The doctrine of separability of arbitration agreements recognises that an arbitration clause contained in a broader agreement is separate and valid despite the invalidity of the rest of the agreement. The doctrine also raises a fundamental question: what is the governing law of the separable arbitration agreement as compared to the remainder of the contract in which it is found?
This question has vexed commentators and practitioners alike for a number of years because there is a myriad of different views. The Supreme Court of the United Kingdom gave some much needed guidance in its recent decision of Enka v Chubb (see summary here). The Court determined that the parties’ explicit choice of law in respect of the overall contract also applies to the arbitration agreement. However, the Court stated that in the absence of an explicit choice, there is a presumption in favour of the law of the seat of the arbitration, being the law that has the “closest connection” to the arbitration agreement. This approach has not been adopted by courts in other jurisdictions, such as in Singapore (see here).
The diversity of approaches to the governing law of the arbitration agreement may, depending on the enforcing court, lead to inconsistent outcomes around the world. This could also result in an outcome that is contrary to the parties’ intention when selecting the governing law of their overall contract.
While the doctrine of separability and the question of the applicable law governing the arbitration agreement may seem more theoretical than practical, determining with precision the proper governing law of the arbitration agreement is important. This is not only to ensure the proper interpretation of arbitration agreements, but also to assess the substantive validity of arbitration agreements and the recognition of arbitral agreements.
Article II(1) of the New York Convention obliges national courts to:
“… recognize an agreement in writing under which the parties undertake to submit to arbitration all or any differences which have arisen or which may arise between them in respect of a defined legal relationship, whether contractual or not, concerning a subject matter capable of settlement by arbitration”.
Article II(3) of the New York Convention obliges national courts to refer the parties to arbitration, upon request of one of the parties, “unless it finds that the said agreement is null and void, inoperative or incapable of being performed.” Unlike Articles V(1)(a) and V(2)(a) of the New York Convention, Article II provides no guidance as to what law governs the question of the validity of the arbitration agreement or the arbitrability of disputes referred to arbitration under it.
Because of this lacuna, national courts have taken inconsistent and sometimes idiosyncratic approaches to assessing the validity of arbitration agreements and arbitrability of disputes. In some cases, national courts determine these questions by reference to the law governing the arbitration agreement chosen by the parties, and in other cases, national courts ignore the parties’ choice of law and determine these questions by reference to the law of the enforcing court. In this post, we will examine three decisions of the Australian and Chinese courts where these issues have arisen.
The conflicting Australian case law
Article II of the New York Convention is incorporated into Australian law under section 7 of the International Arbitration Act 1974 (Cth) (Arbitration Act). According to section 7, Australian courts are obliged to enforce arbitration agreements and stay court proceedings in favour of the parties’ choice to submit their dispute to arbitration. That obligation is, however, subject to the following exceptions: a court is not obliged to enforce the arbitration agreement if “the arbitration agreement is null and void, inoperative or incapable of being performed” or if “the proceedings [are not] capable of settlement by arbitration.” Like the New York Convention, the Arbitration Act is silent as to the law to be applied by the court in determining whether the arbitration agreement is “null and void, inoperative or incapable of being performed” or whether the dispute is not “capable of settlement by arbitration.” This lack of clarity has given rise to two competing and conflicting lines of authority.
In Recyclers of Australia Pty Ltd v Hettinga Equipment Inc (2000) 100 FCR 420, the Federal Court of Australia considered an agreement governed by Iowa law. The Court determined that:
“[t]he question arising under s 7(2) [of the IAA] is whether the proceeding involves the determination of a “matter” that, under the arbitration clause, is capable of settlement by arbitration. As the arbitration clause and the sale agreement are to be governed, construed and interpreted “under the law of the State of Iowa’”, the issue of whether any of the matters involved in the proceeding are arbitrable under the clause is to be determined in accordance with the law of Iowa” (at para 10).
