Part I of this two-part blog post summarized the recent judgment of the Singapore Court of Appeal (“SGCA” or the “Court”) in Sanum Investments Ltd. v Government of the Lao People’s Democratic Republic (“Sanum v Laos”). This Part II provides some comments on that judgment and its significance, including its impact on future Singapore court proceedings involving investment arbitration awards and future disputes involving PRC treaties.
This is the first judgment by Singapore’s highest court involving the curial review of an investment award, and it lays down several clear, important markers for future cases. First, the SGCA confirmed that the interpretation and application of treaties to which Singapore is not a party, when the Singapore courts exercise their functions under the IAA to review jurisdictional issues, are justiciable matters when the Court reviews awards under the IAA. This is clearly correct, and is consistent with the position taken by courts in other jurisdictions, including the United Kingdom, United States, Netherlands and elsewhere.
Second, the SGCA confirmed that the de novo standard of review applies to investment awards under Singapore law. Commentators have, for some time, argued that a more deferential standard of review should be applied by courts in reviewing investment awards as compared to commercial arbitration awards. The stated justification is that such cases typically involve complex questions of public international law, particularly on issues of jurisdiction, on which courts may not necessarily have particular experience or expertise (see e.g. T. Weiler and T. Cheng, “The Virtue of Judicial Restraint: Two Comments on Laos v Sanum”, Global Arbitration Review, 12 March 2015). The SGCA categorically rejected this view, finding that Singapore case law requires a de novo review of an arbitral jurisdiction in investment awards.
This conclusion is arguably mandated by the IAA, which does not draw any distinction between different types of awards and does not contemplate multiple standards of review for jurisdiction. It is also arguably consistent with the approach of other jurisdictions which appear to apply a de novo standard of review for jurisdictional awards (see e.g. Czech Republic v European Media Ventures SA,  EWHC 275 (Comm.); BG Group v Argentina 134 S. Ct. 1198 (2014)). One may query, however, whether categorical distinctions of this sort are sufficiently nuanced and whether there may be circumstances in which a more deferential standard of review would be appropriate with regard to particular issues (for example, where jurisdictional issues are intertwined with the merits of a dispute).
An even thornier question, which may arise in future cases but was not in issue in Sanum v Laos, is whether or not a particular issue goes to jurisdiction or admissibility. Investment tribunals and national courts have, for some time, recognized a distinction between jurisdictional and admissibility issues (see e.g. SGS v The Philippines (ICSID Case No Arb/02/6, Decision on Objection to Jurisdiction, 29 January 2004). Admissibility issues generally pertain to whether aspects of the claim can be heard, and commentators widely agree that, in principle, the arbitral tribunal’s decision on admissibility should be final (just like its decision on the merits) and not subject to the same standard of review by curial courts as jurisdictional issues: see e.g. BG Group v Argentina 134 S. Ct. 1198 (2014); G. Born, International Commercial Arbitration, (Kluwer, 2014), pp. 936-938).
The distinction between jurisdiction and admissibility, however, is not always easy to draw. Unlike issues that clearly go towards admissibility of claims, e.g. time bars, or issues that clearly go towards jurisdiction of the tribunal, e.g. the existence of an arbitration agreement – there are other issues which are more difficult to characterize. For example, there is some divided authority on whether pre-conditions to arbitration are jurisdictional or admissibility issues (although the better majority view is that they are more appropriately characterized as admissibility issues): see G. Born and M. Scekic, Pre-Arbitration Procedural Requirements: A Dismal Swamp, in D. Caron and others (eds), “Practising Virtue: Inside International Arbitration” (Oxford, 2015). It is important that the Singapore courts are careful, in future cases, to distinguish between jurisdictional decisions, which may be reviewed de novo, and decisions by tribunals who have jurisdiction on admissibility issues.
As for the applicability of PRC-concluded BITs to Macau, the SGCA was placed in the difficult position of having to deal with an exchange of diplomatic correspondence by the contracting state parties to a BIT that purported to agree on the territorial scope of their treaty, but which was made only after the Award was rendered (and only a few days before the filing of the court challenge).
Here, the Court adopted a finely nuanced line of reasoning that relied on the “critical date” doctrine to discount the weight to be applied to the 2014 NVs. The overall reasoning and result was correct, given the particular facts of the case, including a complete absence of any evidence to show that the PRC and Laos ever addressed the territorial scope of application of the BIT in any of its bilateral engagements prior to the 2014 NVs.
