Dear ODG,
A decision released this week from the Ontario Court of Appeal will interest those of you following issues of unconscionability, boilerplate, and employee/consumer protection. Heller v Uber Technologies is a class action brought on behalf of Uber drivers in Ontario for various employment standards, including minimum wages. The primary issue, if the case ever gets there, will be whether these drivers are "employees" for the purposes of Ontario's Employment Standards Act ("ESA"). Uber sought a preliminary motion to stay these proceedings in favour of arbitration, to be conducted in the Netherlands, as set out in Uber's services agreements. The Court denied the stay, ruling that this arbitration clause was unenforceable for two reasons: 1) the arbitration clause was an attempt to contract out of the ESA's enforcement provisions; and, 2) the arbitration clause was unconscionable anyway. Together with last year's Supreme Court decision, Douez v Facebook, our courts are finally coming to grips with the problems of employee/consumer protection under digital boilerplate. I share a few thoughts on the direction unconscionability is taking in this context.
The Court acknowledged that the state of unconscionability in Ontario is confused--both a four-factor and a two-factor test have some currency, though without one's spectacles they're awfully similar. The resulting analysis had the typical failing of common law unconscionabilty decisions--since no one knows what either inequality of bargaining power or substantive unfairness mean, judges just call it like they see it. The Court was animated by the fact that Uber's proposed arbitration method included various "filing" and "administrative fees" so that "the cost of initiating the arbitration process alone is US$14,500", not including the costs of travel or counsel (para 59). The drivers' incomes are generally in the range of CA$400-600/wk. It's difficult to see this as anything other than an attempt by Uber to cost its drivers out of a dispute resolution mechanism, when there are near-free tax-funded options available in various closer jurisdictions. And so the Court found that the arbitration clause was grossly unfair.
In the absence of a clear unconscionability doctrine, some judges seem to douse about with their "common sense" until they find an injustice, while others plug their ears and declare that there is no evidence of such a thing. Compare, at para 68 of Heller: "it should be self-evident that Uber is much better positioned to incur the costs associated with the arbitration procedure that it has chosen and imposed on its drivers"; with the dissent in Douez at para 164: "there is no evidence regarding the 'relative convenience and expense of trial' in California as compared to British Columbia". Similarly, in the Superior Court decision that Heller just overturned, the judge simply ignored that the agreement provided for no independent dispute resolution mechanism with a practical price tag.
That there was an inequality of bargaining power was apparently conceded by Uber, costing the court the opportunity to say what sets this employer-employee situation apart from every other.
Regardless of the necessarily confused application of a confused doctrine, the Court takes unconscionability in an important direction. As lawyers beating the boilerplate drum have been saying for years, these are contractual clauses that take on public import by affecting significant civil rights on a mass scale. Last year, Douez acknowledged the importance of public policy but arguably limited its scope to forum selection clauses affecting privacy rights. Here, at para 63, the court saw "no reason in principle why the same approach ought not to be taken to the Arbitration Clause in this case", so that "the Arbitration Clause should be subject to a broader analysis when it comes to the issue of validity, especially in a situation where it is part of a contract of adhesion". Arbitration clauses are like forum selection clauses--and the court also notes that employees are like consumers, at para 71. What a "broader analysis" entails is never quite clear, but it includes an account of the public importance of employee protection. The logic of the broader approach to unconscionability lurking in Douez is filling its proper space. (A biased observer might even see this extension of the two-stage Pompey test as a vindication of my argument in "When Is a Contract Not a Contract?: Douez v Facebook Inc. and Boilerplate", (2018) 60:3 Canadian Business Law Journal 428.)
The other successful ground of appeal, the finding that the arbitration clause was an attempt to contract out of the ESA, turned on a technical and rather vigorous interpretation of the ESA in conjunction with arbitration legislation. The Court made two important steps here that could change the landscape of employment protection in Ontario: first, that on a preliminary motion the court should presume that the plaintiffs will be able to prove they are employees. Second, that the enforcement provisions of the ESA are an "employment standard" under the ESA, and therefore cannot be contracted out of. This gives such plaintiffs the benefits of various procedural protections at an early stage. In Ontario, forum selection clauses, choice of law clauses, and of course arbitration clauses, all might now be litigated in court on the presumption that plaintiffs are employees, rather than being sent on unproductive holidays to exotic jurisdictions at the first blush. Here too the Court, at para 50, was influenced by the expansive language of Douez, concluding thus: