Searching over 5,500,000 cases.

Buy This Entire Record For $7.95

Official citation and/or docket number and footnotes (if any) for this case available with purchase.

Learn more about what you receive with purchase of this case.

Melissa Ferrington, Cheryl Schmidt v. Mcafee


April 6, 2011


The opinion of the court was delivered by: Lucy H. Koh United States District Judge

United States District Court For the Northern District of California


Before the Court is Plaintiffs' motion for final approval of a class action settlement, and Plaintiffs' counsel's motion for attorneys' fees. A fairness hearing on the motions was held on 19 December 15, 2012. After considering the briefs filed by the parties and the papers filed by the 20 objector, and hearing oral argument on the matter, the Court DENIES final approval of the class 21 settlement, and DENIES Plaintiffs' counsel's motion for attorneys' fees for the reasons discussed 22 below. 23


This is a class action against McAfee, Inc. ("McAfee") and Arpu, Inc. ("Arpu") arising out 25 of a partnership between the two companies. McAfee is a provider of computer security software 26 whose products may be purchased and downloaded from the McAfee website. Arpu is a company 27 that places online advertisements that enable consumers to purchase products "with a single click, 28 using credit card information already on file." In 2007, Arpu partnered with McAfee to place ads 2 on McAfee's website that would appear after a customer completed a purchase of a McAfee 3 product. These ads were targeted to entice a customer to purchase an Arpu product using the 4 billing information that had been provided to McAfee for the purchase of the McAfee product. If a 5 customer chose to subscribe to the product or service offered in the Arpu ad, McAfee transmitted 6 the customer's billing information to Arpu for use in the purchase of the Arpu product. 7

2009, respectively. After they completed their transactions, but before they downloaded the 10

computer screens. Believing that clicking on "Try It Now" would download the McAfee software

they had just purchased, Plaintiffs clicked on the button. They later learned that clicking on "Try It 13

Named Plaintiffs in this action, Melissa Ferrington and Cheryl Schmidt, purchased

McAfee's anti-virus program from the McAfee website on August 18, 2009, and November 30, 9

McAfee product, an Arpu pop-up ad with a button reading "Try It Now" appeared on their

Now" authorized McAfee to transfer their billing information to Arpu, enrolled them in a 30-day 14 free trial of a non-McAfee product called PerfectSpeed (alternatively "Arpu software" or "the 15 software"), and authorized Arpu to charge them a $4.95 monthly subscription fee after the 16 expiration of the free trial period. Plaintiffs never downloaded the PerfectSpeed software, and 17 were not aware of this transaction until they noticed charges listed as "TB *PERFECTSPD ON 18

MCAF 202-4461821" on their credit or debit card statements. 19

No. 12). Both complaints alleged that McAfee sold customers' debit/credit billing information to 23

("UCL"), Cal. Bus. & Prof. Code § 17200, and the Consumer Legal Remedies Act ("CLRA"), Cal. 25

McAfee filed a motion to dismiss the FAC (ECF No. 25), which was granted in part and

27 denied in part on October 5, 2010. ECF No. 57. The Court denied Defendant's motion to dismiss 28

A.Procedural Background

Named Plaintiffs Ferrington and Schmidt filed the original Complaint on April 6, 2010

(ECF No. 1), and subsequently filed a First Amended Complaint ("FAC") on May 13, 2010 (ECF 22

Arpu, without authorization and asserted claims under the California Unfair Competition Law 24

Civ. Code § 17500. 26

Plaintiffs' UCL claim, but granted Defendant's motion to dismiss the CLRA claim. McAfee filed 2 an Answer to Plaintiffs' FAC on October 19, 2010. ECF No. 58. Plaintiffs sought leave to file a 3

Second Amended Complaint ("SAC") on April 7, 2011, re-pleading their CLRA claim and adding 4 a breach of contract claim (ECF No. 75). 5

6 as well as the motion for certification of a settlement class and preliminary approval of a class 7 settlement, with the signed settlement agreement attached. ECF Nos. 87-88. 8

On July 13, 2011, Plaintiffs filed their Third Amended Complaint ("TAC"), ECF No. 84-1,

The class as defined in the TAC was:

All persons in the United States who purchased software, products or services from McAfee, Inc. and subsequently accepted an Arpu pop-up advertisement offer presented at

the conclusion of the McAfee transaction and were charged by Arpu, Inc. for unused and

unclaimed software, products and services after McAfee transferred their credit/debit

card and other billing information to Arpu.

United States District Court

For the Northern District of California

TAC ¶ 5, ECF No. 84-1 (emphasis added). Notably, in every version of the complaint, despite 13 other minor differences, the class was consistently defined to include only those individuals who 14 had not downloaded the Arpu software. 15

16 and Arpu by an objector to this lawsuit. On February 16, 2011, Ken Pochis filed suit against Arpu, 17

McAfee, and an additional defendant, Iolo Technologies, Inc., for violations of RICO, the 18

Electronic Funds Transfer Act ("EFTA"), the Electronic Communications Privacy Act, and various 19 state law claims including the UCL and the CLRA. The Pochis class*fn1 was defined as "All persons 20 residing in the United States who, during the four year period preceding the date of filing of this 21

Class Action Complaint, did not provide their credit card, debit card, or billing information directly 22 to Arpu but whose credit card, debit card, or billing information was obtained by Arpu." Pochis 23

