Sibiga c. Fido Solutions inc.
no. de référence : 2016 QCCA 1299
Sibiga c. Fido Solutions inc.2016 QCCA 1299
COURT OF APPEAL
CANADA
PROVINCE OF QUEBEC
REGISTRY OF
MONTREAL
No:
500-09-024648-149
(500-06-000636-130)
DATE:
AUGUST 10, 2016
CORAM:
THE HONOURABLE
FRANCE THIBAULT, J.A.
NICHOLAS KASIRER, J.A.
GUY GAGNON, J.A.
INGA SIBIGA
APPELLANT – petitioner
v.
FIDO SOLUTIONS INC.
ROGERS COMMUNICATIONS PARTNERSHIP
BELL MOBILITY INC.
TELUS COMMUNICATIONS COMPANY
RESPONDENTS – respondents
JUDGMENT
[1] On appeal from a judgment of the Superior Court, District of Montreal (the Honourable Mr Justice Michel Yergeau), dated July 2, 2014, that dismissed the appellant's motion for authorization to institute class action proceedings.
[2] For the reasons of Kasirer, J.A., with which Thibault and Gagnon, JJ.A. agree, THE COURT:
[3] ALLOWS the appeal and SETS ASIDE the judgment of the Superior Court;
[4] GRANTS the appellant's motion seeking authorization to institute the class action;
[5] ASCRIBES to Inga Sibiga the status of representative for the purpose of exercising the class action on behalf of the following class:
All consumers residing in Quebec who were charged international mobile data roaming fees by the respondents at a rate higher than $5.00 per megabyte after January 8, 2010.
Tous les consommateurs qui résident au Québec et à qui les intimés ont chargé des frais d'itinérance pour les données à un taux excédant 5,00 $ par mégaoctet après le 8 janvier 2010.
[6] IDENTIFIES the following as the principal questions of fact and of law to be treated collectively in the action:
Does the disproportion between the international mobile data roaming fees charged to the class members and the value of the service provided by the respondents constitute exploitation and objective lesion under section 8 of the Consumer Protection Act (the “CPA”)?
Are the respondents' international mobile data roaming fees excessively and unreasonably detrimental to consumers such that the contractual clauses allowing them to charge such fees are abusive under article 1437 C.C.Q.?
Must the class member's obligations be reduced and if so, by how much?
Are the class members entitled to punitive damages, and if so, what amount must the respondents pay?
[7] IDENTIFIES the following as the principal conclusions that relate to the aforementioned questions:
AUTHORIZE the class action of the petitioner and class members against the respondents, with costs;
GRANT the petitioner's motion to obtain the status of representative of all class members;
DECLARE that the international mobile data roaming fees charged by respondents amount to exploitation under section 8 of the CPA;
DECLARE that the international mobile data roaming fees charged by the respondents are excessively and unreasonably detrimental to consumers or adhering parties and are therefore not in good faith under article 1437 C.C.Q.;
REDUCE the obligations of the petitioner and class members to pay the respondents for the international mobile data roaming services charged to their fair market value;
ORDER respondent Fido to compensate the petitioner for the amount overcharged;
ORDER the collective recovery of all damages owed to the class members for the amount overcharged;
ORDER the collective recovery of all the punitive damages to be paid to all the class members;
ORDER the respondents to pay each member of the class their respective claims, plus interest at the legal rate as well as the additional indemnity provided for by law in accordance with article 1619 C.C.Q.;
THE WHOLE with costs at all levels, including the cost of all exhibits, experts, expertise reports and notices.
[8] DECLARES that, except in the case of exclusion, members of the class will be bound by any and all judgments relating to the class action in the manner provided by law;
[9] FIXES the time limit for requesting exclusion from the class at sixty (60) days from the date of publication of the notice to members, from which time the members of the class who have not requested exclusion therefrom will be bound by any and all judgments that are rendered in the class action;
[10] ORDERS the publication of the notice to members within sixty (60) days from this judgment in the La Presse, Le Soleil and The [Montreal] Gazette newspapers;
[11] REMANDS the file to the Chief Justice of the Superior Court for determination of the judicial district in which the class action will proceed and for appointment of the judge charged with hearing the case;
[12] With legal costs on appeal; costs in first instance to follow suit.
FRANCE THIBAULT, J.A.
NICHOLAS KASIRER, J.A.
GUY GAGNON, J.A.
Mtre Bruce W. Johnston
Mtre Yves Lauzon
Mtre Andrew E. Cleland
Trudel, Johnston & Lespérance
For the Appellant
Mtre Pierre Y. Lefebvre
Mtre Eleni Yiannakis
Mtre Noah Michael Boudreau
Fasken Martineau DuMoulin sencrl
For the Respondents Fido Solutions Inc. and Rogers Communications Partnership
Mtre Yves Martineau
Mtre Matthew Angelus
Stikeman Elliott sencrl, srl
For the Respondent Telus Communications Company
Mtre Christine A. Carron, Ad.E.
Mtre Frédéric Wilson
Norton Rose Fulbright Canada sencrl, srl
For the Respondent Bell Mobility Inc.
Date of hearing:
September 1, 2015
REASONS OF KASIRER, J.A.
[13] Inga Sibiga appeals the dismissal of her motion for authorization of a class action. She alleged that international roaming fees charged by wireless mobile phone service providers to Quebec consumers, like herself, were abusive, lesionary and so disproportionately high as to amount to exploitation under applicable rules in consumer protection legislation and the Civil Code of Québec.[1]
[14] In reasons both well-researched and compellingly written, the motion judge acknowledged that the prima facie case standard set forth by the Supreme Court in Infineon[2] and Vivendi[3] governed the outcome of the motion. At the same time, he gave plain voice, within the right confines of judicial propriety, to his sense of the mischief that can result if that threshold is applied incautiously. His preoccupation that a lax approach to the standard can result in authorization of class actions that do not deserve to go to trial is no doubt a legitimate one. “On ne lance pas une procédure aussi coûteuse pour le système judiciaire qu’un recours collectif”, he wrote in paragraph [98], “sur une base aussi ténue”. The Supreme Court has taken pains to say that however liberal the standard at authorization, a class action cannot rest on allegations that are vague or imprecise or be hostage to a plaintiff who is unqualified to represent members of the class. A lack of rigour at authorization can indeed weigh down the courts with ill-conceived claims, creating the perverse outcome that the rules on class actions serve to defeat the very values of access to justice they were designed to champion.
[15] That said, while a judge can refuse a motion for authorization that relies on an overly liberal interpretation of the Infineon standard, it is a mistake in law to refuse authorization by treating that standard as overly liberal in itself.
[16] In my respectful view, by denying authorization in the present case based on what he described as an imprecise and speculative claim, the motion judge neglected to apply the prima facie case standard relevant to this consumer class action. He also erred in his evaluation of adequate representation. In the result, I would grant the motion for authorization and propose, as well, that the Court resist arguments that the class be defined more narrowly at this early stage of the proceedings.
I Background
[17] The appellant is a Quebec consumer who has a wireless telephone contract with Fido Solutions Inc. In September 2012, she used her mobile phone on approximately six occasions to access the Google Map service through the internet while travelling on holiday in the United States. She had chosen not to avail herself of a pre-paid travel plan offered by Fido that would have entitled her to a reduced rate for roaming mobile data services outside of Canada. As a result, she was billed on a pay-per-use basis for 40.82 megabytes (MB) of roaming data used at a rate of $6.14 per MB. According to her monthly account summary, she owed $250.81 for the roaming data used in the United States in addition to the amount of her usual monthly invoice.
[18] The appellant says she was disagreeably surprised at the additional amount charged at the time but paid it without complaint.
[19] In December 2012, she received a mass email from the law offices of Trudel & Johnston announcing that the Montreal firm had undertaken an investigation of international roaming fees charged to Quebec consumers using their wireless mobile devices. The firm had been examining the viability of a consumer class action based on unfair international roaming fees before they met the appellant. The email invited consumers who had received bills they considered to be excessive to contact the firm. The appellant did so shortly thereafter.
[20] With Trudel & Johnston acting on her behalf, the appellant filed a motion for authorization to institute a class action and obtain the status of representative in the Superior Court on January 8, 2013. The named defendants were Fido, Rogers Communications Partnership, Bell Mobility Inc. and Telus Communications Company.
[21] In the motion, the appellant alleges that the respondents had charged international roaming fees to Quebec consumers that are disproportionate and exploitative, in violation of section 8 of the Consumer Protection Act.[4] The contracts between consumers and the named wireless service providers are also alleged to be abusive within the meaning of article 1437 C.C.Q. The core claim of the class action is stated in paragraph 2.19: “[…] the available evidence at this stage clearly demonstrates that the underlying cost of providing international mobile roaming data represents a minuscule fraction of the retail rates charged by the Respondents and that such retail rates are disproportionate, exploitative, and abusive”.
