Tag Archive for: class action

Uber Arbitration Clause Unconscionable

In 2017 drivers working under contract for Uber in Ontario launched a class action.  They alleged that under Ontario law they were employees entitled to various benefits Uber was not providing.  In response, Uber sought to stay the proceedings on the basis of an arbitration clause in the standard-form contract with each driver.  Under its terms a driver is required to resolve any dispute with Uber through mediation and arbitration in the Netherlands.  The mediation and arbitration process requires up-front administrative and filing fees of US$14,500.  In response, the drivers argued that the arbitration clause was unenforceable.

The Supreme Court of Canada has held in Uber Technologies Inc. v. Heller, 2020 SCC 16 that the arbitration clause is unenforceable, paving the way for the class action to proceed in Ontario.  A majority of seven judges held the clause was unconscionable.  One judge held that unconscionability was not the proper framework for analysis but that the clause was contrary to public policy.  One judge, in dissent, upheld the clause.

A threshold dispute was whether the motion to stay the proceedings was under the Arbitration Act, 1991, S.O. 1991, c. 17 or the International Commercial Arbitration Act, 2017, S.O. 2017, c. 2, Sch. 5.  Eight judges held that as the dispute was fundamentally about labour and employment, the ICAA did not apply and the AA was the relevant statute (see paras. 18-28, 104).  While s. 7(1) of the AA directs the court to stay proceedings in the face of an agreement to arbitration, s. 7(2) is an exception that applies, inter alia, if the arbitration agreement is “invalid”.  That was accordingly the framework for the analysis.  In dissent Justice Cote held that the ICAA was the applicable statute as the relationship was international and commercial in nature (paras. 210-18).

The majority (a decision written by Abella and Rowe JJ) offered two reasons for not leaving the issue of the validity of the clause to the arbitrator.  First, although the issue involved a mixed question of law and fact, the question could be resolved by the court on only a “superficial review” of the record (para. 37).  Second, the court was required to consider “whether there is a real prospect, in the circumstances, that the arbitrator may never decide the merits of the jurisdictional challenge” (para. 45).  If so, the court is to decide the issue.  This is rooted in concerns about access to justice (para. 38).  In the majority’s view, the high fees required to commence the arbitration are a “brick wall” on any pathway to resolution of the drivers’ claims.

The majority then engaged in a detailed discussion of the doctrine of unconscionability.  It requires both “an inequality of bargaining power and a resulting improvident bargain” (para. 65).  On the former, the majority noted the standard form, take-it-or-leave-it nature of the contract and the “significant gulf in sophistication” between the parties (para. 93).  On the latter, the majority stressed the high up-front costs and apparent necessity to travel to the Netherlands to raise any dispute (para. 94).  In its view, “No reasonable person who had understood and appreciated the implications of the arbitration clause would have agreed to it” (para. 95).  As a result, the clause is unconscionable and thus invalid.

Justice Brown instead relied on the public policy of favouring access to justice and precluding an ouster of the jurisdiction of the court.  An arbitration clause that has the practical effect of precluding arbitration cannot be accepted (para. 119).  Contractual stipulations that prohibit the resolution of disputes according to law, whether by express prohibition or simply by effect, are unenforceable as a matter of public policy (para. 121).

Justice Brown also set out at length his concerns about the majority’s reliance on unconscionability: “the doctrine of unconscionability is ill-suited here.  Further, their approach is likely to introduce added uncertainty in the enforcement of contracts, where predictability is paramount” (para. 147).  Indeed, he criticized the majority for significantly lowering the hurdle for unconscionability, suggesting that every standard-form contract would, on the majority’s view, meet the first element of an inequality of bargaining power and therefore open up an inquiry into the sufficiency of the bargain (paras. 162-63).  Justice Brown concluded that “my colleagues’ approach drastically expands the scope of unconscionability, provides very little guidance for the doctrine’s application, and does all of this in the context of an appeal whose just disposition requires no such change” (para. 174).

