SUPREME COURT OF CANADA

Date: 20131212 Docket: 34738

Between:

AIC Limited

Appellant

and

Dennis Fischer, Sheila Snyder, Lawrence Dykun, Ray Shugar and Wayne Dzeoba

Respondents

And Between:

CI Mutual Funds Inc.

Appellant

and

Dennis Fischer, Sheila Snyder, Lawrence Dykun, Ray Shugar and Wayne Dzeoba

Respondents

Coram: McLachlin C.J. and LeBel, Rothstein, Cromwell, Moldaver, Karakatsanis and Wagner JJ.

Reasons for Judgment: (paras. 1 to 66)

Cromwell J. (McLachlin C.J. and LeBel, Rothstein, Moldaver, Karakatsanis and Wagner JJ. concurring)

AIC Limited v. Fischer, 2013 SCC 69, [2013] 3 S.C.R. 949

AIC Limited Appellant

v.

Dennis Fischer,

Sheila Snyder, Lawrence Dykun, Ray Shugar and Wayne Dzeoba Respondents

‑ and ‑

CI Mutual Funds Inc. Appellant

v.

Dennis Fischer,

Sheila Snyder, Lawrence Dykun, Ray Shugar and Wayne Dzeoba Respondents

Indexed as: AIC Limited v. Fischer

2013 SCC 69

File No.: 34738.

2013: April 18; 2013: December 12.

Present: McLachlin C.J. and LeBel, Rothstein, Cromwell, Moldaver, Karakatsanis and Wagner JJ.

on appeal from the court of appeal for ontario

Civil procedure — Class actions — Certification — Market timing — Investors suing mutual fund managers for breaching fiduciary duties to investors and negligence for failing to curb market timing activities — Investors seeking certification of action as class proceeding under provincial class action legislation — Whether proposed investor class action meets preferability requirement for certification given settlement payments made to investors following proceedings conducted by Ontario Securities Commission — Class Proceedings Act, 1992, S.O. 1992, c. 6, s. 5(1) (d).

A group of mutual fund managers were the subject of an investigation conducted by the Ontario Securities Commission (“OSC”) into “market timing”, a practice which was alleged to have caused long‑term investors to suffer losses in the value of their investments. The fund managers ultimately entered into agreements with the OSC that paid investors millions in settlement. The settlement agreements anticipated and did not preclude the possibility of civil proceedings against the mutual fund managers. Following the settlement agreements, the investors applied to certify a class action against the fund managers relating to the same market timing conduct. The motion judge found that a class action was not a preferable procedure and denied certification. The Divisional Court reversed the motion judge and granted certification. The Court of Appeal upheld the Divisional Court’s result.

Held : The appeal should be dismissed.

The appeal focuses on one branch of the statutory requirement for certification, the requirement that “a class proceeding would be the preferable procedure for the resolution of the common issues”: Class Proceedings Act, 1992 , S.O. 1992, c. 6, s. 5(1)(d). The question is whether the proposed class proceeding, as compared to the non‑litigation OSC proceedings, is preferable from the point of view of providing access to justice. It is clear that the preferability requirement is broad enough to take into account all reasonably available means of resolving the class members’ claims including avenues of redress other than court actions. In assessing preferability, the court must look at the common issues in the context of the action as a whole. However, the court cannot engage in a detailed assessment of the merits or likely outcome of the class action or any alternatives to it. The party seeking certification of a class action must show some basis in fact for each of the certification requirements.

The preferability inquiry has to be conducted through the lens of the three principal goals of class actions, namely judicial economy, behaviour modification and access to justice, but the ultimate question is whether other available means of resolving the claim are preferable, not if a class action would fully achieve those goals. Access to justice is an important goal of class proceedings. Within the proper scope of the certification process, both substantive and procedural aspects must be assessed in determining whether a class action is the preferable procedure. A class action will serve the goal of access to justice if: (1) there are access to justice concerns that a class action could address; and (2) these concerns remain even when alternative avenues of redress are considered. To determine whether both of these elements are present, it may be helpful to address a series of questions. These questions must not be considered in isolation or in a specific order, but should inform the overall comparative analysis.

The first question is: what are the barriers to access to justice? There are two potential barriers to access to justice in this case. First, an economic barrier arises from the nature of the claim. The individual claims are not large enough to support viable individual actions. The second barrier is related to the first. As a result of the nature of the claim, there is potentially no access to a fair process, geared towards protecting the rights of class members, to seek a resolution of the common issues for what could potentially be a class of over a million members. Thus, traditional litigation cannot achieve either the substantive or the procedural dimensions of access to justice in a case such as this. Although generally the most common barrier to access to justice is an economic one; they can also be psychological or social in nature. A common procedural barrier is that there is no other procedure available to afford meaningful redress.

The second question is: what is the potential of the class proceedings to address those barriers? This analysis is not made in isolation, but within the comparative analysis, for the purpose of assessing the class proceedings’ potential to address the access to justice concerns in comparison to the alternative procedure’s ability to do so. Even though a class action is a procedural tool, achieving substantive results is one of its underlying goals. Consideration of its capacity to overcome barriers to access to justice should take account of both the procedural and substantive dimensions of access to justice. In this case, the proposed class action addresses both barriers. It has the potential to make it economically feasible to advance on behalf of the class a group of individual claims that would otherwise not be economically feasible to pursue in the courts and it provides class members with a fair process to resolve their claims.

The third question is: what are the alternatives to class proceedings? The motions court must look at all the alternatives globally in order to determine to what extent they address the barriers to access to justice posed by the particular claim. There is no realistic litigation alternative in this case. The only alternative procedure that was advanced is the OSC proceedings and settlement agreements, the results of which are already known.

The fourth question is: to what extent do the alternatives address the relevant barriers to access to justice? The question is whether the alternative has the potential to provide effective redress for the substance of the claims and to do so in a manner that accords suitable procedural rights.

The last question is: how do the two proceedings compare? In comparing the two proceedings, the motions court must determine whether, on the record before it, the class action has been shown to be the preferable procedure to address the specific procedural and substantive access to justice concerns in a case. The court must also, to the extent possible within the proper scope of the certification hearing, consider the costs as well as the benefits of the proposed class proceeding in relation to those of the proposed alternative procedure.

In answering the last two questions in this case, investor participation in the process leading to compensation is an important factor to consider and one that weighs heavily in favour of finding that the class proceeding meets the preferability requirement in this case. The regulatory nature of, and the limited participation rights for investors in the OSC proceedings, coupled with the absence of information about how the OSC staff assessed investor compensation support the conclusion that significant procedural access to justice concerns remain which the proposed class action can address. Moreover, the focus and nature of the OSC process reinforce the concerns about whether substantial access to justice was achieved. The substantive outcome of the OSC proceedings cannot be dismissed as irrelevant to the question of whether the OSC proceedings addressed the access to justice barrier that is present in this case or whether the way in which it did so suggests that the class proceeding is not the preferred alternative.

However, the substantive outcome of the OSC proceedings and their impact on the preferability analysis must be examined through the appropriate evidentiary lens. In the rather unusual circumstances of this case, where the OSC proceedings have run their course and the results of those proceedings are known, the comparative analysis cannot ignore the question of whether a cost‑benefit analysis supports the contention that the proposed class proceeding is the preferable way to address the claims here. The record in this case, which shows in detail the results of the completed proposed alternative proceedings, also shows that substantive access to justice concerns still remain. Further, there is no reason to believe that potential additional recovery would be consumed by the costs of the proceedings. The investors provided an appropriate basis to support the view that the class action proceeding would overcome access to justice barriers that subsisted after the completion of the OSC proceedings and that a cost‑benefit analysis supported the conclusion that the class proceedings were the preferable proceeding for the investors to pursue their claims. The motion judge in this case erred in principle in his analysis and this justified appellate intervention in his exercise of discretion to refuse certification. The correct legal principles support the decision to certify the proposed class action.

Cases Cited

Applied: Hollick v. Toronto (City) , 2001 SCC 68 , [2001] 3 S.C.R. 158; referred to: Klay v. Humana, Inc. , 382 F.3d 1241 (2004); Cloud v. Canada (Attorney General) (2004), 2004 45444 (ON CA) , 73 O.R. (3d) 401; Rumley v. British Columbia , 2001 SCC 69 , [2001] 3 S.C.R. 184; Webb v. K‑Mart Canada Ltd. (1999), 1999 15076 (ON SC) , 45 O.R. (3d) 389; Pearson v. Inco Ltd. (2006), 2006 913 (ON CA) , 78 O.R. (3d) 641; Halabi v. Becker Milk Co. (1998), 1998 14681 (ON SC) , 39 O.R. (3d) 153; Pro‑Sys Consultants Ltd. v. Microsoft Corporation , 2013 SCC 57 , [2013] 3 S.C.R. 477; Irving Paper Ltd. v. Atofina Chemicals Inc. (2009), 2009 92127 (ON SC) , 99 O.R. (3d) 358; Hague v. Liberty Mutual Insurance Co. (2004), 2004 95289 (ON SC) , 13 C.P.C. (6th) 1; McCracken v. Canadian National Railway Co. , 2012 ONCA 445 , 111 O.R. (3d) 745; 1176560 Ontario Ltd. v. Great Atlantic & Pacific Co. of Canada Ltd. (2002), 2002 6199 (ON SC) , 62 O.R. (3d) 535, aff’d (2004), 2004 16620 (ON SCDC) , 70 O.R. (3d) 182; Chadha v. Bayer Inc. (2003), 2003 35843 (ON CA) , 63 O.R. (3d) 22, leave to appeal refused, [2003] 2 S.C.R. vi; Caputo v. Imperial Tobacco Ltd. (2004), 2004 24753 (ON SC) , 236 D.L.R. (4th) 348; Committee for the Equal Treatment of Asbestos Minority Shareholders v. Ontario (Securities Commission) , 2001 SCC 37 , [2001] 2 S.C.R. 132; Markson v. MBNA Canada Bank , 2007 ONCA 334 , 85 O.R. (3d) 321; Cassano v. Toronto‑Dominion Bank , 2007 ONCA 781 , 87 O.R. (3d) 401, leave to appeal refused, [2008] 1 S.C.R. xiv.

Statutes and Regulations Cited

Class Proceedings Act , R.S.B.C. 1996, c. 50.

Class Proceedings Act, 1992 , S.O. 1992, c. 6, s. 5(1).

Federal Rules of Civil Procedure , 28 U.S.C. app., r. 23(b)(3).

Securities Act , R.S.O. 1990, c. S.5, s. 127.

Authors Cited

Cullity, Maurice. “Certification in Class Proceedings — The Curious Requirement of ‘Some Basis in Fact’” (2011), 51 Can. Bus. L.J. 407.

Good, Mathew. “Access to Justice, Judicial Economy, and Behaviour Modification: Exploring the Goals of Canadian Class Actions” (2009), 47 Alta. L. Rev. 185.

Iacobucci, Frank. “What Is Access to Justice in the Context of Class Actions?”, in Jasminka Kalajdzic, ed., Accessing Justice: Appraising Class Actions Ten Years After Dutton, Hollick & Rumley . Markham, Ont.: LexisNexis, 2011, 17.

Ontario. Attorney General’s Advisory Committee on Class Action Reform. Report of the Attorney General’s Advisory Committee on Class Action Reform . Toronto: The Committee, 1990.

Ontario. Law Reform Commission. Report on Class Actions , vol. I. Toronto: Ministry of the Attorney General, 1982.

Rubenstein, William B. Newberg on Class Actions , 5th ed. Eagan, Minn.: West, 2011 (WL, CLASSACT, updated June 2013).

APPEAL from a judgment of the Ontario Court of Appeal (Winkler C.J.O., Epstein J.A. and Pardu J. ( ad hoc )), 2012 ONCA 47 , 109 O.R. (3d) 498, 287 O.A.C. 148, 346 D.L.R. (4th) 598, 15 C.P.C. (7th) 81, [2012] O.J. No. 343 (QL), 2012 CarswellOnt 635, affirming a decision of Molloy, Swinton and Herman JJ., 2011 ONSC 292 , 104 O.R. (3d) 615, 276 O.A.C. 84, 6 C.P.C. (7th) 139, [2011] O.J. No. 562 (QL), 2011 CarswellOnt 800, setting aside in part a decision of Perell J., 2010 ONSC 296 , 89 C.P.C. (6th) 205, [2010] O.J. No. 112 (QL), 2010 CarswellOnt 135. Appeal dismissed.

James D. G. Douglas , David Di Paolo and Margot Finley , for the appellant AIC Limited.

Benjamin Zarnett , Jessica Kimmel and Melanie Ouanounou , for the appellant CI Mutual Funds Inc .

Allan C. Hutchinson , Peter R. Jervis , Joel P. Rochon and Remissa Hirji , for the respondents.

The judgment of the Court was delivered by

Cromwell J. —

I. Overview

[ 1 ] In order to have a proposed class action certified, the plaintiff must show that there is some basis in fact to conclude that a class proceeding would be the preferable procedure for resolution of the common issues raised in the action: Class Proceedings Act, 1992 , S.O. 1992, c. 6 (“ CPA ”), s. 5(1) (d). The main question on appeal in this proposed investor class action is whether it meets this preferability requirement given that settlement payments were made to investors following proceedings conducted by the Ontario Securities Commission (“OSC”).

[ 2 ] The issue is a thorny one: each of the three levels of court in the proceedings leading to this appeal adopted significantly different approaches. The motion judge found that a class action was not a preferable procedure and denied certification. Given the OSC’s restitutionary mandate, he held that it was not for the court to second-guess the access to justice provided to investors through the settlements or to give much weight to the difference between the mandate and processes of the OSC compared to the courts. The Divisional Court reversed the motion judge and granted certification. Its analysis focused on the level of recovery in the regulatory proceeding as compared to the quantum of damages claimed in the class action. The comparison led the court to conclude that substantial recovery could still be achieved by way of the class action. The OSC proceedings could therefore not be preferable to the proposed class action. The Court of Appeal upheld the Divisional Court’s result, but for substantially different reasons. The Court of Appeal focused on comparing the class members’ procedural rights in class proceedings with the regulatory nature and limited participatory rights of investors in the OSC proceedings. As the Court of Appeal put it, the preferable procedure inquiry must “focus on the underlying purpose and nature of the alternative proceeding as compared with the class proceeding. . . . The CPA mandates that this must be a procedural discussion”: 2012 ONCA 47 , 109 O.R. (3d) 498, at para. 79 .

[ 3 ] The focus of this appeal is on the question of whether the proposed class proceeding, as compared to the OSC proceedings, is preferable from the point of view of providing access to justice. The case provides an opportunity for this Court to elaborate the analytical approach to this question under the CPA , building on the Court’s judgment in Hollick v. Toronto (City) , 2001 SCC 68 , [2001] 3 S.C.R. 158.