The reasoning in Recyclers v Hettinga was cited by the Federal Court subsequently in Casaceli v Natuzzi SpA (2012) 292 ALR 143 regarding a contract and arbitration agreement governed by Italian law. In that case, the Court rejected the Applicant’s argument that the question of arbitrability ought to be determined by reference to Australian, not Italian, law (at para 38).
However, in a more recent decision of the Federal Court of Australia in WDR Delaware Corp v Hydrox Holdings Pty Ltd  FCA 1164, the Court stated: “the question of whether a dispute is arbitrable is to be determined by the application of the nation[al court’s] domestic law alone” (at para 125). While the Joint Venture Agreement in issue was governed by New South Wales law, the Court opined that:
“[t]he issue of arbitrability goes beyond the scope of an arbitration agreement. It involves a consideration of the inherent power of a national legal system to determine what issues are capable of being resolved through arbitration. The issue goes beyond the will or the agreement of the parties. The parties cannot agree to submit to arbitration disputes that are not arbitrable” (at para 124).
Given that the three decisions cited above were all made by the same court, there is no clear precedent as a matter of Australian law as to what law a national court should apply in determining the validity of an arbitration agreement and the arbitrability of disputes referred to arbitration under it.
Approach of the Chinese courts
Australian courts have entertained the notion that the questions of validity of the arbitration agreement and arbitrability of disputes are matters to be settled by reference to the governing law of the arbitration agreement. Chinese courts on the other hand have applied PRC law when asked to refuse enforcement of arbitral awards on the basis that the underlying arbitration agreement is invalid (see Article V(1)(a) of the New York Convention).
For example, in the 2014 case of Application for the Recognition and Enforcement of Foreign Arbitral Awards between Beijing Chaolaixinsheng Sports and Leisure Co Ltd and Beijing Suowangzhixin Investment Consulting Co Ltd, the Supreme People’s Court refused enforcement of an award rendered by a tribunal operating under the auspices of the Korean Commercial Arbitration Board (KCAB). It did so because the reference to foreign arbitration under the arbitration agreement was invalid as a matter of PRC law. In that case, a dispute arose between two Chinese entities, one of which was a wholly foreign-owned enterprise (WFOE) registered in Beijing and owned by a Korean national, in respect of a contract for the operation of a golf course in Beijing. The key issue in the case was whether the status of one of the parties as a WFOE was sufficient to make the agreement ‘foreign-related’ and the reference to foreign arbitration valid as a matter of PRC law. Despite the parties’ choice for KCAB arbitration in the contract, the Supreme People’s Court refused enforcement of the resulting award on the basis that a purely domestic dispute could not be referred to foreign arbitration. The Court concluded that “the applicable law of the underlying contract and its arbitration clause, whether explicitly or [implicitly] agreed by the parties, shall be deemed as PRC law”, and so the question of the validity of the arbitration agreement was to be determined by reference to PRC law (see more detailed summary of the case here).
The lack of clarity in Article II of the New York Convention has certainly given rise to some confusion among national courts. This has led to inconsistent case law regarding the law applicable to determining the validity of the arbitration agreement and the arbitrability of disputes referred to arbitration under it.
As discussed above, there are cases in both Australia and China where the courts have applied local law in determining the validity and arbitrability of disputes arising under contracts and arbitration agreements that, following a choice of law analysis, would not be governed by that local law. It is questionable whether the application of the local law of a recognising or enforcing court to determine the validity of an arbitration agreement is the preferred approach. This may lead to inconsistent outcomes across enforcing courts around the world, potentially undercutting the objectives of the New York Convention to achieve consistency and uniformity in recognising and enforcing arbitral agreements and awards.
Moreover, application of the local law of the enforcing court is likely inconsistent with the intention of the parties to the arbitration agreement. This is because parties would most likely only turn their minds to whether the arbitration agreement is valid according to the law they chose to govern their overall contract, rather than to the various national laws of the innumerable jurisdictions where recognition of the arbitration agreement or enforcement of the award may be sought.