Foreign Public Policy and The Law Governing Arbitration Agreements

The law governing the arbitration agreement is often considered to govern important matters such as the validity and enforceability of the arbitration agreement. Yet, a multiplicity of approaches exists to choosing which law this should be.

Despite this, in practice, most arbitration clauses do not explicitly state what law governs the arbitration agreement. Even the model clauses of major arbitration institutions do not suggest that the parties should deal with this issue explicitly (See, Standard ICC Arbitration clause, LCIA model clauses, SIAC model clause). That said, recent decisions have highlighted the importance for parties to expressly state which law governs their arbitration agreements, especially where the seat of the arbitration differs from the governing law of the contract. In the recently concluded Kabab-Ji saga, for instance, where the parties had not specified which law should govern the Paris-seated arbitration agreement, different conclusions by the French and English Courts in this regard lead to dramatic and inconsistent outcomes: the award was upheld against a non-signatory in Paris, whereas enforcement was denied in England. We reported on this decision here, and the Paris Cour de Cassation recently upheld the Court of Appeal decision (See, Cour de Cassation, decision No. 679 FS-B dated 28 September 2022).

A recent decision by the Singapore Court of Appeal in Anupam Mittal v Westbridge Ventures II Investment Holdings (the ”Mittal case”) once again shows how the determination by courts of what law governs the arbitration agreement could affect its enforceability in ways unexpected by the parties, increasing the importance of an express designation. In the Mittal case, the question before the Singapore Court was what law applies to arbitrability issues at the pre-award phase (the law of the seat or the law governing the arbitration agreement), in the context of an anti-suit injunction sought at the seat of the arbitration in Singapore.

Below, we provide an overview of the reasoning by the Singapore High Court and Court of Appeal and set out the Court of Appeal’s novel reasoning, according to which both the law of the seat and the law of the arbitration agreement should be considered. We provide a critical assessment of this decision and set out some practical considerations flowing from it.

Background

The dispute arose between Mauritius incorporated private equity fund West bridge Ventures

(the ”Defendant”) and, among others, Anand Mittal, the founder of Mumbai-based People Interactive, a company which operated a matrimonial service (the ”Appellant”). The Shareholder Agreement concluded between the parties (the ”SHA”) envisaged that an IPO would be concluded five years from closing, in the absence of which the Defendant, who held shares in People Interactive, could exit its investment and force the Appellant to sell. By 2017, the IPO had not been achieved, and the Defendant sought to divest its interest in the company by selling to a competitor of People Interactive. The Appellant filed a corporate oppression claim before the Indian National Company Law Tribunal (”INCLT”). However, the SHA contained an arbitration clause providing for all disputes arising under the SHA to be resolved by Singapore seated arbitration.

The Defendant therefore applied for an anti-suit injunction in Singapore. Before the Singaporean High Court seized with the request, the Appellant argued that the claims of minority shareholder oppression and mismanagement brought before the INCLT were non-arbitrable under the laws of India, which should be construed as the proper law governing the arbitration agreement as the law governing the SHA, and therefore that there was no breach of the arbitration agreement. The Defendant took the position that the claims are arbitrable under Singapore law, where the arbitration is seated, and thus, the Court proceedings brought before the INCLT were in breach of the parties’ arbitration agreement, thereby warranting the grant by of the anti-suit injunction.

Decision of the High Court

The Singapore High Court granted the injunction and held that the law governing the issue of arbitrability at the pre-award stage is the law of the seat. The notion of arbitrability is captured by public policy under the Singapore International Arbitration Act (“SIAA”), which is based on the UNCITRAL Model Law (See, Section 11(a) of the SIAA: ”Any dispute which the parties have agreed to submit to arbitration under an arbitration agreement may be determined by arbitration unless it is contrary to public policy to do so”). With reference to both the SIAA and the UNCITRAL Model Law, the High Court noted that at both the setting aside and enforcement stages, the relevant public policy is that of the state in which the Court hearing the post-award challenge sits. The High Court found, by analogy, that applying the law of the Court hearing the challenge was also justified at the pre-award phase.

The High Court also noted that applying the law of the arbitral seat (thus in the case at hand, Singapore) is justified because the purpose of the non-arbitrability rule is to set a limit on party autonomy, where such autonomy conflicts with the public policy of the state being asked to give effect to an arbitration agreement. In addition, the Courts of the seat have a supervisory role in the arbitration process and hold the ultimate control over the award. The High Court referred to the UNCITRAL Secretariat’s analytical commentary on draft Article 16 of the Model Law that a tribunal dealing with issues relating to its jurisdiction at the pre-award stage should apply the same law which the seat Court would apply at the post-award stage in setting-aside proceedings. The Court considered that, similarly, at the pre-award stage, the seat Court should also apply its own law to issues of subject matter arbitrability.

