This article was first published on the Arbitration Matters blog, here.
In Petty v Niantic Inc., 2022 BCSC 1077, Justice Mayer stayed a proposed class action in favour of arbitration, except in respect of claims advanced under B.C.’s Business Practices and Consumer Protection Act [BPCPA]. He rejected the Representative Plaintiffs’ arguments that the arbitration agreements were null and void for unconscionability and/or violating B.C. public policy. He also applied the competence-competence principle, holding the arbitral tribunal should decide first as to its jurisdiction over claims based on the Competition Act where the parties’ contract provided for California law.
Background – The Defendants, Niantic Inc. (“Niantic”), Warner Bros Entertainment Inc., Warner Bros Entertainment Canada Inc. and Warner Bros Home Entertainment Inc. (collectively “Warner Bros”), are companies that, among other things, produce and commercialize videogames. Niantic published the popular Pokémon Go and Harry Potter: Wizards Unite games, the latter of which it co-produced with the Warner Bros Defendants. The two Representative Plaintiffs, one residing in B.C. and the other in Alberta, were the Defendants’ customers.
The proposed class action relates to so-called “loot boxes”. These are in-game purchasable items or awards, some of which have resale value. Because a player does not know what “loot” is in the box before buying, the Plaintiffs alleged these are “an unlicensed, illegal gaming system under Canadian law”. Their proposed class action seeks relief for unjust enrichment and damages for breaches of the Competition Act, the BPCPA, the Alberta Consumer Protection Act [ACPA], and the B.C. Infants Act. As far as quantum, one of the representative Plaintiffs alleges she spent $450 on loot boxes. The other alleges he spent a little more than $2,000.
The Stay Application – The Defendants applied for a stay under section 8 of B.C.’s International Commercial Arbitration Act [ICAA] on the basis that the terms of service binding the representative Plaintiffs contained a valid and applicable arbitration agreement. However, the Defendants acknowledged that the claim under section 172 of the BPCPA would be exempt from the stay (pursuant to the Supreme Court of Canada’s holding in Seidel v. TELUS Communications Inc, 2011 SCC 15) and tailored their application accordingly.
The terms of service provided for arbitration under the American Arbitration Association’s Commercial Arbitration Rules and were governed by California law (without reference to its conflict of law rules). The dispute resolution clause was detailed and contained several interesting features:
For the purpose of the stay application, the Plaintiffs acknowledged they entered into the terms of service, including the dispute resolution clause. They nevertheless submitted the arbitration clause was unenforceable.
The Decision – Justice Mayer first concluded that the ICAA applied because the parties had their places of business in different states. He also summarily decided the contractual relationship between the parties was “commercial” such that the ICAA governed.
Based on the Plaintiffs’ submissions, there were three issues for decision:
On the first question, Justice Mayer found the ACPA claim should be stayed.
Subsection 16(1) of the ACPA provides a general rule that a supplier cannot enforce an arbitration agreement in a consumer contract. Subsection 16(3) lists two exceptions: 1) where the parties agreed to submit the dispute to arbitration after it arose (paragraph 16(3)(a)); and 2) where the arbitration agreement allows the consumer to decide, once a dispute has arisen, to pursue an “action in court” rather than arbitrate (paragraph 16(3)(b)).
Justice Mayer concluded paragraph 16(3)(b) applied. He rejected the Plaintiffs’ argument that the opt-out clause was illusory for limiting recourse to small claims court, which the Plaintiffs said would effectively deny access to justice. Justice Mayer allowed that if one were to read the words “action in court” in paragraph 16(3)(b) as preserving the right to bring an action in any Alberta court, the dispute resolution clause would not fall within the exception (because it limited recourse to the small claims court). However, he was unable to reach that conclusion without expert evidence on section 16 ACPA’s interpretation. The Plaintiffs did not tender any. Given the arbitration agreement prima facie applied, the burden was on them to do so. Finally, on this point, Justice Mayer was not persuaded that limiting the Plaintiffs’ choice to small claims court or arbitration prejudiced access to justice.
On the second question, Justice Mayer found that the arbitration clause was not null and void or inoperative, on either unconscionability or public policy grounds.