[22] The appellant claimed that the facts alleged gave her an individual right of action, as a consumer, against Fido, the wireless service provider with whom she was a subscriber, and also gave rise to actions on behalf of class members who contracted with the other respondents. (Rogers and Fido are related companies). The class is described in the motion as follows:
Tous les consommateurs qui résident au Québec et à qui les Intimés ont chargé des frais d'itinérance pour les données à un taux excédant 5,00 $ par mégaoctet après le 8 janvier 2010.
All consumers residing in Quebec who were charged international mobile data roaming fees by the Respondents at a rate higher than $5.00 per megabyte after January 8, 2010.
[23] By way of redress on the merits, the appellant sought a declaration that international mobile data roaming fees charged by the respondents violate the CPA and article 1437 C.C.Q. She sought to reduce the amounts payable for the international roaming fees to an amount equal to their fair market value as well as an order to compensate her and other members of the class for the amounts overcharged. Finally, she asked for collective recovery of punitive damages.
[24] The appellant filed 19 exhibits in support of her motion, but not her own contract with Fido. These exhibits included various pages from the respondents’ individual websites concerning their international roaming rates as well as those of some of their competitors; her own monthly account summary for her wireless phone contract with Fido; a report of the Organization for Economic Cooperation and Development (OECD) on international mobile data roaming; an article from the Globe and Mail newspaper dated June 8, 2011 entitled “Roaming Canadians taking costly wireless hits”; a Telus’ press release entitled “Telus cuts roaming rates up to 60 per cent for its seven million wireless customers” dated June 11, 2011; a UK media article entitled “Why roaming data costs too much” dated March 29, 2011; the European Community Regulation No 717/2007 on Roaming on Public Mobile Telephones within the Community; and the European Union Regulation No 531/2012 on Roaming on Public Mobile Communications Networks within the Union. In turn, the respondents filed exhibits in support of their respective positions, including transcripts of the appellant’s out-of-court examination; affidavits from some of their officers; and further excerpts from webpages.
II Judgment of the Superior Court
[25] After a two and one-half days hearing, the motion judge found that the appellant had not satisfied the requirement in former article 1003(b) C.C.P. that the facts alleged in the motion seem to justify the conclusions sought. He noted that the motion contained no allegation establishing the precise tenor of the contractual obligations assumed by the appellant and by Fido and faulted the petitioner for having failed to produce a copy of her contract (para. [113]). Apart from the fact that the petitioner had paid $6.14 per MB for roaming in the United States, the allegation that the fees were exploitative was, according to the judge, based on nothing more than speculation and hypotheses (para. [114]). While this conclusion alone was sufficient to deny authorization, the judge went on to explain that the appellant was not in a position to represent the members of the class adequately as required by article 1003(d) C.C.P. The circumstances in which the solicitor-client relationship was established here meant that she did not have sufficient control of counsel to act as representative of the other class members. Moreover, her own weak understanding of the class action meant that she could not adequately represent the class (paras [154] and [155]). The judge added that the claims of various members did not raise identical, similar or related questions of law and fact within the meaning of article 1003(a) C.C.P.
[26] He also concluded that the appellant did not have standing to sue and, consequently, could not act as representative of the class. Relying on the judgment of this Court in Agropur,[5] the judge stated that a petitioner in a class action must have a cause of action against each of the defendants sued, subject to exceptions not applicable here (paras [33] to [47]). The fact that the appellant had no contractual relationship with the other defendants meant that she could not seek authorization in a class action on behalf of their clients.
[27] The judge spoke briefly to the definition of the class in connection with article 1003(b) C.C.P. After having noted his concern that the Superior Court was not equipped to conduct the “enquête à caractère public” called for by the motion, he observed that the facts alleged did not justify the broad composition of the class as framed.
[28] He dismissed the motion for authorization, with costs.
III Issues in Appeal
[29] Ms Sibiga raises four general grounds of appeal. She contends that the judge erred, first, in deciding, largely based on jurisprudence now viewed as outdated, that she did not have standing to sue the defendants with whom she has no contractual relationship. Second, the judge erred in finding that the facts alleged in the motion do not justify the conclusions sought because he failed to apply the prima facie standard (art. 1003(b) C.C.P.). Third, he was mistaken in deciding that the appellant was not in a position to represent the members of the class adequately (art. 1003(d) C.C.P.). Fourth, the judge is said to have erred in holding that the recourses of the members of the class did not raise identical, similar or related questions (art. 1003(a) C.C.P.).
[30] The respondents argue that the judge made no mistake that would provide a proper basis for disturbing his decision to deny authorization. They recall that this Court owes deference to the judge’s findings in respect of the relevant criteria.
[31] In the event the appeal were to be allowed, Telus and Bell argue that the class proposed in the motion should be recast so that it is properly aligned both with the evidence presented and the rule of proportionality set forth in article 4.2 C.C.P.
IV Substance of the Appeal
[32] Some preliminary remarks regarding the standard of review on appeal of judgments dismissing motions for authorization are in order.
[33] The respondents are right to say that, barring an error of law, this Court owes deference to the motion judge’s decision, given the inherently discretionary character of his findings relating to the criteria for authorization set forth in article 1003 C.C.P.[6]
[34] While the compass for appellate intervention is indeed limited, so too is the role of the motion judge. In clear terms, particularly since its decision in Infineon, the Supreme Court has repeatedly emphasized that the judge’s function at the authorization stage is only one of filtering out untenable claims. The Court stressed that the law does not impose an onerous burden on the person seeking authorization. “He or she need only establish a ‘prima facie case’ or an ‘arguable case’”, wrote LeBel and Wagner JJ. in Vivendi, specifying that a motion judge “must not deal with the merits of the case, as they are to be considered only after the motion for authorization is granted”.[7]
[35] Since Infineon, our Court has consistently relied upon this standard, invoking it when authorization has been wrongly denied because too high a burden was imposed.[8]
[36] I turn now to a consideration of the four errors alleged and of the respondents’ subsidiary argument that this Court should exercise its authority to redefine the class.
IV.1 Sufficient interest in the action pursuant to article 55 C.C.P.
[37] The appellant submits that the judge erred when he decided that article 55 C.C.P. required her, as representative of the class, to establish a direct cause of action against each of the named defendant mobile telephone service providers. She says the judgment in Agropur,[9] upon which the judge relied, is no longer good law given recent developments in the jurisprudence.
[38] The appellant is correct on this point.
[39] In fairness to the judge, it was not until after the judgment in appeal that the Supreme Court set aside Agropur in definitive terms. In Bank of Montreal v. Marcotte,[10] the Court held that the notion of sufficient interest in article 55 C.C.P. must be adapted to the collective and representative character of a class action. As long as the appellant satisfies the criteria in article 1003, it was open to the judge to authorize the class action even if she herself did not have a direct cause of action against each defendant.
[40] Two comments are in order. First, the judge’s mistaken reliance on Agropur – again, in fairness, it is something of an innocent mistake – does not constitute, in itself, an overriding error, except in respect of the exclusion of Telus and Bell from authorization on this basis alone. It should be recalled that he found the requirements of article 1003 C.C.P. lacking in respect of all of the defendants, including Fido with whom the appellant had a contract. Second, in Marcotte, the Supreme Court was quick to recall that while Agropur has been formally set aside, article 1003(d) still requires that the petitioner be in a position to represent the members of the class adequately, including against defendants, with whom he or she would not otherwise have standing to sue.[11] Telus and Bell say that this caveat needs to be addressed with particular rigour in this case, in particular given the judge’s view that the appellant was unqualified to represent the class adequately under article 1003(d). I shall return to this point below.
[41] I turn now to the appellant’s arguments under article 1003 C.C.P.
IV.2a Did the facts alleged justify the conclusions sought (art. 1003(b) C.C.P.)?
[42] The motion judge held that the class action should not be authorized because the requirement in article 1003(b) was not met. Specifically, the judge decided that the allegation that the roaming fees charged were exploitative and abusive did not justify the reduction in fees or the compensatory and punitive damages claimed by the class.
[43] The judge’s reasons were detailed. They merit close study.
[44] He noted first that the appellant failed to produce her contract with Fido and that, in the circumstances, this omission was fatal to her claim. That contract was, in the judge’s words, a “fait tangible essentiel pour permettre de juger si les conclusions recherchées paraissent justifiées à cette étape et de juger de la représentativité de la requérante” (para. [109]). In respect of the requirements of article 1003(b), the judge was of the view that, absent this contract, any disproportion in the “respective obligations / prestations respectives” of the parties amounting to objective lesion could not be measured for the purposes of section 8 CPA, nor could one determine whether or not the disputed contractual provisions were abusive under article 1437 C.C.Q. The materials presented by the appellant were insufficient to meet the burden of presentation at authorization.
