, by Russia;
- Ukrainian territory controlled by Ukraine and claimed by Russia (unoccupied parts of the Luhansk, Donetsk, Kherson, and Zaporizhzhia oblasts);
- Line of contact on Ukrainian territory claimed by Russia (the line of contact in the above four oblasts);
- Line of contact on Ukrainian territory not claimed by Russia (the line of contact during the offensive towards Kyiv);
- Ukrainian territory controlled but not claimed by Russia (e.g. the areas captured during the offensive towards Kyiv);
- Ukrainian territory controlled and claimed by Russia (e.g. Crimea, occupied parts of the aforementioned oblasts).
The starting point of the interpretation of the term “territory” is that this is strictly a question of treaty interpretation and not one of state territory and international recognition. This is how the arbitral practice approaches the applicability of the RUBIT.[4] Put differently, it is a contractual dispute, not a title dispute. The relevant question is how to interpret the term “territory” in the application of the RUBIT and not whether a given region legally forms part of a country or not (although the two issues certainly overlap, and one may argue that they should coincide). This implies that the RUBIT’s provisions on territorial scope must be construed in accordance with the 1969 Vienna Convention on the Law of Treaties (“VCLT”), customary international law, and arbitral and state practice.
Arbitral and state practice confirms that effective control constitutes territory in the application of BITs. Russia has been the defendant of several arbitral proceedings under the RUBIT. In these cases, listed above, the tribunals consistently concluded that, due to Russia’s effective control, its territory includes Crimea, notwithstanding its occupation’s illegality. These tribunals held that the term “territory” must be interpreted broadly and that de facto territory is the term’s “ordinary” meaning, detaching it from questions of legal title.
The most difficult territorial question is the status of illegally claimed but uncontrolled areas. On September 30, 2022, Russia annexed four Ukrainian provinces (Luhansk, Donetsk, Kherson, and Zaporizhzhia), without controlling substantial parts of each.
Arbitral practice concerning Crimea suggests that the territorial scope of BITs is not a question of control but of international obligations. Effective control is, however, not the only case where a state assumes international obligations over a territory. Although arbitral tribunals have applied the RUBIT based on Russia’s effective control over the area, their reasoning suggests that Russia’s unilateral and unlawful annexation played an important role in this assessment and that its unilateral illegal claim was a crucial consideration in addition to occupation. In Stabil v. Russia, the tribunal noted (¶ 190) that “Russia [had] assumed obligations over Ukrainian investment in Crimea.” It found (¶ 175) that the RUBIT “became opposable to Russia (. . .) upon Russia’s incorporation of Crimea in its territory no later than 21 March 2014 when Russia ratified the Incorporation Treaty and passed the Crimean Integration Law which formally incorporated Crimea as a subject of the Russian Federation.” In Belbek, the tribunal held (¶ 175) that “the term ‘territory’ (…) has a wider meaning capable of encompassing territory for which a State has assumed the responsibility for international relations.” These statements suggest that a state’s unilateral position regarding its own territory may have an important role in assessing this question.
The question of “territory” may also be grasped using the interlinked concepts of good faith, acquiescence and estoppel. The 1969 VCLT does not rule out the use of the principles of interpretation based on customary international law (Linderfalk; Fitzmaurice & Merkouris; Lo; Dörr & Schmalenbach).
By pronouncing the incorporation of the occupied oblasts, irrespective of whether it has gained effective control over their entire territory or not, Russia assumed legal responsibilities related to these territories. This implies that it claimed the right to represent these areas internationally, made their inhabitants Russian citizens, and made Russian law applicable. The question is whether Russia also assumed duties under the RUBIT in relation to these areas and whether Russia can contradict itself and argue that the areas which it declared to be Russian are, in fact, not its territory? Some argue, in the context of Crimea, that, due to the principle of estoppel, the unilateral annexation prevents Russia from raising the objection that the annexed area is not part of its territory.
