1 2 3 4 5 IN THE UNITED STATES DISTRICT COURT 6 FOR THE NORTHERN DISTRICT OF CALIFORNIA 7 8 IN RE TRANSPACIFIC PASSENGER Case No. 3:07-cv-05634-CRB AIR TRANSPORTATION ANTITRUST 9 LITIGATION MDL No. 1913

10 ORDER GRANTING ATTORNEYS’ 11 This Document relates to: FEES AND EXPENSES ALL ACTIONS 12

13 Now pending is Plaintiffs’ Motion for Attorneys’ Fees and Reimbursement of 14 Expenses (“Fees Mot. III”) (dkt. 1307) filed in connection with its third and final 15 settlement. The Court issues this Order to explain its rulings on the amount of fees it shall 16 award and to request additional information from Plaintiffs’ counsel prior to the award of 17 expenses. 18 I. BACKGROUND 19 On May 29, 2019, the Court granted preliminary approval of a third and final 20 settlement between Plaintiffs and ANA, the last remaining defendant in this twelve-year 21 litigation. See Second Am. Order Granting Prelim. Approval (dkt. 1306). The Court 22 granted final approval of the settlement at the motion hearing on November 15, 2019. 23 The Court has already granted two rounds of attorneys’ fees and expenses in the 24 final settlement rounds preceding this third motion for fees. In connection with the first 25 settlement, Plaintiffs’ counsel sought $13,154,166, but the Court awarded $9,000,000 26 based on a net settlement fund of $31,181,800.27 (a 28.9% award). See Order Granting 27 Mot. for Final Approval and Fees (“Fees Order I”) (dkt. 1009) at 4. In connection with the 1 $11,038.071.51, based on a net settlement fund of $48,970,485.79 (a 22.5% award). See 2 Order Granting Mot. for Fees (“Fees Order II”) (dkt. 1252) at 6. Together, these approved 3 settlements constitute 25% of the then-net settlement fund of $80,152,286.06. Id. at 2. 4 In its third fees motion, Plaintiffs’ counsel seeks $18,647,081.15, representing 33% 5 of the net settlement fund of $56,506,306.52 that it achieved pursuant to a settlement with 6 defendant ANA; they also seek reimbursement of litigation expenses totaling $157,898.48. 7 See Fees Mot. III at 1–2. Their gross settlement with ANA, $58 million, is reduced by 8 $935,795 in notice expenses, $400,000 in claims administration expenses, $1,357,098.64 9 in unreimbursed litigation fund expenses, and $50,799.84 in unreimbursed fund expenses, 10 but is supplemented by a $1,250,000 litigation vendor settlement.1 Joint Decl. to Pls.’ Fees 11 Mot. III (“Joint Decl.”) (dkt. 1307) ¶ 83. If the Court granted the third fee request as 12 written, the total fee award (across all three rounds of settlement) would equal 13 $38,685,152.66, or 28.31% of a total net settlement fund of $136,658,592.58. Joint Decl. 14 ¶ 93. 15 II. LEGAL STANDARD 16 A court may award reasonable attorneys’ fees at the conclusion of a class action. 17 Fed. R. Civ. P. 23(h). A district court has discretion to choose either the percentage of 18 recovery method, where the prevailing attorneys are awarded a percentage of the common 19 fund, or the lodestar method, where fees are calculated by multiplying the hours the 20 attorneys reasonably expended on the litigation by the billing rate of the attorneys. See In 21 re Wash. Pub. Power Supply Sys. Sec. Litig.,

, 1295–96 (9th Cir. 1994). 22 Courts in the Ninth Circuit prefer to use the percentage-of-recovery method, but to cross- 23 check the final figure with a lodestar calculation. See Vizcaino v. Microsoft Corp., 290

, 1050–51 (9th Cir. 2002). 25 The Ninth Circuit uses a benchmark of 25% to calculate attorneys’ fees awarded 26 1 Following resolution of a dispute with one of Class Counsel’s litigation vendors, Settlement 27 Class Counsel received $1.25 million, which counsel stated it would subtract from any requested 1 under the percentage of recovery method. See Powers v. Eichen,