[45] The judge also observed that the haphazard manner in which appellant’s counsel sought to identify the wholesale cost paid by the respondents for the international roaming services failed to meet her burden of presentation.
[46] The appellant was obliged to advance tangible facts allowing for a comparison between what she paid for roaming and the cost incurred by Fido to provide the service. While it was known that she paid $6.14 per MB for roaming in the United States, the judge held that evidence of the cost of this service to Fido was lacking. She presented mere hypotheses in support of her claim that the cost of roaming fees to the respondents was only a fraction of what consumers were required to pay in the various countries involved. To allow the class action to proceed on this basis would amount to convening something akin to a multi-jurisdictional commission of inquiry into the fair market value of international roaming fees, a task for which the Superior Court has neither the jurisdiction nor the resources to conduct (para. [121]). He decided that the facts alleged justified neither the claim for lesion nor that for abuse.
[47] On appeal, Ms Sibiga argues that the motion judge erred in three main respects in his analysis of the requirement in article 1003(b) C.C.P.: firstly, he erred in holding that the failure to produce her own contract was fatal to the motion for authorization; secondly, he wrongly evaluated the evidence on the merits, thereby overstepping his role at authorization; and thirdly, his reasons were both unreasonable and incomplete.
[48] Before proceeding with an examination of each of these grounds of appeal, a summary of the law applicable under article 1003(b) C.C.P. is in order.
[49] It should first be noted that there is no disagreement, substantively, as to the applicable “arguable case” standard. The parties recognize, as well, that the judge rightly identified the rule that he need only screen the motion and must not conduct a “procès par anticipation” at this stage (paras [21] and [119]). Appellant Sibiga contends, however, that after properly noting the standard, the motion judge failed to apply it and that this constitutes an error justifying the reversal of the judgment in appeal.
[50] Given the access to justice policy considerations upon which the law of class action rests, LeBel and Wagner JJ. wrote in Infineon[12] that it would be unreasonable to require an applicant to establish anything more than an arguable case at the authorization stage. As some of the history traced in the Supreme Court opinion makes plain, this reflects the lightened evidentiary burden established by the Quebec legislature in 2003 when the requirement of affidavit evidence at the authorization stage was abolished. The purpose of those amendments, it has been usefully written, “was to ensure that the authorization stage be used to filter out only the most frivolous and unsubstantiated claims and to ensure that the authorization process was not being used by judges to render pre-emptive decisions on the merits.”[13]
[51] Courts have recognized access to justice as a “social dimension” to class action law that is relevant to the kind of interpretative task before the judge here.[14] This explains why courts should err on the side of caution and authorise the action where there is doubt as to whether the standard has been met.[15] For the present case, it bears recalling that both consumer law and class action law share this overarching policy concern of access to justice.[16]
[52] The allegations in the motion are presumed to be true, as long as they are sufficiently precise. A motion judge should only weed out class actions that are frivolous or have no prospect of success. To meet this burden, the appellant did not need to prove the elements of the cause of action on the balance of probabilities.
[53] In short, Ms Sibiga had to make an arguable case that the net cost for international roaming in her arrangement with Fido amounted to objective lesion under section 8 CPA and that the contractual provision pursuant to which she committed to pay those fees was abusive. It was not enough, for example, for her to allege that paying $250 for accessing a Google map on several occasions was exploitative or abusive in the abstract. The jurisprudence indicates that objective lesion requires a comparison of what the consumer paid for the service (in this case, $6.14 per MB) and the “wholesale” cost to the merchant of providing this service to a Quebec subscriber accessing the roaming feature in the United States. The judge was right to say that the appellant could not content herself with imprecise allegations. But she was not bound to show objective lesion and abuse on the balance of probabilities, as she will have to do at trial.
[54] As noted, the motion judge’s conclusion that the appellant failed to establish an arguable case cannot be overturned lightly. She must show an error of law or that the judge’s assessment of whether “the facts alleged seem to justify the conclusions sought” is clearly wrong.
[55] I turn now to a consideration of the appellant’s three grounds of appeal concerning article 1003(b) C.C.P.
IV.2.b The failure to produce the appellant’s contract with Fido
[56] Appellant Sibiga argues that the motion judge was mistaken to conclude that he could not evaluate, on a prima facie basis, the exploitative or abusive character of the international roaming fees without examining the whole of the contract.[17]
[57] The respondents answer by saying that the judge was perfectly correct to find that the failure to produce the contract meant that there was an absence of evidence concerning the respective contractual obligations of the parties. In order to determine whether a contractual clause is abusive, they say, the clause must be examined in light of the whole of the parties’ rights and obligations under the contract.[18] Moreover, say the respondents, the determination of whether roaming fees are exploitative pursuant to section 8 CPA requires that the whole of the parties’ respective prestations be considered.[19] They rely on Dubuc v. Bell Mobility,[20] to argue that it was not wrong to hold that the failure to produce the contract precluded authorization of the class action.
[58] The respondents add that the “multidimensional” nature of wireless phone contracts – the fact that the prices of different services in a same package contract might be linked, allowing the consumer to choose the arrangement that meets his or her particular needs – makes the production of the appellant’s particular contract imperative to understanding the price of roaming fees relative to other features of her package. Counsel for Telus gave the example of a consumer who makes numerous long-distance calls but has no intention of using the roaming service. He or she would logically prefer a package with low long-distance rates and, in exchange, would be prepared to pay higher roaming fees. In circumstances like this, it would be wrong not to examine the whole dynamic of the contract to determine whether one aspect of the arrangement was abusive or lesionary, and it would be a mistake to evaluate the fairness of one service, such as roaming fees, in isolation (“en vase clos”, he said), without considering how savings elsewhere made the price of the package fair overall. Accordingly, it is argued, the judge was right to require the production of the appellant’s contract.
[59] In my respectful view, the motion judge erred in law to hold that the production of the contract was necessary to determine if article 1003(b) was satisfied here.
[60] Firstly, this is not a case where a contract between the consumer and Fido needed to be produced because of doubt as to its very existence. The contract is properly alleged, in paragraph 2.71 of the motion, and its existence is not contested. Nor is it a case in which the validity of the contract turns on the verification of its formal aspect, as in the case of a contract made by authentic deed. While there may be some instances in which the production of a contract is necessary when its existence or formal validity is in dispute, here the parties disagree only as to its obligational content.
[61] Secondly, I recognize that in many cases, perhaps most, the best evidence rule for establishing the obligational content of a written contract might well require, at a trial, the production of the document itself. But as has been observed elsewhere, the best evidence rule has a necessarily limited application at certification or authorization given a petitioner’s lightened burden of proof at this essentially procedural, pre-trial phase.[21] At this stage of a class action, the obligational content of the contract can be shown otherwise. I do not share the respondents’ reading of Dubuc in which this Court faulted the petitioner for his failure “d'établir prima facie la teneur des obligations contractuelles liant les parties”[22] and not for failing to produce the contract as such. As in Martin v. Société Telus Communications[23] where my colleague Dutil J.A. observed that the absence of a written contract was made up for by the production of other documents, including bills and pages from the defendant’s website, in our case appellant Sibiga has sought to establish a prima facie case using like materials.
[62] It bears noting that there is something profoundly disingenuous about the respondents’ protestations concerning the absence of the contract. It is of course true that the appellant bears the burden of establishing her case here and eventually at trial. But there is no reason to believe that respondent Fido does not have a copy of the appellant’s contract. I find it hard to accept in these circumstances that the class action would be dismissed before trial because the consumer lost her written contract, especially when one considers that counsel for the respondents, at the hearing on appeal, declined to say whether or not consumers were given written contracts when they agreed to these kind of arrangements as a matter of course.
[63] Ultimately, a determination of whether the contractual arrangement between the appellant and Fido is exploitative or abusive will require the court charged with evaluating the action on the merits to consider the whole of parties right and obligations under the contract. But for the purposes of showing an arguable case, the tenor of the contract could be established by her by other means.
[64] I would add that if, indeed, the wireless phone service agreements relevant to the proposed class action are “multidimensional” and that the relative price of international roaming fees, as against other services furnished by the respondents, reflects a commercial idea of “communicating vessels” (“vases-communicants”), this too can be raised by respondents at trial as a defence on the merits of the case. I hasten to add, however, that it would be both unwise and unfounded to make any suppositions about the links or balancing between and among services offered in the contract at this early stage. Just because a merchant offers one service at a low price does not mean that the high price he or she charges for another service at the same time is not exploitative. Moreover, the international roaming fees are “optional and accessory”,[24] and the fees are billed as an add-on to basic monthly payments. Only at trial, when all the pricing of the different elements of the contract are properly in view, can the defence of the relative price of roaming fees as against other services contracted for by the appellant and other consumers be fully measured.
