Irving v. Western Digital Corp.

2022 BCSC 108 Irving v. Western Digital Corp.

IN THE SUPREME COURT OF BRITISH COLUMBIA Citation: Irving v. Western Digital Corp., 2022 BCSC 108 Date: 20220125 Docket: S-205402 Registry: Vancouver Between: Sheldon Irving Plaintiff And Western Digital Corporation and Western Digital Canada Corporation Defendants Brought under the Class Proceedings Act, R.S.B.C. 1996, c. 50 Before: The Honourable Mr. Justice Branch Reasons for Judgment on Settlement Approval In Chambers Counsel for the Plaintiff: M. Good R. Matheuszik Counsel for the Defendants: A. Borrell B. A. Mitretodis Place and Date of Hearing: Vancouver, B.C. December 15, 2021, Further Written Submissions received December 21, 2021 Place and Date of Judgment: Vancouver, B.C. January 25, 2022 Table of Contents I. INTRODUCTION. 3 II. FACTS. 3 A. Generally. 3 B. The Settlement 3 III. the settlement approval test. 4 IV. FEES. 7 V. NOTICE. 14 VI. HONORARIUM.. 14 VII. CONCLUSION. 14 I. INTRODUCTION [1] This is an application for approval of a settlement and counsel fees in a class proceeding. The case was previously certified as a class proceeding for settlement purposes. II. FACTS A. Generally [2] The plaintiff's allegation is that consumers who bought the defendants' WD RED NAS drives (the "Drives") were misled because the defendants changed the underlying technology in the drives from CMR technology to SMR technology. The plaintiff alleges this was inferior and not

suitable for use in Network Attached Storage ("NAS") devices specifically, and Redundant Array of Independent Disks ("RAID") arrays generally, despite the defendants' marketing that the Drives were intended for use in such devices. [3] The defendants' position is that the Drives were designed and marketed appropriately. In particular, they argue that the type of daily workload associated with a small home office NAS environment is light, leaving sufficient idle time for the drives to perform background data management tasks as needed, irrespective of the technology used. Further, the defendants intended to argue that certification was inappropriate, as a proper determination of the merits of the case would require a consideration of the planned or actual use to which each class member put each Drive. [4] The defendants estimate that the number of Drives sold during the class period is 28,976. B. The Settlement [5] The settlement provides that each class member who makes a claim will receive $6 or $9 for each Drive purchased, depending on the size of the Drive. It is expected that most class members will have purchased several Drives. [6] The settlement also provides for a potential cy-près award, as discussed further below. [7] The defendants agree to pay up to $375,000, and a further $25,000 in administrative expenses. This amount includes any approved class counsel fees to be

paid by the defendants, rather than out of each class member's award. [8] Class counsel seek a fee of $112,500, representing 28% of the total available amount of $375,000. They also seek an honorarium for the representative plaintiff of $1,500. [9] Given the distribution of the different size Drives, the parties estimate that the maximum possible recovery for the class itself, if there is 100% take up, would be $239,604. III. the settlement approval test [10] I reviewed the test for settlement approval in Wilson v. Depuy International Ltd., 2018 BCSC 1192 at paras. 56-61, aff'd 2019 BCCA 440, and more recently in Krishnan v. Jamieson Laboratories Inc., 2021 BCSC 1425 at paras. 11-13. I apply the same test here. [11] I find that the settlement is fair and reasonable in light of the following factors: a) There was a warranty with each Drive that included an arbitration clause which may have prevented the inclusion of non-consumers in the class. b) The plaintiff bore a risk of losing certification in light of the different demands which each class member may have placed on their Drives. c) The plaintiff bore a material risk of losing on the merits. For example, one of the challenges the plaintiff faced was the clear disagreement between the experts as to whether a testing methodology even existed that could determine any issue with the Drives. d) The number of negatively affected

class members may be low, specifically: i. The evidence suggests that there were a limited number of Drives sold in Canada. ii. There was little evidence of Canadian residents who acquired and used a Drive for the purposes for which they were intended, encountered performance issues, and alleged that the Drives were not suitable or resulted in data loss. The defendants provided evidence that only one customer has sought a replacement as a result of the identified issue. e) The class would most certainly have faced a long delay before being able to recovery any amount absent a settlement. f) No class member objected to the settlement, even though direct notice was provided to about 1,000 email addresses of suspected class members. g) The settlement was negotiated at arm's length between experienced counsel. h) The amount that each class member will receive was reasonably determined by comparing the premium charged for drives that had, and did not have, the technical characteristics the class says should have been included in all of the Drives. This premium was determined to be between $10-$13. As such, $6-$9 represents more than half of the premium. This is a reasonable discount given the risks identified above. [12] The parties disclosed that there was a U.S. settlement in parallel litigation that provides for roughly the same individual benefits that each class member will

receive in this proceeding: Malone v. Western Digital Corporation, (July 21, 2021) Case No. 5.20-cv-03584-NC. At the time of the hearing, the settlement had received preliminary approval and received final approval as of December 22, 2021. At a minimum, this case provides a case study of what other counsel were able to achieve in non-arm's length negotiations against the same defendants. The results achieved here compare reasonably well. [13] However, there is one material difference between the settlement proposed here and the U.S. settlement that must be noted. At a minimum, it has an impact on the evaluation of fees, as discussed below. Specifically, the U.S. settlement amount committed to by the defendants is non-reversionary -- meaning this amount will not return to the defendants irrespective of the take-up rate by class members. [14] One might have assumed that the same structure exists under the Canadian settlement, given the focus in the parties' submissions on the $375,000 committed by the defendants. But on a careful reading, this is not the case. Unlike most class action settlements, the defendants' proposed contribution, in the present case, is not defined as the "Settlement Fund", but rather as the "Maximum Settlement Funding". The definition provides as follows: "Maximum Settlement Funding" is CAD $375,000, which is the maximum amount of money the Defendants will

pay to fund the payment of Cash Payments to Claimants, the Cy Pres Donation, Class Counsel Fees, taxes and disbursements, any Administration Expenses in excess of $25,000, plus the separate provision of CAD $25,000 for Administration Expenses. [15] The potential "Cy-près Donation" is to the Law Foundation, and engages pursuant to the following terms: (a) If the total value of the sum of the Cash Payments issues to the Claimants, Class Counsel Fees and Administration expenses is less than $187,500, then the Cy Pres Donation amount to be distributed to the Law Foundation is the difference between CAD $187,500 and the sum of the Cash Payments issued to Claimants, Class Counsel Fees and Administration Expenses. (b) If the total value of the sum of the Cash Payments, Class Counsel Fees and Administration Expenses is more than CAD $187,500, but less than the Maximum Settlement Funding, then the difference between the two will be distributed by way of Cy Pres Donation to the Law Foundation. [16] As such, in effect, the $375,000 (plus $25,000 in Administration Expenses) operates as a ceiling, not a floor. It is therefore inaccurate to refer to this as a "$375,000 settlement". It must more accurately be treated as a "capped process settlement" rather than a "lump sum settlement". [17] A proper appreciation of this structure addresses another concern I raised with counsel, being whether

s. 36.2(1) of the Class Proceedings Act, R.S.B.C. 1996, c. 50 prevents the court from approving a settlement with a reversionary element, whereby residual settlement funds return to the defendants if not claimed. This interesting issue can be left for another day, since what is proposed here is not properly treated as a reversion of committed funds. Rather, the defendants are only agreeing to "pay as you go". There is no commitment of funds unless and until proper individual claims are advanced. In this situation, I find that s. 36.2(1) does not apply. [18] In terms of approving the settlement, I do not find that the fact that this is a process, or "pay as you go", settlement necessarily makes the settlement unfair or unreasonable. The amount made available to each individual class member is reasonable, even if the grand total of the amounts payable is as yet unknown. The process for making claims is relatively simple, which also enhances the reasonableness of the settlement. I approve the settlement as being both fair and reasonable, subject to my adjustment to the fee structure below. IV. FEES [19] Class counsel sought approval for a fee of $112,500, which would be paid by the defendants immediately, whether or not a single class claim is paid. Class counsel argued that their fee proposal was reasonably based off the $375,000 figure, and suggested that they were simply

seeking a fee equivalent to "28% of the settlement value". [20] Had this actually been a $375,000 lump sum settlement, I would have had no difficulty in approving this fee, particularly given that the retainer agreement permitted a maximum fee of 30%. Unfortunately, this submission does not conform to the actual structure of the settlement. [21] In her article, "Judging Fairness in Class Action Settlements" (2010) 28:1 Windsor YB on Access Just. 111 at 133 Catherine Piché notes that U.S. courts are advised to evaluate whether any of the following "hot button indicators" are present in evaluating the reasonableness of a proposed class settlement: If the judge notices "hot button indicators" which show unfairness on the face of settlement, counsel may be asked "hard questions" about the settlement's value to the class. The following elements are considered to be "hot button indicators" in the manual for Complex Litigation 4th, published by the U.S. Federal Judicial Center: - Granting class members illusory non-monetary benefits, such as discount coupons for more of the defendants' product, while granting substantial monetary attorney fee awards; - Imposing such strict eligibility conditions or cumbersome claims procedures that many members will be unlikely to claim benefits, particularly if the settlement provides that the unclaimed portions of the funds will revert to the

defendants; - Treating similarly situated class members differently (for example, by settling objectors' claims at significant higher rates than class members' claims); - Releasing claims of parties who received no compensation in the settlement; - Setting attorney fee based on a very high value ascribed to non-monetary relief awarded to the class, such as medical monitoring injunctions or coupons, or calculating the fee based on the allocated settlement funds; rather than the funds actually claimed by and distributed to class members; and - Assessing class members for attorney fees in excess of the amount of damages awarded to each individual. [Emphasis added, footnotes omitted.] [22] I have a concern about the applicability of Ms. Piche's final point in this case. A hypothetical breakdown of the settlement funds will help to illustrate the concern. If: a) no direct class member claim is advanced or accepted; b) administration Expenses are $25,000; c) fees are approved at $112,500, plus 12% taxes, for a total of $126,000; and d) disbursements are approved at the requested $20,487.24; then e) the Law Foundation would receive $16,012.76 as a cy-près award.[1] [23] In this scenario, class counsel would receive $112,500 as a fee, against a class (indirect cy-près) benefit of only $16,012.76. This example creates an inferred contingency percentage of more than 87%

(112,500/128,512.76). [24] I find that the fee structure must be controlled so as to mitigate the extent to which this structure decreases the reasonableness of the settlement generally, and the fee specifically. [25] The difficulty here is created by a factor I touched on previously in Wilson -- the dangers of having class counsel paid directly by defendants in a manner that undermines the nexus between the results achieved by the class and the results achieved by class counsel. In Wilson, I stated: [125] With the settlement, class counsel seeks a contingency fee of 25%. They indicate that part of the reason they are able to lower their request from 33% is due to the fact that the defendant has proposed to pay $275,000 in fees immediately. This is a direct defendant contribution towards fees. [126] I am prepared to approve a 25% contingency fee with one adjustment and one caution. [ ] [128] [T]reating the $275,000 from the defendant as being paid "on top" of this reasonable fee percentage has the potential to make the actual final fee unreasonable. For example, if only 10 class members come forward, and they receive awards totalling $1,000,000, the total legal fees received would be $525,000 ($275,000 + (25% x $1,000,000)), or effectively 52.5% of the amount going to the class, which is far in excess of that allowed in the normal case under the Law Society Rules, let alone

what a court would approve as reasonable. [129] Counsel did not provide their hours or normal hourly rates, which may in some cases justify approving amounts in excess of standard contingency percentages, in order to ensure that counsel is recovering at least the base value of their time investment, or close to it. [130] In my view, in light of the arrangements already made, the reasonable approach in this case is to treat the $275,000 the defendants have agreed to pay on account of fees as a "down payment" on the approved 25% contingency fee. In other words: a) the $275,000 can be paid to class counsel immediately by the defendants; b) a 25% approved fee percentage can be deducted from each individual award as it is paid to that individual; c) but no additional money should flow to class counsel until after $1,100,000 in claims have been paid (i.e. when the $275,000 already paid represents 25% of the total amount awarded). [131] In terms of the caution, I have grave reservations about the practice of defendants paying legal fees direct to class counsel, for the following reasons: a) It perverts the solicitor-client relationship between the class and class counsel to have a defendant inserting itself in the payment of legal fees. b) It perverts the court's ability to assess the reasonableness of the settlement, since it is almost impossible to determine whether the total amount

payable to the class, or the terms offered to the class, would have been higher or improved if the defendant did not agree to pay those fees to class counsel. c) It loosens the tie between the class and class counsel's interests. Class counsel may be somewhat less concerned about driving quickly and resolutely for the first or last dollar into each class member's hands, because they already have the security of the "bird in the hand" through the payment of at least some portion of their legal fees by the defendant. [132] All of these problems are avoided if class counsel simply seek a pure contingency fee payable from each award made to each class member. In that way, class counsel and the class' interests are tied together as tightly as possible, which ensures the most frictionless operation of the class proceedings statute. This is not to say that there may not be cases where such a direct contribution would be fair and reasonable, but I expect these situations to be unique or extraordinary. [133] That said, with the adjustment made to the fee payment structure, I am prepared to approve the 25% percentage fee proposed here as fair and reasonable. [26] I am prepared to allow the payment of disbursements immediately, since no windfall is associated with such a payment. However, I am not prepared to approve the fee as presently structured. [27] Class counsel asserted that the

settlement and fee approved in Steele v. Toyota, 2015 BCSC 1014 is consistent with the approach proposed here. In that case, the court awarded a fee that was 20% of the total potential amount paid by the defendant under the process settlement. However, the analysis of fees in Steele was very cursory. In particular, I do not see that the court wrestled with the inherent tension created by: a) counsel initially agreeing to a retainer that creates a secure bond between the results achieved by class and counsel; and b) counsel later seeking to unbind themselves from their client class when the defendants make an offer to guarantee counsel their fee, while at the same time refusing to guarantee the class a corresponding benefit. [28] Further, I note that the fee of $980,000 in Steele was less of a concern, as there was a floor amount of $2,300,000 that would be contributed towards class claims, a cy-près award, or fees. As such, the "worst case scenario" was that the approved fee would represent 42% of the total. As noted above, the fee here could represent 87% of the total amount actually committed, a far more significant (and troublesome) percentage. [29] I am influenced here as well by the specific terms of the retainer agreement between class counsel and the representative plaintiff. It provided as follows: The Solicitors' legal fees shall be a maximum of 30% of the total amount

recovered by the Class under any judgments, orders or settlement. [Emphasis added.] [30] In light of the example above, one can see how the approved fee sought here could easily become inconsistent with this clause, particularly if class member take up is low. In essence, class counsel's fee is no longer tied to the "total amount recovered by the Class" at all. Although not necessarily binding, the original retainer should be respected unless there is a compelling reason not to. [31] As a practice point, counsel who may wish to someday seek a fee inconsistent with a contingency arrangement would be well advised to provide for that possibility within the original retainer. It is possible that a well-informed representative plaintiff may choose to investigate whether there is alternate counsel prepared to be governed by a percentage structure not only at the beginning of the case, but through to the end. For example, in Hardwick v. Blue Buffalo Co., 2021 ONSC 5297, counsel's retainer agreement provided for the possibility that the fee requested would be based on a multiplier of the time spent, rather than necessarily on a percentage basis: para.