In dissent, Justice Cote was critical of the other judges’ willingness, in the circumstances, to resolve the issue rather than refer it to the arbitrator for decision: “In my view, my colleagues’ efforts to avoid the operation of the rule of systematic referral to arbitration reflects the same historical hostility to arbitration which the legislature and this Court have sought to dispel. The simple fact is that the parties in this case have agreed to settle any disputes through arbitration; this Court should not hesitate to give effect to that arrangement. The ease with which my colleagues dispense with the Arbitration Clause on the basis of the thinnest of factual records causes me to fear that the doctrines of unconscionability and public policy are being converted into a form of ad hoc judicial moralism or “palm tree justice” that will sow uncertainty and invite endless litigation over the enforceability of arbitration agreements” (para. 237).  Justice Cote also shared many of Justice Brown’s concerns about the majority’s use of unconscionability: “I am concerned that their threshold for a finding of inequality of bargaining power has been set so low as to be practically meaningless in the case of standard form contracts” (para. 257).

The decision is lengthy and several additional issues are canvassed, especially in the reasons of Justice Cote and Justice Brown.  The ultimate result, with the drivers not being bound by the arbitration clause, is not that surprising.  Perhaps the most significant questions moving forward will be the effect these reasons have on the doctrine of unconscionability more generally.

Law on Jurisdiction Clauses Changes in Canada

In 2011 Facebook, Inc. used the name and picture of certain Facebook.com members as part of an advertising product.  In response, a class action was started in British Columbia on behalf of roughly 1.8 million British Columbia residents whose name and picture had been used.  The claim was based on section 3(2) of the province’s Privacy Act.  In response, Facebook, Inc. sought a stay of proceedings based on an exclusive jurisdiction clause in favour of California contained in the contracts of use for all Facebook.com members.

Canadian courts had repeatedly held that “strong cause” must be shown to displace an exclusive jurisdiction clause.  In addition, while there was some ambiguity, the leading view had become that the analysis about whether to stay proceedings due to such a clause is separate and distinct from the general forum non conveniens analysis (para 18).  The clause is not simply an important part of the forum non conveniens analysis – rather, it triggers a separate analysis.

In Douez v Facebook, Inc., 2017 SCC 33 (available here) the Supreme Court of Canada confirms the second of these points: the analysis is indeed separate.  However, by a slim majority of 4-3 the court holds that the “strong cause” test operates differently in a consumer context than in the commercial context in which it was originally formulated.  The court overturns the decision of the British Columbia Court of Appeal and rejects a stay of proceedings, paving the way for the class action to proceed in British Columbia.

The Separate Analysis

All of the judges support the separation from forum non conveniens (paras 17, 20 and 131).  I have found this approach troubling as it has developed and so, while not a surprise, I am disappointed to see it confirmed by the court.  As I understand it, the core reason for the separate analysis is to make sure that the clause is not overcome by a series of less important factors aggregated under the forum non conveniens analysis.  So the separate analysis requires that the “strong cause” to overcome the clause has to involve something closely related or intrinsic to the clause itself.  The best explanation of this view is in Expedition Helicopters Inc. v Honeywell Inc., 2010 ONCA 351 (available here; see in particular para 24).  The problem is that courts, in their search for strong cause, frequently go beyond this and refer to factors that are well established under the forum non conveniens approach.

In its analysis, the court puts almost no emphasis on (and does not really even explain, in the way Expedition Helicopters does) how the separate approach differs from forum non conveniens in terms of how the clause gets displaced.  In places, it appears to actually be discussing forum non conveniens (see paras 29-30 and 155), in part perhaps due to its quite direct reliance on The Eleftheria, an English decision I think is more consistent with a unitary framework rather than a separate approach (a point noted in Expedition Helicopters at para 11).  In Douez, the plurality finds strong cause for two reasons: public policy and secondary factors (para 64).  Leaving public policy aside for the moment, it is telling that the secondary factors are “the interests of justice” and “comparative convenience and expense”.  These are the most conventional of forum non conveniens factors.  If this analysis is followed by lower courts, rather than that as explained in Expedition Helicopters, the separate analysis might end up not being very separate.