Good faith is an elusive legal principle. It serves as the root of both estoppel (Crawford, p. 420) and acquiescence (Reinhold, p. 53-56). In international investment law, estoppel can be understood as responding to inconsistent behaviour (Grenada Private Power, ¶ 208; Karkey Karadeniz v. Pakistan, ¶ 628; Border Timbers v. Zimbabwe, ¶ 411) and can have a narrower and a broader meaning (MacGibbon; Tran Thang Long; Kulick). Under the narrow definition, the principle of estoppel becomes relevant if the other party relied on a representation to its detriment (Amco v. Indonesia, ¶¶ 42-49). According to the broad definition, it suffices if the state acts contrary to its own facts. Although the former is the prevailing position (Besserglik, ¶¶ 423-424; Bolivia v. Chile (Access to the Pacific Ocean), ¶ 158; Bay of Bengal, ¶ 124; Province of East Kalimantan v. Kaltim Prima Coal, ¶¶ 211-12; Duke Energy, ¶ 231; Gruslin v. Malaysia (II), ¶ 20.2; Pope & Talbot v. Canada, ¶ 111), the broader approach also appears in arbitral practice (Rumeli v. Kazakhstan, ¶ 335; Dissenting Opinion of Mr. Bernardo M. Cremades in Fraport v. Philippines (I), ¶ 28), turning into another recognized principle of international law: tacit acceptance. According to the 2006 U.N. Guiding Principles Applicable to Unilateral Declarations of States Capable of Creating Legal Obligations, “[d]eclarations publicly made and manifesting the will to be bound may have the effect of creating legal obligations. When the conditions for this are met, the binding character of such declarations is based on good faith.” It would be reasonable to view the unilateral annexation of Ukrainian territory as implying the assumption of international legal obligations related to sovereignty over these areas, including the protection of BITs. Consequently, whether this implied assumption estops Russia from reneging on international obligations it assumed or if this amounts to acquiescence to these international obligations is a question of semantics.
Arbitral tribunals have also noted that it would violate good faith if Russia could evade the RUBIT notwithstanding its annexation. In Stabil, the tribunal noted, while discussing the principle of good faith, that “Russia cannot at the same time claim that Crimea forms part of its territory and deny the application of a Treaty that it has concluded to protect investments made on its territory, without incurring an inconsistency contrary to good faith and (. . .) consistency” (¶ 170). “Russia (…) has clearly manifested its will to consider Crimea as part of its territory, whilst taking no action to terminate or suspend the Treaty” (¶ 172). “[A] good faith interpretation of the Treaty mandates that Russia’s declaration that Crimea is part of its territory cannot remain without legal consequence to Russia’s Treaty obligations vis-à-vis Ukrainian investors in Crimea” (¶ 174).
The tribunal reached a similar conclusion in Belbek. It found (¶ 265) that “a conclusion that the Treaty no longer applies to conduct occurring in the Crimean Peninsula would . . . relieve the Contracting Parties of their obligation to perform the Treaty in good faith, contrary to the cardinal principle of pacta sunt servanda. It would be to create a legal void . . . that was never contemplated and should not be countenanced.”
V. Mass Claims and Investment Arbitration
Investment arbitration is thought of as being individualistic. Indeed, most investment disputes emerge from individual investor claims. Nonetheless, mass claims are absolutely not unprecedented. They were first recognized in three Argentine bondholder cases (Abaclat v Argentina; Ambiente Ufficio v. Argentina; Alemanni v. Argentina) and subsequently confirmed in Adamakopoulos v. Cyprus. In Abaclat, 180,000 Italian bondholders sued Argentina before the ICSID (the number was later reduced to 60,000). In Ambiente Ufficio, proceedings were launched by 119 claimants. In Alemanni, the claimants were 183 Italian individuals and legal entities. In Adamakopoulos, 956 Greek and Luxembourgish claimants, holders of certain financial instruments and bank deposits, sued Cyprus for financial restructuring measures adopted after the 2009 financial crisis.