[ 4 ] I agree with the Divisional Court and the Court of Appeal that the motion judge erred in principle in his analysis and that this justified appellate intervention in his exercise of discretion to refuse certification. As I see it, the correct legal principles support those courts’ decision to certify the proposed class action. I would therefore dismiss the appeal. However, for reasons that I will develop, I also conclude that the preferability analysis is not solely focused on procedural considerations but must, within the proper scope of the certification process, consider both substantive and procedural aspects.

II. Facts and Proceedings

A. Overview of the Facts

[ 5 ] The appellants are two of the mutual fund managers who were the subject of an investigation conducted by the OSC into “market timing”, a practice which was alleged to have caused long-term investors to suffer losses in the value of their investments. The OSC probe focused on whether the fund managers had taken reasonable steps to protect the funds from harm that could arise from frequent trading market timing, rather than on the “market timers” whose activities directly caused harm to the fund. The fund managers ultimately entered into settlement agreements with the OSC. The appellants AIC Limited and CI Mutual Funds Inc. paid respectively $58.8 million and $49.3 million to their investors as a result of those agreements: motion decision, at para. 94. The settlement agreements anticipated and did not preclude the possibility of civil proceedings against the appellants. The investors who received payments from the appellants are more or less the same people who would ultimately form the class in this proposed class action against the appellants.

[ 6 ] Following the settlement agreements, the respondents applied to certify a class action against the fund managers relating to the same market timing conduct. The action alleged, among other things, that the managers had breached their fiduciary duty to investors and had been negligent by failing to take steps to curb market timing activities. (Although the proposed class action originally named five fund managers as defendants, the claims against three of them have settled so that the appellants in this Court are the only remaining defendants.)

B. Decision on Certification Motion: Ontario Superior Court of Justice, 2010 ONSC 296 , 89 C.P.C. (6th) 205 (Perell J.)

[ 7 ] The judge identified the common issues raised by the proposed class proceeding as being whether the defendants (including the appellants in this Court) had fiduciary duties or duties of care to the proposed class members, and if so whether they had breached those duties. It was common ground on the certification motion that the OSC proceedings and settlement agreements did not bar the claims advanced in the action.

[ 8 ] The main contested issue in the motion was whether the proposed action would be the preferable procedure to resolve the common issues having regard to the purposes of class proceedings: judicial economy, behaviour modification and access to justice. The judge noted that if the class action were the preferable procedure, it should be certified conditional on the court approving a litigation plan.

[ 9 ] The judge focused his preferability analysis on access to justice. There was no dispute that the OSC proceedings and settlement agreements had achieved the goal of behaviour modification, and the judge concluded that if they had achieved access to justice, then they had also served the goal of judicial economy. The motion judge based his conclusion that the proposed class action was not the preferable procedure solely on the existence of the OSC proceedings and settlement agreements. This conclusion followed from two key findings: first, that the OSC proceedings and settlement agreements were properly part of the preferable procedure analysis; and second, that the OSC proceedings were a “genuine alternative that serve[d] the purposes of a class proceeding; namely, access to justice, behaviour modification, and judicial economy” (para. 234). The judge accepted the defendants’ submission that the court “should not second-guess the access to justice provided by the OSC once the court was satisfied that the OSC’s purpose was to obtain restitutionary compensation for the harm suffered by the investors and the process to do so was adequate” (para. 256). He reached these conclusions in spite of his finding “that there [was] some basis in fact for the Plaintiffs’ submission that the investors may not have been fully compensated as a result of the OSC settlement agreements” (para. 101).

[ 10 ] The motion judge therefore denied certification.

C. Ontario Superior Court of Justice, Divisional Court, 2011 ONSC 292 , 104 O.R. (3d) 615 (Molloy J., Swinton and Herman JJ. Concurring)

[ 11 ] The Divisional Court found the motion judge’s analysis of the preferability requirement to be in error in three respects: “(1) he failed to apply the proper low evidentiary burden on the plaintiffs at [the certification] stage; (2) he improperly found that the already completed OSC proceeding was a preferable proceeding for the remaining portion of the plaintiffs’ claims going forward; and (3) he erred in law by considering criteria for approval of a settlement at the certification stage” (para. 33). The court therefore allowed the appeal and certified the class action.

D. Court of Appeal for Ontario, 2012 ONCA 47 , 109 O.R. (3d) 498 (Winkler C.J.O., Epstein J.A. and Pardu J. (ad hoc) Concurring)

[ 12 ] The Court of Appeal dismissed the appeal. It found that although the Divisional Court had not asked itself the right questions, it nonetheless came to the right result. In its view, the Divisional Court should not have focused on whether the OSC settlements provided investors with all or substantially all of the monetary relief they sought: this was an error because, at the certification stage, the amount that would be recoverable in the proposed class proceeding remains unknown, and thus there is no way of determining if the compensation was “substantial” (para. 77). Further, such a determination would be tantamount to making a finding on the merits of the case, which would be a marked departure from the evidentiary burden that applies on a motion for certification. The court was rather of the view that the preferable procedure inquiry had to focus on the purpose and nature of the alternative proceedings as compared with the class proceedings.

[ 13 ] The Court of Appeal thought that important procedural distinctions between the OSC proceedings and the proposed class proceeding supported the conclusion that a class proceeding was preferable for resolving the class members’ claims in this case. First, the OSC’s jurisdiction under s. 127 of the Ontario Securities Act , R.S.O. 1990, c. S.5, was regulatory, not compensatory and, as a result, the remedial powers available to the OSC were insufficient to enable it to fully address the class members’ claims. Second, the OSC proceedings did not provide comparable rights of participation to the affected investors, which was an important access to justice consideration.

III. Brief Summary of the Positions of the Parties

[ 14 ] The appellants complain that the Court of Appeal wrongly equated access to justice with access to court-like procedures to the neglect of other important considerations and that it erroneously focused on theoretical considerations rather than the actual results achieved by the OSC proceedings. These fundamental flaws, the appellants say, led the Court of Appeal into a number of other errors. It wrongly focused on the fact that the OSC jurisdiction is regulatory and not compensatory. It erred in its assessment of investor participation in the OSC proceedings. It failed to take into account the no-fault, timely and no-cost aspects of the OSC process, and the impartiality and independence of the OSC. The appellants also argue that the Court of Appeal simply substituted its own discretion for that of the motion judge, contrary to the applicable standard of review.

[ 15 ] The respondents submit that the Court of Appeal appropriately focused its analysis on the importance of participatory rights coupled with the limited scope and nature of the OSC’s jurisdiction and remedial powers, rather than the outcome of the OSC proceedings. The respondents contend that participation is at the heart of the concept of access to justice. They submit that given the lack of basic procedural rights to investors in the OSC proceedings, the non-binding nature of the settlements, as well as the non-disclosure of how the settlement agreements were achieved, the compensation received by investors was arbitrary and partial and was in no way preferable to a class action.

IV. Analysis

A. Introduction

[ 16 ] The appeal focuses on one branch of the statutory requirement for certification, the requirement that “a class proceeding would be the preferable procedure for the resolution of the common issues”: CPA , s. 5(1)(d). In the circumstances of this case, the question of preferability boils down to quite a narrow issue because there is substantial agreement on a number of points. There is no longer any dispute that the proposed class action meets all of the requirements for certification except for the disputed element of whether it is the preferable procedure. With regard to that requirement, it is no longer disputed that the class action would be a fair, efficient and manageable proceeding or that it would be preferable to any litigation alternatives. It is also common ground that to assess whether the class action would be preferable to any other alternative method of resolving the class members’ claims, the court compares the competing possibilities through the lens of the goals of behaviour modification, judicial economy and access to justice, bearing in mind, of course, that the ultimate question is whether the statutory requirement of preferability has been established. There is also no dispute that the appeal turns on how well the two proceedings can provide access to justice for the investors. This is so because the parties acknowledge that there is no other litigation alternative, that the OSC proceedings accomplished the purpose of behaviour modification, and the motion judge found that either proceeding would serve the goal of judicial economy, a finding that is not contested before this Court.

[ 17 ] While in general all three goals of class action procedures must be weighed in the balance, in the specific circumstances of this case, the question is whether, from an access to justice perspective, certification should be denied on account of results already obtained in a non-litigation proceeding before the OSC.

[ 18 ] I will begin my analysis by outlining the main threads of the jurisprudence dealing with the preferability requirement with particular reference to cases in which it has been argued that some form of non-court, alternative dispute resolution was a preferable procedure to a proposed class action. I will then develop what, in my opinion, is the correct analytical approach to this issue and apply it to this case.

B. The Preferability Requirement — Overview of the Principles

(1) Statutory Provision

[ 19 ] The starting point is the relevant statutory provision. Section 5(1)(d) of the CPA requires the court to conclude that “a class proceeding would be the preferable procedure for the resolution of the common issues”. (See the attached Appendix for s. 5(1) CPA reproduced in full.) Although this provision could be read as requiring a procedure that is capable of producing a formal resolution of the common issues, that reading was rejected by the Court in Hollick . McLachlin C.J., writing for the Court, made clear that the preferability requirement is broad enough to take into account “all reasonably available means of resolving the class members’ claims” including avenues of redress other than court actions (para. 31). An alternative process need not necessarily decide the precise legal and/or factual questions raised by the common issues provided that it effectively resolves the class members’ claims. This broad understanding of the preferability requirement is critical in cases like this one in which individual court actions are not a viable option.

[ 20 ] This understanding of the role of non-litigation alternatives in the comparative analysis under the CPA ’s preferable procedure criterion is different from its American counterpart under the U.S. federal class action regime. The language of Rule 23 of the U.S. Federal Rules of Civil Procedure provides that the court must, among other things, satisfy itself “that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy ”: 28 U.S.C. app., r. 23(b)(3). This wording invites a comparison between the class action and other forms of court action, and has tended to limit reliance on comparison with non-judicial alternatives: see W. B. Rubenstein, Newberg on Class Actions (5th ed. 2011) (WL), at § 4:86.

(2) In Assessing Preferability, the Court Looks at the Common Issues in the Context of the Action as a Whole

[ 21 ] In order to determine whether a class proceeding would be the preferable procedure for the “resolution of the common issues”, those common issues must be considered in the context of the action as a whole and “must take into account the importance of the common issues in relation to the claims as a whole”: Hollick , at para. 30 . McLachlin C.J. in Hollick accepted the words of a commentator to the effect that in comparing possible alternatives with the proposed class proceeding, “it is important to adopt a practical cost-benefit approach to this procedural issue, and to consider the impact of a class proceeding on class members, the defendants, and the court”: para. 29, citing W. K. Branch, Class Actions in Canada (loose-leaf 1998, release 4), at para. 4.690.

(3) The Preferable Procedure Analysis Considers the Extent to Which the Proposed Class Action Serves the Goals of Class Proceedings

[ 22 ] In Hollick , McLachlin C.J. indicated that the preferability inquiry had to be conducted through the lens of the three principal goals of class actions, namely judicial economy, behaviour modification and access to justice (para. 27). This should not be construed as creating a requirement to prove that the proposed class action will actually achieve those goals in a specific case. Thus, when undertaking the comparative analysis, courts must focus on the statutory requirement of preferabilty and not impose on the representative plaintiff the burden of proving that all of the beneficial effects of the class action procedure will in fact be realized.

[ 23 ] This is a comparative exercise. The court has to consider the extent to which the proposed class action may achieve the three goals of the CPA , but the ultimate question is whether other available means of resolving the claim are preferable, not if a class action would fully achieve those goals. This point is well expressed in one U.S. Federal Court of Appeals judgment and it applies equally to CPA proceedings: “Our focus is not on the convenience or burden of a class action suit per se , but on the relative advantages of a class action suit over whatever other forms of litigation [and, I would add, dispute resolution] might be realistically available to the plaintiffs”: Klay v. Humana, Inc. , 382 F.3d 1241 (11th Cir. 2004), at p. 1269, cited in Rubenstein, at § 4:85, fn. 2.

C. Access to Justice as a Goal of Class Proceedings

[ 24 ] There is no doubt that access to justice is an important goal of class proceedings. But what is access to justice in this context? It has two dimensions, which are interconnected. One focuses on process and is concerned with whether the claimants have access to a fair process to resolve their claims. The other focuses on substance — the results to be obtained — and is concerned with whether the claimants will receive a just and effective remedy for their claims if established. They are interconnected because in many cases defects of process will raise doubts as to the substantive outcome and defects of substance may point to concerns with the process. As the Honourable Frank Iacobucci put it, “access to justice must contain both a procedural and a substantive component. I find it difficult to accept that providing injured parties with a process to pursue their claims can be divorced from ensuring that the ultimate remedy arising from the process provides substantive justice where warranted”: “What Is Access to Justice in the Context of Class Actions?”, in J. Kalajdzic, ed., Accessing Justice: Appraising Class Actions Ten Years After Dutton, Hollick & Rumley (2011), 17, at p. 20. While it may be analytically convenient to look at process and substance considerations separately, this must not be done at the expense of an overall assessment of the access to justice implications of the proposed class action.

[ 25 ] The Divisional Court focused its access to justice analysis on substance, relying heavily on its conclusion that there was some basis in fact to believe that the investors were entitled to significantly more than they had received from the OSC proceedings (paras. 4 and 8). The Court of Appeal, on the other hand, focused mainly on process, relying heavily on considerations such as participation rights and remedial jurisdiction. The correct approach, however, must include both substantive and procedural aspects in assessing whether a class action is the preferable procedure. The focus cannot be exclusively on process: a process may be fair but nonetheless not offer a real opportunity to recover compensation for all of the losses suffered. In other words, in some cases even if the process is fair, there will remain significant obstacles to recovery. In addition, an absence of a fair process may also heighten concerns about whether substantive justice has or will be done. Of course, as we shall see, consideration of these aspects must respect the limited scope of the certification process.

[ 26 ] A class action will serve the goal of access to justice if (1) there are access to justice concerns that a class action could address; and (2) these concerns remain even when alternative avenues of redress are considered: Hollick , at para. 33 . To determine whether both of these elements are present, it may be helpful to address a series of questions. These questions must not be considered in isolation or in a specific order, but should inform the overall comparative analysis. I will set out the questions and comment briefly on each.

(1) What Are the Barriers to Access to Justice?

[ 27 ] The sorts of barriers to access to justice may vary according to the nature of the claim and the make-up of the proposed class. They may relate to either or both of the procedural and substantive aspects of access to justice. The most common barrier is an economic one, which arises when an individual cannot bring forward a claim because of the high cost that litigation would entail in comparison to the modest value of the claim. However, barriers are not limited to economic ones: they can also be psychological or social in nature. They may arise from such factors as the ignorance of the availability of substantive legal rights (Ontario Law Reform Commission, Report on Class Actions , vol. I (1982) (“OLRC Report”), at p. 127), ignorance of the fact that significant injuries have occurred (OLRC Report, at pp. 127-28), limited language skills (see e.g. Rubenstein, at § 4:65), elderly age of the claimants (see e.g. Cloud v. Canada (Attorney General) (2004), 2004 45444 (ON CA) , 73 O.R. (3d) 401 (C.A.)), frail emotional or physical state of the claimants (see e.g. Rumley v. British Columbia , 2001 SCC 69 , [2001] 3 S.C.R. 184), fear of reprisals by the defendant (OLRC Report, at p. 128; see e.g. Webb v. K-Mart Canada Ltd . (1999), 1999 15076 (ON SC) , 45 O.R. (3d) 389 (S.C.J.) ) , or alienation from the legal system as a result of negative experiences with it (OLRC Report, at pp. 128-29). A common procedural barrier is that there is no other procedure available to afford meaningful redress.

(2) What Is the Potential of the Class Proceedings to Address Those Barriers?

[ 28 ] The next question concerns the potential of the proposed class action to address the barriers to access to justice which have been identified in the particular case. This analysis is not made in isolation, but within the comparative analysis, for the purpose of assessing the class proceedings’ potential to address the access to justice concerns in comparison to the alternative procedure’s ability to do so.

[ 29 ] A class action may allow class members to overcome economic barriers “by distributing fixed litigation costs amongst a large number of class members . . . [and thus] making economical the prosecution of claims that any one class member would find too costly to prosecute on his or her own”: Hollick , at para. 15 . It may also allow claimants to overcome psychological and social barriers through the representative plaintiff who provides guidance and takes charge of the action on their behalf.

[ 30 ] Through these procedural mechanisms, a class action provides access to the courts for class members. Thus, it is a “procedural tool” ( Hollick , at para. 15 ): it does not guaranty results for class members.

[ 31 ] That being said, class proceedings exist not only to provide access to a procedure, but also to substantive results. The OLRC Report considered the various barriers to litigation and how class actions could play a role in overcoming those barriers. There is no doubt that achieving results for class members was at the heart of these discussions:

In the preceding sections, the Commission has examined the importance of providing increased access to the courts for persons who wish to pursue existing remedies but are unable to do so. The Commission is of the view that many claims are not individually litigated, not because they are lacking in merit or unimportant to the potential claimant, but because of economic, social, and psychological barriers. We believe that class actions can help to overcome such barriers and, by providing increased access to the courts, may perform an important function in society. Quite clearly, effective access to justice is a precondition to the exercise of all other legal rights.

Moreover, empirical evidence indicates that class actions do, in fact, provide access to justice for a broader range of persons. The evidence suggests that individuals are interested in pursuing their claims by means of a class action. In addition, class actions seeking damages do confer a significant monetary benefit, notwithstanding the deduction of lawyers’ fees and administrative costs. [Reference omitted; p. 139.]

[ 32 ] The Report of the Attorney General’s Advisory Committee on Class Action Reform (1990) (“ Report of the Attorney General ”) also acknowledged the underlying goal of class proceedings of providing redress for class members in its opening lines:

A class action is . . . a procedural mechanism that is intended to provide an efficient means to achieve redress for widespread harm or injury by allowing one or more persons to bring the action on behalf of the many. [p. 15]

[ 33 ] The Report of the Attorney General also highlighted the substantive component of access to justice:

Ontarians also live in a society that strives to maximize access to justice for its citizens. Sophisticated and highly evolved rights and obligations are of little value if they cannot be asserted or enforced effectively and economically. Of what value is a right or obligation, or the judicial system itself, if its users must be told that the right is “too small” or “too complex” or “too risky” to justify its enforcement?

A class action can provide the means by which such claims can be asserted and given access to justice. A meaningful procedure can achieve economies in the use of judicial and court resources and provide widespread redress to many individuals who have suffered a loss or injury. [pp. 16-17]

[ 34 ] Thus, class actions overcome barriers to litigation by providing a procedural means to a substantive end. As one author put it in a memorable phrase, a class procedure has the potential to “breath[e] new life into substantive rights”: M. Good, “Access to Justice, Judicial Economy, and Behaviour Modification: Exploring the Goals of Canadian Class Actions” (2009), 47 Alta. L. Rev. 185, at p. 188. Even though a class action is a procedural tool, achieving substantive results is one of its underlying goals. Consideration of its capacity to overcome barriers to access to justice should take account of both the procedural and substantive dimensions of access to justice.

(3) What Are the Alternatives to Class Proceedings?

[ 35 ] The motions court must identify alternatives to the proposed class proceedings. As McLachlin C.J. held in Hollick , “the preferability analysis requires the court to look to all reasonably available means of resolving the class members’ claims, and not just at the possibility of individual actions”: para. 31 (emphasis added). Here, the court considers both other potential court procedures (such as joinder, test cases, consolidation and so on: Hollick , at para. 28 ) and non-court proceedings.

[ 36 ] The motions court must look at all the alternatives globally in order to determine to what extent they address the barriers to access to justice posed by the particular claim: Hollick , at para. 30 . In some cases, non-litigation means of redress will be considered in conjunction with individual actions: see e.g. Hollick , Cloud and Pearson v. Inco Ltd . (2006), 2006 913 (ON CA) , 78 O.R. (3d) 641 (C.A.). In other cases, for example where there is no viable litigation alternative to a class action, the non-litigation means of redress will have to be considered on its own as a potential alternative to the class action: see e.g. Halabi v. Becker Milk Co. (1998), 1998 14681 (ON SC) , 39 O.R. (3d) 153 (Gen. Div.). The nature of the comparison analysis will vary, depending on the nature of the alternatives available for consideration.

(4) To What Extent Do the Alternatives Address the Relevant Barriers?

[ 37 ] Once the alternative or alternatives to class proceedings have been identified, the court must assess the extent to which they address the access to justice barriers that exist in the circumstances of the particular case. The court should consider both the substantive and procedural aspects of access to justice recognizing that court procedures do not necessarily set the gold standard for fair and effective dispute resolution processes. The question is whether the alternative has the potential to provide effective redress for the substance of the plaintiffs’ claims and to do so in a manner that accords suitable procedural rights. This comparison, of course, must take place within the proper evidentiary framework that applies at the certification stage. I will return to that point in a moment.

(5) How Do the Two Proceedings Compare?

[ 38 ] The focus at this stage of the analysis is on whether, if the alternative or alternatives were to be pursued, some or all of the access to justice barriers that would be addressed by means of a class action would be left in place: Hollick , at para. 33 . At the end of the day, the motions court must determine whether, on the record before it, the class action has been shown to be the preferable procedure to address the specific procedural and substantive access to justice concerns in a case. As set out in Hollick , the court must also, to the extent possible within the proper scope of the certification hearing, consider the costs as well as the benefits of the proposed class proceeding in relation to those of the proposed alternative procedure.

D. Evidentiary Considerations

(1) What Is the Evidentiary Burden With Regard to the Preferability Requirement on a Motion for Certification?

[ 39 ] The questions I have just outlined are addressed within the confines of the certification process; the court cannot engage in a detailed assessment of the merits or likely outcome of the class action or any alternatives to it. In Hollick , McLachlin C.J. explained that the evidentiary burden applicable on a motion for certification was low:

I agree that the representative of the asserted class must show some basis in fact to support the certification order. As the court in Taub [ v. Manufacturers Life Insurance Co. (1998), 1998 14853 (ON SC) , 40 O.R. (3d) 379 (Gen. Div.)] held, that is not to say that there must be affidavits from members of the class or that there should be any assessment of the merits of the claims of other class members. However, the Report of the Attorney General’s Advisory Committee on Class Action Reform clearly contemplates that the class representative will have to establish an evidentiary basis for certification: see Report, at p. 31 (“evidence on the motion for certification should be confined to the [certification] criteria”). The Act, too, obviously contemplates the same thing: see s. 5(4) (“[t]he court may adjourn the motion for certification to permit the parties to amend their materials or pleadings or to permit further evidence”). In my view, the class representative must show some basis in fact for each of the certification requirements set out in s. 5 of the Act, other than the requirement that the pleadings disclose a cause of action. That latter requirement is of course governed by the rule that a pleading should not be struck for failure to disclose a cause of action unless it is “plain and obvious” that no claim exists: see Branch, supra , at para. 4.60. [Emphasis added; para. 25.]

[ 40 ] This Court recently reaffirmed these principles in Pro-Sys Consultants Ltd. v. Microsoft Corporation , 2013 SCC 57 , [2013] 3 S.C.R. 477, in the context of the similar British Columbia class actions regime. In his discussion of the standard of proof with regard to the commonality and preferability requirements (para. 101), Rothstein J. indicated that the “‘some basis in fact’ standard does not require that the court resolve conflicting facts and evidence at the certification stage” (para. 102). This reflects the fact that a certification court “is ill-equipped to resolve conflicts in the evidence or to engage in the finely calibrated assessments of evidentiary weight”: Pro-Sys , at para. 102, citing Cloud , at para. 50 ; Irving Paper Ltd. v. Atofina Chemicals Inc . (2009), 2009 92127 (ON SC) , 99 O.R. (3d) 358 (S.C.J.), at para. 119 , citing Hague v. Liberty Mutual Insurance Co . (2004), 2004 95289 (ON SC) , 13 C.P.C. (6th) 1 (Ont. S.C.J.). Further, the “some basis in fact” standard cannot be assessed in a vacuum. As Rothstein J. puts it: “. . . there is limited utility in attempting to define ‘some basis in fact’ in the abstract. Each case must be decided on its own facts” (para. 104).

[ 41 ] Helpful elaboration of the “some basis in fact” standard may be found in the reasons of Winkler C.J.O. in McCracken v. Canadian National Railway Co ., 2012 ONCA 445 , 111 O.R. (3d) 745:

The “some basis in fact” principle is meant to address two concerns. First, there is a requirement that, for all but the cause of action criterion, an evidentiary foundation is needed to support a certification order.

Second, in keeping with the procedural scheme of the CPA , the use of the word “some” conveys the meaning that the evidentiary record need not be exhaustive, and certainly not a record upon which the merits will be argued. This legislative intention is reflected in s. 2(3)( a ) of the CPA , which — although honoured more often in the breach — requires the proposed representative plaintiff to bring a motion for certification within 90 days of the filing of, or the expiry of the time for filing of, a statement of defence or notice of intent. Thereafter, leave of the court is required to bring the motion: see s. 2(3)( b ). [Emphasis added; paras. 75-76.]

[ 42 ] The jurisprudence emphasizes the importance of not allowing the requirement to establish “some basis in fact” to lead to a more fulsome assessment of contested facts going to the merits of the case. For example, in Cloud , the Ontario Court of Appeal indicated that the some basis in fact standard “does not entail any assessment of the merits at the certification stage” (para. 50). Similarly, in Pearson , the Ontario Court of Appeal, having concluded that the representative plaintiff had adduced evidence to show a negative impact on property values concluded that the “some basis in fact” standard with regard to the commonality of the issues had been met. The court pointed out that while the defendant disputed the plaintiff’s evidence, “the certification motion is not the place for resolving that controversy” (para. 76). These evidentiary principles equally apply to the preferability criterion: see e.g. 1176560 Ontario Ltd. v. Great Atlantic & Pacific Co. of Canada Ltd. (2002), 2002 6199 (ON SC) , 62 O.R. (3d) 535 (S.C.J.), at para. 27 , aff’d (2004), 2004 16620 (ON SCDC) , 70 O.R. (3d) 182 (S.C.J. (Div. Ct.)).

[ 43 ] The standard of proof on a motion for certification was at the heart of the appeal in Chadha v. Bayer Inc. (2003), 2003 35843 (ON CA) , 63 O.R. (3d) 22 (C.A.), leave to appeal refused, [2003] 2 S.C.R. vi. The decision makes clear that at the certification stage, the court cannot engage in any detailed weighing of the evidence but should confine itself to whether there is some basis in the evidence to support the certification requirements. In Chadha , the court denied certification on the basis that there was no evidence that the loss component of liability could be proved on a class-wide basis (and thus that there was no common issue). It was not necessary to establish that there was a compelling method to prove such loss, but it was necessary to provide some basis in fact to think that there was some method to do so. The plaintiffs had failed to provide that basis. This Court reached the opposite conclusion in Pro-Sys with regard to the commonality of the issues, because there was “an expert methodology that ha[d] been found to have a realistic prospect of establishing loss on a class-wide basis” (para. 140).

[ 44 ] The limited scope of the factual inquiry on the certification motion means that the motions court will often not be able to compare the potential recoveries in the class action and in the alternative or alternatives to it. For example, in Pro-Sys it was argued that the class proceeding did not meaningfully further the objective of access to justice because the award would likely be distributed cy-près and not to individual class members. Rothstein J., however, rejected this argument noting that it was “premature to assume that the award . . . [would] result in cy-près distribution or that the objective of access to justice [would] be frustrated on this account” (para. 141).

[ 45 ] The limitations imposed by the nature of the certification process are directly relevant in this case. Somewhat unusually, a potential alternative procedure — the OSC proceeding — has run its course and the results of it are known with certainty. This might be seen, as it was by the Divisional Court in this case, as inviting a comparison of those known results with the likely outcome of the proposed class action. Simply put, is there enough left on the table to justify the time and expense of the proposed class proceeding? Viewed from this perspective, “[u]nless it can be said that the plaintiffs have achieved full, or at the very least substantially full, recovery”, they would be entitled to maintain a class action (Divisional Court decision, at para. 8). Based on the record in this case, the Divisional Court concluded that “the plaintiffs’ current claim against AIC and CI, over and above the OSC settlement, [was] $333.8 million” (para. 4). The court qualified this as a “significant amount of money” (para. 8) which entitled the plaintiffs to maintain a class action.

[ 46 ] Although at first glance such an approach would seem to have something to commend it, its allure quickly fades when due attention is given to the nature and limitations of the certification process. The certification process is not the occasion for a searching examination of whether the “plaintiffs have achieved full, or . . . substantially full, recovery” (Divisional Court decision, at para. 8) in comparison to the likely outcome of the class proceeding on its merits. Without that sort of examination, the most that can be done is to assess on the appropriately limited evidentiary record whether the access to justice barriers that may be addressed by a class proceeding remain even after the alternative process has run its course.

