Zaza Okuashvili v. Georgia. Case V 2019/058

The award in Zaza Okuashvili v. Georgia is one of several recent decisions in which arbitral tribunals have addressed jurisdictional objections against claims by dual nationals. The Stockholm Chamber of Commerce (SCC) tribunal decided that an investor who holds the nationality of both Contracting Parties to the Georgia-UK bilateral investment treaty (BIT) qualifies for treaty protection. The tribunal also held that the Georgia-UK BIT, which specifically provides for arbitration only before the International Centre for the Settlement of Investment Disputes (ICSID), enables the investor to access SCC arbitration by virtue of the treaty's Most Favored Nation (MFN) clause. In so holding, the tribunal effectively allowed the investor to circumvent the dual nationality restriction found in Article 25(2)(a) of the ICSID Convention.¹ The award raises novel and complex questions of treaty interpretation in an emerging field of international investment law where tribunals have been asked to determine the standing of dual nationals outside the ICSID regime. The award is significant in that it: (1) includes an intriguing pronouncement on the (non-)application of the customary rule of “dominant and effective” nationality; (2) departs from previous case law on the relationship between nationality requirements in investment treaties and Article 25(2)(a) of the ICSID Convention; and (3) deals with an abuse of rights objection that adds a new perspective on the practice of nationality planning.

Mr. Zaza Okuashvili (or the Claimant) instituted SCC arbitration proceedings in 2019 against Georgia alleging violations of the Georgia-UK BIT. The Claimant argues that he is the ultimate beneficial owner of a group of companies, the “Omega Group,” which has interests in the Georgian tobacco, distribution, printing, and television industries. He further claims that, in 2004, Georgia sent armed men to invade and occupy the Omega Group's premises and that the occupation did not end until he agreed to transfer the license of his TV network to individuals associated with the former Georgian government. Following these events, the Claimant moved to the UK where he naturalized as a British national in 2011. He also holds Georgian nationality by birth and is thus a dual national. His investment in the Omega Group was made when he only held Georgian nationality. He alleges that he repeatedly sought investigation of the 2004 events, but that these efforts have been stymied. For the Claimant, this failure to investigate constitutes a breach of the BIT. Since the BIT offers ICSID as the sole arbitration forum and this mechanism excludes dual nationals, the Claimant relied on the BIT's MFN clause to open the avenue of SCC arbitration through the Georgia-Luxembourg Economic Union (BLEU) BIT. The SCC Arbitration Rules pose no express jurisdictional bar to dual nationals.

Georgia raised four main objections to the jurisdiction of the tribunal: (1) that the BIT's nationality criterion excludes dual nationals from the scope of the treaty; (2) that the BIT's MFN clause could not displace the BIT parties’ exclusive provision for ICSID arbitration; (3) that the Claimant's British (home state) nationality is not “dominant and effective” as required by customary international law; and (4) that the investment made by the Claimant as a national of Georgia is not protected under the BIT and, at any rate, the Claimant's acquisition of British nationality constitutes an abuse of rights.

The first objection targeted the jurisdiction ratione personae of the tribunal. Central to this objection was Article 1(c) of the BIT, which defines a protected investor or national for the purposes of the treaty. This provision, on its terms, does not regulate the standing of dual nationals. It simply provides that protected “nationals” means, for Georgia, “Georgians within the meaning of the law of the Republic of Georgia”; and, for the UK, “physical persons deriving their status as United Kingdom nationals from the law in force in the United Kingdom” (para. 101). Georgia argued that Georgian law should apply to ascertain whether the Claimant, as a dual national, complies with the definition of investor in Article 1(c). The state asserted that since Georgian law prohibited dual nationality at the time the BIT was concluded, Article 1(c) “has the effect of excluding Georgian-British nationals” from the scope of the treaty (id.). Georgia further argued that Article 1(c) should be read in conjunction with Article 8 of the BIT, which states that Georgia and the UK consent to arbitration under the ICSID Convention should a dispute arise “between that Contracting Party and a national . . . of the other Contracting Party” (para. 9). The term “national” under Article 25(2)(a) of the ICSID Convention expressly and definitively excludes dual nationals. In Georgia's view, the exclusive choice of ICSID arbitration “confirms that the Contracting Parties intended to exclude dual nationals from the scope of the Treaty” (para. 102).

