Master Docket No. 12 C 10064. MDL No. 2416. No. 12 C 10135. No. 11 C 5886. No. 12 C 1061.

80 F.Supp.3d 781 (2015)

IN RE CAPITAL ONE TELEPHONE CONSUMER PROTECTION ACT LITIGATION. Bridgett Amadeck, et al., v. Capital One Financial Corporation, and Capital One Bank (USA), N.A. Nicholas Martin, et al., v. Leading Edge Recovery Solutions, LLC, and Capital One Bank (USA), N.A. Charles C. Patterson, v. Capital Management Services, L.P. and Capital One Bank (USA), N.A.

United States District Court, N.D. Illinois, Eastern Division.

Signed February 12, 2015.


The three above-captioned, nationwide class actions were filed against Capital One, its subsidiaries, and its Participating Vendors (collectively, "Defendants"),1 as a result of the Defendants' allegedly using automatic telephone dialing systems or artificial or prerecorded voice messages to contact consumers' cell phones without prior express consent, in alleged violation of the Telephone Consumer Protection Act ("TCPA"), 47 U.S.C. § 227(b)(1)(A). (Dkt. No. 120.) On December 10, 2012, the United States Judicial Panel on Multidistrict Litigation (the "JPML") selected this court to coordinate pursuant to 28 U.S.C. § 1408 the pretrial proceedings in these three class actions, along with other individual lawsuits filed throughout the United States. (Dkt. No. 1.) The cases filed out-side this district were transferred to this district and assigned to this court's calendar. On February 28, 2013, Plaintiffs filed a Consolidated Master Class Action Complaint ("Master Complaint") superseding the complaints filed in the three class actions. (Dkt. No. 19.) On June 13, 2014, after reaching a settlement in principle, Plaintiffs filed an Amended Consolidated Master Class Action Complaint ("Amended Master Complaint"). (Dkt. No. 120 ("Am. Compl.").)

On July 29, 2014, the court granted Plaintiffs' unopposed request for preliminary approval of class settlement,2 (Dkt. No. 129), and entered an Order (Dkt. No. 137) conditionally certifying a settlement class, preliminarily approving the class action settlement, approving the notice plan, and appointing a claims and notice administrator. Since then, the parties have filed memoranda in support of Plaintiffs' motion (Dkt. No. 260) for final approval of the class action settlement. Class Counsel, consisting of the attorneys who collectively represent the class, have also filed a motion for approval of attorneys' fees and for service awards to the class representatives (the "Named Plaintiffs"). (Dkt. No. 175.) Fourteen people out of the more than 17 million settlement class members filed briefs or statements in opposition to the Amended Settlement Agreement and Release ("Settlement Agreement") (Dkt. No. 131 Ex. 1) and Class Counsel's requested fee award. The court, after notice was provided, conducted a fairness hearing on January 15, 2015 to allow any class members who expressed the desire to address the court regarding the settlement to do so. (Dkt. No. 320.)

For the reasons explained below, the court grants the motion for final approval of the class action settlement, (Dkt. No. 260), because under the circumstances and the law the settlement reached in these three consolidated class action cases is fair, reasonable, and adequate. The court grants in part and denies in part Class Counsel's motion for approval of attorneys' fees, and grants Class Counsel's requested incentive awards to the five Named Plaintiffs in the amount of $5,000 each. (Dkt. No. 175.)


I. History of the Litigation

In 1991, Congress enacted the TCPA "to address telephone marketing calls and certain telemarketing practices that Congress found to be an invasion of consumer privacy." Jamison v. First Credit Servs., 290 F.R.D. 92, 96 (N.D.Ill.2013) (Kendall, J.). The "certain telemarketing practices" that drew Congress's legislative action were automatic telephone dialing systems and prerecorded voices. 47 U.S.C. § 227. The two technologies were relatively new in 1991 and greatly improved telemarketers' ability to contact consumers on their phones. In response to the "national out-cry over the explosion of unsolicited telephone advertising," Congress passed the TCPA. See 137 Cong. Rec. 30,817 (1991) (statement of Senator Pressler). The TCPA prohibits callers from using "any automatic telephone dialing system or an artificial or prerecorded voice" to make any non-emergency call to a cell phone, unless they have the "prior express consent of the called party." 47 U.S.C. § 227(b)(1)(A)(iii). The penalties Congress enacted to answer the public outcry are harsh: the TCPA imposes on callers statutory damages of $500 per call, which can be trebled if the court finds the violation to have been willful or knowing. 47 U.S.C. § 227(b).

The calls at issue in these three consolidated class actions were made for the decidedly non-emergency purpose of debt collection. According to the Amended Master Complaint, between January 18, 2008 and June 20, 2014, Capitol One or one of its Participating Vendors (on behalf of Capital One) called class members' cell phones using an automatic telephone dialing system or an artificial or prerecorded voice in connection with an attempt to collect on a credit card debt. (Am. Compl. ¶ 52.)

After Plaintiffs filed their Master Complaint on February 28, 2013, the parties engaged in six months of class-wide discovery "sufficient to engage in meaningful settlement discussions." (Dkt. No. 129 at 13.) On July 2, 2013, November 4, 2013, and January 29, 2014, the parties participated in mediation sessions with retired United States Magistrate Judge Edward A. Infante. The parties also spoke with Judge Infante by phone on two other occasions. (Id.) Capital One and Plaintiffs agreed thereafter to a settlement in February 2014. (Id. at 14.) The Participating Vendors agreed to join the settlement in the months thereafter. (Id.)

On June 13, 2014, Plaintiffs filed their request for an order certifying the proposed class for settlement purposes, preliminarily approving the settlement agreement, approving the notice plan, ordering the dissemination of notice as set out in the Settlement Agreement, and appointing BrownGreer as the Notice and Claims Administrator. (Dkt. No. 121.) Plaintiffs filed an amended motion seeking the same relief on July 13, 2014 and, on July 29, 2014, the court granted Plaintiffs' amended motion. (Dkt. No. 137.)

On August 12, 2014, BrownGreer began implementing the parties' notice plan, which entailed: (1) sending 12,342,000 summary notices via email to all potential class members who had email addresses reflected in Capital One's records; (2) mailing 4,303,218 postcard notices via first class mail to class members who had opted out of receiving email from Capital One, who did not have email addresses on file, or whose emails were undeliverable; (3) running internet banner notices on 40 websites BrownGreer determined class members were likely to visit; (4) establishing a settlement website and toll-free information telephone number dedicated to answering telephone inquiries; and (5) providing notice of the settlement to the officials designated pursuant to Class Action Fairness Act, 28 U.S.C. § 1715. (Dkt. No. 264.)

BrownGreer provided a thorough summary of its execution of the notice plan in two separate declarations provided to the court. (Dkt. Nos. 264, 305.) Here, it is sufficient to note that the notice plan reached 15,983,613 known, unique settlement class members, a figure that represents 96.03% of the known settlement class and 91.22% of the estimated total settlement class.3 (Dkt. No. 305 ¶ 6.) Despite the robust and effective notice plan, only 1,378,534 unique claimants—7.87% of the estimated class—filed claims with the administrator by the filing deadline. (Id. ¶ 14.) 462 class members have submitted valid opt-out requests and another 103 claimants have submitted opt-out requests that are invalid, either because they are incomplete or untimely. (Id. ¶ 8.) BrownGreer estimated that as of December 23, 2014 its total notice and administration costs were $5,093,000. (Id. ¶ 16.) No updated figures have been provided to the court.

II. The Settlement Agreement

The important provisions of the Settlement Agreement provide for both monetary and injunctive relief to class members.

The Settlement Agreement defines the settlement class as follows:

All persons within the United States who received a non-emergency telephone call from Capital One's dialer(s) to a cellular telephone through the use of an automatic telephone dialing system or an artificial or prerecorded voice in connection with an attempt to collect on a credit card debt from January 18, 2008 through June 20, 2014, and all persons within the United States who received a non-emergency telephone call from a Participating Vendor's dialer(s) made on behalf of Capital One to a cellular telephone through the use of an automatic telephone dialing system or an artificial or prerecorded voice in connection with an attempt to collect on a credit card debt from February 28, 2009, through June 20, 2014.

(Settlement Agreement § 2.39.) Plaintiffs estimate that the class includes 17,522,049 members.4

The Settlement Agreement requires Defendants to establish a non-reversionary settlement fund of $75,455,099.5 (Settlement Agreement § 2.42.) After subtracting notice and administration costs ($5,093,000), Class Counsel's requested service awards for the five Named Plaintiffs ($25,000), and Class Counsel's requested fee award ($22,636,530)—all of which will be paid out of the settlement fund—the value of the settlement to class members is $47,700,569. (Dkt. No. 305.); see Pearson v. NBTY, Inc., 772 F.3d 778, 780-81 (7th Cir.2014) (citing Redman v. RadioShack Corp., 768 F.3d 622, 630 (7th Cir.2014) (holding notice costs, administration costs, and attorneys' fees are not part of the value received from the settlement by class members). If all 17,522,049 class members had filed a claim, they would have received $2.72 each. But because only a fraction of class members filed a claim, as is often the case in consumer class actions, the 1,378,534 timely claimants will receive at least $34.60 each, and possibly more if some claimants fail to cash their settlement checks within 210 days, allowing for a second pro rata distribution to class members who filed a claim and deposited their settlement checks on time. (Settlement Agreement § 7.04(e).) If, following the second pro rata distribution, there remain undeposited settlement checks, the remainder of the settlement fund will go to a cy pres recipient. The Settlement Agreement does not identify the recipient of the cy pres award because the parties decided to wait until after the claims period to gauge the potential size of any cy pres award. (Settlement Agreement § 7.04(f).) In their response to objectors, however, Class Counsel agreed to name the Electronic Frontier Foundation. (Dkt. No. 269 at 33.)

