a hand holding a guitar



Arbitration Parties Beware: The Supreme Court of Canada unpacks the Sattva Test for Appellate Review of Arbitration Awards in Teal Cedar Products Ltd v British Columbia


As litigants continue to experience delayed access to justice occasioned by a backlogged judicial system, Canadian courts continue the struggle to protect the legitimacy of commercial arbitration as an alternative dispute resolution process. The Supreme Court of Canada set out a three-step analysis for appellate review of arbitration awards in Creston Moly Corp v Sattva Capital Corp, 2014 SCC 53 ("Sattva") (i) assess the court's jurisdiction; (ii) determine the standard of review; and (iii) apply the standard of review. As easy as it may appear, applying the Sattva test to arbitral awards is not without challenges. In Teal Cedar Products Ltd v British Columbia, 2017 SCC 32 ("Teal"), the Supreme Court of Canada analyzed the components of the Sattva test, defined the three principal categories of questions to be determined at an appellate review (legal, factual, or mixed questions), and set out a fourth category of questions - extricable questions of law. While Teal provides a guide for the application of the Sattva test, more importantly, it underpins the court's endorsement of the efficiency and finality objectives of commercial arbitrations by reminding parties to arbitration agreements of the very narrow scope of appellate review of arbitration awards.


In 2003, the province of British Columbia ("BC") implemented the Forestry Revitalization Act, SBC 2003, c 17 (the "Act"). As a result of the Act, three licenses owned by Teal Cedar Products Ltd. ("Teal Cedar") were affected negatively. As the areas of land covered by two of the licenses were reduced and all three licenses had a lower quantity of allowable harvest, Teal Cedar suffered significant losses. The Act contained a compensation scheme for the value of improvements made to Crown land (the "Improvements Compensation"). Teal Cedar and BC entered into negotiations to settle the compensation owed under the Act including a Settlement Framework Agreement which provided that no interest would be payable under any compensation BC would provide. These negotiations failed and Teal Cedar and BC subsequently entered into an Amended Agreement to submit the dispute to arbitration.


The issues submitted to arbitration were: the valuation method consistent with the Act (the "Statutory Interpretation/Valuation Issue"), whether BC was liable to pay interest regardless of the Agreement (the "Contractual Interpretation/Interest Issue"), and whether Teal Cedar was entitled to compensation for the unaffected license (the "Statutory Application/License Issue").

The arbitrator accepted the depreciation replacement cost method and held Teal Cedar was entitled to interest regardless of the Settlement Framework Agreement which was a product of the failed negotiations. The arbitrator also held that Teal Cedar was not entitled to improvements compensation for the unaffected license.

Judicial History

On appeal, the British Columbia Supreme Court (2012 BCSC 543) upheld the arbitrator's award except in connection with the Statutory Application Issue, which was remitted to the arbitrator and resulted in an additional award in an amount equal to the value of the improvements. A majority of the British Columbia Court of Appeal (2013 BCCA 326) reversed the judge's decision, finding that the arbitrator had erred on both the Statutory Interpretation and Contractual Interpretation Issues, as well as the additional ruling on the Statutory Application issue. The Supreme Court of Canada remanded the appeal back to the Court of Appeal, upon the release of Sattva, to apply the three-step analysis for appellate review of arbitration awards: (a) whether the appellate court has jurisdiction to review the alleged error; (b) if so, whether the standard for the review is reasonableness or correctness; and (c) whether the arbitration award withstands scrutiny under the applicable standard of review. A unanimous Court of Appeal (2015 BCCA 263) held that its previous decision was unaltered by Sattva, and reaffirmed its conclusion that the issues ruled upon by the arbitrator were questions of law subject to appellate review, and that the arbitrator was in error regardless of the standard of review applied.

The Decision

A majority of the Supreme Court of Canada (Moldaver, Côté, Brown and Rowe JJ. dissenting in part) reversed the Court of Appeal decision in part, in light of Sattva. The Supreme Court defined the three principal categories of questions to be determined at an appellate review (legal, factual, or mixed questions) and set out a fourth category - extricable questions of law. Legal questions are questions about what the correct legal test is. Factual questions are questions about what actually took place between the parties. Mixed questions are questions about whether the facts satisfy the legal test. The Supreme Court held that while the application of a legal test to a set of facts is a mixed question, if in the course of that application the underlying legal test may have been altered, then a legal question arises. In the context of a dispute under the Arbitration Act, such a legal question is open to appellate review, assuming the other jurisdictional requirements of that Act are met. The Supreme Court defined the extricable questions of law as a covert form of legal question where a judge's (or arbitrator's) legal test is implicit to their application of the test rather than explicit in their description of the test.