The tribunal rejected Georgia's objection. It first reasoned that Article 1(c) addresses not the circumstances under which “Georgia or the UK may grant or withdraw nationality,” but rather “matters of opposability”—i.e., which individual investors are entitled to invoke the BIT (para. 110). Thus, for the tribunal, “a Georgian-British dual national may be regarded as being a “national” of both of the Contracting Parties,” irrespective of whether Georgian law prohibits dual nationality (para. 108). The tribunal was, in this regard, unwilling to “read into Article 1(c) a tacit, but far-reaching, limitation whereby the nationality of one Contracting Party would in effect be subject to the other Contracting Party's nationality law” (para. 112). The tribunal then held that Georgia's reading of Article 1(c) as a provision that aligns with ICSID's dual nationality restriction “may be viable only if [Georgia] succeeds in establishing that ICSID arbitration is an exclusive forum such that a protected “national” may not resort to other fora by invoking” the MFN clause of the BIT (para. 115).

The second objection was whether the MFN clause in Article 3 of the Georgia-UK BIT entitled the Claimant to access SCC arbitration, despite the provision for ICSID arbitration in the BIT itself. The Claimant submitted that “treatment” under Article 3 covers investor-state arbitration provisions.² The Claimant then argued that the dispute resolution clause in the Georgia-BLEU BIT affords treatment that is “more favourable than that which is afforded under Article 8 of the” Georgia-UK BIT, as the former provides for different arbitration mechanisms and allows him to “launch an arbitration under the SCC Rules while he is unable to do so under the ICSID Convention” (para. 164).³ In response, Georgia argued that the MFN clause “does not detract from the Contracting Parties’ exclusive choice of ICSID arbitration/conciliation by adding other fora” (para. 165). The clause cannot, in other words, “be relied upon to create consent where none exists” (para. 188) with the aim to “cure the Tribunal's lack of jurisdiction” (para. 165). On Georgia's case, therefore, “the Claimant may avail himself of better treatment within the confines of ICSID, namely as to pre-arbitration requirements such as fork-in-the-road provisions, the form and content in which a dispute must be notified to the respondent State, and the cooling-off period thereafter” (para. 189).

A tribunal majority sided with the Claimant. It applied the “primary rule of treaty interpretation” under Article 31(1) of the Vienna Convention on the Law of Treaties (VCLT) and observed that this rule “comprises the principle of effectiveness” (para. 176). This principle requires a treaty to be interpreted so as to give it the fullest weight and effect consistent with its object and purpose. The tribunal considered that the principle of effectiveness “compels” it to extend MFN treatment “in the broadest possible manner” (para. 177). As a result, the MFN clause should cover procedural issues, a reading that, in the tribunal's view, was supported by “the blanket reference” in the clause to Articles 1–11 of the treaty. By upholding jurisdiction to proceed with SCC arbitration, the tribunal effectively enabled the Claimant to bypass the dual nationality limitation found in Article 25(2)(a) of the ICSID Convention. The dissenting arbitrator disagreed with the majority, opining that “consent to an arbitral system as well as to arbitral rules cannot be displaced by an MFN provision,” unless the BIT contains an “explicit agreement” by the Contracting Parties to that effect (Concurring and Dissenting Opinion, para. 62). In his view, the MFN clause did not contain such an agreement.

As to the third objection, Georgia argued that, even if Mr. Okuashvili “come[s] within the scope of the Treaty,” the tribunal lacked jurisdiction based on the customary international law rule of “dominant and effective” nationality (para. 137). This rule, which has originally applied in the context of diplomatic protection claims, provides that a state may not espouse a claim on behalf of a dual national if the person has stronger connections (personal, economic, political, etc.) to the respondent than to the claimant state. Georgia asserted that Mr. Okuashvili's ties with Georgia were stronger and thus he was not entitled to “avail himself of his British nationality” to access the treaty (para. 147). In response, the Claimant contented that “the Treaty is lex specialis and it does not import (or require for its application) a ‘dominant and effective’ nationality test from customary international law” (para. 138).