[65] In my view, the insistence that the contract be produced at the authorization stage was a factor leading directly to the Superior Court’s misapprehension of the materials relevant to the assessment of the prima facie case standard under article 1003(b) C.C.P. In connection with that evaluation, the judge wrote:
[113] Il n’y a dans la requête aucune allégation ni aucun document établissant le cadre des obligations contractuelles assumées par la requérante et par Fido. Les factures produites ne peuvent en être que le reflet. Cela suffit au Tribunal pour conclure que les faits allégués ne paraissent justifier les conclusions recherchées.
[66] Respectfully stated, I am unable to agree with the judge that the contract was absolutely necessary to show the contractual obligations of the parties. Contrary to what the respondents argued, the motion and the documents before the Superior Court did allow for a proper measure of their obligations for the purposes of article 1003(b) C.C.P.
[67] At paragraphs 2.10 to 2.13, of the motion for authorization, the appellant alleges the basis upon which international roaming services were offered and billed to class members. At paragraph 2.50, she alleges, in particular by reference to Exhibit 15, the widely variables rates at which these charges are billed to Quebec consumers in various countries. This exhibit bears reproducing:
[68] While the contract itself was not filed, the evidence produced before the Superior Court included the appellant’s monthly wireless phone bills. From these documents, it is possible to discern the modalities of the plan that she agreed to with Fido; the charges billed for services used; and the manner in which her phone was used on a month-by-month basis. In addition, the appellant produced multiples pages taken from Fido’s internet site which set forth the standard rates for international roaming fees charged to customers. She also prepared the comparative table of fees charged by various wireless service providers reproduced above. The foregoing was sufficient to determine, with the degree of precision required at this stage, the obligational content of her agreement with Fido notwithstanding the absence of the material contract.
IV.2.c Did the judge apply the correct burden of proof pursuant to art. 1003(b)?
[69] Appellant Sibiga argues that the judge erred by evaluating the substance of evidence as would a judge on the merits rather than according to the appropriate prima facie standard pursuant to article 1003(b) C.C.P. Rather than measuring whether there was an arguable case that the fees were exploitative and the contract was abusive according to the standard identified by the Supreme Court in Infineon, the judge is said to have decided evidentiary matters at the authorization stage on the balance of probabilities, a task that is properly within the purview of the trial judge.
[70] The respondents answer that the judge made no such error. He rightly dismissed the action at this early stage because it was based on mere hypotheses and not on hard facts as to the exploitative or abusive character of the roaming fee charges and, in so doing, the judge followed the plain direction of the Supreme Court that the factual allegations cannot be vague, general or imprecise. Moreover, they recall, some evidence to form an arguable case must accompany the motion. This minimal evidence, in particular in respect of the base cost of international roaming service to the respondents, was completely lacking at the authorization stage.
[71] I agree with the appellant. In my respectful view, the judge committed a reviewable error by evaluating aspects of the case on the balance of probabilities, rather than on the basis of the prima facie standard.
[72] It was wrong to characterize the allegations and evidence of the difference between the prices paid by consumers and the cost to the respondents as speculative, imprecise and hypothetical. Even in the absence of proof of the exact cost of paid by the respondents to provide their customers with roaming services internationally, there were allegations and evidence before the judge to sustain an arguable case that Fido appears to have charged the consumer a fee for roaming services that was objectively lesionary and that the agreement in respect of roaming fees was abusive, thereby justifying the remedies sought in the motion for authorization. In light of the burden of presentation facing the appellant, there was evidence to sustain a prima facie case that, after January 8, 2010, consumers were charged roaming fees by the respondents “at a rate higher than $5 per MB” that was exploitative and abusive.
[73] The key problem facing the appellant in her effort to establish an arguable case for objective lesion is that, as is the case in many consumer actions, the appellant was not in a position to know, at this early stage, the amount or amounts of the wholesale cost of the wireless roaming service incurred by Fido or by any of the respondents.
[74] It is true that the appellant failed to bring direct proof of this base cost.
[75] The respondents produced affidavits and exhibits of their own at the authorization hearing but, not surprisingly, did not disclose the wholesale costs they face. Counsel for the appellant sought to obtain information relating to the costs of roaming services through an access to information application to the Canadian Radio and Television Commission (CRTC). This effort was unsuccessful: counsel was told that the CRTC does not possess information regarding the underlying costs because it does not require wireless service providers to file rates relating to international roaming services for approval. It was, of course, not the respondents’ burden to do so but, once again, their position is disingenuous. Had they made this information available to the first judge, he would have been in a position to evaluate the allegations of exploitation and abuse brought by the appellant on behalf of the class immediately. While it was not their burden to disprove the prima facie case, if the wholesale costs did reveal that the roaming prices were, as they suggest, not lesionary, the respondents might well have brought a quick and efficient end to the case rather than taking their chances in testing the appellant’s ability to show a prima facie case.
[76] As the appellant rightly observes, consumers very often face an informational imbalance when they allege objective lesion in that the merchant, and not the consumer, knows the wholesale cost of the good or service in issue. With respect, the judge did not sufficiently consider this fact. In consumer litigation generally, there are different ways in which courts have allowed consumer-plaintiffs to show, by indirect evidence, that the prices charged to them are exploitative based on a disparity between the consumer price and the wholesale price. These include market comparisons as well as other indicators.[25] Such alternate means of proof should be considered by an authorization judge otherwise consumer class actions might never advance to trial.
[77] The judge was harsh with the appellant, noting that in the absence of precise financial evidence of the respondents’ costs: “les avocats de la requérante ont puisé des informations à gauche et à droite sans expliquer en quoi ceux-ci sont préférables à d’autres. Pas un mot sur la méthodologie” (para. [115]).
[78] The indirect evidence that the price of roaming fees for the class was objectively lesionary and abusive was no doubt imperfect and, if measured on the balance of probabilities, was likely fragile. But it was enough to show that the appellant’s claim was not a frivolous one and that, at trial, she would have an arguable case to make on behalf of the class. Indeed in Infineon, where the Supreme Court explained the foundation of the prima facie case requirement, LeBel and Wagner JJ. were careful to say that a petitioner under article 1003 C.C.P. does not need to advance a “sophisticated methodology”, as a general rule, to satisfy the arguable case standard.[26]
[79] Firstly, the appellant filed Exhibit 15, reproduced above, that sets out the retail international mobile data roaming rates per MB. For the United States, for example, the table suggests that Rogers and Fido charge $10.24 per MB without a plan of some sort; Bell charges $6 and Telus charges $5. Competitors charge substantially less according to the facts gathered by the appellant: Vidéotron ($1.03 per MB), Public Mobile and Mobilicity ($1.50) and Wind Mobile ($1). In other words, there was prima facie evidence alleged before the judge suggesting that competitors charged an average of $1.26 per MB whereas Fido charged the appellant $6.14 and other respondents charged amounts ranging from multiples of 3.97 to 8.13 of that average. The table also records dramatic differences in amounts charged by respondents’ competitors for international roaming in France, where, for example, Vidéotron charges $5 per MB and Rogers and Fido charge $31.20. While the respondents’ rates were lower than those of their competitors in some countries outside of North America, the appellant did allege facts that the respondents fees appear to be very high against the prices of their competitors elsewhere. These figures are reproduced, as well, in the motion.
[80] The motion judge did not neglect these facts; instead, he examined them and concluded that the appellant had failed to disclose that Vidéotron charged higher prices outside of the U.S. and that it did not have a network that covered all of Canada, thereby making it a poor comparator. It may indeed turn out to be the case that some of the comparators raised by the appellant prove to be more or less compelling when the proof is adduced and fully examined at trial. But the cases make plain that, at this early stage, the motion judge is not charged with deciding if the action has been made out on the balance of probabilities. Here, the judge should have only asked whether the comparative table and the other materials submitted made an arguable case or, conversely, whether the proposed class action was “untenable”.[27]
[81] In dismissing the action at this early stage, the judge took into account grounds raised by the respondents to defend their prices against the allegation that they were disproportionately high as against the competition. By characterizing the allegations as imprecise and speculative, the judge mentioned, for example, the fact that the appellant had failed to consider that the respondents faced costs for access to international networks that some of their competitors did not face:
[90] Suffit-il par ailleurs d’invoquer les prix de revient ou de détail autorisés par la réglementation de l’Union européenne, alors que les réseaux des grands fournisseurs de services d’Europe, contrairement à ceux des intimées, couvrent le territoire de plusieurs pays, pour convaincre prima facie le Tribunal que les tarifs imposés aux usagers du Québec sont lésionnaires? […]
[94] Mais ce que la requête omet de dire, c’est que Vidéotron n’a pas de réseau national pan-canadien et que d’autres fournisseurs ne font pas affaire au Québec. Elle ne mentionne pas non plus que les frais d’itinérance de Vidéotron vers d’autres pays sont dans bien des cas supérieurs à ceux pratiqués par les intimées.