[ 47 ] Nevertheless, when the results or the limits on recovery of an alternative procedure are known at the time of the certification motion, those uncontested facts cannot be ignored. For example, in Rumley , McLachlin C.J. observed that the alternative procedure was not preferable because, among other things, it limited recovery of any complainant to $60,000 (para. 38). (Although the Court was dealing with the British Columbia Class Proceedings Act , R.S.B.C. 1996, c. 50, in that case, an analogous approach is implicit in the CPA .) In a case involving similar facts and dealing with the CPA , the Ontario Court of Appeal made similar comments with regard to the fact that the alternative procedure capped recovery: Cloud , at para. 92 . The fact that the court considered the limits on recovery in the alternative procedure implies that it took into account the fact that the class action could address the full range of the class members’ claims. Thus, the results and limits may be considered, but within the constraints of the evidentiary basis that is appropriate on a certification motion. In a case where the results of neither the alternative nor the class proceedings are known, the comparative exercise with regard to the substantive access to justice barriers will in general be very limited.

(2) Who Assumes the Burden?

[ 48 ] The party seeking certification of a class action bears the burden of showing some basis in fact for every certification criterion: Hollick , at para. 25 . In the context of the preferability requirement, this requires the representative plaintiff to show (1) that a class proceeding would be a fair, efficient and manageable method of advancing the claim, and (2) that it would be preferable to any other reasonably available means of resolving the class members’ claims: Hollick , at paras. 28 and 31 . A defendant can lead evidence “to rebut the inference of some basis in fact raised by the plaintiff’s evidence”: M. Cullity, “Certification in Class Proceedings — The Curious Requirement of ‘Some Basis in Fact’” (2011), 51 Can. Bus. L.J. 407, at p. 417.

[ 49 ] With regard to the second aspect of the preferability requirement — that is, the comparative analysis — the representative plaintiff will necessarily have to show some basis in fact for concluding that a class action would be preferable to other litigation options. However, the representative plaintiff cannot be expected to address every conceivable non-litigation option in order to establish that there is some basis in fact to think that a class action would be preferable. Where the defendant relies on a specific non-litigation alternative, he or she has an evidentiary burden to raise it. As Winkler J. (as he then was) put it in Caputo v. Imperial Tobacco Ltd. (2004), 2004 24753 (ON SC) , 236 D.L.R. (4th) 348 (Ont. S.C.J.): “. . . the defendants cannot simply assert to any effect that there are other procedures that would be preferable without an evidentiary basis . . . . It must be supported by some evidence” (para. 67). However, once there is some evidence about the alternative, the burden of satisfying the preferability requirement remains on the plaintiff.

E. Application

[ 50 ] There are two potential barriers to access to justice in this case. First, an economic barrier arises from the nature of the claim. The claim advanced here may be referred to as a small claims class action. As the motion judge found, the individual claims are not large enough to support viable individual actions (para. 62). The economic barrier to access to justice, therefore, is that individual claims are not viable to litigate individually. In the context of a small claims class action such as this one, access to justice requires access to a process that has the potential to provide in an economically feasible manner just compensation for the class members’ individual economic claims should they be established. The second barrier is related to the first. As a result of the nature of the claim, there is potentially no access to a fair process, geared towards protecting the rights of class members, to seek a resolution of the common issues for what could potentially be a class of over a million members. Thus, traditional litigation cannot achieve either the substantive or the procedural dimensions of access to justice in a case such as this.

[ 51 ] The proposed class action addresses both of these barriers. It has the potential to make it economically feasible to advance on behalf of the class a group of individual claims that would otherwise not be economically feasible to pursue in the courts and it provides class members with a fair process to resolve their claims. The class action process is geared to protecting the class members’ rights to a significant extent through such mechanisms as the requirement for a representative plaintiff who must “fairly and adequately represent the interests of the class”, produce a workable litigation plan to advance the proceeding on behalf of the class and have no conflict of interests with other class members: CPA , s. 5(1)(e).

[ 52 ] As I have already noted, there is no realistic litigation alternative in this case. The only alternative procedure that was advanced is the OSC proceedings and settlement agreements, the results of which are already known. Thus, the question is to what extent this alternative has addressed the barriers to access to justice and whether those barriers remain now that those proceedings have been completed. This analysis has both a procedural and a substantive component.

[ 53 ] With respect to the procedural component, the appellants submit that the Court of Appeal wrongly focused on the fact that the OSC’s jurisdiction is regulatory and not compensatory and erred in its assessment of investor participation in the OSC proceedings. I do not accept these submissions.

[ 54 ] The main jurisdiction of the OSC under s. 127 of the Securities Act — under which the OSC conducted the probe — is regulatory. Thus, it “is neither remedial nor punitive; it is protective and preventive, intended to be exercised to prevent likely future harm to Ontario’s capital markets”: Committee for the Equal Treatment of Asbestos Minority Shareholders v. Ontario (Securities Commission) , 2001 SCC 37 , [2001] 2 S.C.R. 132, at para. 42 , quoting (1999), 1999 7316 (ON CA) , 43 O.R. (3d) 257, at p. 272, per Laskin J.A. There is no question in this case that the OSC had jurisdiction to approve the settlement agreements or, as the motion judge found, that Commission staff in this case sought to determine the extent of investor losses and to achieve compensation for them through the settlement agreements. Nevertheless, compensation of investors is not the primary focus of the OSC under its s. 127 jurisdiction. Further, there is no way to know how the OSC arrived at the settlement agreements; the details of the methodology used to calculate the amounts having remained confidential throughout. As the motion judge found, “how the OSC came to its calculation is not actually known” (para. 99).

[ 55 ] With regard to investor participation in the OSC proceedings, my view is that the respondents and the Court of Appeal are somewhat off the mark by placing virtually exclusive weight on this consideration in rejecting the OSC proceedings as a preferable alternative. Nevertheless, I agree that investor participation in the process leading to compensation is an important factor to consider and one that weighs heavily in favour of finding that the class proceeding meets the preferability requirement in this case. As the Court of Appeal noted, the OSC proceedings and the procedure by which the settlement agreements were arrived at in this case “provided little to no basis for investor participation” (paras. 58 and 60), whereas class proceedings “allow for the appointment of a representative plaintiff who shares a sufficient common interest with members of the class [and] conducts the litigation on behalf of class members under court supervision and within the presumptive principle of an open court” (para. 61). Moreover, as the motion judge found, nothing was known about how the OSC came to its assessment of compensation. In summary, the regulatory nature of and the limited participation rights for investors in the OSC proceedings, coupled with the absence of information about how the OSC staff assessed investor compensation, support the conclusion that significant procedural access to justice concerns remain which the proposed class action can address. Moreover, the focus and nature of the OSC process reinforce the concerns which I will turn to next about whether substantial access to justice was achieved.

[ 56 ] Turning to the substantive aspect of access to justice, the Court of Appeal found that the motion judge and the Divisional Court had erred by focusing on the substantive outcome of the OSC proceedings, commenting that this “is not a relevant factor in the comparative analysis under s. 5(1)( d ) of the CPA ” (para. 10). In my view, the Court of Appeal took too categorical an approach to this issue in the circumstances of this case. While of course any consideration of the substantive outcome must take place within the evidentiary framework that applies on a certification motion, access to justice as explained earlier is not a purely procedural concept. Access to justice requires access to just results, not simply to process for its own sake. However, I conclude that giving this substantive element the considerable weight that it deserves in this case reinforces the Court of Appeal’s conclusion that this class action should be certified.

[ 57 ] One of the barriers to access to justice in this small claims investor class action is that traditional litigation provides no economically feasible way to recover the investors’ claimed losses. The appellants’ position is that the OSC regulatory proceedings effected significant recovery for the investors at no cost to them. In these circumstances, the substantive outcome of the OSC proceedings cannot, in my view, be dismissed as irrelevant to the question of whether the OSC proceedings addressed the access to justice barrier that is present in this case or whether the way in which it did so suggests that the class proceeding is not the preferred alternative.

