Fraley v. Facebook, Inc.

ORDER GRANTING MOTION FOR FINAL APPROVAL OF SETTLEMENT AGREEMENT RICHARD SEEBORG, United States District Judge. I. INTRODUCTION The proposed settlement class in this action consists of some 150 million members of defendant Facebook, Inc.’s eponymous social network website, whose names and/or likenesses allegedly were misappropriated to promote products and services through Facebook’s so-called “Sponsored Stories” program. The parties now seek final approval of a settlement that will result in small cash payments to the relatively low percentage of class members who filed claims, and cy pres payments of several millions of dollars to certain organizations involved in internet privacy issues. The settlement also requires Facebook to make changes to the Statement of Rights and Responsibilities (“SRRs”) it contends governs use of its site, and to implement additional mechanisms giving users greater information about, and control over, how their names and likenesses are employed in connection with Sponsored Stories. The original settlement agreement proposed by the parties did not win prelimi — 941 — nary approval. The parties responded with a new proposal, earning such approval and triggering notice to potential class members. A number of objectors contend that the updated settlement proposal should not receive final approval for a variety of reasons. Among the objections most vigorously advanced is an argument that the settlement does not appropriately handle issues related to minors. The record leaves no doubt that this settlement was the product of arms-length negotiations and compromise. Although the monetary relief to each class member is relatively small and the percentage of class members who submitted claims is limited, the settlement as a whole provides fair, reasonable, and adequate relief to the class, in light of all the circumstances, including the low probability that a substantially better result- would be obtained through continued litigation. The injunctive relief, while not incorporating all features that some of the objectors might prefer, has significant value and provides benefits that likely could not be obtained outside the context of a negotiated settlement, even if plaintiffs were to prevail on the merits. If “Sponsored Stories” had undisputedly violated the law and represented the gross invasion of class members’ rights as characterized by the complaint, then the adequacy of the settlement would, of course, be viewed through a very different lens. Plaintiffs’ allegations and theories, however, remain largely untested, having only survived a motion to dismiss. Substantial barriers to recovery remained, not the least of which would be the requirement to demonstrate that the complained-of conduct caused cognizable harm. Placing those and other factors discussed below in the balance, the proposed settlement warrants final approval. II. STANDARD OF REVIEW A district court’s approval of a class-action settlement must be accompanied by a finding that the settlement is “fair, reasonable, and adequate.” Fed.R.Civ.P. 23(e). The fairness of a settlement must be evaluated as a whole, rather than by assessing its individual components. Hanlon v. Chrysler Corp., 150 F.3d 1011, 1026 (9th Cir.1998). Crucially, the question whether a settlement is fundamentally fair within the meaning of Rule 23(e) is not the same as asking the reviewing court if perfection has been achieved. See id. at 1027. Although Rule 23 imposes strict procedural requirements on the approval of a class settlement, a district court’s only role in reviewing the substance of that settlement is to ensure that it is “fair, adequate, and free from collusion.” See id. A number of factors guide in making that determination, including: the strength of the plaintiffs’ case; the risk, expense, complexity, and likely duration of further litigation; the risk of maintaining class action status throughout the trial; the amount offered in settlement; the extent of discovery completed and the stage of the proceedings; the experience and views of counsel; the presence of a governmental participant; and the reaction of the class members to the proposed settlement. Id. at 1026 (hereinafter the “Hanlon factors”). Additionally, when (as here) the settlement takes place before formal class certification, settlement approval requires a “higher standard of fairness.” See id. More exacting review of class settlements reached before formal class certification ensures that class representatives and their counsel do not secure a disproportionate benefit “at the expense of the unnamed plaintiffs who class counsel had a duty to represent.” See id. at 1027; see — 942 — also In re Gen. Motors Corp. Pick-Up Truck Fuel Tank Prods. Liab. Litig., 55 F.3d 768, 787 (3rd Cir.1995) (explaining that “[w]ith less information about the class” at the early stage before formal class certification, the court “cannot as effectively monitor for collusion, individual settlements, buy-offs (where some individuals use the class action device to benefit themselves at the expense of absentees), and other abuses”). III. DISCUSSION1 A. Fairness and Adequacy This settlement was achieved through negotiations mediated by a renowned retired federal magistrate judge following months of active, adversarial, litigation. The viability of the pleading had been tested through motion practice, and class certification issues were fully briefed. The parties had engaged in substantial discovery. There is no basis to conclude that the negotiations were anything other than a good faith, arms-length attempt by experienced and informed counsel to resolve this matter through compromise. As such the settlement is entitled to a degree of deference as the private consensual decision of the parties. See Hanlon, 150 F.3d at 1027. 