For today’s Vavilov at 5 entry, I am delighted to present a post from my friend and co-author James Plotkin. James is a Partner at Gowling WLG, and is the co-author of Arbitration Legislation in Ontario: A Commentary. His excellent post focuses on the effect of Vavilov on the world of private arbitration. Enjoy!
Vavilov at 5 – Vavilov’s Impact on Private Arbitration
Anyone reading this knows all about the seismic shift Vavilov brought to administrative law. But its aftershocks have rippled through other areas over the last five years. As far as I can tell, domestic arbitration is among the most affected of all. Having enjoyed each learned installment in this series, I am honoured that Professor Mancini let me take us slightly off track into the world of private arbitration.
Since Vavilov, lower courts have wrestled with whether its holding on appeal clauses altered the standard of review on appeals from arbitral awards. Five years before Vavilov, the Supreme Court in Sattva Capital Corp. v. Creston Moly Corp. held that reasonableness applies to appeals on questions of law.[1] It directly ported administrative law rationales for deference from Dunsmuir. The Court then doubled down on Sattva a few years later in Teal Cedar Products Ltd. v. British Columbia.[2]
As far as judicial climate, Sattva came out a few years after Alberta Teachers and Newfoundland Nurses, and two years before Edmonton East, arguably the highwater mark of curial deference in judicial review of administrative action. Teal Cedar and Edmonton East were released a mere five months apart. One could speculate that this might have supported the Court’s seemingly reflexive move toward deference in Sattva and Teal Cedar.
But Vavilov rocked the boat, with its renewed focus on legislative intent and the notion that when the legislature says “appeal,” it intends for an appellate standard of review to apply. Dozens of lower court decisions have since gone in both directions on whether Vavilov displaced Sattva such that the appellate standard of review now applies to appeals from arbitral awards. By my informal count, the majority say “yes”.[3] So do the only intermediate appellate court decisions on point,[4] and a concurring group of three Supreme Court judges in Wastech Services Ltd. v. Greater Vancouver Sewerage and Drainage District.[5] Unfortunately, despite at least two leave applications that have attempted to put this issue before the Supreme Court, it has not yet taken up the opportunity to clarify this discord in the jurisprudence.[6]
About four-and-a-half years ago, Professor Mancini (he was just Mark back then) and I authored the audaciously titled “Inspired by Vavilov, Made for Arbitration: Why the Appellate Standard of Review Framework Should Apply to Appeals from Arbitral Awards”. “Why audacious?”, you might ask.
Those in the arbitration world, in Canada and beyond, generally want to both be and appear “pro-arbitration”. For most, this means minimizing court intervention in arbitral proceedings. Mark and I suggested that although this view may make sense in some respects, it is “low resolution”, to the point of being wrong, where legislation permits parties to contract for appeals from arbitral awards, and where the parties take the legislature up on it.
In fairness to the minimal curial intervention crowd, their view makes much more sense in a legislative context that bars appeals. Vavilov’s point about “institutional design choice” directly validates this. That is the case with international arbitration legislation across Canada. The federal government and all provinces and territories have adopted the UNCITRAL Model Law on International Commercial Arbitration, with or without modification. The Model Law, as adopted in Canadian legislation, bars appeals on the merits. It provides a limited set of grounds for setting aside or refusing to recognize and enforce and award. These grounds do not go to the award’s merits per se. They address procedural matters (e.g., jurisdiction, procedural fairness or compliance with the parties’ chosen process), the arbitration agreement’s validity and whether the award deals with non-arbitrable subject matter or offends the jurisdiction’s public policy.
But most of the domestic arbitration statutes in Canada are different. In all but Quebec and Newfoundland and Labrador, the legislation expressly permits appeals.[7] They range on: 1) whether the parties can have a full right of appeal or only an appeal on questions of law; 2) whether the parties have to opt-in or opt-out of appeals in their arbitration agreements; and 3) leave requirements. What they all share, however, is a legislative design choice to permit the parties an opportunity to give the court the last word on the merits of the award.
The argument Mark and I advance in the paper is one from legislative design choice, but also arbitration law’s guiding principle: party autonomy. At base, arbitration is all about the parties’ consent to consign their dispute resolution process to a private tribunal. This is generally, and quite rightly, viewed as choosing to take the dispute out of the public court system. The theme of respecting the parties’ agreement to arbitrate courses through the arbitration jurisprudence.[8] It tends to roar loudest in cases interpreting and applying the legislation’s robust stay provisions. These support arbitration by halting court proceedings over disputes subject to arbitration clauses, with narrow exceptions.
