Chavez v. Netflix, Inc.

Related Cases

75 Cal. Rptr. 3d 413 (2008) 162 Cal. App. 4th 43

Frank CHAVEZ, Plaintiff and Respondent,
v.
NETFLIX, INC., Defendant and Respondent;
Laura Ellis, Objector and Appellant. Frank Chavez, Plaintiff and Respondent,
v.
Netflix, Inc., Defendant and Respondent;
David Meininger et al., Objectors and Appellants.
Frank Chavez, Plaintiff and Respondent,
v.
Netflix, Inc., Defendant and Respondent;
John Vogel, Objector and Appellant.

Nos. A114334, A115395, A115571.

Court of Appeal of California, First District, Division One.

April 21, 2008.

*418 Daniel Brown, for Objector and Appellant Laura Ellis.

Steven F. Helfand, San Francisco, for Objectors and Appellants David Meininger et al.

Kendrick & Nutley, J. Garrett Kendrick, C. Benjamin Nutley, for Objector and Appellant John Vogel.

Gutride Safier Reese, Adam J. Gutride, Seth A. Safier, for Plaintiff and Respondent Frank Chavez.

Wilson Sonsini Goodrich & Rosati, Keith E. Eggleton, Rodney G. Strickland, Jr., Dale R. Bish, Palo Alto, for Defendant and Respondent Netflix, Inc.

MARGULIES, J.

Frank Chavez sued Netflix, Inc. (Netflix) over its practice of advertising that it would send customers "`unlimited'" DVD rentals with "1 Day Delivery" for a flat monthly fee. Alleging that both selling points were false, Chavez sought injunctive relief and damages on behalf of himself and a class of current and former Netflix subscribers. Before the class was certified, Netflix agreed to settle the class action by providing one month of free DVD rental services or upgrades to class members who claimed the benefit. The trial court approved the `settlement and awarded attorney fees of $2,040,000 to be paid by Netflix to class counsel.

The appellants in these consolidated appeals objected to the class action settlement and fee award in the trial court. They contend that the trial court abused its discretion in approving the settlement, affording notice to class members, and determining the amount of fees. Finding no abuse of discretion, we affirm the orders in issue.

I. BACKGROUND

A. The Lawsuit

Netflix offers its members an on-line movie rental service in which for a set monthly fee, the members can order movies on digital video discs (DVD's) via the Internet. Netflix ships the DVD's to the *419 member by first-class United States mail. Based on the level of monthly fee paid, the member is allowed to have a specified maximum number of DVD's on loan at any one time, e.g., three for $17.99 per month (priced at the time the settlement was approved) under the most popular plan. Once the maximum number of DVD's have been mailed to the member, the member obtains a new DVD rental by returning one of the DVD's in a prepaid return envelope Netflix provides. When a returned DVD is received by Netflix, Netflix ships the member another DVD chosen from a priority list the member has created on the Netflix Web site. Netflix informs its member by e-mail when it mails a DVD to the member and when it has received a DVD mailed back by the member.

On September 23, 2004, plaintiff Frank Chavez filed a putative class action lawsuit against Netflix over its practice of advertising "`unlimited'" DVD rentals with "1 Day Delivery" for a flat monthly fee. Chavez alleged that these claims were false and misleading, and that Netflix had breached its contract with him and other members, engaged in fraud and deceit, and committed false advertising and unfair trade practices in violation of California law. Chavez alleged that contrary to its advertising Netflix was employing sophisticated algorithms to prioritize the allocation of its DVD's to its lowest-consuming members with the effect that high-consuming members would receive fewer DVD's per month, reducing the costs Netflix incurred to serve this high-usage group, and increasing its profits. Chavez sought restitution, compensatory damages, punitive damages, and injunctive relief.

Netflix denied the allegations and asserted affirmative defenses. Netflix argued, among other things, that (1) there could be no proof of damages because members were informed by e-mail each time a returned DVD was received or a new one sent, as well as the expected delivery date, so the actual delivery speeds were not hidden; and (2) members were free to cancel at any time so if they did not cancel they must have determined that they were receiving a sufficient benefit for the fees they were paying. After the litigation was initiated, and partly in response to it, Netflix altered the terms of use disclosed on its Web site to inform members that in determining shipping and inventory priorities it gave priority to those members who received the fewest DVD's.

The parties conducted extensive discovery. Netflix produced approximately 86,000 pages of documents, answered more than 200 interrogatories and 59 requests for admissions, and made five of its employees, including three executives, available for deposition by Chavez. Chavez produced documents and answered interrogatories. In September 2005, Chavez moved for class certification. Netflix intended to oppose the motion and believed it had valid defenses to class certification. While the motion was pending, the parties reached agreement to settle the lawsuit.

B. The Original Settlement Agreement

In August and September 2005, the parties engaged in mediation conducted by a retired federal magistrate judge and reached a settlement (the Original Agreement). Netflix agreed to an injunction imposing changes in how it advertised its DVD rental programs and how it described them in the registration process. The changes further expanded upon the unilateral changes that Netflix had made to its terms of use after the suit was filed. Under the Original Agreement, Netflix agreed to provide all persons who were current Netflix members as of October 19, 2005, and who submitted an on-line claim *420 form, a one-level membership upgrade for one month, allowing the current members to receive one additional DVD at a time at no charge. This would allow the typical member to receive approximately four additional DVD rentals during the month. All persons who were former Netflix members as of October 19, 2005, and who submitted on-line claim forms, would be provided a free one-month membership at the three-at-a-time level, which would allow the former member to receive a minimum of three and up to 11 or more rentals at no charge. The Original Agreement included an auto-renewal feature whereby (1) current class members who did not act to cancel the upgraded service at the conclusion of the month would continue with the upgraded service level billed at Netflix's regular subscription rate for the upgraded program, and (2) those in the class of former members who did not act to cancel the service after their free month would continue to be renewed as Netflix members unless and until they canceled their service.

With certain changes, the trial court preliminarily approved the settlement on October 27, 2005. The court approved the provision of notice of the settlement to the class members as follows: Notice was to be given by e-mail at the e-mail addresses Netflix uses to communicate with its members or, `in the case of former members, using the e-mail addresses it had previously used to communicate with the former members. Follow-up mail notice would be sent to those whose e-mail addresses came back as undeliverable. The e-mail notices would summarize the terms of the settlement and the class members' rights to make claims and opt out or object, and the deadlines for doing so. The notice would contain references and hypertext links to a settlement Web site that would contain a more detailed settlement notice, the claim form, the settlement agreement itself, a list of frequently asked questions, and a list of important deadlines.

C. Objections and Amended Agreement

After notice was issued, law firms claiming to represent approximately 450 of the 5.5 million class members submitted briefs challenging the adequacy of the settlement on various grounds. In addition, the Federal Trade Commission (FTC) filed an amicus brief asserting that the auto-renewal feature of the settlement was not in the best interests of consumers. After meeting with the FTC in January 2006, Netflix agreed to eliminate the auto-renewal feature and to provide a second notice to class members offering them a remedy that did not include the auto-renewal feature. Class counsel agreed to incorporate the changes into an amended agreement. Because the Original Agreement was being modified to eliminate the auto-renewal provision, the parties agreed to modify it in other respects as well to address certain other issues raised by the objectors.

In response to the amended settlement agreement, the FTC and 428 of the objectors formally withdrew their objections to the settlement. After holding two hearings on the fairness of the settlement terms, on February 22 and March 22, 2006, the trial court issued an order rejecting the remaining objections and finally approving the settlement.

The order approving settlement directed Netflix to issue supplemental notice to the class. Class members who had registered for the benefit following the original notice were not required to re-register. Class members who had opted out were given a second opportunity to accept the remedy without the auto-renewal feature. Class members who had not responded to the original notice were notified again. Like the original notice, the supplemental notice *421 included a hyperlink to a Web site with more detailed information about the settlement and the procedures for registering or opting out. The supplemental notice went out in May 2006. By the end of the registration period, 697,532 class members had registered for the benefit and 1,234 (approximately 0.2 percent of the class) requested exclusion.

D. Attorney Fees

The trial court preliminarily awarded $1.3 million in fees to class counsel. It reduced class counsel's proposed lodestar amount from $1,198,200 to $805,000 on grounds of duplication, an excessive billing rate for document review, and time spent negotiating the amended settlement agreement. In the court's view, the latter did not provide any tangible benefit to the class because it consisted of correcting aspects of the Original Agreement that should not have been in it in the first place. The court stated that it would apply a multiplier to the reduced lodestar amount based on "the success achieved, the quality of the representation, and most particularly the rate of acceptance of the benefit offered to class members."

The court calculated that the total value of the benefits claimed by class members was $4.29 million as of March 22, 2006. Assuming that contingency fees generally ranged between 20 and 40 percent of total recovery after deducting costs, and that the total recovery in this case would include the value of the class benefit plus the fee award, the trial court made a tentative award of $1.3 million in fees, or 23.3 percent of the total of the class benefit value to date plus the fee award. The court reserved the right to make an additional and enhanced fee award upon a showing of further enrollment by class members for the benefit, with the stipulation that the final award would not exceed $2.5 million.

On July 28, 2006, based on the final number of claims, the trial court recalculated the value of the settlement at $7,293,600. It increased the fee award to $2,040,000, or 21.8 percent of the total settlement value, including fees.

E. The Appeals

Three notices of appeal were filed by four objectors. Appellant Laura Ellis asserts error in the denial of her motion to intervene, and asserts objections to the amended settlement agreement, the notices given to class members, and the attorney fees awarded to class counsel (case No. Al 14334). Appellants David Meininger and John W. Davis object to the amount of fees awarded (case No. Al 15395). Appellant John Vogel objects to the notices given to class members and to the fees awarded to class counsel (case No. Al 15571). Upon motion of the parties, we consolidated the three appeals for briefing, argument, and decision.

II. DISCUSSION

We review the following contentions by appellants: (1) the trial court abused its discretion in denying appellant Ellis's motion for leave to intervene, (2) the amended settlement agreement is an improper "coupon settlement" that fails to promote the purposes of class litigation, (3) notice of the amended settlement agreement was deficient in various respects, and (4) the fees awarded to class counsel were excessive and unsupported.

A. Motion for Leave to Intervene

Ellis argues that she should have been granted leave to intervene in order to gain access to the discovery of documents material to her objections to the settlement agreement and attorney fee award. In particular, Ellis maintains that the denial of her motion prevented her from obtaining *422 access to attorney time records. She argues that the trial court abused its discretion by denying her motion without giving an explanation of its reasons.[1]

The trial court has discretion to permit a nonparty to intervene where: (1) the proper procedures have been followed, (2) the nonparty has a direct and immediate interest in the action, (3) the intervention will not enlarge the issues in the litigation, and (4) the reasons for the intervention outweigh any opposition by the parties presently in the action. (Reliance Ins. Co. v. Superior Court (2000) 84 Cal. App. 4th 383, 386, 100 Cal. Rptr. 2d 807.) We review an order denying leave to intervene under the abuse of discretion standard. (Ibid.)

We note first that Ellis's request to intervene was untimely. The class notice specified that all requests to intervene were to be received by the court by January 5, 2006, and were to have been delivered to class counsel and Netflix's counsel by the same date. Ellis did not file her request until January 6, and she served it by mail on January 5, rather than causing it to be delivered to counsel by that date.

Further, Ellis's motion was wholly unconvincing on the merits. It was supported by a memorandum of points and authorities containing barely two pages of text, and no declarations. The points and authorities asserted, erroneously, that Ellis's appellate rights could only be preserved if she were allowed to intervene. In fact, a class member who timely objects to a settlement has standing to appeal regardless of whether the member formally intervened in the action. (Consumer Cause, Inc. v. Mrs. Gooch's Natural Food Markets, Inc. (2005) 127 Cal. App. 4th 387, 395, 25 Cal. Rptr. 3d 514.) Ellis also claimed, in entirely conclusory fashion, that she could not fully participate in the proceedings and protect her rights as a class member without being allowed to intervene, and that class counsel, as a proponent of the settlement, could not adequately provide the court with a critical analysis of any additional proposed findings. These claims were not backed up by any reasoned explanation of what Ellis could have done to protect herself or contribute to the case as a party that she could not do or contribute as an objector.