1. Odds of recovery In bringing this action, plaintiffs understandably characterized the facts in the light most favorable to them. Under that characterization, Sponsored Stories represents a shocking overreach by Facebook, in which it, without its users’ consent, features users’ names and likenesses in commercial advertising, and makes significant profits by doing so. Facebook, however, has a different description of its program. Users- knowingly choose to indicate that they “like” certain entities or activities on Facebook. Even if users have never read Facebook’s SSRs, they know that their “likes” typically will appear on the “news-feeds” of their Facebook friends, subject to whatever limitations they have imposed through using privacy settings. Sponsored Stories, in Facebook’s view, does nothing more than take information users have already voluntarily disclosed to their “friends,” and sometimes redisplays it to the same persons, in a column that also contains more traditional paid advertising. While Facebook indisputably earns money from the Sponsored Stories program, it contends that its return is actually less than from available alternative types of advertising. Regardless of the degree of benefit to Facebook, however, plaintiffs faced a substantial burden in showing they were injured by the Sponsored Stories. While plaintiffs pleaded a sufficient basis for injury to support constitutional standing, it is far from clear that they could ever have shown they were actually harmed in any meaningful way. Indeed, in seeking class certification and in attempting to quantify the value of the settlement’s injunctive relief, plaintiffs have repeatedly relied primarily on their argument that Facebook benefited, rather than that class members were harmed. Plaintiffs also faced a substantial hurdle in proving a lack of consent, either express or implied. While those issues could not be adjudicated in Facebook’s favor at the pleading stage, there was a significant risk that at. some later juncture, plaintiffs would be found to have consented, or that class certification would prove unwarrant — 943 — ed in light of consent issues. As to the complaint’s assertion (echoed by some of the objectors) that Facebook should have acquired parental consent for members of the minor subclass, the Children’s Online Privacy Act (“COPPA”) stands as a potential preemption hurdle. See 15 U.S.C. §§ 6501 et seq. Plaintiffs faced other significant barriers to class certification and/or to eventual recovery as well, including the fact that many Facebook users often post “profile photos” that are not their own image, while some do not use their own name (although Facebook discourages the latter practice). Additionally, as is always the case, even assuming plaintiffs might ultimately prevail on the merits, it likely would only be after a protracted and very expensive journey. 2. The monetary relief The original settlement agreement presented for preliminary approval called for no monetary distribution to the class. The order rejecting that proffered settlement highlighted the absence of a cash component for class members as a concern and directed the parties to provide further analysis and information in the event any revised agreement contemplated a similar approach. The order did not state or imply that a cash payout to the class would be an absolute prerequisite to finding a settlement fair, reasonable, and adequate. The challenge, given the size of the class, is that even a modest per-class member payment could easily require a total settlement fund in the billions of dollars. As articulated in the prior order, this raises the spectre of whether some class actions are simply too big to settle, notwithstanding the strong policy favoring settlements. The solution the parties devised in this case was to propose a fixed settlement fund of $20 million, to be distributed to class members if the number of claims actually made did not cause the per-person payment amount to be so low as to make such distribution impracticable, in which case all of the funds would instead be distributed to cy pres recipients. The parties were effectively betting on a low rate of claims being filed to permit a cash distribution directly to individual class members. Indeed, so few persons have filed claims that the parties now propose paying $15 per claimant, which will still leave sufficient funds remaining for attorney fees, costs, expenses, and