In the legal world, countless hours are spent choosing, weighing and defining words, expressions and phrases, and linguistic precision is instilled as a virtue from the first day of law school. It is therefore rather concerning when courts use both inaccurate and inapt language to address a specific issue, such as when Canadian courts consider arbitration matters using terms and concepts plucked from administrative law. The latest illustration of this problem is found in the recent decision by the Court of Appeal for British Columbia (BCCA) in lululemon (lululemon athletica canada inc. v. Industrial Color Productions Inc., 2021 BCCA 428).
With this decision, the BCCA adds to the growing list of problematic judgments rendered by Canadian courts relating to arbitration in recent years. As I have previously argued with respect to Canadian domestic commercial arbitration and the possibility of appealing domestic awards, the time has come for Canadian courts to truly recognize commercial arbitration as an independent, alternative dispute mechanism that functions according to its own law, rules, and processes.
Standard of review for set aside applications
In lululemon, the BCCA considered the standard of review for an application to set aside an award under the equivalent of Article 34(2)(a)(iii) of the UNCITRAL Model Law, i.e. the award deals with a dispute not contemplated by or not falling within the terms of the submission to arbitration, or contains decisions on matters beyond the scope of the submission to arbitration. The BCCA confirmed that, consistent with international practice in Model Law jurisdictions, courts considering a set aside application based on a jurisdictional challenge do so on a de novo basis (at paras. 42-43).
In getting there, however, the BCCA took some interesting turns. For example, its analysis begins with the statement that Cargill (Mexico v. Cargill, Incorporated, 2011 ONCA 622) “remains the leading case on the standard of review for applications to set aside awards”. That standard of review, notes the BCCA, “is correctness” (at para. 34). It is of note that the BCCA observed that the provision at issue in Cargill, found in the equivalent Ontario statute, is “identical” to the provision at issue in lululemon (at para. 36). Putting aside the fact that Cargill was a case relating to a NAFTA award, it is puzzling that the BCCA would start its analysis by stating that the leading authority stands for the proposition that the standard of review is correctness, but conclude that the applicable standard of review for set aside applications based on an “identical” provision is, in fact, a de novo review.
Indeed, the BCCA first heavily relies on Cargill to explain that (i) both the language of the provision at issue, as well as domestic and international authorities, support a correctness standard (at para. 38), and (ii) a correctness standard is appropriate in light of the restriction on judicial intervention found in the applicable statute and to preserve the autonomy of the international arbitration process (at paras. 39-40). The BCCA then rather inexplicably concludes that, in light of the statutory language at issue and the common international approach that has developed with respect to the review of awards containing jurisdictional rulings (referring here to Gary Born’s treatise), the set aside application in such cases is to be conducted as a de novo hearing before the reviewing judge (at paras. 41-43).
In making its decision on set aside, the BCCA also firmly rejected any notion that it is appropriate for Canadian courts to consider either Sattva (Sattva Capital Corp. v. Creston Moly Corp., 2014 SCC 53) or Vavilov (Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65), a judgment previously discussed on this blog. These decisions would nominally guide determining the standard of review applicable at administrative law. However, as confirmed by the BCCA, they do not apply to an international arbitration award (at para. 44). Indeed, the Court reiterated that Sattva, which deals with the standard of review on appeal for domestic commercial arbitrations, does not address the issue of set aside applications, either domestically or internationally (at para. 45). Vavilov, an administrative law decision, does not even deal with arbitration at all. Simply put, administrative law standards are not to be used to create a standard of review not provided for in the international arbitration context (at para. 46).
Conflating a de novo review with the standard of correctness
As illustrated above, and despite explicitly recognizing that domestic administrative law concepts have no place in the review of a commercial arbitration award, the BCCA nevertheless felt the need to equate the de novo standard with that of correctness, a principle rooted in domestic administrative law (see para. 43). The concepts of de novo and correctness review are inherently different. A de novo review means that parties are not restricted to the record placed before the tribunal, particularly if the court is considering a set aside application based on a jurisdictional challenge, as recently confirmed in Luxtona Limited (The Russian Federation v. Luxtona Limited, 2021 ONSC 460). (A motion seeking leave to appeal this decision, which was previously discussed on this blog, was filed with the Ontario Court of Appeal on July 22, 2021. The hearing was scheduled to take place on January 17, 2022.) Conversely, an administrative decision reviewed on the correctness standard is still (absent exception) only considered in light of the record before the tribunal, meaning that no new evidence may be submitted. A de novo review is a fresh reconsideration of the issue before the court, while a review on a correctness standard aims to determine whether the previous decision-maker came to the right (i.e. “correct”) decision. It follows that these two types of review are fundamentally different and there is no basis for using them interchangeably.
Even more concerning is the fact that the BCCA is not alone in this false equivalency. The Ontario Superior Court’s decision in Luxtona is replete with references to – and reliance on – domestic administrative law concepts to explain why the standard of review in that particular case was de novo, thus allowing for the presentation of fresh evidence by the parties (see e.g. paras. 10, 22). Luxtona relies on the 2011 Court of Appeal for Ontario decision in Cargill [Mexico v. Cargill, Incorporated, 2011 ONCA 622] where, despite noting that importing and directly applying domestic concepts to international arbitration “may not be helpful” (at para. 30), it nevertheless states that one could view the de novo review standard “as a variant of applying the correctness standard” (at para. 38).
The (continuing) case for the emancipation of arbitration
If Canadian courts truly are to embrace commercial arbitration – both domestic and international – as an independent, effective method of dispute resolution, they need to resist the temptation to make it fit into a familiar mold. This includes referring to terms and concepts that are alien to commercial arbitration law. The standards of reasonableness and correctness are of pivotal importance in Canadian administrative law; however, administrative law concepts are not even entirely applicable to the domestic commercial arbitration context. The Supreme Court of Canada itself noted in Sattva that appellate review “takes place under a tightly defined regime specifically tailored to the objectives of commercial arbitrations” (at para. 104). If this is true of domestic commercial arbitration, it is imperative with respect to international arbitration.
If courts continue to approach the review of commercial arbitral awards through the linguistic lens and rhetorical rubric of domestic administrative law, there are bound to be mistakes – whether conscious or not – that will inevitably hamper the development of Canadian arbitral jurisprudence. Moreover, it is decidedly unhelpful for published decisions to make an explicit link between the de novo and correctness standards, let alone reason through administrative law concepts to come to arbitral law conclusions. Such practices will only stem confusion and signal to parties – and their counsel – that they may present arguments from different areas of law in arbitration cases. Arbitration law, particularly international arbitration law, is a sophisticated and well-developed area of law with its own language, concepts and conventions as documented time and again in a highly respected body of academic commentary that has been cited favourably in Canadian jurisprudence.
Different areas of law are governed by unique principles reflected in particular terms and explained with specific language appropriate to the domain. Canadian courts do not employ, for example, family law terms and concepts to resolve commercial contracts cases, such as resorting to equalization payments to determine damages owing in a commercial breach of contract case. While that example may be extreme, the principle remains. A call for clarity deserves itself to be clear: commercial domestic arbitration and international arbitration are not domestic administrative law, and administrative law language and concepts have no place in arbitration decisions.
* The views expressed by the author in this article are her own, and do not necessarily reflect those of Norton Rose Fulbright Canada LLP.
** The author wishes to thank Charles Feldman for his insightful comments, as always.