Complaint ("Pochis Compl.") ¶ 36, Case No. 11-CV-00721-LHK, ECF No. 1. The Pochis 24

Complaint contained no limitation regarding whether the class member had downloaded the Arpu 25 software or not. Thus, the Pochis class included both those individuals who downloaded the Arpu 26 27

The ad program between Arpu and McAfee also prompted another lawsuit against McAfee software, like Plaintiff Pochis himself, and those who did not download the Arpu software, like the 2 Ferrington Plaintiffs. 3

4 negotiations with Defendants. On July 1, 2011, Plaintiff Pochis and Defendants McAfee, Arpu and 5

Iolo filed a stipulation notifying the court that the parties in the Pochis lawsuit had agreed upon 6 material terms for a class-wide settlement in the Pochis action. Case No. 11-CV-00721-LHK, ECF 7

No. 11. However, the parties in the Pochis case never reached a final agreement and never filed a 8 motion for preliminary approval of the class settlement. In contrast, on July 13, 2011, the 9

The Pochis class counsel,*fn2 like the Ferrington class counsel, entered into settlement

Ferrington Plaintiffs filed a motion for preliminary approval of the class settlement in the instant 10 litigation. Case No. 10-CV-01455-LHK, ECF Nos. 86-87. 11

In anticipation of the preliminary approval hearing, the parties in the Ferrington case

13 provided the Court with the proposed class settlement (the "settlement"). The settlement defined 14 the class as: 15

B.Terms of the Settlement

[A]ll persons in the United States who during the Class Period purchased software from McAfee's website and subsequently accepted an Arpu pop-up advertisement

offer presented at the conclusion of the McAfee transaction and were charged by Arpu for the product, service or software sold in the Arpu pop-up.

See Settlement Agreement, ECF No. 88-1, ¶ 2.24. Notably, the class definition in the settlement is 18 broader than the class definition provided for in the previously filed complaints. In the TAC, the 19 class definition limited the class to those individuals who did not download the Arpu software as a 20 result of the pop-up ad. Specifically, the class definition in the TAC was limited to class members 21 who were charged by Arpu for "unused and unclaimed software, products and services." TAC ¶ 5, 22

ECF No. 84-1. In contrast, the class definition in the settlement agreement contains no such 23 limitation. In the settlement agreement, the class is defined to include both those individuals who 24 had, in fact, downloaded the Arpu software, and those who had not. As a result, as the parties 25 conceded at the preliminary approval hearing, the Ferrington settlement class is now essentially 26 27

identical to the class defined in the Pochis litigation. Prelim. Approval Hr'g Tr. at 25 ("Tr."), ECF 2

No. 112-2. 3

4 with Arpu, to pay claims made by class members. Under the terms of the settlement, class 5 members are given the option to elect between two types of benefits: either a cash benefit, or a free 6 license to McAfee software. The "Cash Settlement Benefit," provides either: (a) for those class 7 members who were billed by Arpu monthly, five dollars for each month the claimant paid charges 8 to Arpu for software, up to a maximum of thirty dollars per claim, and (b) for those class members 9 who were billed annually for an annual software license, five dollars for each prepaid month, up to 10 a maximum of thirty dollars per claim. The "Software Settlement Benefit" provides a software

Defendants set aside $1.2 million, the entire amount generated by the pop-up ad program

license for the current version of McAfee Family Protection (which sells at a retail price of

$49.99/year) for either: (a) six months for those class members who paid charges for fewer than six 13 months, or (b) 12 months for those class members who paid six months or more of charges. See 14

The settlement agreement provides that each class member will receive either the Cash

Settlement Benefit or the Software Settlement Benefit for each separate software for which they 17 can prove they paid. As a further condition of receiving any settlement benefit, the claims process 18 requires claimants to attest that they did not, in fact, download the Arpu software available through 19 the program. Thus, the benefits payable to the class provide a restitutionary remedy only to those 20 class members who did not receive the Arpu software for which they had paid. The class members 21 who actually downloaded the Arpu software are not entitled to any settlement benefits. Any funds 22 that are not paid to the class members will revert back to the Defendants. 23

24 from all federal and state claims that were or could have been raised in this action, upon final 25 approval by the Court. Id. ¶ 10.2.*fn3

In exchange for the benefits described above, class members agree to release Defendants

In addition to the release of claims and the scope of the class definition contained in the

Settlement Agreement, the parties in the Ferrington case also agreed to binding arbitration, subject 3 to this Court's approval, before a neutral third party as to the amount of attorneys' fees and 4 expenses that class counsel would seek from the Court. The parties were unable to agree upon an 5 attorneys' fee award, but were able to agree that the fee award should be between $250,000 and 6

$675,000. The arbitrator issued an opinion on July 14, 2011, determining that class counsel should 7 be awarded the total sum of $350,000 for fees and costs. See Arbitrator's Decision and Order on 8

Attorneys' Fees ("Arbitrator Decision") at 10, Friedman Decl. Ex. 7, ECF No. 106-7. These fees 9 and costs were to be awarded separate and apart from the claims fund made available by 10

Defendants to pay for the benefits to the class. Although the Arbitrator Decision discussed a 11 possible claims rate as high as 50 percent, the parties have indicated that the prevailing rule of

thumb with respect to consumer class actions is 3-5 percent. See Arbitrator Decision at 3; see also 13