, 1256 2 (2000). In some cases, however, the 25% benchmark is “inappropriate.” See Vizcaino, 3 290 F.3d at 1048. Courts must explain why the award is appropriate, based on the facts of 4 the case. Id. The Ninth Circuit has identified five factors pertinent to evaluating the 5 reasonableness of a fee request, including (1) the results achieved, (2) the risks of 6 litigation, (3) the skill required and the quality of the work, (4) the contingent nature of the 7 fee and the financial burden shouldered by the plaintiffs, and (5) awards made in similar 8 cases. See id. at 1048–50. The most important of these factors is the resulting benefit 9 obtained for the class. See Order Granting Fees, In re Capacitors Antitrust Litig., No. 10 3:17-CV-03264-JD,

, at *3 (N.D. Cal. Sept. 21, 2018). 11 Under Rule 23(h), class counsel are also “entitled to reimbursement of reasonable 12 out-of-pocket expenses.” Fed. R. Civ. P. 23(h); see Harris v. Marhoefer,

, 18– 13 19 (9th Cir. 1994); Bergman v. Thelan LLP, No. 3:08-cv-05322-LB,

, at 14 *9 (N.D. Cal. Dec. 9, 2016). “To support an expense award, Plaintiffs should file an 15 itemized list of their expenses by category, listing the total amount advanced for each 16 category, allowing the Court to [assess] whether the expenses are reasonable.” Hayes v. 17 MagnaChip Semiconductor Corp., No.14-cv-01160-JST,

, at *9 (N.D. 18 Cal. Nov. 21, 2016). 19 III. DISCUSSION 20 The Court will grant modified attorneys’ fees as calculated below, and will grant 21 expenses pending the resolution of outstanding questions regarding individual firm 22 expenses and the litigation fund. 23 A. Modified Attorneys’ Fees 24 The Court notes its decision-making process for the prior rounds of attorneys’ fees 25 and discusses the Ninth Circuit’s five-factor test, supporting empirical data, and lodestar 26 crosscheck in explaining its award of modified fees in this third round. 27 In its first Fees Order, the Court emphasized that the case “involved two rounds of 1 from the Court), a grueling discovery process (involving 65 depositions and almost 7 2 million pages in documents), and summary judgment (requiring a 60-page omnibus 3 Opposition brief and resulting in an Order keeping the majority of claims in the case).” 4 See Fees Order I at 3–4. The Court agreed that this was a “heavily litigated, complicated 5 case” and acknowledged Plaintiffs’ cited study from 2008 showing that awards of 30% 6 were given in 11 of 16 antitrust cases with recoveries of less than $100 million.2 Id. at 4. 7 The Court noted that the settlement process “demanded . . . risky, challenging, and as-yet 8 uncompensated work.” Id. (internal citation omitted). 9 In its second Fees Order, the Court again considered the factors and returned to an 10 overall benchmark rate of 25%. Fees Order II at 6. While Plaintiffs’ challenging motion 11 practice, protracted and grueling discovery, and difficult defense against summary 12 judgment made this a complex case worth a near-30% fee in the first round, the second 13 round “required far fewer hours [of work] than the work that came before it.” Fees Order 14 II at 5. Following the first round, Plaintiffs’ work included defending the Court’s denial of 15 summary judgment in the Ninth Circuit following ANA’s interlocutory appeal, preparing 16 and defending briefs for final approval of the first round of settlements, and engaging in 17 settlement discussions and mediations with second-round-settling defendants. Id. This 18 totaled only 5,539.55 hours of work between rounds one and two of settlement, compared 19 to 98,364.36 hours of work preceding the first round of approval and fees. Id. at 13–14. 20 Also, prior to the first round of settlements, Plaintiffs contributed over $2 million to the 21 litigation fund,3 but had not needed to contribute additional funds since being granted $3 22 2 Relevant to this final round of attorney fees, the same study stated that 6 or 7 out of 9 antitrust 23 cases yielding recoveries between $100 million and $500 million awarded attorneys’ fees of 30% or more. See Robert H. Lamde & Joshua P. Davis, Benefits from Private Antitrust Enforcement: 24 An Analysis of Forty Cases,