In our article, Mark and I agree that respect for the parties’ agreement to arbitrate is key. We just say that the court should respect the entire agreement to arbitrate. When the parties bargain for an appeal, this means honouring that choice.
Parties who agree to arbitration with a right of appeal send a message: we want to push the courts away in the first instance and allow the arbitral tribunal to do its work. At the same time, we want the ability (but not the obligation) to take a “judicial off-ramp” and pull the courts in on the back end. The statutes allow the parties to sculpt their dispute resolution process this way. That legislative intent, and party intent, must be respected.
I pause to note that it should be lost on no one, especially this crowd, how remarkable a legislative design choice this is. The court’s jurisdiction is a pure question of law on which party agreement is irrelevant.[9] Except here. Domestic arbitration statues are the only example I know of where the legislature effectively delegates to private persons the ability to give (in the case of an opt-in jurisdiction) or take away (in the case of an opt-out jurisdiction) the court’s appellate jurisdiction.[10] This is how strongly party autonomy pervades these legislative schemes.
The question then becomes whether the word “appeal” in an arbitration statute (and an arbitration agreement) means “appeal”, importing the appellate standard of review framework. We say the argument from party autonomy points to “yes”. Recall the Court in Vavilov’s note that “there is no convincing reason to presume that legislatures mean something entirely different when they use the word “appeal” in an administrative law statute than they do in, for example, a criminal or commercial law context.”[11] Far from rebutting the presumption of consistent expression, the arbitration context supports it for the reasons outlined above and unpacked in our article.
The main arguments against our position are those from finality and expertise. These formed the basis for the Supreme Court’s decision in Sattva. In that case, the Court considered the then-in-force B.C. domestic arbitration statute. It held that reasonable applies to appeals from arbitral awards on questions of law.[12] The reasoning expressed in Sattva and its successor, Teal Cedar, is that parties who choose arbitration favour finality and efficiency, which is consistent with a deferential standard of review.[13] Likewise, arbitrators are presumptively chosen for their expertise in the subject-matter of the parties’ dispute.[14] Sattva and Teal Cedar have been uncritically applied to appeals arising under domestic arbitration statute across Canada, despite the differences in how they deal with appeals—and by extension how they weigh the (sometimes competing) values of finality, party autonomy and judicial oversight.[15]
Mark and I offer a number of responses. I will not detail them all here, but the main headline is this: parties who contract for an appeal are obviously more concerned with having a judicial second look than finality, and likewise intended to subject any actual or presumed expertise to judicial scrutiny. This is a totally valid articulation of party autonomy. In other words, to say parties who build an appeal into their arbitration agreement favour finality over court intervention requires you to disbelieve your lying eyes. This is especially so when parties opt into an appeal (or opt to have the appeal as of right rather than with leave of the Court).[16]
Incidentally, Sattva kind of gets hoisted by its own petard on this point. Although it dealt (very briefly) with the arbitral standard of review, most people know it as the Court’s controlling case on contract interpretation. Sattva directs courts (and tribunals) to “read the contract as a whole, giving the words used their ordinary and grammatical meaning, consistent with the surrounding circumstances known to the parties at the time of formation of the contract”.[17] In this interpretation matrix, the surrounding circumstances, though part of the analysis, “must never be allowed to overwhelm the words” of the contract.[18]
The ordinary and grammatical meaning of the word appeal entails correctness on questions of law and palpable and overriding error on questions of fact and inextricable questions of mixed fact/law; it does not mean reasonableness review. From a textual perspective, the arbitration statutes (and arbitration agreements made under them) tend to use the nomenclature and taxonomy of appeals, dividing them into questions of law, mixed fact/law and fact.[19] This is a clear signal that the standard of review framework that divides appeal issues into those categories should apply. I have also mused elsewhere that Vavilovian reasonableness review maps poorly onto appeals from arbitral tribunals on pure questions of general law, which is what most arbitration appeals are about. But that is a topic for another day. Suffice to say that none of these arguments appear to have been before the Court in Sattva or Teal Cedar.
These are the main reason we say Vavilov displaced Sattva on the arbitral standard of review issue, even though Vavilov is an administrative law case that does not mention arbitration. There are others, but you will have to read the article for the rest!