Mem. In Supp. of Pl's Mot. for Attys' Fees, Expenses and Serv. Payments at 3, ECF No. 115. 14

legal or equitable or otherwise, that actually were, or could have been, asserted in the Litigation, 15 based upon any violation of any state or federal statutory or common law or regulation, and any claim arising directly or indirectly out of, or in any way relating to, the claims that actually were, or 16 could have been, asserted in the Litigation, that Plaintiffs have had in the past, or now have, related in any manner to the Released Parties' products, services, software or business affairs; (b) any and 17 all other claims, liens, demands, actions, causes of action, obligations, damages or liabilities of any nature whatsoever, whether legal or equitable or otherwise, that Plaintiffs have had in the past or 18 now have, related in any manner to any and all products, services, software or business affairs of Defendants, the Released Parties, and, in their representative capacities, all of Defendants' past and 19 present officers, directors, agents, designees, attorneys, employees, parents, subsidiaries, 20 associates, divisions, affiliates, heirs, and all successors or predecessors in interest, assigns or legal representatives. Plaintiffs expressly understand and acknowledge that it is possible that unknown 21 losses or claims exist or that present losses may have been underestimated in amount or severity. Plaintiffs explicitly took that into account in entering into this Agreement, and a portion of the 22 consideration and the mutual covenants contained herein, having been bargained for between Plaintiffs and Defendants with the knowledge of the possibility of such unknown claims, were 23 given in exchange for a full accord, satisfaction, and discharge of all such claims. Consequently, 24

Plaintiffs expressly waive all provisions, rights and benefits of California Civil Code section 1542 (and equivalent, comparable, or analogous provisions of the laws of the United States or any state 25 or territory thereof, or of the common law). Section 1542 provides: "A general release does not extend to claims which the creditor does not know or suspect to exist in his favor at the time of 26 executing the release, which if known by him must have materially affected his settlement with the debtor." (c) Each and every term of this section shall be binding upon, and inure to the benefit of 27 the Released Parties, and any of their successors and personal representatives, which persons and 28 entities are intended to be beneficiaries of this section.

C.Preliminary Approval

The Court held a hearing on the preliminary approval of the class settlement on August 4,

2011. At the hearing, Plaintiff Pochis, represented by James Patterson, expressed concern with 4 several aspects of the claims process in the Ferrington settlement. Specifically, Plaintiff Pochis 5 argued that the claims form was too onerous for class members because it required claimants to 6 provide proof of purchase in order to receive a minimal amount of money. Tr. at 9, 26. Plaintiff 7

Pochis also expressed concern that the $1.2 million pay out would not compensate class members 8 who downloaded the software. Tr. at 10. At the conclusion of the hearing, Plaintiff Pochis's 9 counsel stated that "[i]f we can work out the claim form, which it sounds like the court is telling 10 them to make a lot of the changes that we had problems with, then we probably wouldn't object to

13 those who downloaded the Arpu software and were not eligible for the settlement benefits had to 14 engage in additional steps to download the software. According to the additional filings, the 15 individuals who downloaded the software were sent an e-mail that indicated that they would be 16 billed at the end of the 30-day free trial period, and that McAfee was partnering with a third party. 17

Schuck Decl. ¶¶ 7-8; Ex. C, ECF No. 97. Customers had to affirmatively take extra steps to 18 download the software. Id. ¶ 8. Thus, the Ferrington class counsel argued that this subclass of 19 individuals was not actually harmed by the program. On August 16, 2011 the Court conditionally 20 certified a class and granted preliminary approval of the class settlement. ECF Nos. 100-101. 21

23 on November 17, 2011, Plaintiff Pochis filed an objection to the final approval of the Ferrington 24 settlement. Plaintiff Pochis primarily objected to the proposed settlement because: (1) the 25 settlement denies benefits to a "large number of class members," and (2) the class representatives 26 do not fairly and adequately protect the interests of the class. See Objection of Ken Pochis to 27 28 the settlement." Tr. at 60.

The Ferrington Plaintiffs filed additional documentation after the hearing showing that

D.The Pochis Objector

Despite the issues discussed, and apparently resolved, at the preliminary approval hearing,

Proposed Settlement 6-9, ECF No. 109. Additionally, Plaintiff Pochis raised concerns regarding 2 the class notice, the claims process, and the attorneys' fees awarded. 3

5 between the Plaintiffs and Defendants, the Federal Rules of Civil Procedure have vested with the 6 court the duty to oversee the final approval of all certified class settlements. Because class 7 settlements necessarily bind individuals who have not participated in the litigation, "judges have 8 the responsibility of ensuring fairness to all members of the class presented for certification." 9

II.Legal Standard and Analysis

Although the proposed settlement before the Court is a privately negotiated agreement

Staton v. Boeing Co., 327 F.3d 938, 952 (9th Cir. 2003). "Especially in the context of a case in 10 which the parties reach a settlement agreement prior to class certification, courts must peruse the 11 proposed compromise to ratify both the propriety of the certification and the fairness of the settlement." Id. 13

First, the district court must assess whether a class exists under Federal Rule of Civil Procedure 23(a) and (b); "[s]uch attention is of vital importance, for a court asked to certify a 15 settlement class will lack the opportunity, present when a case is litigated, to adjust the class, 16 informed by the proceedings as they unfold." Amchem Prods. Inc. v. Windsor, 521 U.S. 591, 620 17