In the same vein, and echoing the view taken by the UK Supreme Court in Enka, the High Court considered that applying the same law to arbitrability issues at both pre- and post-award stages was desirable because it avoids potential inconsistencies that might otherwise arise from applying the law the arbitration agreement to the question of pre-award subject matter arbitrability.  The High Court referred to several examples, such as where a seat Court, if it applied a law other than the law of the seat, at the pre-award phase, might grant a stay of proceedings at the seat on the basis that the dispute is arbitrable based on foreign law (i.e., the law that applies to the jurisdiction agreement), despite it not being arbitrable at the seat. However, that same award would later be set aside by that same Court in set-aside proceedings, where the Court is mandated by the Model Law to apply its own law.

The High Court also considered that applying the law of the seat is more consistent with the policy of promoting international commercial arbitration; it is Singapore’s public policy to promote international arbitration and giving effect to a foreign non-arbitrability rule is inconsistent with this objective. Finally, the High Court considered that the weight of authority also leaned in favor of this conclusion.

Court of Appeal Decision

Although it maintained the anti-suit injunction, the Singapore Court of Appeal adopted a novel approach to determining arbitrability in the pre-award context, in what appears to be an attempt to give more prominence to foreign public policy.

In its decision dated 6 January 2023, the Court of Appeal adopted a ‘composite approach’ to determining what law should govern arbitrability in the pre-award context, requiring that the Court be satisfied that the subject matter of the dispute is arbitrable under both the law of the seat and the law of the arbitration agreement.

a) Law Governing Arbitrability

The Court of Appeal considered that the UNCITRAL Model Law did not specify what law applies in the pre-award phase and that the SIAA’s reference to public policy was phrased broadly enough for the Court to consider public policy of foreign jurisdictions.

The Court of Appeal disagreed in particular with the High Court’s finding that the law which governed the arbitration agreement was irrelevant to the issue of arbitrability. The Court explained the arbitration agreement taken together with the law which governs it, determines what parties have consented to arbitrate, and the law of the arbitration agreement

”deals with matters of the validity of the agreement and is, in that sense, anterior to the actual conduct of the arbitration” (See, para. 53).

The application of the law of the seat without consideration for the law of the arbitration agreement also ignores there are circumstances in which the public policy of a foreign jurisdiction ”could impact the parties or the arbitration in some way” and that the Court of the seat should give it effect (See, para. 49). For instance, if parties agree to arbitrate a range of questions that include, for instance, custody of a minor, and they agree that the arbitration agreement is governed by a law under which this issue is non arbitrable then, “surely the question of custody simply cannot be arbitrated regardless of what the seat law or any other law provides” (See, para. 53).

The Court of Appeal expressly suggests that it might be particularly important to give effect to foreign policy where the only tie to the country of the Court hearing the claim is the seat. For instance, in the present case, the parties are mainly Indian, the disputes arise under a contract containing an Indian choice of law clause and an Indian law provides that oppression disputes are non-arbitrable.

Therefore, according to the Court of Appeal, a dispute cannot proceed to arbitration in Singapore if it is contrary to foreign public policy, provided that foreign law governs the arbitration agreement, even if Singapore law considers the matter arbitrable. Conversely, if an arbitration agreement does a priori come into effect because it is arbitrable under the foreign law that governs it, but the subject matter happens to be non-arbitrable under the law of the seat, that would be an additionalobstacle to the enforceability of the arbitration agreement by reason of Art 34(2)(b)(i) of the Model Law.

The Court of Appeal noted, however, that an arbitration agreement is ”foreign” only if it is indisputably governed by foreign law, and this would ”in almost all cases” only be the case when expressly chosen by the parties as its governing law.

b) Law Governing the Arbitration Agreement

However, when turning to what law governs the arbitration agreement in this case, the Court of Appeal concluded that the law of the seat, Singapore law, governs the arbitration agreement, and not Indian law. The Court applied the three-step test established by Singaporean Courts in BCY v BCZ [2017] 3 SLR 357 (“BCY”), to determine which law governs the arbitration agreement. In the absence of an express choice by the parties, the Court is to look at any implied choice of law, with the starting point for determining the implied choice of law being the law of the contract (unless there are indications to the contrary). If no express or implied choice can be found, then the Court must consider which system of law has the closest and most real connection to the arbitration agreement.

The Court of Appeal’s application of this test is illustrative of the uncertainty that can arise where the law governing the arbitration agreement is not specified: the Court of Appeal found that the choice of Indian law as governing the contract was not an express choice of law governing the arbitration agreement, and that, while it forms the starting point for an impliedchoice, in this case such an implied choice was negated by the fact that that designating Indian law would frustrate the parties’ stated intention to arbitrate all disputes arising from the arbitration agreement. In that light, the Court found that the law of the seat had most ”real and substantial connection with the arbitration clause”, and was therefore the law of the arbitration agreement. On this basis, the Court of Appeal held that Singaporean law applied to the arbitration agreement, the subject matter was arbitrable, and the anti-suit injunction was upheld on this basis.

Reflection

One may wonder whether the justifications for applying the law of the seat by the High Court have been adequately rebutted by the Court of Appeal.