17. With such a disclosure, the representative plaintiff would at least be going into the case with their eyes wide open. [32] I am not purporting to establish a rule that direct fee payments by a defendant are never justified. However, it is difficult to see where and when such agreements would be in the best interests of the class, or promote the proper administration of justice. [33] Class counsel argues that the benefits achieved here are not strictly economic, as the settlement also achieved behaviour modification. They rely on the following settlement term: 7.1 Disclosure of Shingled Magnetic Recording Technology in WD RED NAS Drives The Defendants will continue to disclose the use of shingled magnetic recording technology on the westerndigital.com product page for all WD RED NAS Drives with shingled magnetic recording technology for two years following the Effective Dare. [34] I agree that any behaviour modification achieved is properly considered in relation to a fee request: Coburn and Watson's Metropolitan Home v. BMO Financial Group, 2018 BCSC 1183 at para.

81. However, the weight to be attributed to that factor here is marginal, given that the settlement simply provides for a continuation of a disclosure that is already ongoing. That said, the litigation could obviously have had some influence on the defendants' decision making, and will affect its decision making going forward. [35] In conclusion, I would control the fee award as follows in order to ensure a better alignment between the results achieved for the class and counsel's fee: Class counsel is entitled to a fee paid by the defendants of up to $112,500, calculated as 40% of the total paid (1) to class members, or (2) as a cy-près donation. This fee award may be paid to class counsel as such payments are made. [36] The award of 40% of the payments made to or on behalf of the class is designed to ensure that class counsel receives approximately 28% of the total, including fees (which is how contingency awards are normally calculated).[2] This creates a better alignment with the original retainer agreement. The approved structure will also ensure that counsel remain motivated to maximize the take-up rate, another important objective. [37] I recognize that this is a small settlement, but that is no reason to so dramatically unbind the class' result from counsel's result. [38] This adjustment should not be viewed as a criticism of class counsel. They should be commended for

pushing a low value case through to a satisfactory resolution for the class. But their reward for such efforts should be more closely aligned to the reasonable result achieved. V. NOTICE [39] During the hearing, I expressed a concern that there was no element of the proposed notice program that would allow for the availability of the settlement to potentially come to the attention of persons not already in the parties' databases. The parties considered this concern and, as a result, the defendants agreed to pay up to $2,000 on top of the settlement for a 60-day Google AdWords cost-per-click campaign. The 60-day campaign will coincide with the 60-day claims period. With this enhancement, I find that the proposed notice program is reasonable. VI. HONORARIUM [40] Class counsel sought an award of $1,500 as an honorarium for the class representative. This is a modest request given the awards that have been made in other cases, and I approve same. VII. CONCLUSION [41] The Settlement Agreement and honorarium are approved, with the noted adjustment to the notice program. Fees are approved on the adjusted formula set out above. "The Honourable Mr. Justice Branch" [1] $187,500 - $25,000 - $126,000 - $20,478.24 = $16,012.76. [2] For "mathletes" reviewing this judgment, a precise figure would be 38.8%, derived as follows. Assume x is counsel's fee, and y is the amount received by the

class. We want to derive x/y (the percentage of the class amount the lawyer should receive). We are seeking to have x/(x+y) be 0.28, such that counsel receives 28 percent of the total. If we multiply both sides of this equation by x+y, we have: x = 0.28x + 0.28y. Therefore 0.72x = 0.28y (by subtracting 0.28x from both sides of the equation). As such determining the appropriate proportion of the class only aware results inx/y = 0.28/0.72 = .388. The complexity of this calculation provides one more illustration why it is preferable for counsel to simply be paid a fee that is a percentage of the total amount a defendant is willing to pay, rather than segregating class counsel's fee from the class award.