(1997). Second, the district court must carefully consider "whether a proposed settlement is 18 fundamentally fair, adequate, and reasonable," pursuant to Federal Rule of Civil Procedure 23(e); 19 recognizing that "[i]t is the settlement taken as a whole, rather than the individual component parts, 20 that must be examined for overall fairness." Hanlon v. Chrysler Corp., 150 F.3d 1011, 1026 (9th 21

Cir. 1998) (citations omitted). With due respect to the rights of the parties to compromise and 22 reach a private resolution to litigation, the Court has serious concerns regarding the settlement 23 agreement proposed by the parties. Ultimately, the Court finds that the agreement does not pass 24 muster under either the class certification requirements presented in Rule 23(a), or under the 25 standard for class settlement approval under Rule 23(e), and so declines to grant Plaintiffs' motion 26 for final approval of the class settlement. 27

A.Class Certification

2 certification is necessary in a settlement context. Amchem Prods., Inc., 521 U.S. at 620. The 3

The Supreme Court has instructed that "undiluted, even heightened, attention" to class "threshold task" of approving a certified class settlement is to ensure that the proposed settlement 4 class satisfies all the requirements of Rule 23(a): (1) numerosity, (2) commonality, (3) typicality, 5 and (4) adequacy of representation. Hanlon, 150 F.3d at 1019. Additionally, the proposed class 6 must also be maintainable under Rule 23(b)(1), (2), or (3). Id. at 1022. A proposed class must 7 meet each of the Rule 23(a) requirements, and failure to meet any one of the requirements is fatal 8 to certification of the class. Here, because the proposed class fails the commonality, typicality, and 9 adequacy of representation requirements, the class is not certifiable for settlement purposes. 10 1.Commonality

Rule 23(a)(2) requires that there be "questions of law or fact common to the class."

"Commonality exists where class members' 'situations share a common issue of law or fact, and 13 are sufficiently parallel to insure a vigorous and full presentation of all claims for relief.'" Wolin v. 14

Jaguar Land Rover N. Am., LLC, 617 F.3d 1168, 1172 (9th Cir. 2010) (quoting Cal. Rural Legal 15

Assistance, Inc. v. Legal Servs. Corp., 917 F.2d 1171, 1175 (9th Cir. 1990)). "The existence of 16 shared legal issues with divergent factual predicates is sufficient, as is a common core of salient 17 facts coupled with disparate legal remedies within the class." Id. (quoting Hanlon, 150 F.3d at 18

The parties' proposed settlement consists of two distinct subclasses: those who downloaded

20 the Arpu software ("the downloaders"), and those who did not ("the non-downloaders"). The 21 parties have asserted that there are no records indicating which of the approximately 57,000 class 22 members fall into each of the subclasses. It is not even possible to know, therefore, what 23 percentage of the settlement class is entitled to recover the settlement benefits, and what percentage 24 is not entitled to recover anything under the terms of the agreement. Fundamentally, the situations 25 of the two subclasses are distinct because they raise different claims. Those who did not download 26 the Arpu software claim that they were charged for something that they never received. In 27 contrast, the downloaders are in a different position. Some class members, like Pochis, may have 28 been misled into believing that the software was provided by McAfee; never intended to purchase 2 the software despite the download; or did not intend to transfer their credit or debit card 3 information to a third party. On the other hand, other downloaders may have knowingly and 4 purposefully downloaded the Arpu software with the intention to pay the monthly charges after the 5 30-day free trial period ended. Thus, the claims of the subclasses are not common. 6

8 or defenses of the representative parties are typical of the claims or defenses of the class." Fed. R. 9


The typicality inquiry under Rule 23(a)(3) requires that Plaintiffs establish that "the claims Civ. P. 23(a)(3). "The test of typicality is whether other members have the same or similar injury, 10 whether the action is based on conduct which is not unique to the named plaintiffs, and whether

other class members have been injured by the same course of conduct." Hanon v. Dataproducts

Corp., 976 F.2d 497, 508 (9th Cir. 1992) (internal citations and quotation marks omitted). 13

"Typicality refers to the nature of the claim or defense of the class representative, and not to the 14 specific facts from which it arose or the relief sought." Id. (internal citation and quotation marks 15 omitted). 16

Christi Hall, are all members of the subclass who did not download the Arpu software. None of the 18 class representatives is a member of the downloader subclass. Because the class representatives 19 are entitled to recover under the terms of the settlement, while the remaining class members may 20 not, the claims of the named Plaintiffs are not typical of the claims of the class as a whole. 21

Accordingly, Plaintiffs have not established the typicality requirement. 22

24 interests of the class." "Without adequate representation, a court order approving a claim-25 preclusive class action settlement agreement cannot satisfy due process as to all members of the 26 class." Hesse v. Sprint Corp., 598 F.3d 581, 588 (9th Cir. 2010) (citing Phillips Petroleum Co. v. 27

The class representatives, Melissa Ferrington, Cheryl Schmidt, Christopher Bennett, and


Rule 23(a)(4) requires that "the representative parties will fairly and adequately protect the

Shutts, 472 U.S. 797, 812 (1985)). "Resolution of two questions determines legal adequacy: (1) do 28

the named plaintiffs and their counsel have any conflicts of interest with other class members and 2

(2) will the named plaintiffs and their counsel prosecute the action vigorously on behalf of the 3 class?" Hanlon, 150 F.3d at 1020. 4

