[Update: the paragraphs on the Singapore High Court’s analysis of the arbitration clause and its implications has been updated to more closely reflect the language at paragraph [17] of the Singapore High Court’s decision.] 

A party may rely on a valid and enforceable arbitration clause to obtain a stay of court proceedings commenced by the counterparty to the arbitration agreement. The applicant party may even obtain an anti-suit injunction to prevent a party to the arbitration agreement from commencing court proceedings (see Power of the Singapore Court to grant permanent anti-suit injunction in aid of arbitration proceedings as well as English Court has power to issue an anti-suit injunction in support of non-existent arbitration).

On the other hand, if an arbitration clause is defective, then a party may resist the stay application on the basis that the arbitration agreement is unenforceable or that the dispute between the parties does not fall within the scope of the arbitration agreement. These issues of enforceability and scope of the arbitration agreement are determined  by the governing law of the arbitration agreement (see The Laws Governing an Arbitration as well as the case of Piallo GmbH v Yafriro International Pte Ltd, [2013] SGHC 260, which is analysed here).

What happens then when parties do not expressly provide for the governing law of the arbitration agreement? The legal analysis can get complicated.

The Singapore High Court case of FirstLink Investments Corp Ltd v GT Payment Pte Ltd and others, [2014] SGHCR 12 had to deal with these two issues concurrently. The Plaintiff, FirstLink Investments Corp Ltd, had commenced Singapore court proceedings against the three defendants. The 1^st Defendant applied for a stay of those proceedings on the basis of a Stockholm Chamber of Commerce (“SCC”) arbitration clause between the Plaintiff and itself. The Plaintiff then resisted the stay on the basis that the arbitration clause was null, void, inoperative or otherwise incapable of being performed.

The arbitration clause read,

“Any claim will be adjudicated by Arbitration Institute of the Stockholm Chamber of Commerce. You and GTPayment agree to submit to the jurisdiction of the Arbitration Institute of the Stockholm Chamber of Commerce. Both parties expressly agree not to bring the disputes to any other court jurisdictions, except as agreed here to the Arbitration Institute of the Stockholm Chamber of Commerce[.]”

Further complicating matters, the governing law of the underlying agreement referred to the rules of the SCC rather than the laws of a state.

“16. General.

This Agreement is governed by and interpreted under the laws of Arbitration Institute of the Stockholm Chamber of Commerce as such laws are applied to agreements entered into and to be performed entirely within Stockholm.”

The Singapore High Court held that the arbitration agreement was a valid and enforceable agreement that was governed by Swedish law.

Threshold question – applicable standard for determining validity of an arbitration agreement

The learned Assistant Registrar Shaun Leong (“Leong AR”) cited his earlier decision in The “Titan Unity” [2013] SGHCR 28 for the proposition that,

“…an applicant for a stay of court proceedings pursuant to section 6(1) of the IAA must satisfy on a prima facie basis the pre-condition of  showing the existence of an arbitration agreement, without which the court would have no jurisdiction to grant a stay. Where this jurisdiction is invoked, the court must grant a stay unless the agreement is shown to be “null and void, inoperative or incapable of being performed”…” (see paragraph [6])

In other words, “it would be sufficient for the purposes of a stay application that the court be satisfied of the agreement’s validity on a prima facie basis without having to descend to a full review” (see paragraph [7]).

Leong AR rejected the Plaintiff’s reliance on English cases for the converse position requiring a more extensive review. He cited Titan Unity for the reasons why it was not appropriate to follow English cases authorities on this issue and drew a distinction between the UK Arbitration Act 1996, which was not based on the UNCITRAL Model Law, and the Singapore International Arbitration Act (Cap. 143A) which was. Furthermore, Commonwealth case law (from Model Law jurisdictions) on the issue far outweighed the English position.

Parenthetically, the English position can be seen in the case of Joint Stock Company “Aeroflot Russian Airlines” v Berezovsky & Ors [2012] EWHC 1610 (Ch) (see Stay of Proceedings – Standard of Proof that Arbitration Clause is not Null and Void for a detailed analysis).

Determination of governing law of arbitration agreement – departure from English position

The Singapore High Court adopted the English Court of Appeal’s three-stage inquiry in SulAmérica Cia Nacional De Seguros S.A. and others v Enesa Engenharia S.A. [2012] 1 Lloyd’s Rep 671 to determine the governing law of the arbitration agreement, namely – (i) the express choice of the parties; (ii) the implied choice in the absence of an express choice; and (iii) where the parties had not made any choice, the law in which the arbitration agreement had its closest and most real connection with.