He conducted a detailed unconscionability analysis as the majority of the Supreme Court of Canada instructed in Uber Technologies Inc. v. Heller, 2020 SCC 16 [Uber]. Per the analysis proposed in Brown J.’s concurring opinion in Uber, Justice Mayer also considered whether the arbitration agreement was void on public policy grounds.
Dealing first with unconscionability, Justice Mayer set out its two requirements: 1) inequality of bargaining power; and 2) an improvident bargain.
In arguing for an inequality of bargaining power, the Plaintiffs relied heavily on the Supreme Court of Canada’s decision inDouez v. Facebook, Inc. , 2017 SCC 33 [Douez]. In Douez, the Court held that a forum selection clause in Facebook’s terms of service was unconscionable. Justice Mayer distinguished Douez based on the difference between Facebook’s services and those of the Defendants:
“ I find noteworthy the Supreme Court’s comments that unlike a standard retail transaction there were few comparable alternatives to Facebook, a social networking platform with extensive reach and
that a party must accept the company’s terms or choose not to participate in its “ubiquitous social network”: Douez at para. 56. This observation was part of the context that the Supreme Court considered in determining the inequality in bargaining party between Ms. Douez and Facebook: at paras. 56–57.
 Unlike the nature of the service at issue in Douez (communication and social networking) and
Uber (an employment relationship), there is no evidence that use of the games Pokémon Go and Harry Potter: Wizards Unite, or the ability to purchase loot boxes within those games, are important elements of everyday life which make the plaintiffs particularly dependant or vulnerable in terms of their need to access the game platforms. The games themselves are free and the user has the choice whether to purchase loot boxes.”
He also refused to find that the Plaintiffs could not understand the arbitration agreement; he held the dispute resolution clause at issue, specifically the arbitration opt-out provision, was much clearer than the arbitration agreement in Uber. In that case, the Court found Mr. Heller would not have been able to “appreciate the financial and legal implications of agreeing to arbitrate under the ICC Rules or under Dutch law”.
On the improvident bargain prong of the unconscionability test, the Plaintiffs listed nine reasons why the arbitration agreement effectively precluded access to justice (see paragraph 65). Among these were: 1) the fact that the terms of service were a contract of adhesion the Defendants drafted for their benefit; 2) the arbitration agreement attempts to contract out of the Competition Act (something Justice Mayer dealt with separately); and 3) there is no right to discovery where the claim does not exceed $10,000. Justice Mayer disagreed.
He recalled Uber’s holding that an agreement is improvident where “the potential for undue advantage or undue disadvantage created by the inequality of bargaining power has been realized” (Uber, paragraph 74). He noted that not all advantages and disadvantages are “undue”, and not all adhesion contracts are unconscionable. Of particular relevance was the fact that the Defendants would cover all administrative and arbitrator fees as well as reimburse the customer’s legal fees if he or she prevails in the arbitration. Justice Mayer also made reference to the expedited procedure mandated in the arbitration agreement for claims under $10,000. The Plaintiffs failed to provide any evidence that a lack of discovery would prejudice them.
Justice Mayer also addressed the argument that the arbitration agreement contained a class action waiver, which the B.C. Court of Appeal found unconscionable in Pearce v. 4 Pillars Consulting Group Inc., 2021 BCCA 198 [Pearce]. Justice Mayer did not take this decision to mean class action waivers are automatically unconscionable. He also noted that the majority of the Supreme Court of Canada in Seidel v. TELUS Communications, 2011 SCC 15 declined to determine whether class action waivers writ-large are unconscionable.
Turning to the public policy submission, Justice Mayer referred back to his reasons for concluding the arbitration agreement was not unconscionable. Unlike the clause in Uber, the provision in the terms of service did not present an “insurmountable economic or procedural barrier to the plaintiffs”. He referred again to the opt-out, the parties’ ability to pursue either arbitration or small claims litigation in their home jurisdiction, the Defendants’ legal fee reimbursement and payment for the arbitrator and administrative fees, and the fact that the AAA rules provided for a timely decision.
On the third question—whether the arbitral tribunal could decide the Competition Act claims—Justice Mayer applied the competence-competence principle and deferred the question to the arbitral tribunal.