5 represented in this action by both named plaintiffs and class counsel. Yet, no class representatives 6 have been appointed to represent the interests of the downloader subclass whose claims are 7 released as a result of this settlement agreement. Moreover, counsel has not been assigned to 8 represent the interests of this subclass. Such a state of affairs presents a conflict of interest: the 9 subclasses have conflicting interests regarding how the settlement fund should be allocated 10 amongst them. Thus, the Court cannot say that the interests of all class members have been

The subclass of individuals who did not download the Arpu software has been capably

adequately represented by the named plaintiffs and class counsel in negotiating the terms of the settlement agreement. See id. (explaining that the Supreme Court "instructs [the district court] to 13 give heightened scrutiny to cases in which class members may have claims of different strength"); 14

Cir. 1995) ("[A] judge must focus on the settlement's distribution terms (or those sought) to detect 16 situations where some class members' interests diverge from those of others in the class. For 17 example, a settlement that offers considerably more value to one class of plaintiffs than to another 18 may be trading the claims of the latter group away in order to enrich the former group."). 19

Accordingly, Plaintiffs have failed to establish the requirements necessary for the Court to certify 20 the proposed class*fn4 for settlement purposes. 21

Plaintiffs must also establish that the settlement complies with Rule 23(e), which requires

23 court approval for the settlement of any class action. In order to be approved, a settlement must be 24

B.Fair, Adequate, and Reasonable Settlement

Inc., 666 F.3d 581 (9th Cir. Jan. 12, 2012). The Court takes no position on this issue at this time, but notes that should the parties decide to pursue a negotiated settlement of class claims, they 28 should take into account the controlling Ninth Circuit law when crafting a settlement.

"fundamentally fair, adequate and reasonable." Torrisi v. Tucson Elec. Power Co., 8 F.3d 1370, 2

1375 (9th Cir. 1993). The factors in a court's fairness assessment will naturally vary from case to 3 case, but courts generally weigh: (1) the strength of the plaintiff's case; (2) the risk, expense, 4 complexity, and likely duration of further litigation; (3) the risk of maintaining class action status 5 throughout the trial; (4) the amount offered in settlement; (5) the extent of discovery completed and 6 the stage of the proceedings; (6) the experience and views of counsel; (7) the presence of a 7 governmental participant; and (8) the reaction of the class members of the proposed settlement. 8

Although several of the Churchill factors -- including the risk and expense of continued

10 litigation, the fact that through discovery the parties have developed a factual record, and the

Churchill Vill., L.L.C. v. Gen. Elec., 361 F.3d 566, 575 (9th Cir. 2004); Torrisi, 8 F.3d at 1375. 9

experience and views of capable counsel favor granting the settlement -- favor granting final

approval, the Ninth Circuit has indicated that applying these factors mechanically is not sufficient 13 to determine whether final approval is appropriate. In re Bluetooth Headset Prods. Liability Litig., 14

654 F.3d 935, 947 (9th Cir. 2011). Ultimately, the Ninth Circuit has explained that the district 15 court, in determining whether a certified class settlement is "fair, adequate, and reasonable," must 16 ensure that the agreement is not the product of fraud, overreaching by, or collusion among, the 17 negotiating parties. Id.; Class Plaintiffs v. City of Seattle, 955 F.2d 1268, 1290 (9th Cir.), cert. 18 denied sub nom. Hoffer v. City of Seattle, 506 U.S. 953 (1992). When the class settlement precedes 19 formal, adversarial class certification, the approval of the settlement requires a higher standard of 20 fairness. Hanlon, 150 F.3d at 1026; see also In re Bluetooth, 654 F.3d at 946 ("Prior to formal 21 class certification, there is an even greater potential for a breach of fiduciary duty owed the class 22 during settlement. Accordingly, such agreements must withstand an even higher level of scrutiny 23 for evidence of . . . conflicts of interest than is ordinarily required under Rule 23(e).").

26 adequate] * * * is the need to compare the terms of the compromise with the likely rewards of 27 litigation." In re TD Ameritrade Accountholder Litig.,266 F.R.D. 418, 422 (N.D. Cal. 2009); 28

1.Release of Claims

"Basic to [the process of deciding whether a proposed settlement is fair, reasonable and Rodriguez v. W. Publ'g Corp., 563 F.3d 948, 965 (9th Cir. 2009). Settlements in which the class or 2 a significant subclass will receive no benefit have been rejected by courts because such agreements 3 are not fair, adequate, and reasonable for all class members. See, e.g., In re TD Ameritrade 4

Accountholder Litig., 266 F.R.D. at 423; Molski v. Gleich, 318 F.3d 937, 953-54 (9th Cir. 2003),

overruled on other grounds by Dukes v. Wal--Mart Stores, Inc., 603 F.3d 571, 617 (9th Cir. 2010), 6 rev'd by Wal-Mart v. Dukes, Inc., 131 S.Ct. 2541 (2011); Mirfasihi v. Fleet Mortg. Corp., 356 F.3d 7

Cir. 2000). Similarly, disparate treatment between class members increases the likelihood that the 9 settlement agreement does not meet the Rule 23(e) standard. See Hanlon, 150 F.3d at 1027; In re 10

13 the Arpu software will receive some form of compensation. The parties explained at the fairness 14 hearing that the settlement benefits are intended to compensate class members for charges that 15 were improperly posted to their credit or debit cards and for which they received no product. Thus, 16 those individuals who downloaded the software are not entitled to the negotiated restitutionary 17 remedy and will receive nothing from the class settlement. 18