The Settlement Agreement also requires Capital One to institute a protocol under which it uses an automatic dialer to call a customer's (or debtor's) cell phone number only in cases where the individual provided the cell phone number on his or her credit application. Capital One will further refrain from calling a cell phone number unless either the cell phone number is linked to the customer's name based on third party research or Capital One has made contact with the customer on the cell phone number within the past 90 days. (Dkt. No. 262 ¶ 21.) As discussed below, this change would bring Capital One's use of an automatic dialer within Plaintiffs' interpretation of the statutorily undefined term "prior express consent," which is excluded from the prohibitions of the TCPA, 47 U.S.C. § 227(b)(1).

The court, at final approval hearing on January 15, 2015, (Dkt. No. 320), heard from Class Counsel, counsel for Capital One, and counsel for objector Jeffrey Collins. Although the court invited specific objectors by name and anyone else present in the courtroom to speak, no other objector addressed the court even though some of the objectors had previously indicated their intent to address the court at the final approval hearing.6


I. Approval of a Proposed Settlement in Class Actions

A court may approve a settlement that would bind class members only if, after proper notice and a public a hearing, the court determines that the proposed settlement is "fair, reasonable, and adequate." Fed. R. Civ. P. 23(e)(3). Under Seventh Circuit law, a district court must, in evaluating the fairness of a settlement, consider "the strength of plaintiffs' case compared to the amount of defendants' settlement offer, an assessment of the likely complexity, length and expense of the litigation, an evaluation of the amount of opposition to settlement among affected parties, the opinion of competent counsel, and the stage of the proceedings and the amount of discovery completed at the time of settlement." Synfuel Techs., Inc. v. DHL Express (USA), Inc., 463 F.3d 646, 653 (7th Cir.2006) (quoting Isby v. Bayh, 75 F.3d 1191, 1199 (7th Cir.1996)).

"The `most important factor relevant to the fairness of a class action settlement' is the first one listed: `the strength of plaintiff's case on the merits balanced against the amount offered in the settlement.'" Synfuel, 463 F.3d at 653 (quoting In re Gen. Motors Corp. Engine Interchange Litig., 594 F.2d 1106, 1132 (7th Cir.1979)). Furthermore, "[i]n conducting this analysis, the district court should begin by `quantifying the net expected value of continued litigation to the class.' To do so, the court should `estimate the range of possible outcomes and ascribe a probability to each point on the range.'" Id. (quoting Reynolds v. Beneficial Nat'l Bank, 288 F.3d 277, 284-85 (7th Cir.2002)).

"Federal courts naturally favor the settlement of class action litigation." Isby, 75 F.3d at 1196. Nevertheless, the Seventh Circuit has warned that "the structure of class actions under Rule 23... gives class action lawyers an incentive to negotiate settlements that enrich themselves but give scant reward to class members, while at the same time the burden of responding to class plaintiffs' discovery demands gives defendants an incentive to agree to early settlement that may treat the class action lawyers better than the class." Thorogood v. Sears, Roebuck & Co., 627 F.3d 289, 293 (7th Cir.2010) (emphasis omitted). District courts must therefore "exercise the highest degree of vigilance in scrutinizing proposed settlements of class actions." Synfuel, 463 F.3d at 652. This court has endeavored to do that.

II. Attorneys' Fees in Class Actions

"In a certified class action, the court may award reasonable attorney's fees ... that are authorized by law or by the parties' agreement." Fed. R. Civ. P. 23(h). In determining a reasonable fee, the court "must balance the competing goals of fairly compensating attorneys for their services rendered on behalf of the class and of protecting the interests of the class members in the fund." Skelton v. Gen. Motors Corp., 860 F.2d 250, 258 (7th Cir.1988), cert. denied, 493 U.S. 810, 110 S.Ct. 53, 107 L.Ed.2d 22 (1989). To determine the reasonableness of the sought-after fee in a common-fund case, "courts must do their best to award counsel the market price for legal services, in light of the risk of nonpayment and the normal rate of compensation in the market at the time." In re Synthroid Mktg. Litig., 264 F.3d 712, 718 (7th Cir.2001) (Synthroid I). The probability of success at the outset of the litigation is relevant to this inquiry. See Florin v. Nationsbank of Ga., N.A., 34 F.3d 560, 565 (7th Cir.1994).

In Synthroid, the Seventh Circuit held that the "market rate for legal fees depends in part on the risk of nonpayment a firm agrees to bear, in part on the quality of its performance, in part on the amount of work necessary to resolve the litigation, and in part on the stakes of the case." Synthroid I, 264 F.3d at 721. The Seventh Circuit has further explained that "[t]he object in awarding a reasonable attorney's fee ... is to give the lawyer what he would have gotten in the way of a fee in arm's length negotiation, had one been feasible." In re Cont'l Ill. Sec. Litig., 962 F.2d 566, 572 (7th Cir.1992). See also In re Trans Union Corp. Privacy Litig., 629 F.3d 741, 744 (7th Cir.2011) (recognizing that "[s]uch [an] estimation is inherently conjectural").

The Federal Rules of Civil Procedure allow the court, in a certified class action, to "award reasonable ... nontaxable costs that are authorized by law or by the parties' agreement." Fed.R.Civ.P. 23(h). The Seventh Circuit has instructed that district courts must exercise their discretion to "disallow particular expenses that are unreasonable whether because excessive in amount or because they should not have been incurred at all." Zabkowicz v. W. Bend Co., Div. of Dart Indus., Inc., 789 F.2d 540, 553 (7th Cir.1986) (quoting Henry v. Webermeier, 738 F.2d 188, 192 (7th Cir.1984)).


I. Approval of the Class Settlement in This Litigation

Applying the five factors identified in Synfuel, the court concludes that the settlement is "fair, reasonable, and adequate," and therefore meets the requirements of Rule 23.

A. Potential of Class Members' Recovery through Continued Litigation Balanced Against Settlement Amount Offered

As noted above, "[t]he most important factor" in determining whether a proposed settlement satisfies Rule 23 is the "strength of [Plaintiffs'] case on the merits balanced against the amount offered in the settlement." Synfuel, 463 F.3d at 653 (citations omitted). The Settlement Agreement, as it stands, requires Defendants to pay $75.5 million into the settlement fund out of which all eligible class members who made a timely claim will receive their pro rata share, and maybe more if fellow claimants are delinquent in depositing their checks. At the time the court granted preliminary approval, Defendants stated that the settlement constituted the largest cash sum in the 22-year history of the TCPA, (Dkt. No. 129 at 7). That fact, though true, is slightly deceiving because the size of the settlement amount is attributable mainly to the large size of the class—17.5 million people. The recovery per class member—excluding administrative costs, Named Plaintiff awards, and attorneys' fees—is a relatively diminutive $2.72. The court does not have the necessary data to compare this proposed settlement to other TCPA actions based on the recovery per class member. There are, however, a number of benchmark settlements to which the court can compare the recovery per claimant. The recovery per claimant here is $34.60. That number falls within the range of recoveries in other TCPA actions but, as Judge Davila noted in discussing a similarly sized settlement last year, it falls on the "lower end of the scale." Rose v. Bank of Am. Corp., Nos. 11 C 2390 & 12 C 4009, 2014 WL 4273358, at *11 (N.D.Cal. Aug. 29, 2014). The settlement also falls far short of the $500 statutory recovery available for each phone call, of which there were many: Capital One or its Participating Vendors made approximately 1.9 billion phone calls in alleged violation of the TCPA. (Dkt. No. 324 at 11:3.) So if Plaintiffs were to litigate their claims successfully through trial, Capital One would be on the hook for a minimum of $950 billion. Moreover, Plaintiffs' recovery could possibly be as high as $2.85 trillion if Plaintiffs proved the violations were knowing or willful.

But a settlement is a compromise, and courts need not—and indeed should not—"reject a settlement solely because it does not provide a complete victory to plaintiffs." In re AT & T Mobility Wireless Data Servs. Sales Litig., 270 F.R.D. 330, 347 (N.D.Ill.2010) (St.Eve, J.). This is especially true when complete victory would most surely bankrupt the prospective judgment debtor. It also bears mention that the $34.60 per claimant recovery in this case does not seem so miniscule in light of the fact that class members did not suffer any actual damages beyond a few unpleasant phone calls, which they received ostensibly because they did not pay their credit card bills on time.

More importantly, $34.60 per claimant is not insignificant considering Capital One's counsel's estimate that Plaintiffs will recover nothing through continued litigation. (Dkt. No. 267.) The court recognizes that Plaintiffs would indeed face myriad hurdles by proceeding to trial.