a distribution to the cy pres recipients. As a means of providing recompense to any genuinely injured parties, this approach certainly would not be ideal. The relatively low amount offered undoubtedly discouraged class members from filing claims. In a sense, adding a direct payment component to the settlement, did very little to buttress its overall fairness.2 That said, the monetary component of the settlement on balance is fair, reasonable, and adequate. As suggested above, there is slim indicia class members suffered any pecuniary harm as the result of appearing in Sponsored Stories, even assuming consent or other defenses did not bar recovery in the first instance. Additionally, even if the monetary benefit to Facebook were held to be an appropriate measure of potential recovery, the record suggests that under plaintiffs’ best case scenario, they would be able to show Face-book’s profits attributable to the alleged — 944 — misappropriation were in the range of $73 million, or approximately $.60 per class member. Given those numbers, a $20 million settlement, and payments of $15 each to those class members who filed claims, is a reasonable compromise.3 The only factor pulling in an opposite direction is the theoretical availability of statutory damages of $750 per violation of California Civil Code § 3344. Were plaintiffs’ chances of overcoming all the hurdles of litigation through final judgment significantly higher, the possibility of a statutory damage award might require a different result. Another important consideration, however, is that the adequacy of this settlement should not be evaluated against some theoretically available judgment, but against what plaintiffs could reasonably expect to recover. Given the class size, it is not plausible that class members could recover the full amount of the statutory penalties in any event, as such a judgment would pose due process concerns and threaten Facebook’s existence.4 Under all of these circumstances, the fact that monetary relief is going to only a small percentage of class members and in a very modest per-claimant sum does not undermine the fairness, reasonableness, and adequacy of the settlement as a whole. 3. The injunctive relief Plaintiffs have provided expert analysis that attempts to quantify the value of the injunctive provisions of the settlement in monetary terms. While assigning a dollar value to that relief might be relevant to the attorney fées application, there is no need to calculate such a figure when evaluating the over-all fairness, reasonableness, and adequacy of the settlement.5 The question is not whether a dollar value can be associated with the injunctive relief, but whether that relief benefits the class and at least ameliorates some of the alleged concerns raised by the complaint. The parties have shown that the injunctive provisions provide at least some meaningful benefits to the class members. Facebook has agreed both to provide greater disclosure and transparency as to when and how member’s names and profile pictures are re-published, and to give them additional control over those events. Additional injunctive provisions have been tailored to address the minor-subclass and the parental consent and control concerns related thereto.6 From the perspective of those seeking maximum privacy and other rights for Facebook members, the injunctive relief provisions leave much to be desired. To be — 945 — sure, any concern of misappropriation, or lack of consent, or commercial exploitation, could be eliminated entirely with provisions that ended Sponsored Stories, or set up an “opt-in” rather than a “opt-out” system, or even provided for members to be paid for use of their names and likenesses. The objectors advocating for greater restrictions to be imposed, however, give insufficient recognition to three points. First, in evaluating a settlement, the question is not whether it is perfect, or even whether a better result could be envisioned. As explained in Hanlon, Of course it is possible, as many of the objectors’ affidavits imply, that the settlement could have been better. But this possibility does not mean the settlement presented was not fair, reasonable or adequate. Settlement is the offspring of compromise; the question we address is not whether the final product could be prettier, smarter or snazzier, but whether it is fair, adequate and free from collusion. 