As reported earlier, the US Supreme Court has recently adjudicated on the issue of the standard of review in relation to arbitration agreements in international investment arbitration.
It is a fact that the majority of the Court has decided that deference should be given to arbitral tribunals to examine questions of procedural conditions, as it characterized the issue of litigation before the domestic courts of Argentina for 18 months before initiating arbitral proceedings. It seemed to the majority of the Court that the issue pertains to whether a duty to arbitrate arises, and not whether such a duty exists at all.
In determining the issue at hand, the majority found that the United Kingdom – Argentina Bilateral Investment Treaty (the “BIT”) is not different from a contract, and did not refer to the customary rules of treaty interpretation, as codified by the Vienna Convention on the Law of Treaties (“VCLT”), while a reference to the VCLT was made by the District Court of Appeals.
In general, the US Department of State supports the policy that the VCLT has to be treated as the authoritative guide to current treaty law and practice (See United States v Yousef, 2003). Moreover, Courts in the United State have stated that the Convention enjoys the status of customary international law, meaning that it should also be considered to be “law of the land”, in other words to form part of US national law (See Gonzalez v Gutierrez, 2002). Nevertheless, the US Supreme Court has refrained from citing it as a source of interpretation of treaties, doing so only sporadically (Sale v Haitian Ctrs Council, Inc, 1993).
The majority of the Court refers to the grammatical interpretation of the BIT, as required by Article 31 VCLT. However, even if the majority refers to the text and structure of the dispute settlement provision found in the BIT, it does not seem to embrace the view that a grammatical interpretation of the BIT would favour the argument that a resort to the domestic courts of Argentina for 18 months is a condition to the consent of Argentina to arbitration, given the fact that the relevant provision (Article 8 BIT) stipulates that an arbitration can be requested only after the domestic courts have had a first crack on the dispute.
The latter reasoning, even if not adopted by the majority of the court, is employed by Chief Justice Roberts, who is also joined by Justice Kennedy in this respect. It is interesting, in this respect, to analyse the opinion of the dissenters. It is reminded that an offer to arbitration by a State remains incomplete until the investor decides to comply with the offer. Until then, the offer remains inchoate. Issues pertaining to the question whether there is an arbitration agreement are, according to the dissenters, for the Court to determine and not for the arbitrators. For Chief Justice Roberts, the rule in the dispute settlement clause in the BIT is not relevant to the question when the duty to arbitrate arises, but whether such duty has arisen. The dissenters criticize the majority for not elaborating on the question what the terms “after a period of eighteen months has elapsed” really mean, and for failing to recognize that the really pertinent question is not whether the requirement of litigation before the domestic courts is a condition to the performance of the contract, but whether the requirement is a condition to the formation of a contract, namely in that case the formation of the arbitration agreement.
It is interesting that the dissenting opinion is in line with the dissenting opinions of Brigitte Stern and John Christopher Thomas QC, in recent cases.
In Impregilo v Argentina, Brigitte Stern suggested that a State can shape its consent to international adjudication as it sees fit, by providing the conditions it deems necessary, rationae personae, rationae materiae, rationae temporis and rationae voluntatis. The condition of litigation before the domestic courts falls under the last category of the abovementioned conditions. Besides, there is no inherent right of access to a jurisdictional recourse. If these criteria are not met, then there is no consent by the State to international arbitration. Hence, in case an investor does not fulfil all of the necessary criteria imposed by the State, an agreement to arbitration should not be considered to be reached.
In Hochtief v Argentina, John Christopher Thomas QC suggests that this form of procedural requirement found in a dispute settlement clause is a standing offer to arbitrate, under certain criteria. If the consent of the investor to arbitration does not match the offer of the State, then there is no meeting of the minds and the consents do not match, and therefore an agreement to arbitrate is not formed.
Similar decisions which found that provisions asking for a period of litigation before domestic courts were to be considered as conditions to a State’s consent to arbitrate (see ICS Inspection and Control Services Ltd v Argentina Republic, 2012; Daimler Financial Services v Argentine Republic, 2012; Wintershall Aktiengesellschaft v Argentine Republic, 2008). These cases have also been cited by the dissenters in their opinion in the present case.
For the above-mentioned reasons, it can be concluded that the decision of the US Supreme Court does not entirely clear up the issue of litigation before the domestic courts, as the interpretations vary and tribunals and courts have reached contradictory judgements. Hence, only future can tell which interpretation will form good law.
________________________
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.
Dear Georgios,
Thank you for your excellent analysis of the impact of the dissent in BG v Argentina. The decision and following comment prompt two questions. First, what difference would it have made if the Court did refer to the VCLT, would the outcome have been different? Second, is it a good thing or a bad thing for the dissenters to cite other arbitral decisions as precedent?
Regards,
Michael J Bond
Thank you for this highly interesting analysis Georgios.
One can say that the Supreme Court was right in concluding that compliance with a procedural requirement does not ordinarily constitute a jurisdictional prerequisite or condition for the coming into existence of an agreement to arbitrate. The key question however is whether in this particular case, the procedural requirement in question also doubled as a substantive requirement for the coming into existence of an agreement to arbitrate. Situations where what is prima facie a purely procedural requirement doubles as a substantive requirement are not unknown to international law. Take for instance, the exhaustion of local remedies requirement that was in issue in Loewen v. United States of America. The general rule that the exhaustion of local remedies is a procedural requirement and not a substantive condition for the coming into existence of a breach of international law, does not apply to the offence of denial of justice in whose context the requirement of exhaustion of local remedies is both a procedural requirement as well as a substantive requirement for establishing that the offence has been committed in the first place – been completed. Notably no BIT conferring jurisdiction over claims of denial of justice need mention the full exhaustion requirement explicitly as it is generally accepted to form part and parcel of the substantive definition of the delict.
From all appearances, what the BG Group case presented was a classic situation in which a requirement that would ordinarily be classified as a merely procedural could (depending on one’s interpretation of Art 8 of the UK/Argentina BIT) well have constituted in addition, a condition precedent for the coming into existence of an agreement to arbitrate. The key difference in the majority and dissenting opinions was apparently in their interpretations of Art 8. Having regard to how Art 8 was couched (and leaving aside the separate question whether a resort to the Argentinian courts would have been futile), it would have been possible to conclude that there was no valid acceptance of Argentina’s standing offer to arbitrate and thus no agreement to arbitrate between BG and Argentina.
Would a reference to the VCLT have made any difference? Perhaps. Considering that the VCLT requires reference not just to the text and structure of a treaty, but also its context in interpreting its terms. In interpreting Art. 8, the dissenting opinions seem to have given more weight to the context of investment treaty arbitration (and offer and acceptance of an offer to arbitrate in that particular context) than to domestic arbitration/contract law. But this is not to say that a majority decision that gave more weight to the international law context would necessarily have come to a different conclusion than the present majority (considering other considerations that could well come into play in that context). Indeed only the future will tell.