First, at a trial, Plaintiffs would have the burden to effectively rebut Capital One's chief defense that the class members' consented to be contacted on their cell phones. Capital One argues that it obtained consent to call from each class member because every version of Capital One's standard cardholder agreement contained provisions expressing that Plaintiffs consented to receive calls through an autodialing technology. (Dkt. No. 267 at 2.) Plaintiffs admit that they agreed to the terms of their cardholder agreements, but argue they did not agree to be contacted "in violation of the TCPA." (Dkt. No. 262.) Many class members, however, even provided their cell phone numbers to Capital One as their primary contact numbers. (Id.) Under an FCC order in 2008 implementing the TCPA, autodialed collection calls to "wireless numbers provided by the called party in connection with an existing debt are made with the `prior express consent' of the called party," and are therefore permissible. In re Rules and Regulations Implementing the Telephone Consumer Prot. Act of 1991, 23 F.C.C.R. 559 ¶ 9 (2008) ("2008 TCPA Order"); 47 U.S.C. § 227(b)(1)(A)(iii). The FCC's same 2008 TCPA Order, however, states that "prior express consent is deemed to be granted only if the wireless number was provided by the consumer to the creditor, and that such number was provided during the transaction that resulted in the debt owed." 2008 TPCA Order ¶ 10. Plaintiffs interpret the FCC's 2008 TCPA Order to mean that the cell phone number must have been provided during the origination of the credit relationship, i.e., during the transaction. (Dkt. No. 262 at 21.) As United States District Judge J.P. Stadtmueller commented in a recent opinion, however, the 2008 TCPA Order is "far from clear." Balschmiter v. TD Auto Finance LLC, 303 F.R.D. 508, 517, 2014 WL 6611008, at *8 (E.D.Wis.2014). Furthermore, in this district, the only district judge to have addressed the issue held that a caller is entitled to summary judgment against a TCPA claim when it can show the plaintiff provided a cell phone number as a contact number. See Greene v. DirecTV, No. 10 C 117, 2010 WL 4628734, at *3 (N.D.Ill. Nov. 8, 2010) (Kocoras, J.). The parties' disparate interpretations of the 2008 TCPA Order are reflective of the split opinion among practitioners and the courts, a split that at least injects uncertainty into this litigation and will continue to warrant caution by plaintiffs and defendants until clearer guidance is provided. See, e.g., Baird v. Sabre, Inc., 995 F.Supp.2d 1100, 1106 (C.D.Cal.2014) (noting the FCC's series of TCPA Orders are not "model[s] of clarity").

Second, should Plaintiffs proceed to trial, there would be manageability concerns that may pose serious obstacles to class certification, thus depriving Plaintiffs of the benefits of a class action. Rule 23(b)(3) requires that "questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy." Fed. R. Civ. P. 23(b)(3). In assessing predominance, a court must analyze "the likely difficulties in managing a class action," id. 23(b)(3)(D), which "encompass[ ] the whole range of practical problems that may render the class action format inappropriate for a particular suit." Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 164, 94 S.Ct. 2140, 40 L.Ed.2d 732 (1974). Identifying consenting class members and the precise timing and nature of that consent would require Capital One to locate documents and analyze call recordings for nearly all of the 17.5 million class members. These individual determinations do not always comport with Rule 23(b)(3)'s manageability requirement and have caused some courts to reject class certification on numerous occasions. See, e.g., Balschmiter, 303 F.R.D. at 527-29, 2014 WL 6611008, at *19-20; see also Jamison, 290 F.R.D. at 107 (denying certification of TCPA litigation where "parties would need to scour [defendant's] records" to determine consent).

Third, without the prompt and final resolution a settlement provides, Plaintiffs run the risk that forthcoming FCC orders may extinguish their claims. There are three sets of petitions currently before the FCC, all of which would eliminate or reduce Capital One's TCPA liability in this case. The first is the FCC's definition of an autodialer. Although the TCPA defines an autodialer as "equipment which has the capacity (A) to store or produce telephone numbers to be called, using a random or sequential number generator; and (B) to dial such numbers," 47 U.S.C. § 227(a)(1), the FCC has expanded the definition to cover predictive dialers that can "store or produce telephone numbers," even if they do not "us[e] a random or sequential number generator." See In re Rules and Regulations Implementing the Tel. Consumer Prot. Act of 1991, 18 F.C.C.R. 14014, 14091-93 (2003). The FCC is considering petitions seeking to exclude from the TCPA predictive dialers used for non-telemarketing purposes, such as debt collection. (See Dkt. No. 267 (collecting petitions).) The second and, perhaps, more pressing set of petitions to the FCC ask the FCC to clarify how and when consent may be expressed by consumers. See Michael O'Rielly, FCC Commissioner, TCPA: It is Time to Provide Clarity, FCC Blog (Mar. 25, 2014) (available at (asserting that "the FCC needs to address this inventory of petitions as soon as possible," and "answer ... whether consent can be inferred from consumer behavior or social norms"). The final set of petitions seeks to clarify that a caller does not violate the TCPA when it makes autodialed calls to another cell phone subscriber by mistake. (See Dkt. No. 267 (collecting petitions).) If the FCC were to issue orders favoring callers in connection with any of the issues discussed above, Plaintiffs claims would be completely barred or materially limited.

In light of Capital One's potentially meritorious defenses and the legal uncertainty concerning the application of the TCPA, the court concludes that Plaintiffs would probably face an uphill battle proceeding to trial and, once there, obtaining relief. The settlement provides value that is fair considering the very real possibility that Plaintiffs may recover nothing if they were to proceed further with the litigation.

B. Likely Complexity, Length and Expense of Litigation

In Synfuel, the Seventh Circuit instructed that the likely complexity, length, and expense of continued litigation are relevant factors district court should consider in determining whether a class action settlement satisfies Rule 23. Synfuel, 463 F.3d at 653. All of these factors when considered in this litigation strongly weigh in favor of approval of the proposed settlement. Although the parties have conducted limited discovery for the purpose of evaluating settlement, they would need to engage in significant additional discovery of Capital One's millions (or billions) of call records, if the litigation were to proceed further. This would likely require each side to retain experts to analyze the mountains of data. There would be significant motion practice, and any judgment in favor of Plaintiffs would be further delayed by any appeal taken from the entry of a final judgment.

C. Scant Opposition to Settlement

The Seventh Circuit has held that the amount of opposition to a settlement among affected parties is yet another factor district courts should consider in deciding whether to approve a class action settlement. Synfuel, 463 F.3d at 653. Only 565 class members have requested to be excluded from the settlement, representing approximately 0.0032% of all class members.7 Of the approximately, 17.5 million class members, the court has received 14 timely objections to the Settlement Agreement and only 9 of those objections take issue with the value of the settlement; the other 5 objectors limit their concerns to Class Counsel's requested fee award. Such a low percentage of opposition favors a finding that the settlement is fair, reasonable, and adequate under Rule 23. See, e.g., In re AT & T Mobility Wireless Data Services Sales Tax Litigation, 789 F.Supp.2d 935, 965 (N.D.Ill.2011) (citations omitted) (finding opt-out or objection by 0.01% of class members was "remarkably low" and supported the settlement).

D. The Experience and Views of Counsel

Under Synfuel, the opinion of competent counsel is relevant to determining whether a class action settlement is fair, reasonable, and adequate under Rule 23. Synfuel, 463 F.3d at 653. The court accepts that Class Counsel in this case are experienced litigators, especially in the TCPA context, and that they strongly support the settlement. (Dkt. No. 262 at 20.) Even though Class Counsel may be considered biased because they stand to benefit from approval, under Synfuel, this factor weighs in favor of approval.

E. Stage of the Proceedings and the Amount of Discovery Completed

The final factor the court is to consider under Synfuel concerns the stage of the proceedings and the amount of discovery completed at the time of the settlement. Synfuel, 463 F.3d at 653. The parties in this case engaged in substantial motion practice and discovery in two of the individual class actions before the JPML transferred the cases to this court. Class Counsel have analyzed a complete set of the contractual language Capital One offers as the basis for class members' consent to be contacted. And prior to the mediation proceedings before retired Magistrate Judge Infante, the parties exchanged discovery over a six-month period sufficient to engage in "meaningful settlement discussions." Although this settlement-directed discovery is not identical to the type of full discovery that counsel may desire "to evaluate the merits of plaintiffs' claims," Armstrong v. Bd. of Sch. Dirs. of City of Milwaukee, 616 F.2d 305, 325 (7th Cir.1980), the court is not convinced that extensive formal discovery, when measured against the cost that would be incurred, would place the parties in a proportionally better position than they are now to determine an appropriate settlement value of this litigation. Evaluating the merits of Plaintiffs' claims would, as discussed above, require an arduous scouring of Capital One's records for individual plaintiffs, undermining the cost-saving benefits of the settlement. The court finds that the parties have completed a sufficient amount of discovery to be able to place value on their respective positions in this case. The final Synfuel factor weighs in favor of settlement.

F. Objections Presented Are Not Well Founded Under the Applicable Law

For the reasons explained above, the factors set out by the Seventh Circuit in Synfuel support approving the Settlement Agreement in this case. Notwithstanding the court's conclusion that the Settlement Agreement is fair, reasonable, and adequate under Rule 23, there are 14 class members who have filed timely objections, although only a subset of those 14 take issue with the amount of the settlement. (Dkt.Nos.144, 152, 184, 189, 193, 196, 199, 202, 215, 225, 227, 228.) These objections collectively state a number of arguments the court will discuss briefly below.