The tribunal's take on this objection is somewhat intriguing. It began the analysis by noting that “[i]t is an extremely delicate question whether, and, if so, to what extent BITs are to be seen as incorporating rules of diplomatic-protection law regarding nationality of claims, such as dominant nationality, continuing nationality, etc.” (para. 151). The tribunal further noted that previous tribunals have answered this question in an inconsistent manner.⁴ As such, the tribunal continued, “answering it in the context of one case may have far-reaching implications for other cases and other BITs” (para. 152). The tribunal then considered that, ostensibly “[f]or reasons of judicial economy,” Georgia's objection should be resolved “on the facts of the case—that is to say, by assuming without deciding that the Claimant must clear the “dominant and effective nationality” rule (id.). By relying on the commentaries of the International Law Commission's (ILC) Draft Articles on Diplomatic Protection,⁵ the tribunal found that the “factual evidence” to be considered in determining “which nationality is predominant compared to the other” includes “habitual residence, . . . financial interests, . . . family ties in each country [and] participation in social and public life” (para. 153). The tribunal ultimately held that, based on the connections the Claimant had with the UK as opposed to Georgia, his British nationality was dominant and effective and rejected the objection.

The fourth and final objection related to the way in which the Claimant had used his two nationalities. Georgia asserted that, because Mr. Okuashvili “was exclusively a Georgian national” at the time his investment was made, he “must properly be categorised as a domestic investor who is not entitled to invoke the protections afforded by the Treaty” (para. 269). To substantiate this argument, Georgia relied, inter alia, on the BIT's preamble, which encourages the Contracting Parties “to create favourable conditions for greater investment by nationals and companies of one State in the territory of the other State” (para. 271). Relatedly, Georgia contended that the Claimant's acquisition of British nationality was an “attempt to internationalise his dispute with the Respondent” and as such his claim “constitute[d] an abuse of right” (para. 269). According to Georgia, the claimant acquired British nationality at a time when the dispute with Georgia was foreseeable, “this being conduct broadly similar to “chang[ing] nationality in order to secure jurisdiction under a BIT” (para. 280).

The tribunal acknowledged the “unusual feature of the case,” that is, that the Claimant's investment was made when he only held Georgian nationality and was therefore “purely domestic” (para. 283). Yet the tribunal disagreed with Georgia that the BIT does not protect “investments initially made by host-State nationals who later come to acquire the nationality of the other Contracting Party” (para. 274). According to the tribunal, such a limitation could not be found in “the plain, unqualified formulation of the express provisions of the Treaty,” which do not refer to investments “made” by nationals of the home state party (para. 275). It added that the BIT's definition of “investment” supports this conclusion since it “expressly provides that “the term ‘investment’ includes all investments, whether made before or after the date of entry into force of this Agreement” (para. 276). As to Georgia's related abuse of rights argument, the tribunal remarked that the principle of abuse of rights would, if applicable, “compel it to decline or to exercise jurisdiction,” since the principle is “a generally accepted principle of general international law” (para. 272). However, the tribunal disagreed that the Claimant had resorted to abusive nationality practices “of the type castigated in cases such as Phoenix Action v. Czech Republic and Philip Morris v. Australia” (para. 281).⁶ In this respect, the tribunal found that “the alleged Treaty breaches had not arisen, let alone crystallized, before he became a UK national,” nor was there evidence indicating that the Claimant could have foreseen the events leading to the dispute (para. 281).

* * * *

The tribunal's position on Georgia's objections failed to pay adequate attention to the wording of the BIT as well as its object and purpose, with the effect of significantly expanding the scope of application of the treaty. The tribunal also turned a blind eye to crucial factual aspects of the case and declined to address an interpretive question of considerable practical importance. We shall first note that all previous tribunals facing dual nationality objections have examined, upon a request from the respondent state, whether broad definitions of investor are subject to the customary rule of “dominant and effective” nationality.⁷ Georgia asked the tribunal to answer the very same “extremely delicate question” but the tribunal refused to answer it, thereby disregarding the parties’ “lengthy and erudite pleadings” on the matter (paras. 147, 151). On the pretext of “judicial economy,” the tribunal preferred instead to deal with Georgia's objection solely on the facts, “by assuming without deciding” that the Claimant must clear the “dominant and effective” nationality test (para. 152). After assessing the connections between the Claimant and his two states of nationality, the tribunal found that the Claimant's British nationality satisfied that test and thus dismissed Georgia's objection (paras. 153–62).