Finally, Ellis's argument that as a party she would have sought discovery of attorney time records was not made in the trial court. She does not explain how such discovery would have been relevant to her objection that the fee award should be no more than 20 to 25 percent of the settlement value actually distributed to the class. In any event, the trial court could not have abused its discretion by failing to consider an argument that was never made to it.

There is no basis for inferring that the trial court did not give reasoned consideration to Ellis's motion. The court had ample grounds to deny the motion, on both procedural and substantive grounds, and it did not abuse its discretion in doing so.

B. Improper "Coupon" Settlement

At the February and March 2006 hearings on the fairness of the proposed settlement, only two objectors—Meininger and Davis—objected to the substantive terms of the settlement, as opposed to its notice or attorney fee provisions. Neither objector has pursued their challenge on this *423 appeal. Only Ellis, who did not appear and challenge the merits of the settlement at the 2006 hearings, now claims that the terms are so substantively unfair and unreasonable that the trial court abused its discretion in approving them. Ellis asserts that the amended settlement agreement constitutes a disfavored "coupon" settlement, i.e. it provides only a free service of nominal value and no cash payment to class members. According to Ellis, the absence of cash payments and restrictions on the use of the free upgrade or free month of Netflix service shortchanges class members and renders the settlement little more than a promotional opportunity for Netflix that provides no deterrence against the type of wrongdoing in which it engaged.

The trial court has broad discretion to determine whether a class action settlement is fair and reasonable. (Dunk v. Ford Motor Co. (1996) 48 Cal. App. 4th 1794, 1801, 56 Cal. Rptr. 2d 483 (Dunk).) Our review on appeal is limited to determining whether the record discloses a clear abuse of discretion by the trial court. (In re Microsoft I-V Cases (2006) 135 Cal. App. 4th 706, 723, 37 Cal. Rptr. 3d 660.) When the following facts are established in the record, a class action settlement is presumed to be fair: "(1) the settlement is reached through arm's-length bargaining; (2) investigation and discovery are sufficient to allow counsel and the court to act intelligently; (3) counsel is experienced in similar litigation; and (4) the percentage of objectors is small." (Dunk, at p. 1802, 56 Cal. Rptr. 2d 483.)

The trial court found that all four of the factors referred to in Dunk were present here. First, the agreement was the result of arm's-length bargaining between the parties. The parties participated in two formal mediation sessions with a highly respected former federal magistrate judge. (See Dunk, supra, 48 Cal.App.4th at pp. 1802-1803, 56 Cal. Rptr. 2d 483.) Negotiations between counsel for the parties continued between the formal sessions, before substantive agreement was reached at the second mediation session. Further negotiations followed Netflix's meeting with the FTC in January 2006, culminating in the amended agreement. Second, before settling, the parties engaged in extensive discovery, including written discovery, document production, and depositions of key Netflix employees. Class counsel had previously undertaken its own prefiling investigation that included numerous interviews and wide-ranging research regarding Netflix's delivery and allocation practices, as well as its advertising and marketing materials. By the time the settlement was reached, all of the critical facts regarding Netflix's disputed policies and practices were on the table. The trial court's finding that "investigation and discovery [were] sufficient to allow counsel and the court to act intelligently" was thus well supported by the record. Third, class counsel and Netflix's counsel both had substantial experience litigating consumer class actions and other complex cases. Finally, the percentage of objectors in this case is small by any measure. Nearly 700,000 class members have registered for the class benefit while only 1,234 members (0.2 percent of the class) opted out, including some who did so only because they sympathized with Netflix and believed the litigation was abusive. (Cf. 7-Eleven Owners for Fair Franchising v. Southland Corp. (2000) 85 Cal. App. 4th 1135, 1152-1153, 102 Cal. Rptr. 2d 777 [response of absent class members was "overwhelmingly positive" where only 1.5 percent elected to opt out].)

Only one objector, Ellis, continues to argue that the settlement is unreasonable. Ellis makes no claim that any of the factors *424 supporting a presumption of fairness is not present in this case. Instead, Ellis bases her entire argument on the premise that this is a coupon settlement and that such settlements are, in general, inherently suspect and improper. In fact, these premises are neither entirely accurate nor particularly useful for evaluating the fairness of the specific settlement terms before us. Although the settlement reached in this case may be classified as a variant of the coupon settlement, it does not in fact share all of the attributes of the category. In a pure coupon settlement, the class members would receive a coupon, voucher, or discount that would partly defray the cost of making a new purchase of goods or services from the defendant. In many cases, the coupon might induce the member to make a purchase he or she would not otherwise have made, which may actually produce a net benefit for the defendant. That is not the case here. The Netflix class members are not being offered a discount that requires them to make new purchases. They are being offered an opportunity to obtain a limited number of rentals at no charge. While it is possible that some existing customers might be induced by the free rentals to purchase a higher level of service and some past customers might be induced to resume their lapsed subscriptions, the potential for Netflix to actually benefit financially from the settlement is much reduced compared to a pure coupon discount program. Ellis's generic discussion of the evils of coupon settlements completely ignores the distinguishing features of this settlement.

The claim that coupon settlements are inherently suspect or improper is also not persuasive. Ellis relies on a law review article and a handful of cases not decided under California law.[2] She also asserts that the federal Class Action Fairness Act of 2005 (CAFA) (28 U.S.C. § 1712), although inapplicable to this proceeding, is "highly suspicious" of coupon settlements because it requires the court to hold a special hearing to determine their value. But while the valuation of coupon settlements may pose special challenges, neither CAFA nor any of the authorities Ellis cites hold that coupon settlements are per se improper. Notably, Ellis does not discuss or distinguish California cases in which coupon settlements have been found to be fair and reasonable. (See, e.g., In re Microsoft I-V Cases, supra, 135 Cal.App.4th at pp. 711-713, 37 Cal. Rptr. 3d 660; Wershba v. Apple Computer, Inc. (2001) 91 Cal. App. 4th 224, 247, 110 Cal. Rptr. 2d 145; Dunk, supra, 48 Cal.App.4th at pp. 1804-1805, 56 Cal. Rptr. 2d 483.)