First, some objectors argue that class members are not receiving enough money in light of the available statutory damages. As the court discussed above, a class-wide recovery in line with the statutory awards is unrealistic and would ultimately result in class members finding their place in line among Capital One's unsecured creditors in a bankruptcy proceeding. Moreover, as discussed above, the strength of Plaintiffs' case did not warrant a settlement anywhere close to the statutory award, which is what Plaintiffs would have sought had they prevailed at trial on the liability issue.

Second, certain objectors complain that they should be able to make claims against the settlement fund for every call they received, consistent with the framework of the TCPA. Although the court inquired of counsel about the possibility of a call-based claims process as well, the court ultimately accepts the representations of Class Counsel and Capital One that it is unlikely that a material portion of the class had an average call volume greater or lesser than any other class member. The court also accepts Class Counsel's representation that a call-based claims process would be extremely costly to administer and is inadvisable given the fact that increased administration costs would result in a corresponding decrease in the money available to the class.

Lastly, the court rejects the complaints of the objectors who have any issue with the notice and claims process. The court agrees with Class Counsel that the notice provided by BrownGreer was state of the art and well-tailored to reach the maximum number of class members. Lead counsel for Capitol One represented to the court that percentage of class members reached through the notice process was the highest he had ever seen. (Dkt. No. 324 at 35:16-20.) Submitting a claim, in this court's experienced view, was exceedingly easy for the class members. Each class members needed only to complete a short online form or return a postcard.

Accordingly, the court finds that none of the objections to the total amount of the settlement or its administration are well-founded and, for the reasons explained in detail above, the court grants the motion (Dkt. No. 260) for final approval of the class action Settlement Agreement.

II. Attorneys' Fees

Although certain of 14 timely objectors contend that Class Counsel's requested fee is excessive, the court need not engage in a lengthy analysis of each objector's argument because the Seventh Circuit has directed district courts, when deciding whether requested fees are excessive, to estimate the contingent fee that the class would have negotiated with Class Counsel at the outset of the litigation, "had negotiations with clients having a real stake been feasible." In re Trans Union Corp. Privacy Litig., 629 F.3d 741, 744 (7th Cir.2011). The court endeavors to approximate such a market fee below.

A. Class Counsel's Requested Fees

Class Counsel in this case represent that they have spent 4,268 hours in professional time over a three-year period litigating and settling this case on a contingent fee basis. (Dkt. No. 252.) They seek for their efforts an award of attorneys' fees equal to 30% of the $75,455,099 settlement fund or $22,636.530. They do not seek additional payment or reimbursement for any of their expenses on top of the requested fee award, nor do they seek compensation for the injunctive relief barring Capital One from calling an individual's cell phone without prior express consent.

To justify their request, Class Counsel argue that their requested fee is less than the 33.3% fee "consistently" awarded in TCPA and non-TCPA class action litigation in this district. (See Dkt. No. 176 at 19-20 (collecting cases).) They further argue that the substantial risk they assumed in prosecuting the litigation on a purely contingent basis supports a 30% fee because there is a reasonable chance that this case, if litigated, could result in no recovery at all for the class. Class Counsel urge the court to adopt their preferred approach to calculating fees based on a percentage of the settlement fund rather than through a lodestar analysis, and to calculate that percentage from the total settlement fund inclusive of administrative costs, cy pres awards, and incentive awards.

B. Fee Calculation Method

Although the court granted limited discovery regarding Class Counsel's lodestar data, the court agrees with Class Counsel that the fee award in this case should be calculated as a percentage of the money recovered for the class. It has long been the law in the Seventh Circuit that in common fund cases like this one, district courts have discretion to choose either the lodestar or a percentage approach to calculating fees. Florin v. Nationsbank of Ga., N.A., 34 F.3d 560, 566 (7th Cir.1994) ("It bears reiterating here that we do not believe that the lodestar approach is so flawed that it should be abandoned .... We therefore restate the law of this circuit that in common fund cases, the decision whether to use a percentage method or a lodestar method remains in the discretion of the district court. We recognize here... that there are advantages to utilizing the percentage method in common fund cases because of its relative simplicity of administration."). Ultimately, the district judge's job is to approximate the market rate between willing buyers and willing sellers that would have prevailed had the parties negotiated the rate at the outset of the representation. Silverman v. Motorola Solutions, Inc., 739 F.3d 956, 957 (7th Cir.2013) (citations omitted). Although the court need not adopt the calculation method that is most prevalent in the marketplace as it existed at the time of the hypothetical negotiation, see Cook v. Niedert, 142 F.3d 1004, 1013 (7th Cir.1998), such an approach is more efficient for the court and more likely to yield an accurate approximation of the market rate.

Here, had an arm's length negotiation been feasible, the court believes that the class would have negotiated a fee arrangement based on a percentage of the recovery, consistent with the normal practice in consumer class actions. An ex ante agreement based on lodestar requires a client to monitor counsel, and the class-member "clients" here had little incentive to do so. There are approximately 17.5 million class members in this case, the prospective relief is minimal, and none of the class members suffered tangible damages beyond the inconvenience of receiving one or more debt-collection calls to their cell phones on ostensibly overdue credit card bills. The class would not have negotiated a compensation scheme that required a level of monitoring the class members were not interested in or capable of providing. Instead, the class would have chosen the compensation scheme that required the least monitoring to align the incentives of the class and its counsel—the percentage method. The court will therefore apply the percentage method as well.

The court does not, however, agree with Class Counsel's assertion that "it is appropriate—and the norm in the Seventh Circuit—to include administrative and notice costs when calculating fees based on a percentage-of-the-fund." (Dkt. No. 269 at 17.) The Seventh Circuit has instructed district courts that the "ratio that is relevant to assessing the reasonableness of the attorneys' fee that the parties agreed to is the ratio of (1) the fee to (2) the fee plus what the class members received." Redman v. RadioShack, 768 F.3d 622, 630 (7th Cir.2014). Administration and notice costs, although paid through the settlement fund, are not benefits to the class and thus not part of "what the class members received." Id. Class Counsel argue that the Seventh Circuit's recent instruction in Redman applies only in cases involving a fund that must be monetized from coupon redemptions. The settlement in this case, by contrast, is a non-reversionary cash fund. (Dkt. No. 269 at 18.) The court does not agree with Class Counsel's limited interpretation of Redman. In Pearson v. NBTY, Inc., 772 F.3d 778 (7th Cir.2014), decided two months after Redman, the Seventh Circuit applied its holding in Redman to a reversionary cash fund and clarified that costs incurred as part of the settlement do not shed light on the fairness of the split between Class Counsel and class members. Id. at 781 (citing Redman, 768 F.3d at 630). The Seventh Circuit also extended its analysis to cy pres and service awards because neither award directly benefits the class, or at least the whole class, and therefore should not be included in the court's assessment of the settlement's value to the class. Id. at 784.

In order to evaluate the fairness of Class Counsel's fee request consistent with the Seventh's Circuit's recent guidance, the court must recalculate the percentage fee sought by Class Counsel. After subtracting administration and notice costs ($5,093,000) and the Named Plaintiff service awards ($25,000), the total money available to split among the class and Class Counsel is $70,337,099. Class Counsel seeks 22,636,530 of that total, or slightly above 32%.

C. Class Counsel's Requested 32% Fee Exceeds the Market Rate

The Seventh Circuit has instructed district courts to approximate the market rate that would have prevailed at the outset of the litigation had negotiations between Class Counsel and "clients having a real stake" been feasible. Trans Union Corp. Privacy Litig., 629 F.3d at 744. The Seventh Circuit, however, has not expressed a preference for a particular method of determining that market fee. The market-mimicking approach is, as the Seventh Circuit acknowledged, "inherently conjectural." Id.; see also Synthroid I, 264 F.3d at 719 ("[I]t is indeed impossible to know ex post the outcome of a hypothetical bargain ex ante.") In Synthroid I, however, the Seventh Circuit explained that it is possible to learn about "similar bargains" and set forth three "guides" or "benchmarks" to help district courts estimate the market fee: (1) actual fee contracts between plaintiffs and their attorneys; (2) data from similar common fund cases where fees were privately negotiated; and (3) information from class-counsel auctions. Synthroid I, 264 F.3d at 719. At least two judges presiding in this district, and one judge from another district employing the Seventh Circuit's market-mimicking approach, have applied these three benchmarks to determine ex post the market contingent fee. See In re AT & T Mobility Wireless Data Services Sales Tax Litigation, 792 F.Supp.2d 1028, 1033-1034 (N.D.Ill.2011); In re Trans Union Corp. Privacy Litig., No. 00 C 4729, 2009 WL 4799954, at *10-13 (N.D.Ill. Dec. 9, 2009) (Gettleman, J.), rev'd on other grounds, 629 F.3d 741 (7th Cir.2011); In re Cabletron Sys., Inc. Sec. Litig., 239 F.R.D. 30, 40-45 (D.N.H.2006) (Smith, J.). This court will likewise analyze each benchmark to estimate the prevailing market rate for TCPA class action litigation generally, and then adjust that rate based on the risk of nonpayment in this case.