[82] Had the issue been whether or not, on the balance of probabilities, the claim that roaming fees were disproportionately high, the judge would have been entitled to weigh these defences. But at this stage, the fact that competitor Vidéotron does not have a pan-Canadian network does not make the class action a frivolous one nor does it render the claim less than arguable. Similarly, the fact that wireless service providers offer lower cost international roaming plans, and that appellant Sibiga declined to purchase one of those plans, was not decisive at this stage for the purpose of establishing an arguable case. The judge appears to have placed great importance on this fact, in particular the “decision” of the appellant to forgo the pre-paid plan (paras [12], [14] and [121]). The relative cost of these two options to the consumer may, when all the evidence is in, reveal the fees paid by the appellant and persons in her situation, to be more or less exploitative. It may well be right to raise it as a defence on the merits. But at the authorization stage, this fact does not neutralize evidence that, prima facie, suggests that the pay-per-use roaming fees are not disproportionate.
[83] By considering grounds of defence at this early stage, the judge thus trenched on the work of the trial judge. This Court has been clear in its direction to motion judges that the time to weigh such defences as against the allegations in the motion for authorization that are assumed to be true is, as a general rule, at trial.[28] Speaking of the defence of immunity that the Attorney General sought to raise at authorization in a class action in Carrier,[29] my colleague Guy Gagnon, J.A. wrote for the Court:
[37] Au moment de l'autorisation, alors que la suffisance de la preuve n'est appréciée que de manière prima facie, règle générale, il sera prématuré de conclure qu'une défense d'immunité s'applique en faveur de l'État. Ce qui n'est qu'un moyen de défense parmi d'autres, celui de l'immunité ici invoquée par l'intimé ne peut, lors de l'examen portant sur l'autorisation, être érigée au rang de moyen de non-recevabilité. À moins de convenir que la demande à sa face même est frivole, manifestement vouée à l'échec ou encore que les allégations de faits sont insuffisantes ou qu'il soit « incontestable » que le droit invoqué est mal fondé, il me paraît, outre ces circonstances, qu'il n'est pas souhaitable en début d'analyse de décider de la valeur absolue d'un tel moyen de défense.
[84] The judge had other evidence before him that was relevant to his measure of the prima facie case brought by the appellant. The judge was dismissive of the exhibit produced from a British publication, ZDNet UK, in which an executive at Three, a UK wireless service provider, declared that the underlying costs of international data transport bear no relation to retail rates. For the judge, the declaration was a dubious value because the appellant neglected to consider, as the article itself mentioned, that wireless service providers have significant infrastructure costs (para. [88] and [98]).
[85] In this instance again, the judge weighed the probative value of this evidence of low wholesale costs rather than simply asking if it constituted the basis for an arguable case. His criticism here does not relate to the vague or imprecise character of the allegations but rather to the evidentiary value of this exhibit. The judge was not wrong to read the document as a whole, nor was he mistaken to observe that the article raises an issue – the high costs of infrastructure necessary to support roaming services – that, for some observers, justifies high roaming rates charged to consumers. But the judge was mistaken to discount the exhibit simply because it expressed two divergent points of view. By denying any value to the opinion in the article that the rates are too high, he lost sight of the fact that the burden on the plaintiff in a class action is merely to establish an arguable case. It is at trial, not at the authorization stage, where the evidentiary debate as to whether infrastructure costs render the prices charged for roaming fees exploitative or not will take place. The infrastructure argument is another ground of defence but, here again, the measure of importance in establishing objective lesion should have been postponed to a later stage. As my colleague Bélanger, J.A., wrote recently: “si, par malheur, le juge de l’autorisation se retrouve devant des faits contradictoires, il doit faire prévaloir le principe général qui est de tenir pour avérés ceux de la requête pour autorisation, sauf s’ils apparaissent invraisemblables ou manifestement inexacts.”[30]
[86] It appears to me to be clear that the judge required more than an arguable case of objective lesion. He engaged the motion and its supporting evidence on the merits in concluding that, at this early stage, the facts alleged did not seem to support the conclusions sought. This amounts to a refusal to apply the Infineon standard to the interpretation of article 1003(b) which, in my respectful view, amounts to an error of law.
IV.2.d Did the judge err in evaluating the evidence of a prima facie case?
[87] As noted, the judge concluded that the appellant had failed to produce any tangible fact establishing prima facie proof of objective lesion.
[88] In my respectful view, the judge set aside elements of the record that allowed – indeed required – the court to conclude that an arguable case had been established that international roaming fees were exploitative and abusive.
[89] I disagree with the judge that there was no evidence that allowed him to discern, at least at this stage, the arguable case because of an absence of evidence about the costs incurred by merchants for providing international roaming services. The prices charged by service providers for discounted packages do suggest, in themselves, that amounts far less than $5 per MB for roaming charges remained a viable business model for the respondents. Paragraphs 2.56 to 2.60 of the motion assert that Fido offers consumers daily or monthly U.S. and international roaming “passes”, as attested by the pages of the company’s websites filed as an exhibit. The rates in the U.S. charged to consumers range from $0.20 to $1.50 per MB, substantially less than what the appellant paid and the $5 threshold in the class definition. Similarly, the motion alleges that plans for roaming outside the U.S. and Canada offer prices to consumers that are anywhere from 5.2 to 10.4 times less expensive than the standard rates.
[90] The substantial difference in pricing between pay-per-use roaming and roaming under the pre-paid plan is suggestive, at least for the purposes of applying the Infineon standard, that the respondents have charged consumers who use the pay-per-use option an exploitative amount. Let me explain by reference to the appellant’s situation. We know she paid $6.14 per MB for roaming in the U.S. over about a week. The cost to her was $250.81, but we do not know how much the base cost was to Fido. We do know, however, according to the excerpts of the Fido webpages, that had she purchased a 31-day plan for a fixed fee of $30, the effective roaming rate would have been $1.50 per MB, and had she bought a plan for $50, the rate would have been just 50 cents per MB. Had she paid $100, the rate would have been only 20 cents per MB.
[91] That these options are part of a viable business model for Fido raises the inference – at least for the purposes of the prima facie case test – that $6.14 per MB on a pay-per-use basis is exploitative for the appellant. The same reasoning would apply to other consumers in the class who paid an amount “in excess of $5” would be too. Contrary to what the judge wrote, this inference provides a tangible basis for concluding that the standard rates charged to consumers above $5 are objectively lesionary.
[92] The evidence in the file of plans (or “forfaits”) at which the service providers offered international roaming fees for a fraction of the amount charged to the appellant is thus prima facie evidence that the market value of this service was far lower that what she paid. There may be an explanation – the discounted plans may be loss leaders, the profits on upfront fees may offset costs over a large group, or plans may make better business sense in some parts of the world as opposed to others – but these defences, among others, should be measured at trial. At this stage, it was enough to say that a prima facie case was made out, recalling that liability is not decided at authorization.
[93] The failure to draw this inference reflects a palpable error. When considered alongside the prima facie evidence raised by the appellant’s other exhibits, notably the OEDC report, newspaper articles and press releases, I am satisfied that the requirement of article 1003(b) has been met here.
[94] Furthermore, it appears the judge also failed to consider the whole of the record before him. While he rightly mentioned, at paragraph [85] of his reasons, the existence of a declaration of a Telus official “affirmant que les fournisseurs de services pourraient aisément réduire les frais d’itinérance internationale de moitié et faire tout de même des profits”, he failed to explain why this did not, in itself or alongside other evidence, constitute prima facie evidence of objective lesion.
[95] The judge also misinterpreted, in my respectful view, the conclusions sought in the motion for authorization in respect of the OEDC report when he wrote, at paragraph [82], that “le Tribunal, à ce stade-ci, ne peut conclure non plus que le Canada, selon le rapport d’un groupe de travail de l’OEDC, présente les tarifs d’itinérance les plus élevés.” While this assessment may be true, and could possibly be relevant to a defence at trial, it is not the issue raised in the motion which is whether “the Respondents have charged and continue to charge Quebec consumers international mobile data roaming rates that are clearly disproportionate, exploitative and abusive”. In answering this question, it is not necessary to prove that Canadian service providers have the highest rates, simply that their rates are disproportionate against true costs.
[96] Given the tenor of the motion and of the exhibits, it was imprudent and indeed mistaken to dismiss the consumer class action based on article 1003(b).
IV.3 Adequate representation of the class (art. 1003(d) C.C.P.)