, 911 tbl. 7B (2008). But importantly, because the study’s “cases were not randomly selected, it is difficult to generalize from [their] 25 conclusions.” Id. at 908. 3 Co-lead counsel established a Litigation Fund to fund the prosecution of the action. Plaintiffs 26 had (prior to the first settlement) contributed $2,252,790 in assessments to the Litigation Fund and used $1,877,660.12 from the Fund to pay for necessary litigation costs and expenses prior to the 27 first fees motion. See William Decl. to Fees Mot. I (dkt. 987) ¶¶ 79–80. Class Counsel incurred 1 million for “future expenses” following the first round. Id. Finally, the Court 2 acknowledged empirical data indicating that “courts are less willing to go above the 3 twenty-five percent benchmark when using the percentage of recovery method in larger 4 settlements.” Id. at 6. 5 In this third round of fees, the situation has changed again. Now, Plaintiffs’ counsel 6 have twice received compensation for their work, and have worked only about 5,132.25 7 hours since the Court granted final approval and fees in the second round of settlements.4 8 Since that time, Plaintiffs’ counsel engaged primarily in trial preparation, including 9 preparation of an opposition to ANA’s motion in limine regarding the admissibility of 10 ANA’s previous guilty plea.5 See Joint Decl. ¶ 61. Plaintiffs’ trial preparation involved 11 “producing expert reports, exchanging exhibit and witness lists with ANA, and the many 12 other tasks associated with trial preparation.” Id. ¶ 82. Plaintiffs’ counsel also participated 13 in extensive mediation with ANA to reach the current settlement. See id. The parties 14 settled approximately one month before trial. Id. ¶ 63. 15 1. Five-Factor Test 16 Analyzing counsel’s work with respect to the five factors in part requires deciding 17 how much of counsel’s work to consider at this stage. Despite the rigors of trial 18 preparation, the work Plaintiffs’ counsel performed in this last round favors a downward 19 adjustment in the award because it was not nearly as extensive as the work done in prior 20 rounds. Furthermore, Plaintiffs’ settlement in this round was not necessarily due to 21 proportionally greater effort on their part. See Alexander v. FedEx Ground Package Sys., 22 Inc., No. 05-CV-00038-EMC,

, at *1 (N.D. Cal. June 15, 2016) (“It is 23 not one hundred fifty times more difficult to prepare, try, and settle a $150 million case 24 than it is to try a $1 million case.”) (internal citations omitted). 25

