The recent award in Ahron G. Frenkel v. Republic of Croatia has already succeeded in dividing the very tribunal that rendered it—never an encouraging omen. It is unlikely to cease causing a stir in the near future, given its immediate impact on the claimant and its longer-term contribution to the perennial headaches induced by the doctrine of res judicata in investment arbitration. This post addresses four principal matters: the circumstances giving rise to the dispute in Frenkel v. Croatia; the competing analyses of res judicata advanced by the majority of the tribunal and the dissenting opinion; the importance of finality vis-á-vis a rigorous interpretation of the applicable law; and what the Frenkel award adds to the debate on the wider question of whether investment arbitration aspires to systemic coherence or rests on formal distinctions alone.
Factual Background and Procedural History
The facts of the case are fairly commonplace. Ahron G. Frenkel, an Israeli national, initiated arbitration proceedings against Croatia under the 2000 Israel-Croatia bilateral investment treaty (“BIT”) following the alleged derailment of his investment on Mt. Srđ. Croatian authorities, he claimed, had stymied the project through a medley of revocations, refusals, and bureaucratic evasions, all set to the backing track of political interference. So far, so ICSID.
This was not Mr. Frenkel’s first outing, though. His companies—Elitech B.V. and Razvoj Golf d.o.o., entirely owned by him—had previously brought near-identical claims under the 1998 Croatia-Netherlands BIT, which were dismissed in 2023. Mr. Frenkel returned undeterred, arguing that his claim rested on a different treaty and a slightly different set of facts. Croatia objected, maintaining that the parties, the relief sought, and the grounds were the same in substance if not in appearance, and that the dispute was therefore precluded. The majority of the Tribunal agreed, finding Mr. Frenkel and his companies to be in privity of interest, declaring the claims inadmissible, and labelling the renewed attempt an abuse of process.
As is often the case when res judicata makes an appearance in investment arbitration, a dissenting opinion follows close behind—and the one attached by Stanimir Alexandrov is rather difficult to dissent, if you’ll pardon the pun.
The Dissenting Opinion: Formalism and the Right to Present the Case
The first and most fundamental issue is the question of identity of parties. The majority assumed that because Mr. Frenkel wholly owns Elitech, he is the same as Elitech. That assumption is incorrect in law: it has long been recognised that a company has a separate legal personality from its shareholders, irrespective of how many shares the latter owns. A 100% shareholder may have aligned interests with a company, but alignment is not equivalence.
Second, the majority treated the Elitech and Frenkel arbitrations as involving identical causes of action. The two treaties—Croatia-Netherlands and Croatia-Israel BITs—certainly contain similar language; but does similarity of wording amount to identity of obligation? The Vienna Convention on the Law of Treaties makes clear that context matters, as do party intentions and interpretative practices. One must not conflate uniform drafting with uniform meaning. Indeed, if the mere replication of phrases across BITs were sufficient to establish identity of causa petendi, nearly every arbitration brought under an investment treaty could fall within the shadow of another.
Third, and more decisively, the Elitech tribunal found that it lacked jurisdiction over certain claims. That finding must logically preclude any claim of res judicata: a tribunal cannot extinguish a claim it has no power to hear. This is a matter of both logic and fairness. The doctrine of res judicata rests on the idea that an issue has been heard and decided on its merits: if jurisdiction was never affirmed, there is no judgment on the merits, and thus no bar to re-litigation. The Claimant, Alexandrov notes, was effectively denied the opportunity to present parts of his case—yet is now told that he is precluded from presenting them again.
A further concern lies in the Tribunal’s appeal to consistency in arbitral jurisprudence. That consistency is desirable is not disputed; but consistency cannot become a substitute for sound legal reasoning. Deference to precedent—especially in a system without formal stare decisis—must never come at the expense of examining the particularities of a case. Uniformity is not always justice, and respect for earlier awards cannot justify mechanical transposition. Each case, each claimant, is entitled to a hearing, not only in form but in substance.
A Question of Abuse of Process?
The claim of abuse of process, too, must be handled with care. It is fashionable in some circles to see treaty-based arbitration as a haven for opportunistic investors, and there is little doubt that some cases give weight to that view. However, it does not follow that every instance of repeated litigation is an abuse. In this case, one may legitimately argue that Mr. Frenkel sought not to multiply remedies but to be heard at all. One might criticise the practice of treaty shopping as undesirable, but it is not unlawful per se. Finally, one must ask whether the triple identity test—of parties, cause, and relief—was truly met: on parties, it appears that identity was assumed rather than proved; on cause, the treaties are not the same, even if their words are; and on relief, the claims may have resembled one another, but the basis of the claims—the party bringing them, the rights invoked, and the procedural posture—was different. These differences are not cosmetic: they go, in fact, to the heart of what res judicata is meant to capture.
The Fine Line Between Finality and Formalism
It would be a serious mistake to assume that the majority in Frenkel v. Croatia acted without due regard to principle, precedent, or fairness. On the contrary, Mónica Pinto and Zachary Douglas are both lawyers of high standing and unquestionable integrity. The temptation, always present in investment arbitration, is to treat every instance of res judicataas an unwelcome guest; but finality in adjudication is no less a component of the rule of law than is access to justice, and it is not unquestionable that Mr. Frenkel, while formally distinct from Elitech B.V., was made the functional equivalent by underlying interests, control, and conduct. As to the identity of cause, one must certainly be cautious not to treat BITs as if each were hermetically sealed; in fact, since Maffezini, it is hard to argue that each treaty is its own universe. The Croatia-Netherlands and Croatia-Israel BITs, while formally distinct, provided prima facie the same protections. Moreover, it cannot be overlooked that the Elitech tribunal rendered a reasoned award after full proceedings. That some claims were dismissed on jurisdictional grounds is true, but others were determined squarely on the merits.
In arbitration, it is no small matter when two arguments both rest on solid logic. The majority prized finality, while the dissenting arbitrator adopted a much more formalistic approach. Arguably, it is the latter that persuades, for in holding firm to the distinction between shareholder and company, it preserves the claimant’s right to be heard; and that, ultimately, is justice’s first requirement.
The Bigger Picture: System or Mechanism?
There is, however, a question that lurks behind many arbitral awards, often shrugged off as academic until it barges into the hearing rooms: is investment arbitration a system or a series of unrelated disputes dressed up in similar clothes? If it is truly a system based on a collective concerted effort of its actors, then the majority in Frenkel were right to peer behind the corporate veil and read the triple-identity test with a realist’s eye. Such a system would demand consistency, discourage treaty shopping, and elevate the elusive jurisprudence constante from polite fiction to functional truth. If, however, arbitration is no more than a mechanism—a switch tripped by bilateral treaties with no central spine—then the test must be applied as written: formally, rigidly, even obtusely if needed. The result may be untidy and possibly not suited to current multinational structures, but that is not an arbitrator’s concern. Either view has its virtues, but one cannot have both realism and formalism in equal measure.
________________________
To make sure you do not miss out on regular updates from the Kluwer Arbitration Blog, please subscribe here. To submit a proposal for a blog post, please consult our Editorial Guidelines.