The parties put forward conflicting expert evidence on whether an arbitral tribunal bound to apply California law (the law applicable to the terms of service) would apply to the Competition Act. Justice Mayer referenced the prima facie approach to competence-competence espoused in Dell Computer Corp. v. Union des consommateurs, 2007 SCC 34. Since the arbitral tribunal’s jurisdiction to apply the Competition Act in the circumstances was a question of mixed fact and law requiring more than a superficial review of the record, Justice Mayer found the question was best left for arbitration. Justice Mayer also cited Williams v Amazon.com, Inc., 2020 BCSC 300 for the proposition that “the prospect that an arbitrator may lack jurisdiction to award damages under s. 36 of the Competition Act is not a ground for finding the arbitration clause at issue to be void, inoperative or incapable of performance”.
In the result, all but the claims under the BPCPA were stayed in favour of arbitration.
First, the dispute resolution clause in this case differed drastically from that at issue in Uber. By creating an arbitration opt-out, eliminating the administrative costs for small-dollar disputes, explicitly providing customers the ability to arbitrate or got to small claims court “at home”, and allowing asymmetrical recovery of legal fees (in favour of the weaker party), the drafters of the Niantic arbitration agreement were careful to avoid the unconscionability hazard Uber constructed. Although forcing arbitration of consumer disputes in lieu of class proceedings remains a controversial topic in Canada and beyond, arbitration agreements of this sort represent meaningful attempts to preserve consent to arbitrate without creating the sort of practical barrier to dispute resolution crying out for an unconscionability (or public policy breach) finding.
Second, the opt-out provision itself is interesting. On one view, a clause like this looks good but achieves little. After all, people rarely, if ever, read terms of service for online/digital products they purchase or interact with. Although a court might be reticent to take judicial notice of that fact, who among us can say they read every provision of every click-wrap or browse-wrap agreement that binds them?
On another view, this is no answer at all. The law remains that these forms of contract are, in principle, binding. That being so, a provision allowing a customer to opt out of arbitration altogether if done within a reasonable time is a powerful argument against unconscionability. A consumer who considers arbitration onerous or inappropriate has a unilateral right to opt out of it and sue in court. One could argue that although consumers are weaker parties, that weakness does not absolve them of the responsibility to read (as long as the relevant terms are written clearly and in simple language). This is consistent with the Uber majority’s focus on the fact that, in that case, Mr. Heller could not have been expected to understand the financial implications of arbitration, even if he read carefully (Uber, paragraph 93). In contrast, the arbitration opt-out provision in the terms of service in this case is clear. The customer is able to know what he or she is getting into and, more importantly, how they can get out of it at their discretion.
Third, whether consumer contracts are “commercial” for the purpose of article 1 of the Model Law is subject to debate. A similar question arose in Uber as to whether an employment contract is “commercial”. Notably, the Analytical Commentary to the Model Law, a recognized and important instrument for its interpretation, states that “labour or employment disputes and ordinary consumer claims” are not commercial “despite their relation to business”.
However, the Model Law is just that, a model. Justice Mayer summarily decided the ICAA applied because of an addition the B.C. Legislature included in its 2018 amendments to the statute. Subsection 1(6) ICAA provides a non-exhaustive list of contracts deemed “commercial” for the purpose the ICAA. Two of the enumerated categories are “a trade transaction for the supply or exchange of goods or services” and “licensing”. With respect to “trade transactions”, there is some authority that this excludes consumer contracts (see: Gary Born, International Commercial Arbitration, vol. I, International Arbitration Agreements (2nd ed. 2014), at p. 309, as cited inUber at paragraph 27. It is also awkward to think of consumers as having a “place of business” in reference to a consumer transaction, lending further support to the view that the ICAA should not apply.
The reasons are unclear on whether the items contained in a loot box are sold or licensed (and it is unclear whether this was on the record), although Justice Mayer does say they were “purchased”. Given loot boxes are arguably digital goods, the transaction between Niantic and a player is either a sale or a license. Based on the foregoing, if the transaction is a sale, the ICAA’s application is doubtful at best. But if the transaction is a license, the ICAA’s application seems clear. Justice Mayer’s reasons do not go into detail on this point, nor do they indicate whether this was a contentious issue (or whether the parties raised it at all). If not, future jurisprudence should do so.