26 4 The total work submitted (which included work done through rounds one and two) for all Plaintiffs’ counsel was 103,903.91 hours. Lebsock Decl. to Fees Motion II (dkt. 1228) ¶ 90. The 27 total work submitted in this round (including work done through rounds one, two, and three) was 1 It is still sensible, however, to consider the work that Plaintiffs’ counsel has done 2 prior to the second settlement and award of attorneys’ fees in order to see how it fits 3 among the factors. See Pearl Decl. (dkt. 1307) ¶ 34 (endorsing a lodestar figure that 4 covers the entire case, since “work done at an early stage was still significant to the 5 settlement that resulted with the final settling defendant[.]”). 6 All of the following are notable examples of counsel’s work with respect to ANA: 7 (1) Plaintiffs responded to ANA’s motion to dismiss6 (2) Plaintiffs successfully moved to 8 compel the deposition of ANA’s CEO Osamu Shinobe in 20147; (3) Plaintiffs fought 9 ANA’s individual motion, and others’ motions for summary judgment on the filed rate 10 doctrine, as well as later appeals8; (4) Plaintiffs opposed ANA and EVA’s Petition for Writ 11 of Certiorari to the Supreme Court9; (5) Plaintiffs fought another summary judgment 12 motion by ANA regarding the Satogaeri Class10; and (6) Plaintiffs engaged in a “major 13 responsive effort” to combat ANA’s challenges to Plaintiffs’ class certification for fuel 14 surcharges and Satogaeri fares.11 15 This work speaks to two factors: counsel’s risk and the contingent-fee nature of 16 counsel’s work. Some activities, such as compelling ANA’s CEO’s deposition and 17 opposing the motions for summary judgment, occurred prior to the first award of fees. 18 Others, such as opposing ANA’s Petition for Writ of Certiorari and fighting summary 19 judgment on the Satogaeri class, occurred between the first and second rounds of 20 settlement. Still others, such as responding to ANA’s challenges to class certification, 21 occurred following the second settlement and award of fees. These phases of litigation 22 demonstrate different levels of risk. The earlier activities—constituting the bulk of the 23 litigation— were done totally on a contingency basis and were therefore the riskiest. The 24 6 See Mot. To Dismiss by ANA, China Airlines, and Thai Airways (dkt. 304). 25 7 Joint Decl. ¶ 30 (citing dkt. 867). 8 Id. ¶¶ 44, 48. See ANA’s Mot. For Summ. J. (dkt. 724); Omnibus Opp. to Summ. J. Mot. (dkt. 26 869). 9 Id. ¶ 53. 27 10 Id. ¶ 55. See Order Denying Summ. J. (dkt. 1194). 1 later activities occurred after counsel received fees and $3,000,000 in a future litigation 2 fund, which seemed to have obviated all or most risk, to the extent that Plaintiffs would not 3 continue to spend more than was in the fund.12 Though Plaintiffs’ fees expert asserts that 4 all work to date was done on a contingency fee basis, see Pearl Decl. at 11, that is not the 5 practical effect of the litigation, given the fees and reimbursements counsel have already 6 received. Accordingly, the factors of risk and the contingent nature of the work (along 7 with counsel’s financial burden) militate in favor of a lower fee. 8 The balance tips the other way when considering two other factors: counsel’s skill 9 level and quality of work, and the results achieved (the most important factor). Each step 10 of the 12-year litigation evidences counsel’s skill and high quality of work.13 And because 11 it was not possible to know, in earlier rounds, that counsel’s work with respect to ANA 12 would result in the settlement ultimately achieved, it probably makes sense to consider the 13 strength of all work retroactively. It is also the case that counsel achieved excellent results 14 for the class—this last settlement, with one defendant, was the single largest settlement in 15 the entire litigation. 16 As to the final factor, considering awards made in similar cases, it is helpful to 17 consult the empirical data below, which suggests that a fee percentage lower than 18 Plaintiffs’ request is appropriate. It is nonetheless worth keeping in mind the complexity 19 and duration of this antitrust action when comparing fee percentages in other cases. See, 20 e.g., Order Granting Fees, In re Capacitors,

, at *4 (granting 21 $16,725,000 (25%) settlement in a “multi-year, international price-fixing cartel case 22 against 22 sprawling Defendant corporate families based almost entirely in Japan.”). 23 12 The Court reasoned, in granting the second round of fees and expenses, that despite a 24 $232,386.51 balance in Plaintiffs’ litigation fund at the time, Plaintiffs theoretically still had either $1,147,324.60 or $717,211.94 in the litigation fund (depending on the inclusion of a recent 25 $430,112.66 invoice, and rejecting outright Plaintiffs’ untimely February 27, 2015 Nathan Associates, Inc. invoice of $914,938.09). See Fees Order II at 3–4. Despite Plaintiffs’ hypothesis 26 at the time that they would require “at least $1,000,000” in litigation fund expenses between that time and trial, the risks of loss appear to have been slight. See id. at 3. 27 13 Plaintiffs fairly note that this was “an intrinsically difficult case due to the length and scope of 1 Though the reduced risk and difficulty of achieving later settlements supports a 2 downward adjustment in attorneys’ fees, Plaintiffs’ counsel performed work involving 3 ANA throughout the litigation, leading to the current settlement. This benefit to the class 4 is significant—despite previous awards for the same overall work, and despite the 5 relatively few hours worked since the second round of settlements. 6 2. Empirical Data 7 Nonetheless, empirical data indicates that larger settlement awards are generally 8 associated with smaller percentage fees. Empirical studies are an “important additional 9 data point in the determination of an appropriate award.” Rodman v. Safeway, Inc., No. 10 11-CV-03003-JST,