It could be argued that, by refusing to answer Georgia's legal question, the tribunal failed to comply with its mandate, a move that may constitute a ground for annulment at the seat of the arbitration. What is clear, nevertheless, is that the tribunal's refusal adds a new layer of uncertainty in the field that will be reviewed by tribunals currently deciding claims by dual nationals. One may also wonder if the tribunal would have taken a different approach had the facts shown that the Claimant's Georgian nationality was the predominant one. It is likely that, in this scenario, the tribunal would have ultimately examined, under the interpretive rules set out in the VCLT, whether the BIT should be read as incorporating the rule of “dominant and effective” nationality. The findings on the other objections strongly suggest that the tribunal would, at any rate, have answered this question in the negative. Georgia's argument that Article 1(c) of the BIT makes a renvoi to Georgian domestic law prohibiting dual nationality is of particular relevance in this context. The tribunal rejected this argument, finding that “if the Contracting Parties intended to limit the circle of persons entitled to rely upon the Treaty, those limitations would have been placed in Article 1(c)” (para. 110). The tribunal was, therefore, unwilling to “read into Article 1(c) a tacit, but far-reaching, limitation” that would exclude dual nationals from the scope of the BIT by virtue of Georgian law (para. 112). On its express terms, Article 1(c) also does not incorporate the customary rule of “dominant and effective” nationality. In line with this reasoning, we can safely conclude that the tribunal, at least implicitly, sympathized with the view that general international law on dual nationality does not trump the BIT's explicit language.

This is not, however, the only part of the award that deserves scrutiny. The tribunal also unconvincingly resorted to the MFN clause as a premise for rejecting (or more accurately disregarding) Georgia's argument on Article 8 of the BIT. Georgia argued that Article 8 and, by its explicit and exclusive reference to the ICSID Convention, Article 25(2)(a) of the Convention, are part of the context in which the treaty's definition of protected “nationals” must be interpreted. Recall that the term “national” under Article 25(2)(a) of the ICSID Convention expressly and definitively excludes dual nationals. Hence, by incorporating Article 8 in the BIT, so the argument went, Georgia and the UK necessarily excluded the Claimant from the scope of application of the treaty. A similar reading has been endorsed in several cases involving claims by dual nationals, which the tribunal omitted from its analysis.⁸ The tribunal instead found that “[o]n its own terms . . . the Respondent's argument may be viable only if” the Claimant is not entitled to access SCC arbitration through the MFN clause (para. 115). In taking this approach, the tribunal made its personal jurisdiction conditional on the successful application of the MFN clause. This approach seems, however, at cross-purposes with the tribunal's own statement that “[f]or a person or company to avail themselves of [MFN] treatment, they must first come within the scope of the Treaty, as set out notably in” Article 1(c) (para. 205, emphasis added).

Indeed, the determination of whether the Claimant can benefit from the MFN clause depends on the Claimant qualifying as a protected investor under the BIT, but not vice versa. Accordingly, the tribunal should have examined, in isolation from the MFN clause, whether Georgia's argument on the alignment between Articles 1(c) and 8 of the BIT was “viable on its own terms.” The wording of Articles 1(c) and 8, “read harmoniously in the context of the overall treaty text, so they can develop effet utile” (para. 192), permits a conclusion that dual nationals fall outside the BIT's definition of investor. A reading to the contrary would mean that the Contracting Parties have ascribed two different meanings to the term “national” in the BIT. The term “national” would be read to include dual nationals in Article 1(c) and to exclude dual nationals in Article 8 based on the exclusive reference to the ICSID Convention in the latter provision. By finding personal jurisdiction through the MFN clause, the tribunal yielded, albeit indirectly, this conflicting interpretation of the same term in the BIT. In so doing, the tribunal permitted the Claimant to circumvent the jurisdictional hurdle of Article 25(2)(a) of the ICSID Convention, with the effect of rendering the text of Article 8 meaningless.

Another important objection related to the fact that the Claimant's investments in Georgia were made well before he acquired UK nationality on February 22, 2011. As such, the investments were, as the tribunal put it, “purely domestic, Georgian affairs” (para. 283), which, according to Georgia, excluded the Claimant from treaty protection. The tribunal held that “[t]he Treaty's relevant provisions do not indicate that it is confined to protecting investments made from the outset by nationals or companies of the other Contracting Party” (para. 275). The tribunal relied, more particularly, on the definition of “investment” in Article 1(a) of the BIT, which provides that “the term ‘investment’ includes all investments, whether made before or after the date of entry into force of this Agreement” (para. 276). This reading of Article 1(a) is unconvincing. The wording “all investments” in this provision merely serves as a clarification that the BIT covers “every kind of asset,” including “[a] change in the form in which assets are invested.”⁹ Differently put, “the firm, unqualified wording” of Article 1(a) does not indicate that the definition of “investment” extends to assets invested by nationals of the host state (para. 177).