[97] Article 1003(d) C.C.P. directs that the member seeking the status of representative be “in a position to represent the class adequately / en mesure d’assurer une représentation adéquate des membres”. As the judge correctly observed, this is generally said to require the consideration of three factors: a petitioner’s interest in the suit, his or her qualifications as a representative, and an absence of conflict with the other class members.[31] These factors should, says the Supreme Court, be interpreted liberally: “No proposed representative should be excluded unless his or her interest or qualifications is such that the case could not possibly proceed fairly”.[32]
[98] The judge found that the appellant was not in a position to represent members of the class adequately for two reasons. First, she had an insufficient interest in the suit because of the lead taken by counsel in planning and instituting the class action. The judge read the reference to adequate representation in article 1003(d) as an indication by the legislature that the role of the representative must be “au-delà de la simple figuration” (para. [140]) and more than a “simple présence passive” (para. [148]). The lawyers recruited the appellant and, according to the judge’s view of things, she would not have any meaningful authority to withdraw their mandate if she lost confidence in their conduct of the case. Secondly, he found that the appellant lacked competence to act as representative for the class as a whole. Her testimony during the examination on discovery indicated that she has an insufficient understanding of the class action that had been instituted in her name. The judge made special mention of one of her answers suggesting she did not understand the calculation of $5 per MB charged that is the basic measure for membership in the class.
[99] The appellant argues that the judge misapplied the law in respect of article 1003(d), in particular by adopting too narrow an interpretation of the interest and competence requirements identified by the Supreme Court. He was mistaken to conclude that she did not have the interest to represent members of the class by placing too much emphasis on the role the lawyers had in initiating this consumer class action. Second, he is said to have erred in respect of the competence criterion by requiring of her a degree of understanding of the basis for her case that was both unrealistic and unnecessary in connection with a complicated consumer class action.
[100] I agree with the appellant.
[101] The lead role taken by counsel and the circumstances in which the appellant was recruited to represent the class are not incompatible with her status as representative.
[102] While it is not inappropriate to be mindful of possible excesses of what some have described as “entrepreneurial lawyering” in class actions, it is best to recognize that lawyer-initiated proceedings are not just inevitable, given the costs involved, but can also represent a social good in the consumer class action setting. As Perrell J. wrote in one Ontario case, “the entrepreneurial nature of a class proceeding can be a good thing because it may be the vehicle for access to justice, judicial economy, and behaviour modification, which are all the driving policy goals of the Class Proceedings Act, 1992”.[33] Scholars have observed that, within the proper limits of ethical rules that bind all lawyers, courts should recognize that lawyer-initiated consumer class actions can be helpful to meet the access to justice policy goals of the modern law of civil procedure.[34] In my view, the fact that lawyers play an important, even primary role in instituting a consumer class action is not in itself a bar to finding that the designated representative has the requisite interest in the suit.[35] Where the personal stake of a consumer representative is small – here, the appellant was charged $250.81 for roaming, of which only a portion is alleged to be overpayment – it is often unrealistic to insist upon a consumer-initiated class action.
[103] A lawyer-initiated consumer class action is not inherently incompatible with an acceptable solicitor-client relationship, nor does it mean that the client has “no control” over counsel. Article 1049 C.C.P. requires that a lawyer act for the representative. In our case, the appellant retains the authority to walk away from the class action, with permission of the court, and the lawyers cannot unilaterally “dismiss” the client as representative of the class. The judge was wrong to suggest that the fact that the lawyers chose their client here means that the appellant is an inadequate representative. As my colleague Dufresne, J.A. wrote in Fortier[36]:
[147] Cela dit, les juges peuvent déceler, à l’occasion, des indices qui laissent croire que les démarches ayant donné naissance à la requête portent fortement l’empreinte des avocats, mais cela ne discrédite pas nécessairement celui ou celle qui fait valoir une cause d’action qui apparaît suffisamment sérieuse alors que, sans lui, le groupe serait privé de l’exercice d’un droit.
[104] Nothing in the record suggests that the appellant is not a genuine claimant and nothing suggests unethical conduct on the part of her counsel, either in the “investigative” stage of the case or after proceedings were instituted. I see nothing that would disqualify her by reason of the implication of her lawyers. In my view, denying her that status for that reason appears to contradict the policy basis upon which class actions are founded. If lawyers’ role is to be reconfigured in this setting, it strikes me that article 1003(d), as drafted, is not a sound basis for achieving that end.[37]
[105] Is the appellant competent to represent other members of the class adequately?
[106] The judge was harsh in his evaluation of the appellant’s comprehension of the class action. She misunderstood “un élément capital du syllogisme élaboré par les avocats” in that she did not grasp the means of calculating the $5 per MB threshold for membership in the class action (para. [155]). For the judge, the appellant’s mistake on this point “touche à l’essence” of the class action, and signalled that she did not understand “le raisonnement développé par les avocats au dossier” (para. [157]). She could not therefore offer adequate representation to members of the class.
[107] Here again, respectfully stated, I find myself unable to agree with the judge.
[108] It is best to recognize, as does the appellant herself in written argument, that she may not have a perfect sense of the intricacies of the class action. This is not, however, what the law requires. As one author observed, Quebec rules are less strict in this regard that certain other jurisdictions: not only does the petitioner not have to be typical of other class members, but courts have held that he or she “need not be perfect, ideal or even particularly assiduous”.[38] A representative need not single-handedly master the finery of the proceedings and exhibits filed in support of a class action. When considered in light of recent Supreme Court decisions where issues were equally if not more complicated, this is undoubtedly correct: in Infineon, for example, the consumer was considered a competent representative to understand the basis of a claim for indirect harm caused down the chain of acquisition for the sale of computer memory hotly debated by the economists; in Vivendi, the issue turned on the unilateral change by the insurer of in calculations of health insurance benefits to retirees and their surviving spouses; in Marcotte, the debate centered on currency conversion charges imposed by credit card issuers. It would be unrealistic to require that the representative have a perfect understanding of such issues when he or she is assisted, perforce, by counsel and, generally speaking, expert reports will eventually be in the record to substantiate calculations of what constitutes exploitative roaming fees.
[109] To my mind, this reading of article 1003(d) makes particular sense in respect of a consumer class action. Mindful of the vocation of the class action as a tool for access to justice, Professor Lafond has written that too stringent a measure of representative competence would defeat the purpose of consumer class actions.[39] After reviewing the law on this point, my colleague Bélanger, J.A. observed in Lévesque v. Vidéotron, s.e.n.c.,[40] a consumer class action, that article 1003(d) does not impose an onerous burden to show the adequate character of representation: “[c]e faisant, la Cour suprême envoie un message plutôt clair quant au niveau de compétence requis pour être nommé représentant. Le critère est devenu minimaliste”. In Jasmin v. Société des alcools du Québec,[41] another consumer action, Dufresne, J.A. alluded to the Infineon standard and warned against evaluations of the adequacy of representation that are too onerous or too harsh, echoing an idea also spoken to by legal scholars.[42]
[110] In keeping with the “liberal approach” to the interpretation of article 1003(d), especially suited with the consumer class action, it suffices here that the appellant understand, as she has alleged, that she was billed a disproportionate amount for roaming because of the unfair difference between the amount charged and the real cost of the service to the respondent Fido. She must know that, like herself, others in the class, whether roaming in the U.S. or elsewhere, were also disproportionately billed, either with her own service provider or others who offer like services to Quebecers. She of course must see that her claim raises common questions with others in the class and that she is prepared to represent their interest and her own going forward.
[111] The examination before plea suggests that she had this understanding:
Q. [12] O.k. And when you say «the reason for being here», can you expand on that a little?
A. Sure. Well, basically what has happened to me and what, I know that it happens quite often to a lot of people, is that I was charged very a large sum of money for roaming fees when I went away for a week to the United States last year and I basically, I mean, you know, I've used my phone a certain way, I knew that I was going to, you know, get charged a little bit but when I received my bill, it was, I was extremely shocked to see, you know, the actual, the bill that I was expected to pay. And this is why, when I heard from Bruce approached me about this, that's why I immediately came forward because it was, I think a lot of people basically have this situation happening to them every month. So that's why I'm here.
[112] In addition, the appellant’s testimony indicates that she has a clear understanding of wireless services and international roaming data. She understands too that in order to succeed, she will have to establish the cost of roaming services and is prepared to join counsel in making efforts to obtain this information.
[113] In my respectful view, the judge failed to apply the liberal standard called for by the Supreme Court, both by misapprehending the consequences of counsel’s initiatives and by requiring a level of understanding of the claim that is too harsh for a consumer class action. This is not an instance in which the adequacy of the representative is compromised in a manner that, to revert to the Infineon standard quoted above, “could not possibly proceed fairly”. Indeed, neither the judge nor the respondents in their arguments on appeal advance any serious suggestion that the fairness of the class action was threatened by the recognition of the appellant as class representative. Moreover, if ever the appellant were considered to no longer be in a position to represent the class members properly, the law provides a mechanism whereby she could be replaced by another member of the class at a later stage in the proceedings.[43]
[114] The judge’s finding that article 1003(d) was not satisfied must be set aside.