However, the Supreme Court narrowed the scope of extricable questions of law to preserve finality in commercial arbitration and deference to factual findings. The Supreme Court cautioned courts in identifying extricable questions of law because mixed questions, by definition, involve aspects of law. Courts must be vigilant in distinguishing between an allegation that a legal test may have been altered in the course of its application (an extricable question of law), and an allegation that an unaltered legal test should have, when applied, resulted in a different outcome (a mixed question). This is because of counsel's motivation to strategically frame a mixed question as a legal question in order to gain jurisdiction in appeals from arbitration awards or a favourable standard of review in appeals from civil litigation judgments. The Supreme Court noted that while a question of statutory interpretation is normally characterized as a legal question, contractual interpretation questions do not offer such simple characterization. Contractual interpretation involves factual, legal, and mixed questions, and characterizing the nature of the specific question before the court requires delicate consideration of the narrow issue actually in dispute. In general, contractual interpretation remains a mixed question, not a legal question, as it involves applying contractual law (principles of contract law) to contractual facts (the contract itself and its factual matrix).

Applying the Sattva test, the Supreme Court held that the Statutory Interpretation Issue in this case involved two categories of questions: (a) questions about the broad category of methods that are acceptable under the terms of the Revitalization Act; and (b) questions about the specific or preferable method, within that broad category of acceptable methods, that should ultimately be applied. The first are matters of statutory interpretation and therefore questions of law reviewable by courts. The second are inextricably linked to the evidentiary record at the arbitration hearing, where various experts opined on the virtues of conflicting valuation methodologies, and therefore are mixed questions if not pure questions of fact not reviewable by courts.

The Supreme Court held that courts have no jurisdiction to review the Contractual Interpretation Issue, as the arbitrator was best situated to weigh the factual matrix in his interpretation of the parties' agreement regarding the payment of interest. The fact that the arbitrator may have placed significant weight on that evidence in interpreting the agreement does not engage a legal question conferring jurisdiction on the courts under the Arbitration Act, as that did not alter the underlying test he applied in this case. Further, courts have no jurisdiction to review the Statutory Application Issue, as it engaged whether the arbitrator correctly applied the valuation methodology to a license - a mixed question beyond appellate review.

The Supreme Court reaffirmed the principle in Sattva that the standard of review on legal questions arising from an arbitrator's analysis of statutory interpretation issue is reasonableness, which is almost always the applicable standard when reviewing commercial arbitration awards. According to the Supreme Court, the preference for a reasonableness standard also dovetails with efficiency and finality, which are key objectives of commercial arbitration. It would be an error to claim that all statutory interpretation by an arbitrator demands correctness review simply because it engages a legal question. The Court highlighted the distinction between arbitral awards and a civil litigation judgment, where the nature of the question resolves the standard of review, with factual and mixed questions reviewed for palpable and overriding error and legal questions (including extricable questions of law) reviewed for correctness.

The Supreme Court found that the reasonableness standard was not negated in this case in light of the nature of the question at issue and the arbitrator's presumed expertise. Therefore the Court of Appeal erred in holding that that the standard of review should be correctness for the Statutory Interpretation Issue. While the nature of the question (legal, mixed, or fact) is dispositive of the standard of review in the civil litigation context, it is not in the arbitration context. The Supreme Court found the arbitrator's decision on the question of law reasonable, in that, it fell within a range of possible, acceptable outcomes which were defensible in respect of the facts and law, and the decision was justified, transparent and intelligible.


Arbitration can be one of the quickest and flexible ways parties use to resolve disputes, as an alternative to the delays occasioned by overburdened trial courts. The Supreme Court of Canada has endorsed the efficiency and finality objectives of commercial arbitrations. While arbitration has many benefits, parties should be aware of the very narrow scope of the court's jurisdiction to review arbitral awards before entering into arbitration agreements. Parties should address their minds to the fact that courts have no jurisdiction to review contractual interpretation issues arising in arbitrations, as the arbitrator is best situated to weigh the factual matrix in the interpretation of the parties' agreement, provided the arbitrator did not alter the applicable legal test in the course of application. Courts also have no jurisdiction to review application of legal principles to the context of a specific case. Further, unlike civil litigation judgment where the standard of review is fixed (factual and mixed questions reviewed for reasonableness and legal questions, including extricable questions of law, reviewed for correctness), pure questions of law in arbitral awards are almost always reviewed for reasonableness, absent any negating factor.