150 F.3d at 1027. Second, some of the objectors appear to take it as a given that Facebook violated the law and/or that any settlement representing less than a complete vindication of all of plaintiffs’ positions cannot be fair, reasonable, and adequate. As noted above, the strength of that view remains untested, and Facebook has offered defenses, which although similarly nascent, stand as potentially significant impediments to recovery. Additionally, the implication underlying many of the arguments is that any imposition on the privacy interests of Facebook members is per se wrongful. As Facebook points out, however, it is a platform for sharing information, which members join voluntarily. Members are not charged any fees for Facebook’s services, which cost the company hundreds of millions of dollars to provide. While it does not follow that Facebook has carte blanche to exploit material belonging to, or regarding, its members in any fashion whatsoever, neither is it foreclosed from adopting SRRs that are not as “pro-member” or “pro-privacy” as some might like. Finally, as the parties correctly observe, the proffered settlement provides some benefits to the plaintiff class that would be difficult, if not impossible, ever to obtain through a contested judgment, even if plaintiffs were eventually to prevail on the merits. While a court might have some discretion to craft specific injunctive provisions, the settlement process has resulted in Facebook agreeing to implement various tools and procedures that address plaintiffs’ concerns in a more nuanced manner than would likely emerge from any victory at trial. Accordingly, while plaintiffs’ efforts to assign a multi-million dollar value to the injunctive relief is less than persuasive, those provisions of the settlement plainly provide actual benefits to the class. Going forward, operation of the Sponsored Stories program will be more transparent, and Facebook users will have a greater ability to see how and when their activities result in generation of Sponsored Stories, and to limit recurrences. The minor subclass, and parents of minors, will have further opt-out options. The injunctive relief, while not as robust as some would prefer, contributes to the conclusion that the settlement as a whole is fair, reasonable, and adequate. .4. Cy pres payments With the cash payouts increased to $15 and plaintiffs’ fee award substantially reduced as explained in the accompanying order, there will remain several million dollars to be distributed out of the settlement fund. As previously noted, it is well946 settled that a class action settlement providing for distribution of monetary relief in the form of cy pres payments may be appropriate where “the proof of individual claims would be burdensome or distribution of damages costly.” Nachshin v. AOL, LLC, 663 F.3d 1034, 1038 (9th Cir.2011) (quoting Six Mexican Workers v. Ariz. Citrus Growers, 904 F.2d 1301, 1305 (9th Cir.1990)). In this instance, implementing a cy pres component is realistic in light of the implications of the class size discussed above. While it has proven feasible to distribute a portion of the settlement fund by way of $15 payments to those class members who submitted claims, capping Facebook’s obligation facilitated settlement, and the entire payout would have gone to cy pres recipients had the number of filed claims precluded direct distribution. Indeed, it is evident that absent availability of a cy pres component, it simply might not have been feasible to settle this action, a result which plainly would conflict with the strong policy favoring settlements. Additionally, cy pres is a well-accepted method for distributing unclaimed settlement funds in any event. See Nachshin, 663 F.3d at 1036. Thus, while the circumstances are somewhat unusual, cy pres distribution is an acceptable approach to providing relief to the class in this action. The cy pres recipients selected by the parties also satisfy the requirement that there be “a driving nexus between the plaintiff class and the cy pres beneficiaries.” Id. at 1038; see also, Dennis v. Kellogg Co., 697 F.3d 858, 865 (9th Cir.2012). The recipient organizations focus on consumer protection, research, education regarding online privacy, the safe use of social media, and the protection of minors — the very issues raised in plaintiffs’ complaint.7 To be sure, the somewhat amorphous nature of the harm plaintiffs allegedly suffered makes it more difficult to say that providing funds to these recipients is precisely aligned with the “purpose” of the lawsuit and the plaintiff class. See Nachshin, 663 F.3d at 1039. The nexus, however, is sufficiently direct — these are not merely “worthy” recipients with “noble goals,” but organizations and institutions with demonstrated records of addressing issues closely related to the matters raised in the complaint.8 5. The release The release provided in the settlement agreement contains typically broad lan — 947 — guage to ensure that it applies to all appropriate persons (e.g. “affiliated and related entities, predecessors, successors and assigns and to all appropriate claims, known and unknown. The release is expressly limited, however, to claims arising from “display of any Class Member’s name, nickname, pseudonym, profile picture, photograph, likeness, or identity in a Sponsored Story.”9 As such, the release does not represent overreaching, or present a concern that class members are relinquishing more than would be warranted.10 B. The Objections Seventeen objections were submitted in conformance with the requirements of the preliminary approval order, made on behalf of 29 class members. An additional 87 statements purporting to be objections were filed on behalf of 95 class members.11 Over a third of the “objections” actually assert opinions that Facebook’s conduct was not improper and/or that the