1. Class Counsel's Contingent Fee Agreements

The first benchmark is any actual agreement between plaintiffs and attorneys in this case. This was a useful starting point in Synthroid because one group of sophisticated plaintiffs had negotiated a fee agreement at the outset and set the opening price. Synthroid I, 264 F.3d at 720. That, however, is not the case here. Like most consumer class actions, the only fee agreements are Class Counsel's retainer agreements with the Named Plaintiffs, which provide for contingent fees ranging from 33.3% to 40% of the settlement fund. (Dkt. No. 176 at 22.) These retainer agreements are of little value to determining the market rate because named plaintiffs are less often sophisticated buyers of legal services and more often "the cat's paws of the class lawyers." In re Trans Union, 629 F.3d at 744. Moreover, the agreements here were between Class Counsel and five individual plaintiffs who, individually, did not have "a sufficient stake to drive a hard—or any—bargain with the lawyer[s]." Continental, 962 F.2d at 572. The court therefore finds that Class Counsel's contingent fee agreements with the Named Plaintiffs do not inform the court's estimation sufficiently as to what Class Counsel would have received in an ex ante negotiation with the entire class, has such a negotiation occurred.

2. Data on Fee Awards in Other Cases

The second and third Synthroid benchmarks concern data from similar common fund cases where the parties set fee schedules ex ante, either through a private negotiation or a judicially conducted "auction." Because data from pre-suit negotiations and auctions tend to be sparse—and with regard to TCPA class actions, nonexistent—district courts have also examined empirical data analyzing fee awards in other class actions where fees were awarded at the end of the case. See, e.g., AT & T Mobility, 792 F.Supp.2d at 1033; Trans Union, 2009 WL 4799954, at *11-13. The Seventh Circuit has relied on the same empirical data to determine the "norm" for fee awards, see Silverman, 739 F.3d at 958, and this court will do so as well. While large-scale empirical studies necessarily include ex post fee awards from other circuits that may not be reflective of the market price at the time those cases were filed, the awards likely affect the price at which national class action lawyers are willing to provide their services going forward. See, e.g., Trans Union, 2009 WL 4799954, at *11 (noting awards in class actions generally may influence expectations of a lawyer and client engaging in negotiation); Nilsen v. York County, 400 F.Supp.2d 266, 282-83 (D.Me.2005) ("Other courts' awards necessarily affect the expectations of lawyers and, therefore, what they might agree to in voluntary negotiation.")

i. Empirical Studies

In 2004, Theodore Eisenberg and Geoffrey Miller examined two data sets covering class actions from 1993 to 2002 and found that the mean fee award from settlements in the $38 to $79 million range was 16.9% and the median was 15.5%. Theodore Eisenberg & Geoffrey P. Miller, Attorney Fees in Class Action Settlements: An Empirical Study, 1 J. Empirical Legal Studies 27, 73 (2004). In 2010, Eisenberg and Miller updated their study in to analyze class action settlements from 1993 to 2008. Theodore Eisenberg & Geoffrey P. Miller, Attorney Fees in Class Action Settlements: 1993-2008, 7 J. Empirical Legal Studies 248 (2010). The updated study found that the mean award from settlements in the $38.3 to $69.6 million range, the third highest decile, was 20.5% and the median was 21.9%; the mean award from settlements in the $69.6 to $175.5 range, the second highest decile, was 19.4% and the median was 19.9%. Id. at Tab. 7.

In the same year as Eisenberg and Miller published their updated 2010 study, Brian Fitzpatrick, who filed a declaration in this case on behalf of Class Counsel, (Dkt. No. 270), published a paper analyzing every federal class action settlement in 2006 and 2007. Brian T. Fitzpatrick, An Empirical Study of Class Action Settlements and Their Fee Awards, 7 J. Empirical Legal Studies 811 (2010). Fitzpatrick found that the mean award from settlements in the $72.5 to $100 million range was 23.7% and the median was 24.3%. Id. at 839.

All three studies confirm Eisenberg and Miller's original finding of a scaling effect whereby the percentage fee decreases as the class recovery increases. See 1 J. Empirical Legal Stud. At 28 ("[A] scaling effect exists, with fees constituting a lower percent of the client's recovery as the client's recovery increases.") In Eisenberg and Miller's 2010 study, they found that settlements in the top decile by total recovery yielded a median and mean fee percentage that was less than one-third of the median and mean percentage fee in settlements in the lowest decile. 7 J. Empirical Legal Studies at 264. Fitzpatrick found a similar, although less pronounced scaling effect: settlements in the top decile by recovery yielded a mean fee of 18.4% and a median of 19%, whereas settlements in the lowest decile yielded a mean fee of 28.8% and a median of 29.6%. 7 J. Empirical Legal Studies at 839; see also Silverman, 739 F.3d at 958 (observing that the empirical data show that the percentage of the fund awarded to counsel declines as the size of the fund increases).

Accordingly, if past awards are reflective of the market for this case and its $75.5 million negotiated settlement fund, and if the published empirical data discussed above accurately reflect the fees awarded in TCPA class actions, it is fair to conclude that class members would have negotiated an across-the-board fee somewhere between 20% and 24% of the settlement fund. Class Counsel's requested 32% fee (or 30% fee, depending on the denominator) exceeds that across-the-board range.

ii. The Court's TCPA Class Action Settlement Analysis

To assist the court in determining whether the findings from the above empirical studies are indeed representative of the across-the-board percentage fees awarded in TCPA class actions, the court requested class counsel in another settled TCPA case, Wilkins v. HSBC Bank, No. 14 C 190 (N.D.Ill.), pending on this court's docket, to submit data from other finally approved TCPA class action settlements since 2010. Class counsel in HSBC, many of whom also represent class members in this case, diligently compiled publicly available summary information contained in 73 TCPA class action settlements approved since 2010. See HSBC, No. 14 C 190 (N.D.Ill.) (Dkt. No. 109-1). The court has now analyzed the data from 72 of the cases—one case lacked publicly available fee information—and has attempted in the table below to recreate the Eisenberg-Miller and Fitzpatrick summaries for TCPA class action settlements. Like the empirical analyses discussed in the previous section, the table below reports the mean and median fee percentage, as well as the standard deviation, for each total recovery decile of the TCPA class action settlements provided by class counsel in HSBC.

            Recovery         Mean      Median      Standard      Number
              Range         Fee (%)    Fee (%)     Deviation     of Cases

           Less than         31.2       31.5          3.1           8

           $345,000          33.1       33.3          2.8           7

           $510,000          37.1       33.0         18.1           7
           $1.1 Million

           $1.1 million      29.4       33.3          6.3           7
           $1.6 million

           $1.6 million      30.7       30.7          4.1           7
           $2.6 million

           $2.6 million      26.1       33.0         10.2           7
           $4.6 million

           $4.6 million      24.1       25.0          9.8           7
           $7.0 million

           $7.0 million      25.8       25.0          4.6           7
           $9.8 million

           $9.8 million      23.7       25.0          8.6           7
           $15.9 million

           $15.9 million     17.2       17.7          4.8           8
           $39.9 million

[Editor's Note: The preceding image contains the reference for footnote8] HSBC, No. 14 C 190 (N.D.Ill.) (Dkt. No. 109-1).

Because this court lacks the technical expertise of Eisenberg, Miller, or Fitzpatrick, and because the sample size of cases (72) is quite small, the statistics set forth above are but an informal analysis.9 The data similarly fail to provide a meaningful benchmark for a case like this one, where the $75.5 million recovery begins to approach what many courts consider a "megafund." Despite these shortcomings, the available TCPA data offer two important insights. First, the across-the-board percentage awards in TCPA class actions roughly track the fee awards in other types of cases, after controlling for class recovery amount. Second, TCPA class actions exhibit the same relationship between fee awards and recoveries as other types of cases: that is, the percentage of the fund awarded to counsel generally declines as the size of the fund increases.

iii. Competitive Fee Structures Negotiated Ex Ante

The analysis desired by Seventh Circuit authority is not at an end, however, because the second and third benchmarks from Synthroidex ante arrangements and judicially overseen "auctions"10—reveal that sophisticated parties engaged in an pre-filing fee negotiations rarely agree to a single, across-the-board percentage fee structure, and rarely pay a percentage of the recovery equal to the benchmark established by past awards. The court has not uncovered any data about ex ante fee arrangements or auctions in the consumer class action context, let alone data on TCPA class actions. Data from published opinions in securities and antitrust cases do exist where district courts utilized a competitive approach to negotiate a fee structure on behalf of the class at the outset. As far as the court can tell, there are at least fourteen class action cases—twelve securities actions and two antitrust actions—where district court judges have selected lead counsel and negotiated a fee structure using a competitive process. See In re Oracle Sec. Litig., 131 F.R.D. 688 (N.D.Cal.1990); In re Wells Fargo Sec. Litig., 157 F.R.D. 467 (N.D.Cal.1994); In re Amino Acid Lysine Antitrust Litig., 918 F.Supp. 1190 (N.D.Ill.1996) (Shadur, J.); In re California Micro Devices Sec. Litig., 168 F.R.D. 257 (N.D.Cal.1996); In re Cendant Corp. Litig., 182 F.R.D. 144 (D.N.J.1998); In re Network Assocs., Inc., Sec. Litig., 76 F.Supp.2d 1017 (N.D.Cal. 1999); Sherleigh Assocs. LLC v. Windmere-Durable Holdings, Inc., 184 F.R.D. 688 (S.D.Fla.1999); In re Lucent Techs. Inc. Sec. Litig., 194 F.R.D. 137 (D.N.J. 2000); In re Bank One Shareholders Class Actions, 96 F.Supp.2d 780 (N.D.Ill.2000) (Shadur, J.); Wenderhold v. Cylink Corp., 188 F.R.D. 577 (N.D.Cal.1999); In re Auction Houses Antitrust Litig., 197 F.R.D. 71 (S.D.N.Y.2000); In re Quintus Sec. Litig., 201 F.R.D. 475 (N.D.Cal.2001); In re Commtouch Software Ltd. Sec. Litig., No. 01C-00719, 2001 WL 34131835, Order Re Lead Plaintiff Selection and Class Counsel Selection (N.D.Cal. June 27, 2001); In re Comdisco Sec. Litig., 141 F.Supp.2d 951 (N.D.Ill.2001) (Shadur, J.) (Memorandum Opinion entering attached Apr. 6, 2001, Memorandum Order).