Most importantly, Ellis failed to perform any analysis of the settlement terms to try to overcome the presumption of fairness to which they are entitled. Ellis cites one of the federal cases she relies on for the proposition that the most important factor in evaluating the fairness of a settlement is the strength of the plaintiffs case weighed *425 against the amount of the settlement. (See Synfuel Technologies, Inc. v. DHL Express (USA), supra, 463 F.3d at p. 653!) Nowhere in her 14-page discussion of coupon settlements, does Ellis attempt such a comparative analysis. In fact, the benefit provided by the settlement—free DVD rentals worth $6 to current subscribers and $16.99 to former subscribers—directly addresses the harm alleged in the complaint, which was Netflix's alleged failure to deliver as many DVD's as promised. While the dollar value of the settlement per class member is small, it must be remembered that the damages allegedly caused by Netflix's allocation and delivery policies were hardly unlimited either, and plaintiffs would have encountered considerable difficulties in trying to prove their amount.

Other than suggesting that a cash settlement would have had more value to class members and more deterrent value, Ellis fails to explain why the settlement terms are not fair and reasonable in relation to the range of possible results further litigation might have produced, including no class certification and/or zero or minimal recovery of damages by class members. The issue before the trial court was not whether the settlement agreement was the best one that class members could have possibly obtained, but whether it is "fair, adequate, and reasonable." (Dunk, supra, 48 Cal.App.4th at p. 1801, 56 Cal. Rptr. 2d 483.) On that question, we find nothing in Ellis's arguments to overcome the presumption of fairness that applies in this case.

C. Notice Issues

Appellants Ellis and Vogel argue that the judgment must be vacated because the notice to class members failed to comport with the requirements of due process. Ellis complains that (1) changes to the settlement agreement required a third round of notices to be sent out to class members, and (2) the notices failed to apprise class members of the value of the settlement or of the amount of damages claimed by plaintiffs. Vogel asserts that (1) the "`summary notice'" sent out to class members was deficient, and (2) the notice improperly discouraged class members from opting out or objecting to the settlement.

1. Ellis's Objections

Ellis argues that a third round of notices was required because the second notice failed to call attention to a change made by the amended settlement agreement to lengthen from 90 to 180 days the time period after the effective date of the agreement by which Netflix would provide current subscribers with their service upgrade benefit. Vogel and others had argued that if all the upgrades were provided in 90 days, the service might be degraded due to the greater demand on Netflix's DVD inventory. Lengthening the period to 180 days would alleviate the potential strain on inventories. The change also required Netflix to provide the subscriber with 10 days' notice before the benefit was to be provided so that class members did not have to remain enrolled as subscribers while waiting for their one-month benefit. Vogel withdrew his objection following these amendments.

We are satisfied that these changes improved the settlement, and that no notice of them was therefore required. (See, e.g., In re Integra Realty Resources, Inc. (10th Cir.2001) 262 F.3d 1089, 1111 [no notice required of change expanding rights of class members]; Denney v. Deutsche Bank AG (2d Cir.2006) 443 F.3d 253, 271 [an additional opt-out period is not required with every shift in the marginal attractiveness of the settlement].) In any event, the long-form notice of the settlement *426 agreement stated that the service upgrade would be provided within 180 days after the effective date, and the email notice linked to a Web site that contained a redline comparison between the original and amended agreements showing the changes.[3] Ellis's contention that another round of notices was required is without merit.

Ellis's claim that the notices were deficient for failing to provide a dollar estimate of the overall value of the settlement in relation to the damages sought by plaintiffs is also without merit. Ellis cites no authority requiring that such notice be given. The notice gave sufficient information to allow each class member to decide whether to accept the benefit he or she would receive under the settlement, or to opt out and pursue his or her own claim. (See Oswald v. McGarr (7th Cir.1980) 620 F.2d 1190, 1197 [notice should contain sufficient information to enable a class member to determine whether to accept the offer to settle, the effects of settling, and the available avenues for pursuing his claim if he does not settle].) No more than that was required. The class-wide damages claimed by one side in the litigation, which the opposing party hotly contests, does not in any event provide very useful information for evaluating the fairness of the overall settlement, much less for enabling an individual class member to decide whether to opt out.

2. Vogel's Objections

Vogel argues that the "`summary notice' " sent to class members did not fairly apprise them of the proposed settlement or of the options open to them to intervene or object, opt out, or accept the settlement. In particular, Vogel asserts that the notice failed to discuss (1) the release of claims contemplated by the settlement, (2) the fact that the final judgment would bind all members of the class who did not opt out, (3) the procedure for opting out, (4) the fact that class members who did not request exclusion could enter an appearance through counsel, and (5) the date of the final approval hearing. According to Vogel, these omissions violated Civil Code section 1781, subdivision (e)[4] and California Rules of Court, rules 3.766(d)[5] and 3.769(f).[6]

*427 We note first that the trial court "`has virtually complete discretion as to the manner of giving notice to class members.'" (7-Eleven Owners for Fair Franchising v. Southland Corp., supra, 85 Cal. App.4th at p. 1164, 102 Cal. Rptr. 2d 777.) Here, the summary notice in fact included all of the following: (1) a brief explanation of the case; (2) a statement that the court would exclude the class member from the case if they mailed a letter by a specified date; (3) a statement that by signing up for the benefit, the class member waived the right to bring a separate lawsuit concerning the released claims; (4) a statement that class members who wished to object to the settlement could file legal papers in the court by a specified date; and (5) a statement and hyperlink directing the class member to the settlement Web site to get more information about how to accept the benefit, opt out of the settlement, or object to it in court. The long-form notice on the Web site provided detailed information on how to exercise each option, as well as the date of the final settlement hearing, and a statement of the class member's right to intervene or attend the settlement hearing in person or through an appearance by counsel. The long form also stated: "As a Class Member, you will be bound by any judgment or other disposition of the Litigation, even if you do not submit a claim or take advantage of any of the Class Benefits." A "Frequently Asked Questions" section of the Web site included answers to such questions as, "Do I have to participate in this settlement?" and "What happens if I opt out or exclude myself from the settlement?," with the answers containing a hyperlink to a specific section of the longform notice.[7]