One might view the Court’s interpretive approach at one isolated part of the judgment, in testing obiter whether the 2014 NVs would constitute subsequent agreement or practice under Article 31(3) VCLT by reference to whether they demonstrated any “pre-existing agreement,” as being in tension with the classical approach to treaty interpretation. The interpretation of treaties is not necessarily determined by a “fixed ‘original intent’, but must rather be determined by taking into account a broader range of considerations, including certain later developments,” and in this context the parties’ subsequent agreement or practice can be an authentic means of interpreting a treaty (see e.g. Second Report of the Special rapporteur, Mr. Georg Nolte at the 66th session of the ILC (2014) A/CN.4/671, at pp. 50-51; Yearbook of the International Law Commission (1966), vol. II, p. 221, para. 15).
The Court’s approach, however, is explicable on the particular facts of Sanum v Laos, not solely because of the timing issues, but also because the PRC-Laos BIT, like many investment treaties, specifically creates legal rights and obligations that are intended for the benefit of third parties to the treaty. And here, where such rights and expectations have accrued to particular persons claiming under the BIT, the Court should weigh such accrued third party rights in the balance in deciding how much evidential weight to accord the 2014 NVs. Indeed, under Article 31(3) VCLT, the Court need only “take into account” any such subsequent agreement or practice. The Court’s approach is consistent with an emerging view that the interpretation of treaties which establish rights for other states or actors may be less susceptible to post-hoc “authentic” interpretation by its parties in the context of a dispute (see e.g. Sempra Energy International v. Argentine Republic (ICSID Case No ARB/02/16, Award, 28 September 2007, para. 386)).
The implications for future determinations of the applicability of PRC BITs to Chinese Special Administrative Regions remain unclear. While clarifying that post-hoc documents that retroactively alter the scope of those treaties will be not be given effect, the Court also confined its findings to the features of the particular PRC-Laos BIT. The Court appeared to accept obiter that, had the PRC-Laos BIT, like NAFTA, permitted the States to issue binding interpretive statements, the 2014 NVs might have resulted in a different outcome. The effect of binding interpretive statements are far from non-controversial, having on occasion sparked rebuke from Tribunals that have seen it as a way for Contracting States to avoid liability (see Pope & Talbot v Canada, Award in Respect of Damages, 31 May 2002). Commentators have also noted that some interpretative statements are tantamount to retroactive amendments, which may undermine the rule of law (see e.g. Kaufmann-Kohler, Interpretive Powers of the Free Trade Commission and the Rule of Law in Gaillard and Bachand, Fifteen Years of NAFTA Chapter 11 Arbitration (JurisNet LLC, 2011), at pp.190-192). Given the increasing number of newer generation investment treaties that permit Contracting States to issue binding interpretive statements, this will continue to be a fertile topic for debate.
Notably, the PRC Ministry of Foreign Affairs has recently issued a statement criticizing the judgment, and repeating its position that PRC treaties do not apply to Macau and Hong Kong. Given how much of the Sanum judgment was focused on the particular facts of the case, it remains unclear how future tribunals or courts will approach the territorial scope of different PRC treaties in different contexts, in light of the state practice of the PRC, and the conduct of other states in response—although the Court’s careful forensic analysis of the various pieces of extrinsic evidence will no doubt be of considerable influence.
Finally, with respect to the “amount of compensation” clause, the SGCA read Article 8(3) purposively in light of the principle of effective interpretation, and was careful to give Article 8(3) a meaning that would not render nugatory the investor protections under the BIT. The Court did not find the ordinary meaning of the words in the BIT to be dispositive, as the PRC-Laos BIT was at best ambiguous. This approach may be contrasted to the decision of the Svea Court of Appeal in Juan Ignacio and others v Russia, Case No T 9128-14, 18 January 2016, which, in setting aside the award in the Renta 4 case, took a highly textual approach to interpreting a similar clause in the Russia-Spain BIT. In contrast, the SGCA focused more comprehensively on the object, purposes and other provisions of the BIT (in particular its fork-in-the-road clause), while being careful to limit its conclusions to the PRC-Laos BIT.