The Consumer Context

The majority (which is comprised of two decisions: a plurality by three judges and a separate solo concurrence) considers the unequal bargaining power and potential for the relinquishing of rights in the consumer context to warrant a different approach to the “strong cause” test (para 33).  In part, public policy must be considered to determine whether the clause is to be given effect.  As a matter of law, this may well be acceptable.  But one of the key features of the plurality decision is the basis on which it concludes that strong cause has been shown on the facts.  It reaches this conclusion because the contract is one of adhesion with notable inequality of bargaining power and because the claim being brought relates to “quasi-constitutional rights” (para 58), namely privacy.  If these factors are sufficient, then a great many exclusive jurisdiction clauses in standard form contracts with consumers are subject to being defeated on a similar basis.  Lots of consumer contracts involve unequal bargaining strength and are in essence “take it or leave it” contracts.  And it may well not be that difficult for claims to be advanced, alongside other claims, that involve some form of quasi-constitutional rights (the breadth of this is untested).  This possibility that many other clauses do not provide the protection once thought is likely the most notable dimension of the decision.

The Dissent

The dissent would not modify the “strong cause” test (paras 125 and 171).  It stresses the need for certainty and predictability, which are furthered by exclusive jurisdiction clauses (paras 124 and 159).  The dissent concludes the clause became part of the contract, is clear and is not unconscionable.  It reviews possible factors which could amount to strong cause and finds none of them present.  It is critical of the majority for its use of public policy as a factor in the strong cause analysis.  If the clause is enforceable – and in its view it is, even with the inequality of bargaining power – then it is wrong to rely on the factors used by the plurality to find strong cause (para 173).  In the immediate aftermath of the decision I think the dissent has the better of the argument on whether strong cause has been shown in this particular case.

Territorial versus Subject Matter Jurisdiction

The proposed class action relies on a statutory provision.  That statute contains a provision (section 4) that provides that the British Columbia Supreme Court must hear and determine claims under the statute.  The British Columbia Court of Appeal concluded that this provision addresses subject matter jurisdiction and not territorial jurisdiction (para 14).  The dissent agrees with that view (para 142).  In contrast, the plurality conflates the two types of jurisdiction.  While it accepts that the provision is not one which overrides jurisdiction clauses (para 41), in the public policy analysis it is concerned that in litigation in California the plaintiff class would have no claim (para 59).  But as the dissent points out, it is open to the California courts to apply the statute under its choice of law analysis (paras 165-66).  No evidence was adduced to the contrary.  Section 4, properly interpreted, does not prevent that.  Even more worrying is the analysis of Justice Abella in her solo concurring decision.  She concludes that section 4 deals with territorial jurisdiction and so overrides any jurisdiction clause to the contrary (paras 107-08).  This is a remarkable interpretation of section 4, one which would see many other provisions about subject matter jurisdiction instead read as though they addressed territorial jurisdiction (which she does in footnote 1 in para 109).


The split between the judges as to what amounts to strong cause sufficient to set aside an exclusive jurisdiction clause is the most dramatic aspect of the decision.  They see what is at stake very differently.  On one view, this is a case in which consumers should not be deprived of important statutory rights by a clause to which they did not truly agree.  On another view, this is a case in which contracting parties should be held to their agreement as to the forum in which any disputes which arise should be resolved because, even though the contract involves consumers, the agreement is not unfair and has not been shown to deprive them of any substantive rights.  This debate will now play out across a wide range of consumer contracts.

Supreme Court of Canada Allows Courts to Sit Extraterritorially

In Endean v British Columbia, 2016 SCC 42 (available here) the Supreme Court of Canada has held that “In pan-national class action proceedings over which the superior court has subject-matter and personal jurisdiction, a judge of that court has the discretion to hold a hearing outside his or her territory in conjunction with other judges managing related class actions, provided that the judge will not have to resort to the court’s coercive powers in order to convene or conduct the hearing and the hearing is not contrary to the law of the place in which it will be held” (quotation from the court’s summary/headnote).

The qualifications on the holding are important, since some of the earlier lower court decisions had been more expansive in asserting the inherent power of the superior court to sit outside the province (for example beyond the class proceedings context).  I am concerned about any extraterritorial hearings that are not expressly authorized by specific statutory provisions, but I do appreciate the utility (from an efficiency perspective) of the court’s conclusion in the particular context of this dispute.  It remains to be seen if attempts will be made to broaden this holding to other contexts.