In our view, the summary notice addressed each subject required by Civil Code section 1781 and California Rules of Court, rule 3.766. The summary notice and long-form notice together provided all of the detail required by statute or court rule, in a highly accessible form. The fact that not all of the information was contained in a single e-mail or mailing is immaterial. The manner of giving notice is subject to the trial court's virtually complete discretion. Using a summary notice that directed the class member wanting more information to a Web site containing a more detailed notice, and provided hyperlinks to that Web site, was a perfectly acceptable manner of giving notice in this case. (See Browning v. Yahoo! Inc. (N.D.Cal.2006) 2006 WL 3826714 at *8-9 [approving two-tiered notice system using summary e-mail and long-form notice posted on Web site].) The class members conducted business with defendant over the Internet, and can be assumed to know how to navigate between the summary notice and the Web site. Using the capability of the Internet in that fashion was a sensible and efficient way of providing notice, especially compared to the alternative Vogel apparently preferred—mailing out a lengthy legalistic document that few class members would have been able to plow through.[8] We find no abuse of discretion or deprivation of due process in the trial *428 court's approval of the form and content of the notice given in this case.,

Vogel claims that the notice program discouraged class members from opting out. He stresses the fact that the Original Agreement included a provision requiring class counsel to "make every reasonable effort to encourage Class Members to participate and not to optout." While acknowledging that this provision was removed from the amended settlement, Vogel claims that "its effects had already cast their pallor over the notice program." Specifically, Vogel claims the notice program was designed to discourage opt-outs because although class members could file their claims online, they had to send a first-class letter in order to opt out.[9]

We note first that the most direct comparison is not that between the procedures for opting out and those for filing a claim, but between those for opting out and those required for staying in the class. There is clearly no legal impediment whatsoever to making it harder to opt out than to stay in. In fact, requiring class members to take affirmative steps to opt in has been held to be contrary to state and federal class action law and policy. (See Hypertouch, Inc. v. Superior Court (2005) 128 Cal. App. 4th 1527, 1543-1550, 27 Cal. Rptr. 3d 839, and cases cited therein.) Nor does Vogel make any claim that requiring class members to opt out by mail is so burdensome that it violates their due process rights. He admits that it is "unremarkable" that courts have allowed this method. Thus, the only issue presented here is whether due process requires that the procedure for opting out be just as convenient as the procedure for filing a claim. We reject that contention.

As an initial matter, Vogel waived the claim by failing to make any objection in the trial court about the absence of an online opt-out procedure. (See Wershba v. Apple Computer, Inc., supra, 91 Cal. App.4th at pp. 236-237, 110 Cal. Rptr. 2d 145.) That legal theory was not encompassed by any of Vogel's trial objections and, had it been timely raised, the parties could have responded to it by modifying the opt-out method or creating a record of their reasons for not wanting to do so. Vogel cannot complain that no evidence supports respondents' decision to require mailed opt-out notice when that state of the record most likely resulted from the fact that he failed to raise the issue in the trial court.

In any event, Vogel's present objection is not persuasive on its merits. He cites no authority requiring that opt-out and claim-filing procedures be equivalent, and we have found none. The two procedures perform very different functions. Any rule mandating equivalence between them would unduly hamper the courts in ensuring that the parties adopt the procedures best suited to the particular circumstances of each case. For example, it is in the interests of class members that procedures for claiming benefits be as simple and convenient as possible. That should not mean that as a matter of due process every improvement in claims procedures should automatically require conforming changes to otherwise permissible opt-out procedures.

Furthermore, in the event of any future dispute over whether a class member is or is not bound by the judgment, or is or is not eligible for benefits under the settlement agreement, requiring mailed notice *429 offers protections for both sides that an on-line system may not be able to match. It allows both sides to maintain a paper record of the transaction that might not be possible, or could more easily be falsified, with a purely on-line procedure.

Although we assume the parties could have developed an on-line system for opting out in this case, we decline for these reasons to second-guess their choice not to do so.

The trial court did not abuse its discretion in approving the notice and opt-out procedures utilized in this case, and those procedures did not violate the due process rights of absent class members.

D. Attorney Fee Award

All appellants attack the attorney fee award, relying on the various grounds discussed below.

1. Ellis

Ellis argues that the trial court (1) overvalued the benefits to the class by assuming that the face value and the actual value of the settlement benefits were the same, (2) failed to provide any concrete justification for the calculation of the lodestar, (3) improperly adjusted its lodestar analysis, and (4) capriciously awarded attorney fees to some objectors and denied them to others without clear criteria.

The trial court valued the settlement by multiplying the prices Netflix then charged for the services to be offered free to former subscribers ($18.00 for one month's free membership) and current subscribers ($6 for a one-month upgrade) times the number of class members who had enrolled for each benefit. Based on Leslie, supra, Ellis rests her objection to this methodology on the generic observation that "settlement coupons very frequently possess an actual value much less than their purported face value." This observation may apply to some true coupon-based settlements, in which the class members receive coupons that will cover part of the purchase price of specified products, if they are purchased within a particular time period. The actual benefit is less than the face amount of the coupon because, among other reasons, many class members will not be able to use the coupon during the time period allowed, the coupons cannot typically be transferred or aggregated, and the products for which the coupons may be used may not be ones the member would have purchased without the coupon. (Leslie, supra, at pp. 1004, 1016-1026.) Ellis does not explain why any of these factors would apply here or what quantitative impact they might have. Free DVD rental services will automatically be made available to every class member who has filed a claim for them, without any additional purchase. The member or former member must simply pick out the DVD's and return them in order to obtain the value of the benefit. Given this type of benefit program, the trial court's chosen methodology for estimating the class benefit was not an abuse of discretion.