These cases confirmed that the collective nature of mass claims is not a hurdle to arbitral proceedings. Although the ICSID Convention and Arbitration Rules are silent on this issue, there is nothing preventing arbitral tribunals from accommodating the special needs of mass actions with applicable rules of procedure, especially because Article 44 of the ICSID Convention authorizes tribunals to decide unsettled procedural questions (Ambiente, ¶ 146). Arbitral practice highlights two important propositions. First, if a tribunal has jurisdiction over individual claims, it may also have jurisdiction over their aggregation. There is no requirement of specific consent: states are not required to specifically consent to jurisdiction over mass claims, as long as they consented to jurisdiction over individual claims (Abaclat, ¶ 490; Ambiente, ¶ 146; Alemanni, ¶¶ 284, 286-87; Adamakopoulos, ¶¶ 269-270). Second, there needs to be a substantial link between the aggregated claims (Abaclat, ¶¶ 540-41; Alemanni, ¶¶ 287-88, 292; Adamakopoulos, ¶¶ 210-21). Third, although mass claims may raise issues of manageability, and unmanageable aggregations may be inadmissible, tribunals usually find mass claims manageable (Adamakopoulos, ¶¶ 224, 259).
For the purposes of the present analysis, it is sufficient to conclude that mass claims are admissible and feasible and that investment tribunals have a practice of entertaining them. Ukrainian war victims can meet the requirements of interconnectedness and manageability by creating “pockets” of claims and submitting them in separate collective proceedings. For instance, residents expelled from a specific occupied area can claim compensation for lost property. Similarly, the residents of a town damaged by shelling could jointly claim compensation for the indirect expropriation of their homes. They may establish an entity (a company or association) and assign their claims to this entity or establish a joinder of parties and sue jointly.
The above matters were entertained by ICSID tribunals, while the RUBIT, in investor-state arbitration matters, stipulates the jurisdiction of the Arbitration Institute of the Chamber of Commerce in Stockholm or ad-hoc arbitration under UNCITRAL rules. This is an irrelevant distinction. Neither the ICSID Convention nor ICSID Arbitration Rules, address mass claims and arbitral tribunals have dealt with them under this “regulatory silence.” There is no reason to doubt that the same approach could be adapted to the rules of the Stockholm Arbitration Institute or UNCITRAL.
Conclusion
International investment law is not susceptible to providing a quick, comprehensive vehicle to use frozen Russian assets to compensate Ukrainian victims. Instead, it can be used mainly to claim compensation for damages to tangible and intangible property, excluding personal injury, pain and suffering, and death. Nonetheless, investment arbitration could be a powerful tool that opens a path to remuneration. BITs are not a vehicle of compensating war damages but a mechanism for protecting foreign investments. Simultaneously, the existence of war crimes does not rule out or even limit the application of BITs, which partially overlap with compensation for war damages. This overlap can be used to achieve compensation.
The magic of international investment law can turn immune sovereign acts into non-immune commercial matters. Once the magic of an arbitral award occurs, the claim becomes enforceable against frozen Russian assets, which are, under international law, otherwise untouchable. To accomplish this, victims need to walk an arduous path to turn their claims into an arbitral award. The proposed enforcement strategy is a jig-saw puzzle made up of myriads of pockets of congenial claims. However, a meticulously constructed jig-saw puzzle still provides a fuller picture.
*Csongor István Nagy is a professor of law at the University of Szeged and a research professor at the Center for Social Sciences of the Hungarian Research Network. He is a recurrent visiting professor at the Central European University (Budapest/New York/Vienna) and the Sapientia University of Transylvania (Romania), and an associate member at the Center for Private International Law at the University of Aberdeen, Scotland. Currently, he is a CICL visiting fellow at the University of Michigan.
[1] An exception is Canada, which opened the way to the confiscation (forfeiture) of frozen Russian assets and the use of the proceeds for the reconstruction of Ukraine, the restoration of international peace and security and the compensation of victims. Sections 5.4 and 5.6 of 1992 Special Economic Measures Act. The provision on confiscation was introduced by Section 439 of the 2022 Budget Implementation Act. [2] An exception is the European Convention on Human Rights, which, however, does not enable large-scale claims. Furthermore, Russia withdrew from the Convention as from 16 September 2022. The European Court of Human Rights remains, however, competent to entertain cases concerning actions (or omissions) that occurred up until this date. [3] For an exception, see e.g. Mayorga. [4] See also Happ & Wuschka. Ackermann, p. 80. For a criticism of the approach of the arbitral tribunals, see Dumberry, Krumbiegel.