Vavilov has proven to be a gamechanger, in administrative law and arbitration law. As someone who practices in both, it awoke me[20] from the very dogmatic slumber that led to Sattva’s entrenchment of the reasonableness standard. Among his many insightful comments in his contribution to this series, one of Justice Stratas’s observations in particular struck me:
“Vavilov is ideologically neutral and neither favours nor restrains the administrative state. If a government wants plenty of powerful administrative tribunals doing important things and legislates to that effect, Vavilov will enforce that legislative will. Similarly, if a government wants to carve back the administrative state and legislates to that effect, Vavilov will enforce that legislative will too. If a government wants to adjust the standard of review, Vavilov will not stand in the way.” (Emphasis added)
Likewise, ideology should not stand in the way of Vavilov applying to appeals from arbitral awards. Justice Stratas’s observation applies mutatis mutandis. The legislatures have spoken. “Appeal” means appeal, and nothing about the arbitration context should make it otherwise!
[1] 2014 SCC 53.
[2] 2017 SCC 32.
[3] See for example: Burwell et al v. Wozniak, 2024 ONSC 5851; Broadband Communications North Inc. v. 6901001 Manitoba Ltd., 2021 MBQB 25; Clark v. Unterschultz, 2020 ABQB 338; Kumer v. MTCC No. 775, 2021 ONSC 1181; Buffalo Point First Nation et al. v. Cottage Owners Association, 2020 MBQB 20. For cases taking the opposite view, see for example: Ontario First Nations (2008) Limited Partnership v. Ontario Lottery And Gaming Corporation, 2020 ONSC 1516; Cove Contracting Ltd. v. Condominium Corporation No. 012 5598 (Ravine Park), 2020 ABQB 106.
[4] Northland Utilities (NWT) Limited v. Hay River (Town of), 2021 NWTCA 1; Travelers Insurance Company of Canada v. CAA Insurance Company, 2020 ONCA 382.
[5] 2021 SCC 7 (per Brown and Rowe JJ. (Côté J. concurring).
[6] March of Dimes Canada v. Escape 101 Ventures Inc. (SCC Docket 40439); Tall Ships Landing Development Inc. v. Corporation of the City of Brockville (SCC Docket 40611).
[7] There is case law under the Newfoundland and Labrador statute suggesting merits review is available on a reasonableness standard, although there is no appeal provision proper.
[8] See for example: TELUS Communications Inc. v. Wellman, 2019 SCC 19, paras. 130-138.
[9] 744185 Ontario Inc. v. Canada, 2020 FCA 1, para. 53, citing, inter alia, Hupacasath First Nation v. Canada (Foreign Affairs and International Trade Canada), 2015 FCA 4, paras. 37-38.
[10] There are exceptions. The statutes in force in the prairies do not allow the parties to contract out of appeals on pure questions of law with leave of the Court. Some of the statutes also remove the ability to contract out of limited appeals in family matters. Also, The opt-out jurisdictions that permit appeals on questions of fact or mixed fact and law require the parties to opt in for those grounds.
[11] Vavilov, para. 44 (emphasis added).
[12] That iteration of the B.C. domestic arbitration statute, as well as the current version, only permits appeals on questions of law, either as of right if provided for in the arbitration agreement or with leave of the court if the parties’ agreement does not exclude appeals. Under the B.C. legislation, appeals on questions of mixed fact/law and fact are never permitted. As noted above, this is not so in other Canadian domestic arbitration statutes.
[13] Teal Cedar, paras. 1, 45 and 74.
[14] Sattva, paras. 102-106.
[15] One decision has spoken to this idea. See: Esfahani v Samimi, 2022 ABKB 795.
[16] The default under most of the domestic arbitration statutes (B.C., Alberta, Saskatchewan, Manitoba, Ontario, Nova Scotia (commercial matters only) and New Brunswick) is appeals on questions of law only, and with leave of the court. From there, the parties can either opt out of appeals entirely (except under the Albera, Saskatchewan and Manitoba statutes), remove the leave requirement for appeals on questions of law and also agree to appeals on questions of fact or mixed fact/law (except for B.C.). P.E.I., and the statutes in the Territories are all opt-in. P.E.I. permits opt in on questions of law only, and with leave of the court. The legislation in the Territories allows appeals up to and including a full right of appeal.
[17] Sattva, para. 47.
[18] Sattva, para. 57.
[19] See for example: Arbitration Act, 1991, SO 1991, c 17, s. 45.
[20] Admittedly with some gentle Socratic prodding from my brilliant mentor, Alyssa Tomkins.