Regarding the lack of any "concrete justification" for the calculation of the lodestar amount, Ellis is simply wrong on this point. The court carefully went through the hours claimed by class counsel, and explained the basis for the reductions it was making in hours billed and billing rates. Ellis fails to offer any reasoned argument explaining where the court went wrong in its application of the lodestar method.

The court cited "quality of representation," "success achieved" and, most importantly, the "rate of acceptance of the benefit offered to class members," as factors justifying an upward adjustment of the lodestar. Ellis suggests that "quality of *430 representation" is inappropriate because it duplicates the factors justifying the hourly rates the court used for its lodestar calculation. According to Ellis, "success achieved" and "rate of acceptance of the benefit" are also duplicative of one another. Ellis's arguments in this regard depend on unwarranted assumptions about the nature of the factors the court mentioned.

First, a lodestar enhancement based on "quality of representation" by definition involves considerations not captured by counsel's hourly rates. (See Ketchum v. Moses (2001) 24 Cal. 4th 1122, 1139, 104 Cal. Rptr. 2d 377, 17 P.3d 735 [court can award multiplier for an exceptional quality of representation when representation "far exceeds the quality ... that would have been provided by an attorney of comparable skill and experience billing at the hourly rate used in the lodestar calculation"].) Nothing in Ketchum v. Moses requires the trial court to recite an express finding that class counsel's representation "far exceed[ed]" the level of representation that comparably skilled attorneys would have provided. (See also In re Marriage of Arceneaux (1990) 51 Cal. 3d 1130, 1133, 275 Cal. Rptr. 797, 800 P.2d 1227 ["[a] judgment or order of a lower court is presumed to be correct on appeal, and all intendments and presumptions are indulged in favor of its correctness"].)

Second, "success achieved" and the "rate of acceptance of the benefit" are not necessarily identical. The former is a general category that includes all of the positive results achieved by the litigation. This could include changes in company policies that were not part of the settlement, the dollar value of settlement benefits, and the absolute size of the class of persons who are eligible for the benefit. This category would also encompass other measures of the litigation's success including the `early stage at which it produced benefits, and the availability or unavailability of less costly means for achieving the same benefits. In contrast, the "rate of acceptance," which the trial court stressed was the most important factor in its analysis, focuses on a single, very specific factor that measures one aspect of the overall success achieved—the degree to which the settlement benefits were in fact of interest to class members. The two categories are not wholly or largely duplicative of one another. The court did not err by mentioning both factors while placing primary reliance on the more specific measure.

The various factors cited by the court in its fee determination thus were not duplicative of one another, and the court did not abuse its discretion in relying on them.[10]

Finally, Ellis complains that the trial court acted arbitrarily in granting attorney fees to some objectors and denying fees to others. The trial court stated that "some of the objectors have proven that they may have made helpful suggestions to class counsel in the first part of 2006," but also that "[i]n light of all the facts known to the Court, none of the objectors was a substantial factor in improving the benefits offered to the class by the settlement." Ellis argues that the two statements illustrate the court's inconsistency and evidence an abuse of discretion.

There is no necessary inconsistency. There was evidence that certain objectors involved in the negotiations surrounding the amended settlement agreement helped to identify problems that were fixed and contributed to additional improvements that benefitted the class. Among other *431 things, these objectors reviewed drafts of the amended settlement agreement, brought media attention to the case, and helped to persuade Netflix to amend the settlement. These kinds of contributions can be of concrete benefit even though they do not individually result in any substantial improvement in settlement benefits. The court may also have felt that certain objectors raised issues that assisted the court in its own deliberations. Ellis fails to meet her burden of demonstrating that the handful of modest awards made to objectors, in amounts far less than that requested by counsel, was an abuse of discretion.

2. Meininger and Davis

Meininger and Davis contend that the trial court abused its discretion by (1) basing the rate of acceptance of the benefit on the number of class members who had taken the first of two required steps in order to obtain the benefit instead of on the number who complete both steps, and (2) permitting the total value of benefits recovered to influence the fee when that amount is due primarily to the size of the class rather than to the de minimis amounts recovered per class member.

Meininger and Davis's reference to a two-step claim process is puzzling. Once a current subscriber has submitted a claim during the claims period (which expired 45 days after the supplemental notices were issued), the class member will automatically receive the class benefit without taking any further action. It is true that former subscribers will be required to complete a Netflix registration process before they can receive their benefits. But instructions on how to do so are to be sent out automatically on a rolling basis to every former subscriber who filed a claim. It is reasonable to presume that the great majority of them, having gone to the trouble of filing a claim, will take the final step to redeem their benefit when invited to do so. In our view, considering the number of former member claims submitted as a measure of that subgroup's acceptance of the settlement benefit is not so unreasonable as to constitute an abuse of discretion. The alternative Meininger and Davis would apparently prefer—postponing any determination of a fee award until 18 months or more after the effective date of the settlement—seems highly unfair and impractical by comparison.

Meininger and Davis also claim the trial court should not have considered the dollar value of the settlement in setting the fee award because the value of the benefit per class member was low in this case. They rely on the following observation from Lealao v. Beneficial California, Inc. (2000) 82 Cal. App. 4th 19, 97 Cal. Rptr. 2d 797 (Lealao): "[Permitting the amount of the recovery to influence the fee is most justified where the amount of the recovery is not due primarily to the size of the class." (Id. at p. 53, 97 Cal. Rptr. 2d 797.) As the Lealao court makes clear at another point in its opinion, fees based on a percentage of the benefits are in fact appropriate in large class actions when the benefit per class member is relatively low, except that the percentage should generally decrease as the number of class members and the size of the fund increases. (Id. at pp. 48-49, 97 Cal. Rptr. 2d 797.) This is based on a recognition that beyond a certain point a larger number of identical claims does not typically require greater efforts by counsel. (Ibid.) We find nothing in Lealao's discussion of this issue to suggest that it was an abuse of discretion for the trial court to apply a percentage figure at the low end of the typical contingency contractual arrangement (21.8 percent) to calculate the multiplier in the context of this settlement.