Ferrington Plaintiffs argue that the settlement is reasonable to all members of the Ferrington

19 class and that the disparity in benefits between the subclasses is attributable to the relative strengths 20 of the subclasses' respective claims. Ferrington Plaintiffs argue that the downloader subclass 21 received the software for which they paid. Moreover, because the downloaders were required to go 22 through additional steps before downloading the product, their purchases were not inadvertent. 23

Thus, the members of the subclass who downloaded the software were not harmed, and therefore, 24 have no viable claims against the Defendants that would justify settlement benefits. From 25

Ferrington Plaintiffs' perspective, the terms of the compromise for this subclass are perfectly 26 justified because their claims are worthless. 27 28

781, 782-3 (7th Cir. 2004); Crawford v. Equifax Payment Servs., Inc., 201 F.3d 877, 880-82 (7th

General Motors, 55 F.3d at 807 ("One sign that a settlement may not be fair is that some segments of the class are treated differently from others.").

Under the terms of the settlement agreement here, only the subclass who did not download

2 does not necessarily mean that the downloaders suffered no compensable harm. The downloader 3 subclass may still have claims arising out of the McAfee program that will be extinguished through 4 the class settlement here. For example, the Pochis Complaint alleges that he and other class 5 members were misled into downloading the software, believing it to be a McAfee product and 6 believing it to be a free 30-day trial. Pochis Compl. ¶ 18-19. Pochis claims that he gave McAfee 7 his debit card information, but never intended McAfee to transmit this information to third parties. 8

While the Court agrees that the claims of the downloader subclass are relatively weak, that

Id. ¶¶ 19; 30-32. Additionally, Pochis claims that when he realized that his debit card was being 9 charged for the software that he had purchased and downloaded, he attempted to contact McAfee 10 regarding the disputed charge. Id. ¶ 33. Pochis claims that McAfee responded that it had no

affiliation with the Arpu software, and "that it would never had [sic] given his billing information

to a third party." Id. Pochis claims that when he attempted to discontinue the monthly charges, he 13 was unable to contact anyone at Arpu to help him and that the only way to discontinue the charges 14 was to cancel his debit card. Id. ¶ 35. 15

16 defendant, violates EFTA, among other causes of action. EFTA establishes the rights, liabilities, 17 and responsibilities of participants (consumers, financial institutions, and intermediaries) in 18 electronic fund transfer systems. In re Easysaver Rewards Litig., 737 F. Supp. 2d 1159, 1181 (S.D. 19

Cal. 2010) (citing 15 U.S.C. § 1693). EFTA governs, for example, automatic direct payroll 20 deposits, transactions at automated teller machines, and debit card purchases at retail or online 21 shops. See id. (citing 15 U.S.C. § 1693a(6), (7)). EFTA provides that a "preauthorized electronic 22 fund transfer from a consumer's account may be authorized by the consumer only in writing, and a 23 copy of such authorization shall be provided to the consumer when made." 15 U.S.C. § 1693e(a). 24

Similar claims have been recognized as viable in other cases alleging similar factual

25 circumstances. For example, in Easysaver, the district court denied, in part, defendants' motion to 26 dismiss plaintiffs' EFTA and CLRA claims. In Easysaver, plaintiffs were redirected to a third 27 party website offering a discount on future purchases. Rather than receiving a "free gift," however, 28

Pochis has alleged that the scheme perpetrated by McAfee and Arpu, as well as a third

customers were enrolled in a separate discount scheme for which they were charged monthly. 737 2

F. Supp. 2d at 1164. Like Pochis, in Easysaver, efforts to have the charges reversed were 3 unsuccessful and the plaintiff had to close her bank account to stop the unauthorized charges. See 4 id. In Easysaver, the district court concluded that the EFTA claim survived a motion to dismiss 5 because the electronic monthly payment was either not pre-authorized or violated the provision of 6

1182-83. Additionally, plaintiff's EFTA claim that the defendant continued to withdraw funds 8 after plaintiff called to cancel her membership also survived a motion to dismiss. It does not 9 appear that the EFTA claim turns on whether Plaintiffs receive a benefit from the transaction, but 10 rather whether the authorization and notice provisions of the statute are met. See id.

EFTA requiring the customer be provided with a copy of the authorization to deduct funds. Id. at 7

Thus, EFTA provides for statutory damages, which, while minimal, may still provide potential relief for those who downloaded the Arpu software and are barred from seeking relief 13 from Arpu and McAfee*fn5 under the terms of the Ferrington settlement. See 15 U.S.C. § 1693m(a). 14

It is troubling, at the very least, that the subclass of individuals who downloaded the Arpu software 15 will necessarily be releasing all of their claims without any compensation despite the pending 16

Pochis lawsuit, and despite the potential viability of some of their claims. "Such [claims] would 17 not be a sure bet, but colorable legal claims are not worthless merely because they may not prevail 18 at trial." Mirfasihi, 356 F.3d at 783. 19

Additionally, the actions of Defendants in the Pochis litigation appear to undermine the Ferrington Plaintiffs' arguments that the claims of the downloader subclass are worth no 21 consideration. The parties in the Pochis litigation engaged in negotiations and agreed upon 22 material terms for a class-wide settlement. Even without the benefit of knowing the terms of the 23 settlement agreement in the Pochis litigation, the Court can safely say that the Pochis plaintiffs 24 were likely to settle for some consideration. This certainly suggests that, regardless of the strength 25

of the claims, extinguishing the Pochis class's claims for no consideration would be unfair and 2 unreasonable. 3