The Singapore High Court considered that with regard to stage two of the inquiry (implied choice), the English courts had created a rebuttable presumption that the express substantive governing law (the governing law of the underlying agreement) was the parties’ implied choice of the governing law of the arbitration agreement. However, where no substantive law was chosen, the choice of the seat of the arbitration would be “overwhelmingly significant” and that would likely be the governing law of the arbitration agreement. The learned AR considered this to be the effect of the trio of English cases – SulAmérica, Asranovia Ltd & Ors v Cruz City 1 Mauritius Holdings [2012] EWHC 3702 (Comm), and Habas Sinai Ve Tibbi Gazlar Istihsal Endustrisi AS v VSC Steel Company Ltd [2013] EWHC 4071 (Comm).

For a more detailed analysis of the cases of SulAmérica and Arsanovia (see The Proper Law of the Arbitration Agreementand Case Update: (1) Lack of substantive jurisdiction in respect of one respondent affects award as against the other respondent; (2) Substantive jurisdiction not affected by finding of liability under a different agreement respectively).

The Singapore High Court disagreed with the English position to the following extent: where parties had expressly provided for the governing law of the underlying agreement (e.g. England and Wales) but had chosen a different seat for the arbitration (e.g. Singapore), the Singapore courts would not infer or assume that parties intended for the law of the underlying contract to take precedence over that of the law of the seat of the arbitration (see paragraph [13]). The Singapore High Court reasoned that,

“…this court takes the view that it cannot always be assumed that commercial parties want the same system of law to govern their relationship of performing the substantive obligations under the contract, and the quite separate (and often unhappy) relationship of resolving disputes when the more commercially sensible viewpoint would be that the latter relationship often only comes into play when the former relationship has already broken down irretrievably. There can therefore be no natural inference that commercial parties would want the same system of law to govern these two distinct relationships. The natural inference would instead be to the contrary. When commercial relationships break down and parties descend into the realm of dispute resolution, parties’ desire for neutrality comes to the fore; the law governing the performance of substantive contractual obligations prior to the breakdown of the relationship takes a backseat at this moment (it would take the main role subsequently when the time comes to determine the merits of the dispute), and primacy is accorded to the neutral law selected by parties to govern the proceedings of dispute resolution. …”

Furthermore, given the importance of the choice of the seat in an arbitration, the Singapore High Court agreed with the English Court of Appeal’s decision in C v D [2007] EWCA Civ 1282 that it would be “rare” for the governing law of the arbitration to be different from that of the law of the seat (see paragraph [14]).

Finally, the Singapore High Court reasoned that parties would conceivably demand that the law of the seat of the arbitration be the governing law of the arbitration agreement in order to achieve consistency between the lex arbitri (the law of the seat governing the arbitration’s procedures) and the governing law of the arbitration agreement (determining the validity of the arbitration agreement). This was because “commercial parties would not…select a place to be the seat if they do not at least have the notional confidence that the supervisory court would recognise and give effect to the arbitration agreement in the first place” (see paragraph [15]).

Arbitration Agreement was valid

The Plaintiff had sought to argue that the arbitration agreement was non-enforceable because the substantive law of the underlying agreement applied to the arbitration agreement and it “would make no sense” for a system of rules (SCC) applying to the conduct of an arbitration to govern the substantive obligations of the parties (see paragraph [10]). The Singapore High Court disagreed on the basis that the choice of SCC arbitration, in this case, entailed that parties had chosen Swedish law to govern the arbitration agreement.

On the basis that parties chose SCC (Swedish Chamber of Commerce) arbitration, the Singapore court held that the seat of arbitration was Sweden and that the governing law of the arbitration agreement was Swedish law. Leong AR reasoned at paragraph [17] that:

1. “[I]n the absence of any express clause prescribing a difference place in which the arbitration proceedings will be conducted”, the choice of SCC arbitration meant that the place of arbitration was Sweden.
2. This choice “evinces an objective intention [on the part of the parties to the arbitration clause] to elect the lex arbitri of Sweden as the curial law applicable to the arbitration“.
3. This is because section 46 of the Swedish Arbitration Act 1994 provides that the Act shall apply to arbitration proceedings which take place in Sweden.
4. And “in the absence of factors pointing to the contrary, it follow[ed] naturally that parties have selected Sweden as the seat of the arbitration”. This was especially so given that the Act also provided that in the absence of parties’ agreement on a proper law, an arbitration agreement shall be governed by the law of the country in which the proceedings shall take place.
5. There was no indication in the underlying agreement that parties intended to have any non-Swedish law governing the arbitration agreement.

The Singapore High Court therefore held that the arbitration agreement was valid because there was no submission that the arbitration agreement was invalid under the laws of Sweden.