*432 3. Vogel

Vogel complains that the trial court (1) failed to support its lodestar analysis with a sufficiently detailed breakdown of the reductions it made for duplicative legal services or for services charged at excessive hourly rates, (2) erred in establishing the multiplier by using as a benchmark the percentage of the fees awarded divided by a sum including both the class benefit and the amount of the fee award, and (3) compounded the latter error by allowing additional fees based on new claims made after the settlement agreement was amended.

The law firm representing the class submitted a declaration from counsel and exhibits supporting a lodestar amount of $1,198,176. The declaration stated that the firm's two partners had spent 1,416 and 1,320 hours respectively prosecuting the litigation, and it included an exhibit showing that the partners' respective hourly billing rates of $450 and $425 were in line with what partners at Silicon Valley firms were charging for their services. The firm did not provide billing records showing the number of hours the partners spent on particular activities or days. Based on its knowledge of the litigation, the court made a reduction of 200 hours in the total number of hours claimed to reflect its estimate of time spent by both attorneys on activities, such as court appearances, that could have been handled by one of them alone. This would have reduced the lodestar amount by approximately $85,000 to $1,113,176. The court made two further reductions totaling approximately $308,000 to arrive at a lodestar of $805,000: (1) an unspecified amount for activities such as document review that could have been done by attorneys or paralegals with lesser expertise than the firm's partners at an assumed blended rate of $250 per hour, and (2) a further unspecified amount for-time spent responding to the FTC's objections.

We start from the proposition that the "`experienced trial judge is the best judge of the value of professional services rendered in his court, and while his judgment is of course subject to review, it will not be disturbed unless the appellate court is convinced that it is clearly wrong.'" (Serrano v. Priest (1977) 20 Cal. 3d 25, 49, 141 Cal. Rptr. 315, 569 P.2d 1303, quoting Harrison v. Bloomfield Building Industries, Inc. (6th Cir.1970) 435 F.2d 1192, 1196.) Further, detailed time sheets are not required of class counsel to support fee awards in class action cases. (Wershba v. Apple Computer, Inc., supra, 91 Cal.App.4th at pp. 254-255, 110 Cal. Rptr. 2d 145.) The court may award fees based on time estimates for attorneys who do not keep time records. (Margolin v. Regional Planning Com. (1982) 134 Cal. App. 3d 999, 1006-1007, 185 Cal. Rptr. 145.)

It follows from these authorities that the trial court has wide discretion in making reductions based on its estimate of time spent on activities that are noncompensable in whole or in part. Here, the trial judge had been assigned to this litigation from its inception, and was familiar with the nature, extent, and reasonableness of class counsel's litigation services. The court made sizeable but reasonable reductions in the lodestar amount, based on reasons that it clearly explained and that Vogel does not question. Although it might have been better for the court to provide a separate breakdown of each of the three reductions it made in order to arrive at the $805,000 figure, Vogel offers no persuasive argument that it was required to do so.

To establish a benchmark for determining the enhanced lodestar amount, the court used the percentages that a hypothetical enhanced fee would represent of *433 the sum of the fee plus the aggregate value of the benefits claimed by class members under the Original Agreement ($4.29 million). It viewed the resulting number as being equivalent to a contingency fee percentage that might be specified in the typical contingent fee contract. For illustrative purposes using the $4.29 million figure, the court plugged different hypothetical fee amounts into this formula that would translate into contingency fee percentages of 20, 25, and 40 percent, which the court believed encompassed the 20 to 40 percent range of contingency fee contracts found in the marketplace. It established class counsel's initial award in an amount ($1.3 million) that would translate into a contingency fee percentage, approximately 23 percent, that was close to the low end of the 20 to 40 percent range. That still left open the issue of whether additional class member enrollments for the benefit after the supplemental notice was issued should result in additional fees and, if so, how much. The court held that such new benefit claims would result in an additional increment of fees to be set based on the amount of fees necessary to emulate a 20 percent contingency fee on the additional value of the benefits claimed, subject to a maximum fee award of $2.5 million. In other words, the additional fee would be set at a level such that the amount of the fee would constitute 20 percent of the sum of the fee plus the value of additional benefits claimed.

We find no error or abuse of discretion in the court's methodology. If a contingency fee contract provides that the attorneys are to receive, for example, 25 percent of the plaintiffs recovery, the plaintiff who recovers $100,000 keeps $75,000 and pays $25,000 to his attorneys. If we did not already know the contingency fee percentage set by the parties' fee contract, that number could be calculated by dividing the amount received by the attorney ($25,000) by the sum of the amount received by the client and the amount received by the attorney ($100,000). That is the same formula the court used to calculate a benchmark for enhancing the lodestar amount in this case.

Vogel appears to be arguing that it was error for the court not to use the exact same percentage-of-the-benefit method discussed in Lealao, supra, 82 Cal.App.4th at pages 25-36, 97 Cal. Rptr. 2d 797, which would typically look at the straight ratio of proposed fees to class benefits and compare that to the percentage of fees awarded in common fund cases. (See Lealao, at p. 36, 97 Cal. Rptr. 2d 797.) In our view, the Lealao court did not purport to mandate the use of one particular formula in class action cases. The method the trial court used here and that discussed in Lealao are merely different ways of using the same data—the amount of the proposed award and the monetized value of the class benefits—to accomplish the same purpose: to cross-check the fee award against an estimate of what the market would pay for comparable litigation services rendered pursuant to a fee agreement. (See Lealao, at pp. 47-50, 97 Cal. Rptr. 2d 797.) It is not an abuse of discretion to choose one method over another as long as the method chosen is applied consistently using percentage figures that accurately reflect the marketplace.[11]

*434 Vogel also objects to the additional fees awarded under the 20 percent formula for new benefit claims made after the supplemental notice was issued. He equates this to rewarding class counsel for time spent by them responding to the FTC's objections to the original settlement, hours the court had already found noncompensable in its lodestar analysis.