In fact, the tribunal's finding fails to reconcile with other provisions of the BIT relied by Georgia. The Preamble, for instance, clearly states that the treaty is designed “to create favourable conditions for greater investment by nationals . . . of one State in the territory of the other” and to foment the “reciprocal protection . . . of such investments.”¹⁰ Article 8(1) of the BIT also refers to “legal dispute[s] arising between [a] Contracting Party and a national or company of the other Contracting Party concerning an investment of the latter in the territory of the former.”¹¹ This means that, on its express terms, the BIT, as any other investment treaty, only covers “foreign” investments, that is, investments made by a national of the home state into the host state. It is therefore difficult to disagree with the dissenting opinion that having “acquired his wealth in Georgia” and “having taken part of this wealth out of Georgia to the UK to buy property there and to expand his Georgian business,” the Claimant's “profile does not fit the objective to the Preamble of the BIT, which is to create favourable conditions for greater investment by UK nationals in Georgia” (Concurring and Dissenting Opinion, para. 7). Previous tribunals have endorsed a similar approach.¹²

This leaves the tribunal's finding that the Claimant's acquisition of UK nationality did not amount to an abuse of rights. It should first be noted that abuse of rights objections based on the practice of nationality planning have almost exclusively focused on legal persons. A company can, for instance, alter its organizational structure by incorporating a legal entity in a contracting party to the targeted investment treaty, thereby becoming a protected home state national. While most tribunals have considered corporate nationality planning as legitimate, some have begun to put limits to this practice.¹³ Reaching back to case law on these limits, the obvious touchstone is Philip Morris v. Austria. In that case, the tribunal held that “the initiation of a treaty-based investor-State arbitration constitutes an abuse of rights . . . when the investor has changed its corporate structure [i.e., its nationality] to gain the protection of an investment treaty at a point in time when a specific dispute was foreseeable.”¹⁴ The tribunal also held that jurisdiction ratione temporis should be denied if an investor seeking access to an investment treaty does not hold the nationality of the home state at the time of the alleged events on which its claim is based.¹⁵

In line with this jurisprudence, Georgia argued that “the Claimant “could foresee” “confrontations” with the Respondent when he acquired British nationality” in 2011 (para. 280). The Claimant responded that the actions and omissions resulting in a breach of the BIT took place in 2015. The tribunal found no “evidence establishing that the events which occurred from 2015 onward were foreseen by the Claimant in a calculated decision to be naturalized as a British citizen (unlike in Philip Morris)” (para. 281). However, certain parts of the award narrating “a number of significant events” that occurred well before the Claimant became a British national suggest otherwise (para. 279).

For example, in Section III.C.1 of the award, the tribunal explains that, according to the Claimant, “the Respondent sent armed men to invade and occupy the premises of Omega Group companies in 2004” (para. 68), and that “he has repeatedly sought investigation of the 2004 events, but that these efforts have been stymied” (para. 69). More crucially, it clarifies that, for the Claimant, “the Respondent's alleged failure to investigate the 2004 occupation of the Omega Group constitutes a breach of the Treaty” (para. 69). Also important is the Claimant's assertion that the former Prime Minister of Georgia, Bidzina Ivanishvili, said that Georgia “should compensate the Omega Group for the 2004 events” (para. 70). This evidence suggests, at a minimum, that the dispute could have been foreseen by the Claimant in 2004, when he was not yet a UK national. Under this view, the Claimant could satisfy the foreseeability test established in Philip Morris v. Australia. The narrative of the 2004 events also indicates that the Claimant was not a national of the UK at the time of the alleged breach of the treaty, which should have resulted in the tribunal lacking jurisdiction ratione temporis under the BIT.

The award in Zaza Okuashvili v. Georgia illustrates that, while corporations have long been strategists in matters of nationality, individuals are increasingly adopting this role, seizing opportunities that states have inadvertently created for them. Investors like Mr. Okuashvili now enjoy the benefit of having different passports that can be used to make and operate the investment and, when it becomes convenient, to access an investment treaty. This decision can also be considered as creating an incentive for investors with one nationality to “internationalize” their claims through the acquisition of a second nationality to benefit from the investment treaty regime. These practices are the result of broad definitions of individual investors and a permissive approach toward claims by dual nationals. States that find these practices objectionable are advised to narrow the personal scope of their treaties. In the meantime, it remains to be seen whether the current (and future) arbitral tribunals deciding claims by dual nationals will follow the approach adopted in this case.