The court has also held that “A video link between the out-of-province courtroom where the hearing takes place and a courtroom in the judge’s home province is not a condition for a judge to be able to sit outside his or her home province. Neither the [class proceeding statutes] nor the inherent jurisdiction of the court imposes such a requirement. The open court principle is not violated when a superior court judge exercises his or her discretion to sit outside his or her home province without a video link to the home jurisdiction” (quotation from the court’s summary/headnote).

This aspect of the decision concerns me, since my view is that the open court principle requires that members of the Ontario public and the media can see the proceedings of an Ontario court in an Ontario courtroom.  It is a hollow claim that they can fly to another province to watch them there.  The separate concurring decision appreciates this aspect of the case more than the majority decision, though it too stops short of requiring a video link.  In its view, “While the court should not presumptively order that a video link back to the home provinces be set up where the court sits extraprovincially, members of the public, the media, or counsel can request that a video link or other means be used to enhance the accessibility of the hearing. If such a request is made, or the judge considers it appropriate, a video link or other means to enhance accessibility should be ordered, subject to any countervailing considerations” (quotation from the court’s summary/headnote).

A Court’s Inherent Jurisdiction to Sit Outside its Home Territory

Another step in the evolution of the common law on this issue has been taken by the Court of Appeal for Ontario in Parsons v Ontario, 2015 ONCA 158 (available here).  The court disagrees in some respects with the earlier decision, on the same issue, of the British Columbia Court of Appeal in Endean v British Columbia, 2014 BCCA 61 (available here) (discussed by me over a year ago here).  It may be that in light of this conflict the Supreme Court of Canada will end up hearing appeals of either or both decisions.

People infected with the Hepatitis C virus by the Canadian blood supply between 1986 and 1990 initiated class actions in each of Ontario, Quebec and British Columbia.  These actions were settled under an agreement which provided for ongoing administration of the compensation process by a designated judge in each of the three provinces.  In 2012 the issue arose as to whether the period for advancing a claim to compensation could be extended.  Rather than having three separate motions in each of the provinces before each judge to address that issue, counsel for the class proposed a single hearing before the three judges, to take place in Alberta where all of them would happen to be on other judicial business.  In the face of objections to that process, motions were brought in each province to determine whether such an approach was possible.  The initial decision in each province was that a court could sit outside its home province.  The Quebec decision was not appealed but the other two were.

The Court of Appeal for Ontario has now released its decision on the appeal, and the three judges are quite divided.  They divide even over a preliminary issue, namely whether the order made below is “final” or “interlocutory” for purposes of the appeal route.  If it is the former, the appeal is properly brought to the Court of Appeal, but not if it is the latter (in which case the appeal would be to the Divisional Court).  The judges split 2-1 in deciding the order is final.

Turning to the merits, the judges remain divided.  Justice LaForme upholds the order below.  He concludes the court has the inherent jurisdiction to sit outside Ontario and that it can do so without violating the open court principle, even in the absence of a video link to an Ontario courtroom (for spectators and perhaps some lawyers).  Justice Lauwers agrees that the court has the inherent jurisdiction to sit outside Ontario, but that doing so without a video link back to Ontario would be a violation of the open court principle.  He reverses the order below, but only to the extent that he insists on such a link.  Justice Juriansz agrees with the result reached by Justice Lauwers but his reasoning is quite different.  He relies on Ontario’s Rules of Civil Procedure which allow for a motion to be heard by video-conference.  In his view, the proposed hearing outside of Ontario falls within these rules if there is a video link back to an Ontario courtroom.  No resort to inherent jurisdiction is required and the open court principle is not impaired.

I remain somewhat skeptical that the court has the jurisdiction to sit outside the province.  I would rather see such a process addressed by statute rather than through invocation of the court’s inherent powers.  I am also concerned that Justice Juriansz’s approach is something of a fiction, using the video-conference rules to in essence pretend that the hearing is actually being held in the courtroom to which the video feed is transmitted.  I consider such a video link essential, but for me it goes to the question of the open court principle and not to jurisdiction.

A side note: this is my first post in many months.  My sense, and that of many of my colleagues in Canada, is that we have had a dearth of interesting developments in private international law over the past year.