4 negotiated class settlement contained a provision that provided benefits for only one subclass. The 5 other subclass received nothing, while surrendering all its members' claims against the defendant. 6

The Seventh Circuit reversed the district court's grant of final approval of the class settlement, in 7 part because the agreement released all of the claims of a subclass for essentially no consideration. 8

A similar set of circumstances arose in the settlement presented in Mirfasihi, in which a

See 356 F.3d at 783-85. The Seventh Circuit concluded that while the subclass that was to receive 9 no consideration may not have had strong claims, the claims were not worthless, and therefore the 10 settlement could not stand. Id. at 783, 786. Likewise here, the claims of the downloader subclass 11 are not so meritless that releasing the claims for no consideration is fair and reasonable.

14 does not adequately protect the interests of all class members. These signs include (1) "when 15 counsel receive a disproportionate distribution of the settlement," and (2) "when the parties arrange 16 for fees not awarded to revert to defendants rather than be added to the class fund." In re 17

Bluetooth, 654 F.3d at 947; see also Mirfasihi, 356 F.3d at 785. Several of these red flags are 18 present in the settlement agreement here and convince the Court that final approval of the proposed 19 settlement agreement is not warranted. 20

First, class counsel*fn6 will receive a disproportionately large percentage of the total amount

21 that Defendants have agreed to pay to settle this lawsuit. The parties selected a third party neutral 22 to resolve the fee dispute and provide a binding arbitration award between $250,000 and $675,000. 23

The parties apparently justified the attorneys' fee range based in part on the assumption that in a 24

2.Other Factors

In addition, the Ninth Circuit has identified other subtle signs that the proposed settlement consumer class action the participation rate could be as high as 50 percent. The arbitrator 2 determined that, in light of the range of the fee award, the total fees accrued by Plaintiffs, and the 3 specific facts and circumstances of the case, a total attorneys' fee award of $350,000 was 4 appropriate.*fn7 See Arbitrator Decision at 10. 5

6 class is $1.2 million. On April 5, 2012, the parties provided updated claims information. See 7

The total amount that Defendants have agreed to set aside for claims to members of the

Parties' Updated Claims Information, ECF No. 117. Confusingly, in the parties' updated claims 8 information, the parties represent that there are 57,238 class members. Id. at 2. By contrast, in the 9 briefing on the motion for final approval, the parties asserted that the class size was approximately 10

claims information states that the claims rate is approximately 3.8%. Parties' Updated Claims

Information at 2, ECF No. 117. However, if the actual class size is 66,000 members, the claims 13 rate could be lower (3.3%). The class notice date was September 19, 2011, and the deadline to 14 submit claim forms was January 14, 2012. See Settlement Website, 15 Nonetheless, the Court notes that the participation rate (and 16 thus the claims pay out) could theoretically increase because, pursuant to section 2.7 of the 17 settlement agreement, the claims period is open until 30 days after final settlement approval. 18

However, based on the notice to the class and the deadline for submitting claim forms, it is unlikely 19 that the claims rate will significantly increase. 20

21 the class is $67,825 ($43,675 in cash $23,350 valued for 12 month licenses $800 valued for 6 22 month licenses) in comparison to the $350,000 attorneys' fee award. Thus, the attorneys' fee 23 award is more than 83% percent of the total amount paid by Defendants to settle this lawsuit. The 24

66,000 members. See Mot. for Final Approval at 3, 13, 16, ECF No. 102. The parties' updated

The parties' April 5, 2012 updated claims information states that the total claims payable to

attorneys' fee award far exceeds the Ninth Circuit "benchmark" of 25% for a reasonable fee award, 2 and represents a disproportionate distribution of the settlement funds. Cf. Six (6) Mexican Workers 3

Moreover, the parties should have foreseen that the settlement would produce a situation in

5 which the attorneys' fees comprised a disproportionate distribution of the settlement. The parties 6 conceded at the hearing that they did not expect the claims rate to reach the 50 percent benchmark 7 cited in the arbitrator's fee decision. Instead, Plaintiffs asserted that a participation rate of 3 to 5 8 percent would be a "good claims rate" in a consumer class action. Under the settlement terms, 9 class members are entitled to receive $5 to $30 in cash depending upon how many months the class 10 member was charged for the Arpu software. Assuming a 3 percent claims rate, the total amount

paid to the class would be anywhere from $8,550 (57,000 class members x 3% participation x $5) 12 to $51,300 (57,000 class members x 3% participation x $30). Assuming a 5 percent claims rate, 13 the total paid to the class would be anywhere from $14,250 (57,000 class members x 5% 14 participation x $5) to $85,500 (57,000 class members x 5% participation x $30). Alternatively, 15 class members also have the option to elect a McAfee product, valued up to $49.99, as an in-kind 16 benefit. Assuming a 3 to 5 percent participation rate based solely on the election of the McAfee 17 product,*fn8 the value to the class of the settlement benefit would be between $42,750 (57,000 class 18 members x 3% participation x $25) to $142,500 (57,000 class members x 5% participation x $50). 19

Thus, assuming a 3 to 5 percent participation rate, the negotiated settlement was likely to produce 20 class benefits worth somewhere between $8,550 and $142,500. 21