The data from these securities and antitrust cases, where available, do not shed light on the market rate in consumer class actions, but they do illustrate that (1) negotiated fee agreements regularly provide for a recovery that increases at a decreasing rate and (2) negotiated fee agreements frequently result in lower fee awards than those suggested by the empirical data on past awards granted after the fact. In 2006, United States District Judge William Smith conducted a survey of the fee structures adopted in some of the cases listed above. Cabletron, 239 F.R.D. at 43. The findings of Judge Smith's survey are reprinted in part below and supplemented by this court's independent research. Because the case before Judge Smith was a securities class action, Judge Smith applied the negotiated fee schedule of each surveyed case to the settlement amount in Cabletron. This court will not conduct a similar analysis because, as discussed above, the court cannot impute the market rate for attorneys' fees charged in securities and antitrust cases onto a consumer class action. It is sufficient to note that in nearly every case, the presiding judge selected a bid with a declining contingent-fee scale.11 See Cabletron, 239 F.R.D. at 44 ("[T]he competitive fee structures uniformly reflect a downward scaling as the settlement fund increases.").

Instead, the court compares the prevailing fee percentage (i.e., the "blended" rate) in each case to the mean and median set forth in Eisenberg and Miller's 2010 study for the corresponding recovery amount. Such a comparison should help the court determine whether and to what extent the empirical data—which largely reflect past fee awards determined ex post—overestimate the contingent fees parties agree to when they actually negotiate at the outset of a case.

The summary is as follows:

              Case Name                              Total        Actual Fee   Eisenberg & Miller
                                                    Recovery       Award12        Mean/Median
In re Oracle Sec. Litig., No. 90-CV-931
931 (N.D. Cal., Judge Vaughn Walker)              $25 million       22.5%      22.1%(mean)
                                                                               24.9% (median)
In re Wells Fargo Sec. Litig., No. 91-CV-1944
(N.D. Cal., Judge Vaughn                                                       23.8% (mean)
Walker)                                           $13.7 million      22%       25.0% (median)
In re California Micro Devices Sec.
Litig., No. 94-CV-2817 (N.D. Cal.,                                             22.1% (mean)
Judge Vaughn Walker)                              $26 million       15.7%      24.9% (median)
In re Amino Acid Lysine Antitrust
Litig., No. 95-CV-7679 (N.D. Ill.,                                             20.5% (mean)
Judge Milton Shadur)                              $49 million        7.0%      21.9% (median)
In re Cendant Corp. PRIDES Litig.,
No. 98-CV-2819 (D.N.J., Judge                                                  12.0% (mean)
William H. Walls)                                 $341 million       6.0%      10.2% (median)
In re Cendant Corp. Non-PRIDES
Litig., No. 98-CV-2819 (D.N.J., Judge                                          12.0% (mean)
William H. Walls)                                 $3.2 billion       8.7%      10.2% (median)
In re Auction Houses Antitrust Litig.,
No. 00-CV-648 (S.D.N.Y., Judge                                                 12.0% (mean)
Lewis Kaplan)                                     $512 million       5.2%      10.2% (median)
In re Bank One Shareholders Class                                              20.5% (mean)
Actions, No. 00-CV-880 (N.D. Ill.,                $45 million        7.0%      21.9% (median)
Judge Milton Shadur)
In re Network Associates, Inc., No. 99-CV-1729                                 22.1% (mean)
(N.D. Cal., Judge William                         $30 million        8.0%      24.9% (median)
In re Quintus Corp. Sec. Litig., No. 00-CV-4263                                22.8% (mean)
(N.D. Cal., Judge Vaughn                          $10 million       11.3%      22.1% (median)

[Editor's Note: The preceding image contains the reference for footnote12]

Cabletron, 239 F.R.D. at 44; Laural L. Hooper & Marie Leary, Auctioning the Role of Class Counsel in Class Action Cases: A Descriptive Study, Washington, D.C.: Federal Judicial Center, 2001 (supplementing Judge Smith's analysis).

As discussed earlier, the foregoing analysis is merely illustrative and does not purport to approximate the contingent fee for a TCPA class action, had that fee been negotiated at the outset of the case with a "client" having a real stake in the outcome. The analysis does, however, suggest that selecting competent counsel using a competitive process generates a lower percentage-of-the-fund fee arrangement than Eisenberg and Miller's mean and median percentages, which mostly reflect awards granted ex post. The spreads between the negotiated fees and Eisenberg and Miller's estimates vary from 0% to 17% and are especially pronounced for settlements that produced large recoveries. The particular spread depends on the unique facts and risk factors of each case, but the court's finding here is generally consistent with the experience of district court judges who have used a competitive-bid approach to select counsel in the past. See, e.g., In re Comdisco Sec. Litig., 150 F.Supp.2d 943, 947 n. 7 (N.D.Ill.2001) ("[T]his Court's prior experience as well as the bidding results in the present case confirm that the cited mythic norm [of 25 percent to 35 percent] is grossly excessive even where substantially smaller [than $100 million] amounts are at stake.")

The remaining question is whether the findings discussed above apply to a hypothetical ex ante negotiation in the consumer class action context, or merely to securities and antitrust cases. The court believes the findings from securities and antitrust cases provide some guidance regarding the ex ante negotiation in any type of class action. As Eisenberg and Miller concluded in 2004 and again in 2010, "the overwhelmingly important determinant of the fee is simply the size of the recovery obtained by the class," not the subject matter of the litigation. 7 J. Empirical Legal Studies at 250. All of the empirical studies surveying past awards found similar across-the-board percentage awards for securities, antitrust, and consumer class actions. Id. at 264 (Tab.5); 7 J. Empirical Legal Studies at 835 (Tab.8). Additionally, this court's informal analysis discussed earlier in this opinion, and which the court could not have conducted without the diligent assistance of counsel, confirmed that the same conclusion applies to the TCPA subset of consumer class actions. Furthermore, the court has no reason to believe that securities or antitrust cases are any more or less predictable than consumer class actions, such that counsel would be willing to negotiate a scaled fee schedule in one set of cases but not the other. As the Seventh Circuit explained in Silverman, a downward scaling fee arrangement is well-suited to securities litigation because a large portion of class counsel's expenses must be devoted to establishing liability, whereas damages can be calculated mechanically from movements in stock prices. Silverman, 739 F.3d at 959. That applies equally, if not more, to TCPA cases because nearly all of counsel's efforts are devoted to determining liability. Damages are fixed by statute.

3. Court's Estimation of the Market Rate for TCPA Class Actions Exclusive of Risk

The Seventh Circuit acknowledged in Synthroid I that it is, of course, "impossible to know ex post the outcome of a hypothetical bargain ex ante . . . . [b]ut a court can learn about similar bargains." Synthroid I, 264 F.3d at 719 (emphasis original). That is what the court has endeavored to do in the preceding sections and the court now draws the following conclusions. First, given the class's inability to effectively monitor counsel, an ex ante negotiation would have produced a fee arrangement based on a percentage of the recovery. Second, the data available on past awards in TCPA cases and other class actions show that the mean and median recovery for a $75.5 million TCPA case are between 20% and 24% of the settlement fund. Third, an ex ante negotiation between Class Counsel and class members in this case, had individual class members had a real stake in the litigation, would have produced a downward scaling fee arrangement. See Silverman, 739 F.3d at 959 ("The [empirical data] reinforce the observation in the Synthroid opinions that negotiated fee agreements regularly provide for a recovery that increases at a decreasing rate.") Fourth, given the $75.5 million recovery, the downward scaling fee arrangement would have produced an actual percentage award below or toward the bottom of Eisenberg and Miller's 20% to 24% range for similarly sized settlements.