lawsuit is otherwise without merit or abusive.12 Many objectors complain that the monetary relief to each class member is insufficient. Those objectors who predicted there would be no individual payments have proven incorrect, and the actual payments will be larger than originally anticipated. The adequacy of the $15 payments has been addressed above. In particular, the argument that a substantially higher payout would be necessary in light of the theoretical availability of statutory damages fails to give sufficient weight to the reality that it would be virtually impossible for plaintiffs to be awarded, and to collect, the full amount of the statutory damages on a class-wide basis. Objectors also complain about the size of the attorney fee award that plaintiffs are seeking. The revised settlement agreement eliminated the “clear sailing” provision, and Facebook has opposed the fee application, arguing for a substantially smaller award. The size of the actual award will be addressed by separate order, and will not disturb the conclusion that the settlement as a whole is fair, reasonable, and adequate.13 — 948 — Among the most vigorously-pressed objections are those challenging whether the settlement can and does appropriately address issues relating to the minor subclass. Objectors complain that the settlement does not ensure valid parental consent to a minor’s participation in Sponsored Stories before the minor agrees to the SRRs. As an initial matter, these objections would have the Court decide — in plaintiffs’ favor — the merits of the dispute. The law is well-settled, however, that in evaluating a settlement the court is not to “reach any ultimate conclusions on the contested issues of fact and law which underlie the merits of the dispute, for it is the very uncertainty of outcome in litigation and avoidance of wasteful and expensive litigation that induce consensual settlements.” Officers for Justice v. Civil Service Com’n of City and County of San Francisco, 688 F.2d 615, 625 (9th Cir.1982). Thus, “the settlement or fairness hearing is not to be turned into a trial or rehearsal for trial on the merits.” Id. Furthermore, even to the extent some preliminary analysis of the merits may be appropriate, the objectors have not persuasively shown the settlement to be improper. As noted above, COPPA may well preempt claims based on any failure by Facebook to obtain parental consent for members of the minor class. COPPA requires an “operator of a website or online service” to obtain parental consent before it “collects” or “use[sj” the “personal information” of a “child,” but only where the child is “under the age of 13.” 15 U.S.C. §§ 6501(1), 6502(a), 6502(b)(l)(A)(ii) (emphasis added); 16 C.F.R. § 312.5(a)(1). Because COPPA expressly preempts state requirements that are “inconsistent with” this “treatment,” 15 U.S.C. § 6502(d), it could bar any efforts by plaintiffs to use state law to impose a parental consent requirement for minors over the age of 13.14 The suggestion that the laws of other states might provide additional protection for minors fails both because objectors have neither demonstrated that the supposed differences in law are material, nor that choice-of-law principles would permit the application of such law in any event. The remaining objections all generally fall into the category of suggestions as to how the settlement might be made better, particularly from the perspective of the plaintiff class. Once again, however, “the question we address is not whether the final product could be prettier, smarter or snazzier, but whether it is fair, adequate and free from collusion.” Hanlon, 150 F.3d at 1027. Even if every suggestion represents an actual potential “improvement,” and even considering all the suggestions cumulatively, they do not support a conclusion that this settlement is the product of collusion, or otherwise fails to meet the minimum threshold of fairness and adequacy. All other considerations applicable to settlement approval, including such issues as numerosity, typicality, adequacy of representation, notice, are not seriously in dispute.15 The findings made in conjunc — 949 — tion with prehminary approval remain appropriate and are adopted here. IV. CONCLUSION The objections to the settlement are overruled, and the motion for final approval is granted. A separate order will issue on plaintiffs’ motion for an award of attorney fees and for incentive awards to the named plaintiffs. IT IS SO ORDERED. Footnotes [1] The factual and procedural background of this litigation has been set forth in prior orders and will not be repeated here. [2] Far more significant was the effective increase in the total amount going to the class (either directly or constructively through cy pres) that results from the elimination of the “clear sailing” attorney fee provision, and the reduction in the amount of fees even being requested. The fee award will be addressed in a separate order. [3] The implications of raising the payout to $15 per claim have been carefully considered, as has the potential of raising it even further. Although the notice to class members