[115] As a final point, counsel for the respondents argued that given the change in the law relating to standing since Marcotte, the rules on adequate representation in article 1003(d) should be more strictly enforced. In service of this argument, they point to dicta in the judgment of this Court in Marcotte where Dalphond, J.A. suggested that article 1003(d) stood as a protection against unmanageable or unfounded class actions against unconnected defendants.[44] Indeed, one might argue that the adequacy of representation, as well as the common question requirement, might prove to be especially important on the facts of a given case where there are members of the class who, unlike the representative, have no direct cause of action against one or another defendant. But a new reading of articles 1003(a) and 1003(d) C.C.P. cannot be proposed in a manner that would revive the standing debate that Marcotte has put to rest. It might also be recalled in this context that Quebec does not have a typicality test for the representative, and that article 1003(d) should not be interpreted to create one.[45] What is important, in the present case, is that the appellant plainly understood the allegation that, like her, consumers with other service providers paid for that service at unfair rates. And as we shall see in the next section, the common question requirement was met for all members of the class, including those with Telus or Bell contracts.
IV.4 Common questions (article 1003(a) C.C.P.)
[116] For a class action to be authorized, article 1003(a) C.C.P. requires that the recourses of the members raise “identical, similar or related questions of law or fact / questions de droit ou de fait identiques, similaires ou connexes”.
[117] The judge held that the requirement set out in article 1003(a) was not met. He considered the matter in the same section of his reasons in which he decided that the appellant was not an adequate representative for the class and, in particular, that she did not have sufficient interest to sue Telus and Bell, with whom she had no contract.
[118] The judge reasoned that the vast and variable range of services offered by the respondents in their respective contracts with consumers precluded a finding that a client of Fido, like the appellant, is in the same position as a client of Telus or Bell (para. [131]). He noted that it would be possible for roaming prices in one contractual setting to be disproportionately high whereas in another setting, where contractual services differed, that price would be acceptable (para. [135]). Given that the appellant had not shown the tenor of the contractual obligations between Telus and its clients and Bell and its clients, it was impossible to measure the proportionate or disproportionate character of the parties’ contractual prestations. The judge drew the conclusion that it was therefore inherently impossible to identify identical, similar or related questions of law or fact relevant to all members of the class.
[119] The appellant says the judge misread the relevant Supreme Court cases bearing on article 1003(a). He failed to consider that the identification of one common question that would serve to advance the class action was all that was needed at authorization.
[120] The respondents argue that the judge made no mistake. Their different contractual arrangements with consumers meant that international roaming services were offered with such different permutations that no common question emerges based on the record before the judge. Telus adds that the questions could not possibly be common to its clients because the company charged its clients $5 per MB for roaming in the U.S. and the class was limited to Quebec consumers who paid more than $5.
[121] With respect for the contrary view, I agree with the appellant that the judge did not interpret article 1003(a) in accordance with the standard established in Infineon and Vivendi and that this led him to err in finding that the requirement was not met.
[122] In Vivendi,[46] LeBel and Wagner JJ. proposed a “flexible” approach to article 1003(a) according to which the identification of one common question would be sufficient. Their Lordships recognized that variation might exist within the class and that this was not a bar to meeting the common question requirement. Drawing on decided cases in Quebec, they wrote:
[58] […] To meet the commonality requirement of art. 1003(a) C.C.P., the applicant must show that an aspect of the case lends itself to a collective decision and that once a decision has been reached on that aspect, the parties will have resolved a not insignificant portion of the dispute. […] All that is needed in order to meet the requirement of art. 1003(a) C.C.P. is therefore that there be an identical, related or similar question of law or fact, unless that question would play only an insignificant role in the outcome of the class action. It is not necessary that the question make a complete resolution of the case possible.
[123] The judge did not apply this test of a single, significant common question but focussed instead on what he presumed to be disparate contractual arrangement amongst members of the class that, he wrote, precluded him on finding commonality. Again in Vivendi,[47] the Supreme Court warned against this kind of analysis that risks overemphasizing variation between members of the class and losing sight of one or more common questions that will advance the class action. Moreover in Infineon, the Court held that it is not necessary that the member of the class be in the same situation but that it is enough that they be in a sufficiently similar situation such that a common question for which the class action seeks answers can be identified. “At the authorization stage” wrote the Supreme Court, “the threshold requirement for common questions is low”.[48]
[124] Does the motion for authorization propose one or more such common questions?
[125] Among the four principal questions presented by the appellant, at least one meets the above-mentioned standard in my view:
Does the disproportion between the international mobile data roaming fees charged to the class members and the value of the service provided by the Respondents constitute exploitation and objective lesion under [s.] 8 of the CPA?
[126] Broken down to its component parts, this question has two aspects:
· What is the value of international roaming services?
· From what point should charges for international roaming be considered disproportionate?
[127] The answer to the first of these questions turns in large measure on the wholesale price of the service which, as noted above, has not yet been precisely identified. This amount might be expected, of course, to vary from country to country, or at least from one international roaming zone to another. But the answer to this question is central to determining whether the difference between market value for roaming services and the amounts charged to consumers is objectively lesionary beyond the prima facie case made out at the authorization stage. Similarly, in respect of the amount charged to consumers, one can well expect the price of the service will also vary, at least potentially so, from one consumer package to another beyond the $5 per MB threshold already identified in the definition of the class. But again, the identification of the point from which the difference between cost paid and price charged for roaming service is exploitative, if at all, will advance the case for all members of the class.
[128] The resolution of this issue is a common one in that, to quote McLachlin C.J. in Dutton, “it is necessary to the resolution of each class member’s claim”.[49] Contrary to what the Superior Court decided, it is not fatal to the commonality of the question that class members are not identically situated vis-à-vis the respondents. Moreover, as this Court decided in Suroît,[50] in dicta taken up by the Supreme Court in Vivendi, the determination of common issues need not lead to the complete resolution of the case, and it could give rise instead to small trials at the stage of the individual settlement of the claims. That too is not a bar to finding that article 1003(a) has been satisfied where, if anything, the Quebec rules are more flexible than those in other provinces as was noted in Vivendi, and common questions need not give rise to common answers.
[129] In sum, applying the principles identified by the Supreme Court to the question identified in the motion for authorization, the judge committed a reviewable error in his interpretation of article 1003(a) and should have concluded that the appellant had established the existence of a common question that would advance the resolution of the dispute with respect to all the members of the class, and that would not play an insignificant role in the outcome of the case.
IV.5 The definition of the class
[130] As a subsidiary argument made in the event this Court sets aside the motion judge’s decision to deny authorization, certain of the respondents ask that the class be redefined pursuant to article 1005(a) C.C.P. Noting that the judge stated the class as defined in the motion includes Quebec consumers who are charged international roaming fees in potentially dozens of countries for which there is not an iota of proof alleged in the file, they ask for a narrower definition that better reflects concerns expressed by the judge about the unwieldy dimensions of the class and the proposed hearing of the action on the merits. The respondents recall that the person seeking authorization has the burden of demonstrating the appropriateness of the class as defined in the motion under article 1002 C.C.P. and, they say, that was not done here.
[131] The first judge did briefly note that the definition of the class in the motion was lacking. Near the end of his discussion of article 1003(b) C.C.P., he wrote that “[d]ans le cas présent, les faits allégués ne paraissent justifier ni la composition du groupe, ni les conclusions en lésion objective, ni celles voulant que les prix chargés aux abonnés par les intimés soient abusifs au sens du Code civil” (para. [122], emphasis added). He also stated, in the preceding paragraph, that the breadth of the motion would require the Superior Court to embark on the equivalent of a public inquiry involving the determination of the base costs of international roaming fees across the world, an inquiry for which the courts are both ill-equipped and without a legitimate mandate. Telus says the judge was right to conclude as he did: the class is geographically defined in a manner that is disproportionately wide. Moreover, there is no proper justification in fact for setting the threshold price for membership in the class at $5 per MB, a number arbitrarily chosen by counsel for the appellant.
[132] Telus and other respondents propose using the appellant’s own experience as a basis for limiting the class to roaming fees incurred by Quebecers in the U.S., and not in other countries. At the hearing, a further suggestion was considered in setting the threshold price at $6.14 per MB – what the appellant was charged and actually paid – rather than the $5 amount chosen by her lawyers.
[133] Whatever the amount charged, Telus’ interest in limiting the class to consumers who paid roaming fees in the U.S. is plain. Because Telus did not charge more than $5 per MB for roaming in the U.S., this would entitle the company to be excluded as a defendant given the definition proposed by the appellant (“[…] at a rate higher than $5 per megabyte”).