The special master appointed by Judge Gettleman in Trans Union observed, correctly, that determining the criteria for the hypothetical negotiation is the easy part; attaching actual numbers to the hypothetical downward scaling fee agreement is "more art than science." Trans Union, 2009 WL 4799954, at *15. As a starting point, the court applies a slightly modified version of the fee schedule set out by the Seventh Circuit in Synthroid II because Synthroid II is the only consumer class action known to this court where the parties (or in this case the court) estimated a downward scaling fee agreement in a consumer class action.13 Synthroid II, 325 F.3d at 980. As demonstrated in the table below, applying the modified Synthroid II scale to the total recovery in this case yields a result that is largely consistent with the conclusions drawn from the court's analysis in Section II.C.2. The fee structure affords Class Counsel a relatively high rate for the initial recovery consistent with Class Counsel's need to devote most of their efforts to determining liability. The marginal rates diminish as the recovery increases because, notwithstanding the class's desire to incentivize counsel to seek a higher award, the measure of damages depends more on the number of class members (or phone calls) than the additional efforts of counsel. Finally, the modified Synthroid II structure produces an actual percentage fee of 19.97%, which is .03% below Eisenberg and Miller's 20% to 24% range for similarly sized settlements.

                       Application of Modified Synthroid II Structure
          Recovery           Fee Percentage                            Fee
First $10 million                 30%                               $3,000,000
Next $10 million                  25%                               $2,500,000
$20-45 million                    20%                               $5,000,000
Excess above $45 million          15%                               $4,568,265
Total Fee                       19.97%                              $15,068,265

In light of the Synthroid II structure's fit with this court's observations about the TCPA class-action market, and the fact that the Synthroid II structure resembles the fee schedules actually put forth by lawyers in an ex ante negotiations (although it is admittedly less tailored), the court adopts the Synthroid II structure as its estimation of the market contingent fee for a $75.5 million TCPA class action independent of the risks associated with a particular case.

4. Risk of the Litigation

The last factor the Seventh Circuit instructs a district court to consider is the risk plaintiffs' lawyers face of possibly losing the litigation when they undertake class representation. The estimated magnitude of the risk necessarily affects the price at which Class Counsel in this case would have been willing to offer their services in an ex ante negotiation, had such a negotiation occurred. See Synthroid I, 264 F.3d at 721. The Seventh Circuit has explained the risk premium in fee negotiations with the following hypothetical: "[I]f the market-determined fee for a sure winner were $1 million the market-determined fee for handling a similar suit with only a 50 percent change of a favorable outcome should be $2 million." Trans Union, 629 F.3d at 746 (citations omitted).

As this court discussed earlier in this opinion, Class Counsel in this case faced a variety of serious obstacles to success in bringing the lawsuit, and faced the real prospect of recovering nothing. First, it was quite possible that the discovery may have revealed that many class members acquiesced to receiving calls on their cell phones when they agreed to their cardholder agreements with Capital One. Some customers provided Capital One with their cell phone numbers as their primary contact numbers, arguably waiving any right not to receive debt-collection calls on their cell phone from Capital One. Second, at the outset of the litigation there was a serious question whether the Plaintiffs' claims could meet Rule 23's manageability requirement given that Capital One would have to review its records to determine which class members provided consent through cardholder agreements, which class members actually provided their cell phone numbers to Capital One, and whether each class member actually owned their cell phone number at the time Capital One called it using an autodialer. Third, as Capital One has noted throughout this litigation, there are presently petitions before the FCC urging the FCC to (1) revise the TCPA's definition of "automatic telephone dialing system" to exclude dialers like those used by Capital One, and (2) provide a safe harbor for all calls that Capital One inadvertently made to wrong numbers. Consequently, the longer this litigation were to continue, the longer Plaintiffs would be exposed to the possibility that the FCC would take action that might extinguish Plaintiffs' claims.

On the flip side, Capital One's potential monetary liability in this litigation is staggering. Even if each of the 7.5 million class members in this case had only received one phone call a piece and could not prove that any of the calls were made in willful violation of the TCPA, Capital One's exposure is still greater than $8.7 billion. That type of potentially bankruptcy-level exposure is sufficient to compel an in terrorem settlement before a liability determination is made and is accordingly a factor that reduces Class Counsel's risk of non-payment. See AT & T Mobility LLC v. Concepcion, 563 U.S. 333, 131 S.Ct. 1740, 1752, 179 L.Ed.2d 742 (2011) (noting class actions can produce in terrorem settlements).

The precise level of risk faced by Class Counsel more than two years ago, when the cases were filed, is difficult for a district court to determine and quantify after the fact. The Seventh Circuit, however, has criticized at least one district court for failing to make an attempt to do so. See Trans Union, 629 F.3d at 746-48. As hard as the task may be, this court will endeavor to determine an appropriate risk multiplier. After preliminary approval of the Class Settlement and after the court granted limited discovery of Class Counsel's lodestar information in this case and other TCPA cases, Professor Todd Henderson submitted a report on behalf of class member objector Jeffery Collins. Professor Henderson's report calculated that Class Counsel recovered on behalf of the various classes they had represented (and themselves) in about 43% of past TCPA cases.14 (Dkt. No. 294 ¶ 10.) Professor Henderson's determination was based on the limited sample of TCPA cases in which Class Counsel participated. It is, however, the best information available to court. As a result, the court assumes the average TCPA case carries a 43% chance of success, and the question the court ultimately must answer is whether Class Counsel in this case faced a greater or lesser chance of prevailing. Considering the circumstances of this case, the court believes that the class members' consent issues made the representation riskier than a typical TCPA class action, but only slightly so after considering the strong incentives to settlement created by the magnitude of Capitol One's potential liability.

The next question the court now faces is how to adjust the market fee structure the court has determined, as discussed earlier in this opinion, to account for the increased risk Class Counsel faced of losing. Eisenberg and Miller in their 2010 study concluded that "high risk" consumer class actions yield a percentage fee premium of about 6% above the "low or medium risk" cases. 7 J. Empirical Legal Studies at 265 (Tab.8). Absent better information and in light of the court's determination that this case was only slightly riskier than a typical TCPA class action, the court adopts Eisenberg and Miller's risk premium and applies it to the court's estimated market rate. Although Eisenberg and Miller's 6% premium applied to the entire fee award, such an application does not make sense in a case like this one, where the risk existed only with regard to liability, not damages. Each of the potential impediments to establishing Capital One's liability: the class members' alleged consent to be called; Rule 23 manageability issues; and potentially forthcoming FCC orders; only affected Class Counsel's ability to prove their case on liability and consequently their ability to recover any damages. Once the risk resulting from the impediments to establishing liability was overcome and Capital One's liability established, Class Counsel's ability to obtain a large recovery was no longer materially affected by that risk. As discussed above, one of the purposes of a downward scaling fee schedule is to account for cases where the marginal costs of increasing the class's damages recovery are low. Class counsel in an ex ante negotiation must nevertheless be provided an incentive to take the case at the outset and seek the highest award on behalf of the class that is reasonable under the facts and law of the case.

Because all of the risk factors in this case were limited to the question of Capital One's liability, it follows that the risk premium related to Class Counsel's fees should apply only to the attorneys' fees associated with the initial recovery tier negotiated between Class Counsel and the sophisticated class members before the case was filed. In the hypothetical ex ante negotiation, Class Counsel would have desired compensation for their enhanced risk regardless of the eventual level of recovery; the way to affect that desire is by incorporating the risk premium into the attorney fee percentage related to the first recovery tier. Sophisticated class members, by contrast, would have balked at agreeing to a similar adjustment to the second, third, and fourth recovery tiers, because the risk factors present in this case related only to establishing liability and would not have affected Class Counsel's ability to achieve the additional damages recovery reflected in second, third and fourth tiers. The court therefore applies the 6% premium only to the first $10 million in the first tier of the market fee structure. The court's risk-adjusted market contingent fee structure is set forth in the table below, and nets Class Counsel an additional $600,000.

                           Risk-Adjusted Fee Structure
          Recovery               Fee Percentage              Fee
First $10 million                    36%                  $3,600,000
Next $10 million                     25%                  $2,500,000
$20 to $45 million                   20%                  $5,000,000
Excess above $45 million             15%                  $4,568,265
Total Fee                           20.77%                $15,668,265

5. Professor Henderson's Model

Lastly, as discussed earlier, the court granted objector Jeffrey Collins's request for discovery of information from Class Counsel regarding Class Counsel's hours and hourly fees to calculate the lodestar in this case and in Class Counsel's previous TCPA class cases. Collins sent Class Counsel's information to Professor Henderson, who in turn filed a report analyzing the data and proposing an alternative method for approximating the ex ante market rate at the conclusion of a case. Collins's counsel acknowledged at the final approval hearing that no district court or court of appeals has ever adopted Professor Henderson's methodology. (Dkt. No. 324 at 54:17-19.) This court similarly declines to apply Professor Henderson's method of estimating the appropriate fee award in this case. Professor Henderson's model, though not applied, nevertheless merits a brief discussion.