In the abstract, the High Court’s reasoning as to the proper governing law of the arbitration agreement raises certain issues. First, there is a distinction between the juridical seat (place) of an arbitration and the venue where the hearings of the arbitration takes “place” (Leong AR discusses the case of Braes of Doune Wind Farm v Alfred McAlpine [2008] EWHC 426 (TCC), which provides an excellent example of this distinction – see our discussion here). It is not entirely clear how the Swedish Arbitration Act 1996 draws a distinction between them, especially given translational issues (see the unofficial English translation of the Swedish Arbitration Act 1996 here). Second, there is no real reason to infer that the arbitration is to take place in Sweden notwithstanding that the arbitration clause provides for SCC arbitration. For example, ICC arbitrations are seated (legally and physically) variously in Paris, London, Zurich, Hong Kong and Singapore.

However, an argument could be raised that the substantive governing law clause provides some mooring for the High Court’s reasoning. Insofar as it provided that “This Agreement [was to be] entered into and to be performed entirely within Stockholm”, it could be argued that this too applied to any arbitration conducted under the auspices of the SCC pursuant to the arbitration clause (see also the reasoning of the English High Court in Arsanovia at [22] on substantive governing law clause and express choice of governing law of the arbitration agreement).

Can an agreement or even an arbitration agreement be governed by a system of rules rather than substantive law?

For the sake of argument, the Singapore High Court considered that even if the arbitration agreement was governed by the substantive governing law clause (SCC rules), that would not necessarily prove fatal to the validity of the arbitration agreement. The Singapore High Court was willing to be forgiving on the basis that laypersons (and even lawyers) could not be expected to be “avid students of jurisprudence with an acute philosophical understanding of what “law” is”. As such,

“…in so far as parties’ intentions are to be given effect to, the reference to “law” need not necessarily be read as “law” in the conventional Hartian, Dworkinian or Razian sense. At least in the realm of international arbitration, it is not entirely inconceivable that a dispute over the validity of an arbitration agreement may be resolved by rules of law as opposed to national laws”(see paragraph [18])

In this respect, the High Court pointed out that there are non-national systems of law which arbitrators are entitled to turn to in order to determine a substantive dispute e.g. lex mercatoria or transnational principles of national law. This applied with even greater force to a court’s determination of the validity of an arbitration agreement.

The Singapore High Court also noted the SCC had its own body of jurisprudence such that “there does appear to be a clear system of rules and consistent application principles of the SCC which would guide arbitrants in determining the validity of an arbitration agreement” (see paragraph [20] and [21]). This, the High Court held, “may persuade a court to find that, at least on the prima facie threshold, such an arbitration agreement would be valid, but only for the specific purpose of staying court proceedings, which does not preclude a full jurisdictional challenge before the arbitral tribunal, or a complete review of the question by the enforcement court” (see paragraph [21]).

However, the High Court declined to provide a definitive view. Instead, Leong AR was content to consider that it was theoretically possible for “a substantive body of international law instead of a national law” to govern international arbitration agreements (see paragraph [20]).

Conclusion and Take-aways

Leong AR’s introductory paragraph about so-called “midnight clauses” is worth iterating,

“It is not uncommon that commercial parties omit to include in their contracts an express choice of law governing their international arbitration agreements. These “midnight clauses” may be included in the main contract very late in the day along with other standard terms just before the contract is signed, understandably so as most parties would be enthusiastic about concluding the negotiations on the contractual obligations, while failing to direct their minds to a possible breakdown of the commercial relationship and the attendant specifics of the dispute resolution process. …”

This case is a good example of how an arbitration clause:

1. is a separate agreement from the underlying contract and can be governed by a law that is different from the governing law of the underlying contract; and
2. can be affected by a badly drafted governing law clause.

The arbitration and governing law clauses in FirstLink certainly could have been better drafted. However, we would not go so far to characterize the arbitration clause as an example of a pathological arbitration clause. Pathological or defective arbitration clauses generally involve uncertainty as to whether parties intended to have their disputes referred to arbitration and usually make references to non-existent or wrongly named arbitration institutes and rules (for more examples, see here and here).

Commercial parties would be well advised to remember that an arbitration clause is a jurisdiction clause, albeit one that is much more flexible and responsive to parties’ choice and autonomy than simply picking the court of a particular jurisdiction. Nonetheless, parties should be conscious that such arbitration clauses can be tricky and might give rise to unintended consequences if parties pay insufficient attention to drafting them and ancillary provisions like the substantive governing law clause.

For more information on the topic of drafting arbitration clauses and arbitration agreements, see our Guidance Note on “How to Draft an Effective Arbitration Clause and Arbitration Agreement”.