The application of both the law of the arbitration agreement and the law of the seat mandated by the composite approach does appear to prevent the inconsistent outcomes flagged by the High Court (because whenever the subject matter is not arbitrable under either the law of the arbitration agreement or the law of the seat, the arbitration will not be able to proceed) (See, CA Judgment, para 59). That said, this approach nevertheless gives rise to increased uncertainty as to which laws will be considered by the Courts in the absence of a designation of the law that governs the arbitration agreement. It is also inconsistent with established practice of state Courts in different jurisdictions (State Courts in several jurisdictions have applied the lex fori at the pre-award phase, including those of the United States, France, Switzerland, Holland, Belgium, Italy, Austria and Swede, See, para 43 of CA judgement, citing Gary B Born) and likely, the practice of arbitral tribunals, who will be inclined to adopt the approach recommended by the UNCITRAL secretariat.

The High Court’s concern that Singapore’s own public policy of promoting international arbitration by not giving effect to foreign non-arbitrability rules is, in our view, also not adequately addressed by the Court of Appeal, especially where arbitrability is a known guerilla tactic by parties seeking to avoid arbitration agreements. It is also unclear that this attempt to give broader effect to foreign public policy and principles of comity – possibly to the detriment of Singapore public policy – is warranted, when, given that a Court where the award is enforced can always, on the basis of its own law, deny enforcement of the award (the enforcement Court may invoke Article V(2)(a) or Article V(2)(b) of the New York Convention).

One may also wonder whether this novel reasoning will in practice make any difference, as illustrated by the decision itself. On the one hand, the Court of Appeal emphasizes the importance of giving effect to foreign public policy. This seems precisely like the type of case where, according to the Court of Appeal itself, foreign public policy could legitimately be considered given the few ties to the seat, and the many ties to India. On the other hand, however, the Court of Appeal avoids having to do so on the sole basis that the parties intended to arbitrate, and therefore they could not possibly have meant for Indian (foreign) law to apply to the arbitrability issues.

It is difficult to conceive of any case where this ”negation” of the implied choice of law would not exist, except perhaps in the case of an express choice of foreign law governing the arbitration agreement. Indeed, the Court was not prepared to accept that the parties intended for Indian law to govern the agreement as it rendered it invalid, but it would be willing to accept such an absurd outcome had the parties expressly selected Indian law as governing the arbitration agreement (as the Court explained with the custody example given above).

In any case, what is clear is that this case confirms the importance of express designations by the parties of the law that governs the arbitration agreement: it appears that, in this case, had the parties expressly designated Indian law as the law that applied to the arbitration agreement, the anti-suit injunction would have been denied. Conversely, had they been clear that Singapore law applies, the debate could have been avoided altogether.

Takeaways

In that light, a few clear practical recommendations should be set out to prevent parties from exploiting ambiguity regarding the law applicable to the arbitration agreement.

• Parties should, especially where the law of the seat differs from the law of the contract (often as result of negotiation), include an express choice of law for the arbitration agreement, ideally separate from the general choice of law clause governing the main contract. Evidently, parties should ensure that this choice of law will not render disputes non-arbitrable.
• In the absence of an express choice of law for the arbitration agreement, parties may want to ensure that the disputes they submit to arbitration are arbitrable under both the law of the contract and the law of the seat (where these diverge), to avoid any surprise outcome if a law different than what was anticipated is applied to the arbitration agreement or matters of arbitrability.
• Regarding arbitrability specifically, it is wise to look to jurisdictions of likely enforcement of an eventual award with legal advisors, to determine whether enforcement of any award might be denied on the basis of arbitrability in thesefora.
• Choosing an arbitration-friendly seat is also advisable to avoid attempts by parties at escaping arbitration agreements.

Dutch Courts are a very safe forum in this regard. To the extent that Dutch law itself is relevant to the question of arbitrability – for instance if the arbitration is indeed seated in the Netherlands, Dutch Courts, on the basis of Article 1020(3) DCCP, have considered very few matters to be non-arbitrable. Civil law matters generally, including contract and tort claims as well as disputes arising out of shareholder agreements, are generally arbitrable. Examples of non-arbitrable matters include the legal validity of decisions and resolutions of organs of a legal entity and amendments to articles of association (See, e.g., A. Marsman International Arbitration in the Netherlands, with a Commentary on the NAI and PCA Arbitration Rules (Kluwer Law International 2021) at pp. 55-100).

Additionally, when non-arbitrability issues are raised before a Dutch Court having been seized at a pre-award phase of a foreign-seated international arbitration, the Court will likely apply principles of minimal curial intervention as well as a robust pro-arbitration policy. Consequently, the Dutch Courts would likely wait until the arbitral tribunal renders its decision on jurisdiction (and likely act in a similar vein when it comes to a related decision of the supervisory Courts of the seat), reserving review only for the enforcement stage.

This blog was first published on the JusMundi's Blog.