Using Class Counsel's lodestar data, Professor Henderson determined that Class Counsel, who are highly experienced, achieve a recovery for their "clients" in approximately 43% of their TCPA cases. (Dkt. No. 294 ¶ 10.) But after adjusting for Class Counsel's tendency to devote substantially more professional time to their winning cases than to their losing cases, Henderson concluded that Class Counsel have a 64% chance of obtaining recovery for any given dollar of lodestar invested in TCPA litigation. Given a 64% chance of recovery, Henderson determined that Class Counsel need only obtain a weighted average 1.57 lodestar multiplier in successful cases to compensate Class Counsel for their lodestar investment and the contingent risk Class Counsel faces in TCPA class action litigation. (Id.) Applying that multiplier to this case, Professor Henderson concluded that Class Counsel would have represented the class in this case for 4.6% of the total recovery had they been forced to compete for the legal work at the outset of the case.15

Professor Henderson's model may possibly be a good predictor of the going rate in a competitive market of homogeneous plaintiffs lawyers. It does not, however, comport with the Seventh Circuit's guidance requiring the court to hypothetically approximate an ex ante fee negotiation. As a threshold matter, Professor Henderson's model relies exclusively on data relating to cases that were resolved after Class Counsel filed this case. That is not a criticism of Professor Henderson's methodology—he had no choice because he was limited to the data available through discovery. The limitation, however, does undermine the applicability of Professor Henderson's model to this case. Class Counsel did not know, at the outset of the litigation, that they needed only to achieve a 1.57 lodestar multiplier to compensate themselves for the contingent risk, and accordingly could not have relied on that multiplier to formulate their hypothetical ex ante bid for the legal work in this case. Professor Henderson's model also assumes "a hypothetical competitive market for plaintiffs' lawyers' services," without ever establishing that such a market exists. (Dkt. No. 294 ¶ 64.) The court's job is to approximate the market as it existed before the litigation, including the degree of competition. In doing so, the court cannot assume a perfectly competitive market without the benefit of reviewing additional evidence that is absent from this record. Indeed, the joint representation model present in this case and many of the comparable TCPA cases suggests that the market among plaintiffs class action lawyers, at least for large TCPA cases, may not be highly competitive. See also Joseph Ostoyich and William Lavery, Looks Like Price-Fixing Among Class Action Plaintiffs Firms, Law360 (Feb. 12, 2014 11:30 AM),

Ultimately, the court must follow the Seventh Circuit's guidance in approximating the fee that would have been negotiated ex ante in this TCPA case had such a negotiation occurred. Unfortunately for Professor Henderson, his model is not among the methods accepted by the Seventh Circuit. Using the benchmarks set forth in Synthroid I, as explained above, the court concludes that the tiered fee arrangement displayed above, which approximates the agreed-upon negotiated percentage of the attorneys' fees to be taken from a $75.5 million settlement had Class Counsel negotiated with capable, sophisticated class members having a real stake in the litigation, is about 20%. Although the court noted earlier that the fairness of a fee percentage is to be considered against the total value of the settlement to the class less administrative and notice costs, the benchmarks the court used to determine the market rate evaluated fees as a percentage of the total recovery. The court therefore grants Class Counsel $15,668,265 of fees which is equal to about 20.77%, or about one fifth, of the entire $75,455,099 settlement fund.

The court further grants Class Counsel's requested incentive awards for the Named Plaintiffs in the amount of $5,000 each. Incentive payments sufficient to induce Named Plaintiffs to participate in the lawsuit are appropriate in the Seventh Circuit and, given the circumstances in this case, were necessary. Continental, 962 F.2d at 571. Moreover, a $5,000 award is consistent with the awards granted by other courts in this district in similar litigation. See AT & T Mobility, 792 F.Supp.2d at 1041 (collecting cases).

The Settlement Agreement states that Defendants' contributions to the settlement fund are non-reversionary, (Settlement Agreement § 2.42), and that the settlement will "continue to be effective and enforceable by the Parties," in the event that the court declines Class Counsel's fee request or awards less than the amounts sought (id. § 5.03). But the Settlement Agreement is silent on the matter of who, precisely, should receive the additional funds available should the court reduce the requested fee award, as it has done here. For the avoidance of doubt, the court orders that the additional money available as a result of its reduction to Class Counsel's requested fee should go to the class members who made timely claims. After incorporating the court's reduced fee award, the money available to class as result of the settlement is $54,668,834, which results in a payment to each timely claimant of at least $39.66, and possibly more if some claimants fail to deposit their settlement checks within 210 days.


For the reasons set forth above, Plaintiffs' motion for final approval of the class action settlement [260] is granted. The settlement is fair, reasonable, and adequate. Class Counsel's motion for approval of attorneys' fees [175] is granted in part and denied in part. The court awards attorneys' fees and costs in the total amount of $15,668,265 (about 20.77% of the $75,455,099 settlement amount) and incentive awards of $5,000 to each of the five Named Plaintiffs.


1. Capitol One includes defendants Capital One Bank (USA), N.A., Capital One, N.A., Capital One Financial Corporation, Capital One Services, LLC, and Capital One Services II, LLC. The Participating Vendors include defendants Capital Management Services, LP ("CMS"), Leading Edge Recovery Solutions, LLC ("Leading Edge"), and AllianceOne Receivables Management, Inc. ("AllianceOne").
2. Plaintiffs never filed a proper motion for preliminary approval, although they filed two memoranda in support of such a motion. (Dkt. Nos. 121, 129.) They captioned the memoranda as "motions" in the docket text, but the actual headings of the filings reveal that neither is a motion, merely a memorandum. The court ignored Plaintiffs' oversight in light of the need for a standalone order (Dkt. No. 137) granting preliminary approval.
3. The parties estimate that approximately 5% of the settlement class is unknown to Capital One or Plaintiffs. (Dkt. No. 264 ¶ 11.)
4. The court calculated this figure using BrownGreer's number of contacted class members, 15,983,613, in conjunction with its assessment that 15,983,613 represents 91.22% of the total class. (See Dkt. No. 305 ¶ 6.)
5. The settlement fund is actually $75,455,098.74. For the sake of simplicity, the court has rounded the numbers to the closest dollar, as it has done with the other figures discussed in this opinion.
6. The court originally set the final approval hearing for December 9, 2014, but rescheduled the hearing for January 15, 2015 after granting Collins' request for additional discovery. Class Counsel informed all objectors who had previously stated a desire to appear of the date change and, out an abundance of caution, the court's clerk waited in the courtroom designated for the hearing on December 9 to record the appearance of any objector who mistakenly appeared on that date. No objector came to the designated courtroom on December 9, 2014.
7. The court includes invalid and untimely opt-out requests in the total because those requests, although invalid, signal disapproval of the settlement.
8. In cases where any unclaimed portion of the settlement reverted to defendants, the court considered the total recovery to be the amount made available to class members before any reversion.
9. The court has expended considerable time and effort placing the information submitted by HSBC counsel into usable a dataset for this informal analysis. To assist judges in future cases, and to provide a starting point for more adept statisticians, the court will make its underlying dataset available in a separate order on the docket.
10. The Seventh Circuit has repeatedly stated that litigants do not select their own lawyers through auctions because there is no standard of quality of legal services. Silverman, 739 F.3d at 958; In re Synthroid Marketing Litig., 325 F.3d 974, 979-80 (7th Cir.2003) (Synthroid II). To the extent that the term "auction" implies an iterative process where bidders compete exclusively on price, that is not the process described here. The auctions described in this section reflect cases where judges placed themselves in the "clients'" shoes and selected the "best bid" based on the quality of the legal work and the price offered. See Synthroid I, 264 F.3d at 720.
11. In In re Auction Houses Antitrust Litigation, 197 F.R.D. 71 (S.D.N.Y.2000), counsel agreed to the opposite approach, taking no fees for the first $405 million recovered and 25% of everything above $405 million. The government had already established liability and the lawyers (as well as the class and the court) believed that the first few hundred million would come easy. See Synthroid I, 264 F.3d at 721 (discussing fee structure selected in Auction Houses).
12. Unlike Judge Smith's analysis, and in recognition that Class Counsel in this case have not included a request for expenses on top of their overall fee request, the court includes expenses awarded to Class Counsel in calculating the fee award.
13. In Synthroid II, the Seventh Circuit set the third "recovery tier" of the consumer class fee schedule at $20-$46 million because it used the total recovery by third-party payers, $46 million, to benchmark the consumer class scale. Here, the court adopts $20-$45 million as the recovery range for the third tier of the estimated fee scale because fee scales negotiated ex ante, including those surveyed above, generally reflect uniform recovery ranges—in this case, multiples of five.
14. Professor Henderson further determined that after adjusting for the amount of effort Class Counsel invested in each case, about 64% of Class Counsel's total investments were in cases in which they recovered. (Id. ¶ 10.) Because the court is concerned with the riskiness of this case relative to other TCPA cases, however, it adopts Professor Henderson's 43% estimate, unadjusted for Class Counsel's investment savvy.
15. (Multiplier (1.57) × Lodestar ($2,213,769)) ÷ Recovery ($75,455,099) = 4.6%. Professor Henderson's model is more complicated than the court's basic description here. For purposes of this opinion, however, and because the court did not apply Professor Henderson's approach, the court's summary will suffice.


1000 Characters Remaining reserves the right to edit or remove comments but is under no obligation to do so, or to explain individual moderation decisions.

User Comments

Listed below are the cases that are cited in this Featured Case. Click the citation to see the full text of the cited case. Citations are also linked in the body of the Featured Case.

Cited Cases

  • No Cases Found

Listed below are those cases in which this Featured Case is cited. Click on the case name to see the full text of the citing case.

Citing Cases