[ 58 ] That said, however, the substantive outcome of the OSC proceedings and their impact on the preferability analysis must be examined through the appropriate evidentiary lens. As I have explained, the plaintiffs’ burden is to provide “some basis in fact” to think that the class proceedings are preferable to the alternative. In the rather unusual circumstances of this case, where the OSC proceedings have run their course and the results of those proceedings are known, it seems to me that the comparative analysis cannot ignore the question of whether a cost-benefit analysis supports the respondents’ contention that the proposed class proceeding is the preferable way to address their claims.

[ 59 ] The answer to this question, as I see it, is quite straightforward in this case. The respondents have pleaded viable causes of action, the OSC proceedings and settlement agreements were without prejudice to those claims, and the motion judge was satisfied “that there [was] some basis in fact for the Plaintiffs’ submission that the investors may not have been fully compensated as a result of the OSC settlement agreements” (para. 101). The motion judge also found that the class members were not trying to “have their cake and eat it too”:

I do not agree with any arguments that suggest that the Plaintiffs [respondents in this Court] and the investors are being unfair, or perhaps piggish, in eating the cake of the OSC compensation and also having a class action. The investors are not playing “heads I win, tails you lose.” They did not ask the OSC to be their champion, and they did not do anything wrong in accepting the spoils secured by the OSC’s campaign. The putative representative plaintiffs do no wrong in attempting to certify their action as a class proceeding, and I have no reason to believe that they do not genuinely believe that they were under-compensated. [para. 218]

[ 60 ] The motion judge estimated the size of the proposed class against AIC at 264,036 members and 803,903 members against CI. These represent the numbers of settlement payments issued by each appellant pursuant to the OSC settlements: motion decision, at para. 56. AIC’s and CI’s OSC settlement payments were respectively $58.8 million and $49.3 million. The plaintiffs presented expert evidence which estimated that AIC’s investors’ losses could be as low as $6.5 million or as high as $251.0 million, and that CI’s investors’ losses could be as low as $72.1 million or as high as $349.3 million, depending on the method of calculation used. Based on an expert’s preferred method of calculation (he submitted a total of five), losses were estimated at $192.6 million for AIC’s investors and $349.3 million for CI’s investors: motion decision, at para. 94. Thus, in the plaintiffs’ expert’s view, AIC’s investors have received only 31% of the compensation they are entitled to through the settlement agreements, and CI’s investors have received only 14%: motion decision, at para. 94.

[ 61 ] Of course, the certification motion is not the proper setting to delve into the likely success of these claims or to debate the merits of these approaches to calculating the investors’ losses. The record in this case, which shows in detail the results of the proposed alternative proceedings which have run their course, also shows that substantive access to justice concerns still remain. Further, there is no reason to believe that potential additional recovery would be consumed by the costs of the proceedings. In fact, the motion judge found that since the defendants were able to distribute the OSC settlement payments “by reviewing their own records to make decisions about entitlement, causation, and quantification”, that “there is at least a realistic possibility that acceptable procedures could be fashioned by the common issues trial judge to address quantification and distribution issues in a fair, manageable and efficient manner” (para. 208). Accordingly, in my view, the plaintiffs (now respondents) provided an appropriate basis to support the view that the class action proceeding would overcome access to justice barriers that subsisted after the completion of the OSC proceedings and that a cost-benefit analysis supported the conclusion that the class proceedings were the preferable proceeding for the investors to pursue their claims.

[ 62 ] To conclude, I am of the view that the motion judge erred in principle in the preferability analysis. Respectfully, he erred by agreeing with the defendants’ (now appellants’) submission that he should not “second-guess the access to justice provided by the OSC once [he] was satisfied that the OSC’s purpose was to obtain restitutionary compensation for the harm suffered by the investors and the process to do so was adequate” (paras. 256-57). On the contrary, it was precisely his role to compare and evaluate, within the limited scope of the certification motion, the access to justice provided by the two proceedings, both in the substantive and procedural dimensions of the term as part of his overall assessment of whether the plaintiffs had established the preferability requirement on the appropriate evidentiary standard. The fact that the results of the OSC process were known in this case added an element that would often not be present at the certification stage.

[ 63 ] I agree with the Court of Appeal that the motion judge erred in principle by “treating the negotiated payments that were made to investors in the OSC settlements as somehow eliminating the need to compare the purely regulatory function served by the OSC proceedings with the private remedial function to be played by the proposed class action” (para. 80). This led the motion judge to wrongly dismiss as irrelevant important access to justice considerations relating to the regulatory focus, absence of investor participation and the absence of information about how the investor losses were assessed in the OSC process. The process limitations of the OSC proceedings reinforce the concern that access to substantive justice will be better served by the proposed class action.

[ 64 ] I also agree with the Divisional Court and the Court of Appeal that the motion judge erred in principle by relying on the sorts of considerations that would be relevant to approving a settlement of a class action. The certification application, and particularly the preferability aspect of it, is not an appropriate point in the proceedings to engage in any in-depth analysis of either the merits of the plaintiffs’ claims or the likely quantum of recovery. As the Court of Appeal put it, at the certification stage, in most instances, “no reliable yardstick is available because the amount recoverable in the proposed class proceeding would be as yet unknown. Put another way, the preferability analysis should not be reduced to an ex post facto assessment of the adequacy of the award arrived at through the alternative procedure” (para. 77).

[ 65 ] I recognize that a decision by a certification judge is entitled to substantial deference: see e.g. Pearson , at para. 43 ; Markson v. MBNA Canada Bank , 2007 ONCA 334 , 85 O.R. (3d) 321, at para. 33 . Specifically, “[t]he decision as to preferable procedure is . . . entitled to special deference because it involves weighing and balancing a number of factors”: Pearson , at para. 43 . However, I conclude that deference does not protect the decision against review for errors in principle which are directly relevant to the conclusion reached such as, in my view, occurred here: see e.g. Cassano v. Toronto-Dominion Bank , 2007 ONCA 781 , 87 O.R. (3d) 401, at para. 23 , leave to appeal refused, [2008] 1 S.C.R. xiv; Markson , at para. 33 ; Cloud , at para. 39 .

V. Disposition

[ 66 ] I would dismiss the appeal with costs.

APPENDIX

Class Proceedings Act, 1992 , S.O. 1992, c. 6

5. — (1) [Certification] The court shall certify a class proceeding on a motion under section 2, 3 or 4 if,

(a) the pleadings or the notice of application discloses a cause of action;

(b) there is an identifiable class of two or more persons that would be represented by the representative plaintiff or defendant;

(c) the claims or defences of the class members raise common issues;

(d) a class proceeding would be the preferable procedure for the resolution of the common issues; and

(e) there is a representative plaintiff or defendant who,

(i) would fairly and adequately represent the interests of the class,

(ii) has produced a plan for the proceeding that sets out a workable method of advancing the proceeding on behalf of the class and of notifying class members of the proceeding, and

(iii) does not have, on the common issues for the class, an interest in conflict with the interests of other class members.

Appeal dismissed with costs.

Solicitors for the appellant AIC Limited: Borden Ladner Gervais, Toronto.

Solicitors for the appellant CI Mutual Funds Inc . : Goodmans, Toronto.

Solicitors for the respondents: Rochon Genova, Toronto.