Plaintiffs have indicated that the average subscription length for class members was 3.7

22 months. Pls' Mem. In Supp. of Mot. for Prelim. Approval at 10, ECF No. 87. Because the 23 maximum benefits require subscription lengths of six months or longer, much of the class was 24 never eligible for the maximum benefits, and the total settlement award received by the class was 25 unlikely to ever reach $142,500. Thus, the parties should have foreseen that the settlement was 26

v. Ariz. Citrus Growers, 904 F.2d 1301, 1311 (9th Cir.1990). 4

likely to result in a disproportionate attorneys' fee award. In light of the actual disparity between 2 the class award and the requested attorneys' fee award, and the fact that the structure of the 3 settlement was likely to produce such a result, the Court is concerned that the settlement, taken as a 4 whole, is not fair, reasonable, and adequate. See In re Bluetooth, 654 F.3d at 948-49 ("Even when 5 technically funded separately, the class recovery and the agreement on attorneys' fees should be 6 viewed as a 'package deal.'") (citations omitted); Hanlon, 150 F.3d at 1021 (explaining that a 7 subtle sign that the product of the negotiations does not meet the requirements under Rule 23(e) 8 arises "when counsel receive a disproportionate distribution of the settlement"). 9

10 remain after all of the claims are distributed to class members. The parties confirmed at the final 11 approval hearing that the settlement is, in fact, a claims-mad

12 distribution. In other words, only claims are paid, and any remainder will revert to Defendants. A 13 reversion provision, like the one contained in the agreement here, is yet another indication that the 14 proposed agreement does not meet the standards set forth in Rule 23(e). See Mirfasihi, 356 F.3d at 15

785 ("[T]he reversion of unclaimed funds to the putative wrongdoer" is a "questionable feature" of 16 a certified class settlement agreement.). 17

Class action settlements can be a great benefit to consumers because they allow suits for

18 damages in cases that would otherwise be economically infeasible to prosecute. See In re General 19

A second, related concern regarding the settlement arises from the residual funds that

e settlement with no minimum

Motors, 55 F.3d at 784. Especially in a case such as this, where the Ferrington class has 20 meritorious claims and a settlement would allow consumers to be compensated for questionable 21 business practices by the Defendants, the Court is reluctant to deny final approval of the class 22 settlement. Nonetheless, the settlement presents many features that the Ninth Circuit has indicated 23 suggest that the proposed settlement does not meet the standard set forth in Rule 23(e). Granting 24 final approval would require an unknown subset of the class to relinquish its claims against 25

Defendants for no consideration. Moreover, additional features of the proposed settlement, 26 including a low rate of claims participation, a small claims pay out, a disproportionately large 27 28

attorneys' fee award, and a reversion clause, convinces the Court that final approval of the class 2 settlement is not appropriate here. 3

4 class be redefined to exclude those class members who downloaded the Arpu software. Objection 5 of Ken Pochis to Proposed Settlement at 11-12, ECF No. 109. The Court, however, has no 6 authority to do so. Courts do not have the ability to "delete, modify or substitute certain 7 provisions. The settlement must stand or fall in its entirety." Browning v. Yahoo!, Inc., No. 04-8

Therefore, the Court is compelled to deny Plaintiffs' motion for final approval of the class 10 settlement.

As an alternative to denying final approval of the class settlement, Pochis requests that the

CV-01463, 2006 WL 1390555, at *1 (N.D. Cal. May 19, 2006) (citing Hanlon, 150 F.3d at 1026). 9

Nonetheless, guidance may be useful here to aid the parties in attempting to reach

resolution of this case. Many of the concerns with the proposed settlement identified in this Order 13 would be addressed if the class definition had conformed to the class definition in the TAC, which 14 excluded from the definition of the class those individuals who had actually downloaded the Arpu 15 software.*fn9 Alternatively, to the extent the parties wish to reach a global settlement in both this case 16 and the related class action in which Pochis is a named plaintiff, the parties must take care to 17 ensure that all subclasses are adequately represented and to give adequate consideration to all 18 members of each subclass. 19

21 approval of a class settlement, and so does not deny Plaintiffs' motion lightly. Staton, 327 F.3d at 22

952 ("Judicial review also takes place in the shadow of the reality that rejection of a settlement 23 creates not only delay but also a state of uncertainty on all sides, with whatever gains were 24 potentially achieved for the putative class put at risk."). While "[d]enial of final approval of the 25 proposed settlement presents an unusual situation," In re TD Ameritrade Accountholder Litig., 266 26


The Court recognizes that real costs to the parties necessarily flow from the denial of final

F.R.D. at 424, it is necessary here to protect the rights of the absent class members of the subclass 2 of claimants who downloaded the Arpu software and who are essentially releasing their claims 3 against McAfee and Arpu for no consideration. In light of the reservations identified above, the 4

Court cannot find that the proposed class may be properly certified or that the proposed settlement 5 is fair, reasonable and adequate. Accordingly, the motion for final approval is DENIED without 6 prejudice. 7

Additionally, the Order conditionally granting class certification for the purposes of the

8 settlement agreement is set aside. See ECF No. 101. A further case management conference is set 9 for May 23, 2012 at 2:00 p.m. The parties shall come prepared to discuss scheduling and other 10 matters. 11


Buy This Entire Record For $7.95

Official citation and/or docket number and footnotes (if any) for this case available with purchase.

Learn more about what you receive with purchase of this case.