sufficiently preserved the possibility that payments would exceed $10, the, payments cannot be raised to a level that would be unfair to those class members who declined to submit claims for what they believed almost certainly would be $10 or less. The increase to $15 represents an appropriate balance between the interest in distributing as much of the cash directly to class members as possible and the need to avoid creating such unfairness. [4] Paying every class member $750 each would require over $112 billion. [5] Plaintiffs' contention that a dollar value should be assigned to the injunctive relief for purposes of justifying a percentage-based fee award will be addressed in the ruling on the fee application. [6] Objectors led by John Schachter seek leave to file a supplemental brief regarding media reports that Facebook is unilaterally eliminating the Sponsored Stories program. (Dkt. 354). While leave to file the supplemental brief is granted, the news reports to which objectors point do not show that the injunctive relief will be rendered "meaningless,” even to the extent the assertions in those articles could be assumed true. [7] As provided by the settlement agreement and approved herein, the funds available for cy pres distributions will be allocated to the following entities in the percentages specified: Center for Democracy and Technology (10%), Electronic Frontier Foundation (10%), MacArthur Foundation (10%), Joan Ganz Cooney Center (10%), Berkman Center for Internet and Society (Harvard Law School) (6%), Information Law Institute (N.Y.U Law School) (6%), Berkeley Center for Law and Technology (Berkeley Law School) (6%), Center for Internet and Society (Stanford Law School) (6%), High Tech Law Institute (Santa Clara University School of Law) (6%), Campaign for Commercial-Free Childhood (6%), Consumers Federation of America (6%), Consumer Privacy Rights Fund (6%), ConnectSafely.org (6%), and WiredSafety.org (6%). [8] Objector Brodsky complains that Connect-Safely.org is an inappropriate recipient because it receives other funding from Face-book, a fact disclosed by the parties. The mere receipt of some other funding from Facebook does not give rise to a reasonable doubt that the organization lacks sufficient independence to serve as a suitable recipient of cy pres funds. Objectors Boisvert and Frank complain that Santa Clara Law School should not receive any funds because one of plaintiff’s attorneys is a graduate, and the judge previously presiding over this action has ties to the institution. These connections are too tenuous to disqualify the school as a cy pres recipient, particularly in light of the reassignment of this case. [9] The named plaintiffs have agreed to a substantially broader release, which does not bear on the fairness of the agreement as to the unnamed class members. [10] This also answers the concern of some objectors that the injunctive relief does not go far enough to prevent Facebook from engaging in future conduct that raises similar concerns. The release is unlikely to bar claims in such instances. [11] In the absence of any prejudice to the parties, the motion of Janine R. Menhennet, as guardian ad litem for Michael M., to have her objection deemed validly submitted (Dkt. No. 353) is granted. The arguments raised in that objection are cumulative to those of other objectors, and will not be separately addressed. [12] The administrator reports that 6,825 class members exercised their right to opt out of the settlement, a number that may at first blush appear somewhat large, but which as a percentage of the class is miniscule. Give the percentage of “objections” that faulted the suit rather than the settlement, it is likely a significant number of those opting-out may have held similar views. The right to opt-out, of course, also ensures that any class member who may have more tangible damages, or who wishes to pursue the statutory damages claim, may still seek a remedy. [13] Likewise, the individual incentive awards do not make the settlement unfair. Unlike the situation in Radcliffe v. Experian Information Solutions, 715 F.3d 1157 (9th Cir.2013) the awards were not conditioned on support for the settlement. The named plaintiffs here also at least theoretically were at risk of an attorney fee award being entered against them if Facebook prevailed, under the fee-shifting provisions of Civil Code § 3344. Finally, the actual amount of the incentive awards are adjusted in a contemporaneously948 issued order regarding fees, costs, and those awards. [14] Objectors’ reliance on provisions of the California Family Code is similarly unavailing. While Family Code § 6701(a) prevents minors from entering into enforceable "delegation[s] of power” and § 6701(c) limits their ability to contract away rights to "personal property, not in the immediate possession or control of the minor,” neither subsection is implicated by the circumstances here. [15] While the Shane objectors argue that the minor subclass required separate counsel, that is premised on the untested view that the minors’ claims are "substantially stronger” than those of adult class members.