[134] How should the class be defined here?
[135] First, this Court is not bound by the judge’s finding that the facts as alleged do not justify the definition of the class proposed in the motion for authorization. Article 1005(a) C.C.P. only calls on a motion judge to describe the group in the event that authorization is granted. The judge denied authorization here. His comment in paragraph [122] is thus of the order of an obiter dictum.
[136] It is of course true that a motion judge may, as an alternative to denying authorization, redefine the class so that its dimensions are better aligned with the claim as framed by the applicant.[51] As Professor Lafond has had occasion to observe, the exercise of redefining the class on a narrower basis can be preferable to denying authorization altogether given the policy favouring access to justice.[52]
[137] The burden of showing the class to be of the proper size is generally said not to be a heavy one. In Hollick,[53] Chief Justice McLachlin wrote that there must be a rational link between the common questions and the class as identified in the motion. She added that it must be shown that “the class is not unnecessarily broad – that is, that the class could not be defined more narrowly without arbitrarily excluding some people who share the same interest in the resolution of the common issue”. Where the class could be defined more narrowly, wrote the Chief Justice, the motion judge should either disallow certification or allow certification on condition that the definition of the class be amended. Importantly, the class can be redefined in Quebec law not just at authorization, but at later stages in the process as well.[54]
[138] Quebec courts have developed rules for understanding the appropriate definition of the class: the definition must be founded on objective criteria with a rational foundation; the definition of the class must not be circular or imprecise; and it cannot be based on criteria that are dependent on the outcome of the action on the merits.[55]
[139] Is the class as proposed unnecessarily broad and should it be recast here?
[140] There is a temptation in this case to narrow the definition of the class to respond, at this early stage, to objections raised by the judge in respect of the vast inquiry necessary, across many countries, and his doubts that the appellant had a plain understanding of the situation of members of the class in circumstances other than her own. In other words, one might argue that in order to meet proportionality concerns, a class action should only be authorized, in respect of criteria set forth in articles 1003(b), 1003(d) and 1003(a), if the scope of the class is curtailed.
[141] By the same token, at the authorization stage, it seems to me that one should exercise caution before limiting the dimension of the class as stated by the applicant. After all, the consequence of excluding members of the class at this early stage is a serious one. For the reasons that follow, I am of the view that, at this stage, limiting the class to consumers who were charged roaming fees in the U.S. only would run the risk of arbitrarily excluding other potentially legitimate claimants from compensation.
[142] It is true that the evidence in the file, at present, is not robust as to the number of travellers or the costs to the respondents in respect of the different places in which Quebec consumers are said to have paid exploitative or abusive roaming fees.
[143] The appellant has not brought direct proof that Quebec consumers other than herself have been charged disproportionate international roaming fees in the U.S. or elsewhere. In support of the geographically broad class proposed in the motion, the appellant asked the motion judge to infer, based on travel statistics, that other Quebecers using mobile devices internationally also paid roaming fees. She did file a 2010 Statistics Canada report regarding international travel by Canadians in 2010. According to the report, Canadian residents took 45 million “person-trips” to the U.S. that year, of which about 5 million were by Quebecers. Some 545,000 Quebec person-trips were made to Europe and 834,000 to the Caribbean. The motion specifically invites an inference from these statistics that, beyond the appellant’s own experience, a number of these travelling consumers used mobile devices and were charged improper roaming fees (see paras 3.1 to 3.3 of the motion). This strikes me as a reasonable inference at this stage of the proceedings.
[144] Limiting the class to consumers who travelled only to the U.S. would, at this stage, be an arbitrary exercise of discretion. The prima facie case that Quebecers travelling to Europe or elsewhere were improperly charged roaming fees is based on the same reasoning that suggests that travellers to the U.S. other than the appellant were improperly charged: the difference is only in the numbers that, in the case of the U.S., makes the inference more probable. Moreover there seems to be no logical reason why consumers should have to introduce separate class actions for each country in which roaming charges were levied – the respondents seem to suggest, in their written argument, that this is the only way forward. This position strikes me as contrary to the application of the principle of proportionality that a judge is called on to undertake based on article 1003 C.C.P.
[145] Additionally, I note that the respondent wireless service providers themselves do not organize billing for international roaming fees only on a country-by-country basis. Each of Fido, Bell and Telus divide the world into three or four “zones” for this purpose, with each zone regrouping different countries that the service providers, it would appear, consider to be similarly situated for the purposes of fixing roaming fees. The spectre of an unmanageable class action involving far flung Quebec consumers in “200 countries” raised by Telus in its factum may in fact materialize as one class with three or four subgroups based on these zones, a model that has proved viable in other circumstances. And if, at a later date, the judge charged with the action post-authorization does find that there is, for example, insufficient commonality amongst members, he or she has the power at that time to redefine the class.
[146] It would be equally imprudent in my view to follow the argument that the class should be modelled on consumers who paid the same amount for roaming fees as the appellant. While it is true that the $5 per MB proposed threshold is not anchored in her own experience, paragraphs 2.67 to 2.70 of the motion for authorization explain how the appellant and her counsel based this proposal on an investigation that took into account proportionality. The fact that the appellant was unable to explain the $5 amount correctly when examined on discovery does not, in my view, justify redefining the class at $6.14, a number that she understood to be exploitative based on her own experience. As noted above, the key point in our case is that she understood that other consumers, including those that paid less than her, would be in a similar position as she was to make a claim against their service providers based on unfair pricing. While it is possible that a representative’s weak understanding of the dimensions of the class be a relevant consideration for changing its definition, that is not the case here. Excluding consumers who paid between $5 and $6.14 per MB for roaming services would be arbitrary given the presence of a sufficient common question in the motion.
[147] Two further factors suggest that modifying the class would be inappropriate.
[148] First, I think it would be wrong for the respondents to trade on their informational advantage against consumers to secure a narrowing of the class at this stage of the proceedings. I noted above that the appellant has the burden of proof under article 1003 C.C.P., and it is true as well that she has the burden of justifying the description of the class under article 1002. But the respondents have information about the costs they face to provide roaming fees to Quebec consumers in the various parts of the world relevant to the class, as well as, presumably, a clear picture of how many consumers are involved and what those consumers paid themselves for the services. They have not been obliged, as yet, to share this material with the appellant, but they should not be permitted, at the same time, to complain that it would be unfair to them to approve the class because it is overbroad. The comments of Bélanger, J.A. in Lévesque[56] were made in circumstances sufficiently similar to this case to be a helpful guide here:
[28] Le recours proposé ici a ceci de particulier que l’on peut présumer que les intimées possèdent toutes les données nécessaires à l’estimation du nombre d’abonnés concernés par le recours, ainsi que le nombre de locations de « Films pour adultes – Torride » effectuées par ces derniers. En effet, les relevés mensuels produits par l’appelant indiquent, de façon précise, la journée et l’heure de la location de chacun de ces films. Le tarif de 10,99 $ correspondant, à ce que j’en comprends, au montant facturé pour cette catégorie de films.
[29] De plus, nous pouvons raisonnablement présumer qu’un certain pourcentage des quelque 1,8 million d’abonnés des intimées, quoique, inconnu à ce moment, loue des « Films pour adultes - Torride ». C’est le constat qu’a fait la juge et elle a eu raison de le faire. Dans ce contexte très précis, l’identification d’autres membres potentiels ou encore d’une approximation quant à leur nombre devient secondaire. Par ailleurs, les intimées possèdent ces informations et elles étaient en mesure d’apporter des précisions quant à leur nombre au moment de la demande d’autorisation, si elles estimaient que ces données pouvaient être pertinentes au rejet de la demande.
[149] Additionally, an overly strict approach to the definition might serve to undermine the liberal approach that the Supreme Court had advised for interpreting the requirements for authorization of class actions in Vivendi and Infineon. I am struck by the fact that at paragraph [73] of the latter case, LeBel and Wagner JJ. cited with approval observations made by this Court in Guilbert v. Vacances sans Frontière Ltée,[57] regarding the inappropriateness of narrowing the class in a consumer class action where common questions amongst members are judged to be sufficient:
[translation] The fact that the situations of all members of the group are not perfectly identical does not mean that the group does not exist or is not uniform. To be excessively rigorous in defining the group would render the action useless . . . in situations in which claims are often modest, there are many claimants and dealing with cases on an individual basis would be difficult. [p. 517]
[150] While I cannot exclude the possibility that the class will be reconfigured later in these proceedings, it would be arbitrary to do so at this stage.
***
[151] In conclusion, and with renewed respect for the opposing view, I propose that the appeal be allowed, with legal costs, the judgment of the Superior Court be set aside and that the class action be authorized as stated in the appellant’s motion to institute proceedings.
NICHOLAS KASIRER, J.A.