In our view, Vogel is wrong about the court's basis for increasing the fee award based on additional claims. The court was not concerned with the reason the additional claims were made or who deserved credit for causing additional class members to make claims. It seems indisputable, for example, that many of the additional claims were filed not because the FTC's objections had improved the benefit but because class members had additional time to hear about the case and another opportunity to get their claims in. Should such new claims be credited to the Original Agreement that class counsel negotiated or to the amendments for which the FTC is primarily responsible? In our view, rather than get caught up in making such distinctions, the court simply wanted to be consistent in applying the principle that counsel's fees should be based on the monetized value of the benefits class members chose to avail themselves of. The court recognized that enrollment for benefits was "an ongoing process" at the time it issued its fee order, and saw no justification for giving zero weight in its fee formula to some of the class members who enrolled for benefits, merely because they enrolled later in the process than others. Given the theory upon which the court enhanced the lodestar amount in this case, such a distinction would have seemed completely arbitrary.

Finally, we are not persuaded that the 2.5 multiplier that class counsel are to receive is so out of line with prevailing case law as to constitute an abuse of discretion. (See Wershba v. Apple Computer, Inc., supra, 91 Cal.App.4th at p. 255, 110 Cal. Rptr. 2d 145 [multipliers can range from 2 to 4 or even higher]; City of Oakland v. Oakland Raiders (1988) 203 Cal. App. 3d 78, 249 Cal. Rptr. 606 [affirming a multiplier of 2.34].) We also agree with class counsel that the effective multiplier is lower than 2.53 because the lodestar excluded time spent negotiating the amended settlement and services required after the filing of Chavez's fee application.[12]

III. CONCLUSION AND DISPOSITION

The trial court did not abuse its discretion in approving the amended class action settlement agreement, approving the notice given to class members, or determining the amount of fees to which class counsel was entitled. The order approving settlement, order approving fees and expenses, and order approving additional fees are affirmed. Costs in cases Nos. Al 14334, Al 15395, and Al 15571 are awarded to respondents.

We concur: STEIN, Acting P.J., and SWAGER, J.

NOTES

[1] The trial court's April 28, 2006 order awarding fees, after denying fees to most objectors, simply stated that "[e]ach an[d] every motion to intervene is denied." From this, Ellis concludes that the trial court did not consider the merits of her motion to intervene.

[2] The cases Ellis cites are not particularly germane on their facts. (See Acosta v. Trans Union, LLC (C.D.Cal.2007) 240 F.R.D. 564 [free credit report as part of settlement had little or no value because consumers are entitled to one free credit report per agency per year and few take advantage of that right]; Bloyed v. General Motors Corp. (Tex.Ct.App. 1994) 881 S.W.2d 422 [$1,000 coupon toward purchase of new van or truck within 15 months]; Synfuel Technologies, Inc. v. DHL Express (USA) (7th Cir.2006) 463 F.3d 646 [in-kind compensation worth less than cash since some goods or services offered will not be used and will have no cost for defendant].) The law review article is Leslie, A Market-Based Approach to Coupon Settlements in Antitrust and Consumer Class Action Litigation (2002) 49 UCLA L.Rev. 991 (hereafter Leslie), which argues that coupon settlements will be overused as long as class counsel are compensated in cash.

[3] The e-mailed supplemental notices stated: "Additional changes to the settlement are reflected in the Amended Settlement Agreement, lf you wish to compare the Amended Settlement Agreement with the Original Settlement Agreement, the Parties have posted a comparison at www.netflix.com/settlement."

[4] Civil Code section 1781, subdivision (e) provides: "The notice required by subdivision (d) shall include the following: [¶] (1) The court will exclude the member notified from the class if he so requests by a specified date. [¶] (2) The judgment, whether favorable or not, will include all members who do not request exclusion. [¶] (3) Any member who does not request exclusion, may, if he desires, enter an appearance through counsel."

[5] "The content of the class notice is subject to court approval. If class members are to be given the right to request exclusion from the class, the notice must include the following: (¶) (1) A brief explanation of the case, including the basic contentions or denials of the parties; [¶] (2) A statement that the court will exclude the member from the class if the member so requests by a specified date; [¶] (3) A procedure for the member to follow in requesting exclusion from the class; [¶] (4) A statement that the judgment, whether favorable or not, will bind all members who do not request exclusion; and [¶] (5) A statement that any member who does not request exclusion may, if the member so desires, enter an appearance through counsel." (Cal. Rules of Court, rule 3.766(d).)

[6] "If the court has certified the action as a class action, notice of the final approval hearing must be given to the class members in the manner specified by the court. The notice must contain an explanation of the proposed settlement and procedures for class members to follow in filing written objections to it and in arranging to appear at the settlement hearing and state any objections to the proposed settlement." (Cal. Rules of Court, rule 3.769(f).)

[7] The questions were not in fact frequently asked but represented Netflix's effort to anticipate the type of information a class member might want to have readily available.

[8] As the trial court put it: "A succinct notice that alerts the reader to the significant issues and advises them where to go for more, to my mind, is a great deal better than being confronted with ... [a] small typed, single spaced, full notice."

[9] Vogel also complains that the letter had to either be postmarked by December 28, 2005—the middle of the holiday period—or received by January 6, 2006. However, the supplemental notice extended the deadline to June 26, 2006.

[10] For the reasons already stated, we reject the similar arguments made by Vogel, Meininger, and Davis concerning the trial court's asserted reliance on duplicative factors.

[11] Using the percentage of the benefits to class claimants as a benchmark, class counsel's initial award was 30.3 percent of the benefits, and the final fee award was 27.9 percent of the benefits. This is not out of line with class action fee awards calculated using the percentage-of-the-benefit method: "Empirical studies show that, regardless whether the percentage method or the lodestar method is used, fee awards in class actions average around one-third of the recovery." (Shaw v. Toshiba America Information Systems, Inc. (E.D.Tex.2000) 91 F. Supp. 2d 942, 972.)

[12] We express no opinion as to whether class counsel has a right, as claimed, to additional fees for work performed on this appeal.