Can a Court Sit Outside its Territorial Jurisdiction?

In Parsons v The Canadian Red Cross Society, 2013 ONSC 3053 (available here), Winkler CJ (of the Court of Appeal, here sitting down in the Superior Court of Justice) has held that a judge of the SCJ can sit as such outside Ontario.  No authority, it seems, requires the SCJ to sit only in Ontario.

The decision seems to me, at least on an initial reading, largely based on pragmatism.  It seems efficient to so allow and so the court does.  But I have some preliminary sense that there are some larger concerns here that are not being fully thought through.  The place where a court sits seems awfully fundamental to its existence and authority as a court.  In addition, the brushing aside of concerns about the open court principle (see paras 48-50) seems too minimal.

Part of the decision is based on Morguard and the federal nature of Canada (see para 25), so maybe the judge could not so sit outside Canada?

For news coverage of the decision, see this story.

Could this idea get pushed beyond the fairly narrow bounds of this case?  Say a case is started in Ontario and the defendant seeks a stay in favour of Alberta because of all the factual connections to that province.  Could the plaintiff, if otherwise likely to see the proceedings in Ontario get stayed, ask the court to have one of its judges hear the case in Alberta, sitting as a judge of the Ontario court?  That way the plaintiff gets an Ontario judgment and the defendant gets the case heard in Alberta…

Article on Global Class Actions in Canada

Associate Professor Tanya Monestier of the Roger Williams University School of Law has written an article on the willingness of Canadian courts to hear class actions involving a global plaintiff class. It is entitled “Is Canada the New ‘Shangri-La’ of Global Securities Class Actions?” and is forthcoming in 2012 in the Northwestern Journal of International Law and Business. The article is available here from SSRN.

The abstract reads:

There has been significant academic buzz about Silver v. Imax, an Ontario case certifying a global class of shareholders alleging statutory and common law misrepresentation in connection with a secondary market distribution of shares. Although global class actions on a more limited scale have been certified in Canada prior to Imax, it can now be said that global classes have “officially” arrived in Canada. Many predict that the Imax decision means that Ontario will become the new center for the resolution of global securities disputes. This is particularly so after the United States largely relinquished this role last year in Morrison v. National Australia Bank.

Whether Imax proves to be a meaningful precedent or simply an aberration will largely depend on whether the court dealt appropriately with the conflict of laws issues at the heart of the case. No author has yet addressed the conflict of laws complications posed by the certification of global class actions in Canada; this Article seeks to fill that void. In particular, I use the Imax case as a lens through which to canvass the conflict of laws issues raised by the certification of global classes. I look at the difficult questions of jurisdiction simpliciter, recognition of judgments, choice of law, parallel proceedings, and notice/procedural rights that need to be addressed now that global classes have come to Canada.

Parallel Class Actions in Canada

Canadian provincial courts continue to analyze how to manage class actions that include class members from other provinces.  While Canada is a federal country, it is acceptable for the court in a province to certify a class that includes members from other provinces.  A difficulty arises if two provinces are each asked to certify a multijurisdictional class in respect of the same underlying claim.

Currently there are class actions against Merck Frosst in both Ontario and Saskatchewan in respect of Vioxx.  In each of these provinces, the class action regime is “opt-out”, so that the class as defined catches all described members without any specific action on the part of a particular member.  Merck moved to stay the Ontario action on the basis that it should not be subject to two multijurisdictional class actions that involve substantially the same plaintiffs and issues.  In Mignacca v. Merck Frosst Canada Ltd. (an as-yet unreported decision of the Ontario Divisional Court, dated Feb. 13, 2009) the court refused to stay the Ontario action.

The court refused to adopt an approach that would defer to the court that first certified the class action: “a rule of swiftest to the finish line taking all encourages tactics that may well be contrary to the interests of justice” (para. 47).  The court noted that in other cases parallel class actions involving jurisdictional overlap had been resolved through the cooperation of counsel and guidance from the court. 

An unusual element of this case was the Ontario court’s concern about the lawyer representing the plaintiff class in the Saskatchewan proceedings.  It noted that he had five disciplinary violations from 1972 to 2006.  This strengthened the court’s desire to have the Ontario proceedings continue.