Fourth, another important modification the B.C. Legislature made to article 1 of the Model Law is found in subsection 1(8) ICAA. That provision reads: “[t] his Act does not affect any other law in force in British Columbia by virtue of which certain disputes may not be submitted to arbitration or may be submitted to arbitration only in accordance with provisions other than those of this Act”.
Curiously, subsection 1(8) is limited to “any other law in force in British Columbia”. This could be read to exclude other provincial legislation restricting the enforceability of arbitration agreements, like Ontario’sConsumer Protection Act, 2002, should a defendant seek to stay a B.C. proceeding involving a claim under that statute. This might have been relevant had the proposed class included a sub-class of Ontario residents (the Ontario act does not contain the same exception in the ACPA that applied here). However, subsection 1(8)’s drafting does not necessarily create an implied exclusion—that the ICAA does “affect” (i.e., render ineffective) non-B.C. provincial legislation restricting recourse to arbitration. Rather, one could easily read subsection 1(8) as a “for greater certainty” provision, in which case other provincial legislation restricting arbitration would remain unaffected. However, this ambiguity might merit judicial comment.
Fifth, Justice Mayer appears to have accepted the Plaintiffs’ contention that, if the arbitration agreement stands, a customer could never pursue a class proceeding. Given the drafting, it is unclear whether this is correct. The relevant clauses read as follows:
“If you live in the US or another jurisdiction which allows you to agree to arbitration, you and Niantic agree that any disputes will be settled by binding arbitration, except that each party retains the right: (a) to bring an individual action in small claims court and (b) to seek injunctive or other equitable relief in a court of competent jurisdiction to prevent the actual or threatened infringement, misappropriation, or violation of a party’s copyrights, trademarks, trade secrets, patents or other intellectual property rights.
Without limiting the preceding paragraph, you will also have the right to litigate any other dispute if you provide Niantic with written notice of your desire to do so… within thirty (30) days following the date you first accept these Terms (Such notice, an “Arbitration Opt-out Notice”).” (Emphasis added)
If a customer who delivered an Arbitration Opt-out Notice within 30 days could not bring a class proceeding, the emphasized paragraph would have no meaning at all. This is because the first paragraph preserves the right to commence a small claims court action at any time, and the second paragraph begins by saying “[w]ithout limiting the preceding paragraph…”. Though not the only interpretation, the arbitration agreement, read as a whole, appears to say a customer may: 1) opt out of arbitration altogether within 30 days and bring a claim in any court; or 2) bring a small claims action at any time, regardless of whether they delivered an Arbitration Opt-out Notice (i.e., “without limiting the preceding paragraph”). Again, it is unclear whether Justice Mayer had the benefit of submissions to this effect.
Sixth, Justice Brown’s concurring opinion in Uber in which he would have found an arbitration agreement void, not for unconscionability, but for breach of public policy, has taken root in B.C. In addition to this case, the B.C. Court of Appeal considered and analyzed a class action waiver using a public policy lens inPearce. The Court of Appeal in that case found the waiver provision was void, both on the basis of unconscionability and public policy. For an earlier Case Note on Pearce, see: B.C. – unlike agreement to arbitrate, class action waiver not effective to resist class action certification – #241. For a Case Note on the Uber case, see Supreme Court – courts should not refer jurisdiction challenge to arbitrator if real prospect that challenge might never be resolved – #344.
In this episode of the Arbitration Conversation Amy interviews Stephen J. Ware, the Frank Edwards Tyler Distinguished Professor of Law at the Kansas University School of Law. https://youtu.be/efCM0NMkpYcBy Stephen Ware, Amy Schmitz
This article first appeared on the Securities Arbitration Alert (SAA) blog, here. A CFPB taskforce on consumer financial law issued a massive report in early January, with hardly a reference...By George Friedman
I'm very excited to share the news that we've just posted a new course on Arbitrate University from the legendary arbitrators (and arbitration professors) Karl Bayer and Tracy McCormack. There...By Colin Rule