, at *5 (N.D. Cal. Aug. 23, 2018). One seminal 11 study illustrates a strong inverse relationship between the size of a settlement and fee 12 percentage awarded to counsel; for class action recoveries between 2009 and 2013 that 13 were greater than $67.5 million, the average fee percentage awarded decreased to 22.3%. 14 See id.; Theodore Eisenberg et. al., Attorneys’ Fees in Class Actions: 2009–2013, 92

, 948 (2017) (discussing data that may illustrate a “scaling effect” 16 occurring in the award of increasingly large settlements). 17 Another study highlighting fee awards for federal class action settlements from 18 2006 to 2007 illustrates that settlements between $100 and $250 million received a mean 19 fee percentage award of only 17.9% and a median percentage of only 16.9%, (based on a 20 sample of 14); this supports the study’s assertion that “fee percentages are strongly and 21 inversely related to the size of the settlement,” and that drop-offs in percentage awarded 22 become even steeper for settlements above $100 million. See Brian T. Fitzpatrick, An 23 Empirical Study of Class Action Settlements and their Fee Awards, 7 J. Empirical Legal 24 Stud., 811, 837–38, 839 tbl.11 (2010). 25 Judge Koh has noted “persuasive evidence that ‘the median attorney’s fee award in 26 a sample of 68 ‘megafund’ class action settlements14 over a 16-year period was 10.2%.” 27 1 See Alexander v. FedEx Ground Package Sys., Inc., No. 05-CV-00038-EMC,

3351017, at *2 (N.D. Cal. June 15, 2016) (quoting In re High-Tech Emple. Antitrust Litig., 3 No. 11-CV-02509-LHK,

, at *13 (N.D. Cal. Sept. 2, 2015) (awarding 4 10.5% of $435 million combined antitrust settlements by using lodestar multipliers of 2.2 5 and 1.5 and conducting a percentage cross-check)). Judge Koh used an earlier iteration of 6 the Eisenberg study, which covered class actions from 1993 to 2008. See High-Tech, 2015

, at *13; Theodore Eisenberg & Geoffrey P. Miller, Attorney Fees and 8 Expenses in Class Action Settlements: 1993–2008, 7 J. Empirical Legal Stud. 248, 265 9 tbl.7 (2010). 10 However, in the later Eisenberg study, antitrust cases between 2009 and 2013 11 reflected mean and median percentage fee awards of 27% and 30%, respectively. This 12 was, admittedly, in a sample of 19 cases where the mean recovery was roughly $500 13 million but the median recovery was roughly only $37 million—a settlement value very 14 different from the instant case. See Eisenberg, 2009-2013, supra, at 951–52. The 15 Fitzpatrick study, however, illustrates that federal antitrust class actions between 2006 and 16 2007 had mean and median awards of 25.4% and 25%, respectively (from a sample of 23 17 actions). See Fitzpatrick, supra, at 835 tbl.8. 18 The Court notes that while the data illustrates an inverse relationship between 19 settlement size and percentage fees, the Ninth Circuit “has expressly rejected any hard rule 20 that megafund cases are to be treated differently” based on this relationship. See In re 21 Cathode Ray Tube Antitrust Litig., MDL No. 1917,

, at *42 (N.D. Cal. 22 Jan. 28, 2016) (citing Vizcaino, 290 F.3d at 1047). 23 3. Lodestar Crosscheck 24 Beyond addressing the Ninth Circuit’s five factors and looking to empirical data, 25 the lodestar cross-check is the Court’s important last step. The lodestar is particularly 26 27 1 important as the net settlement fund is a megafund, and economies of scale at work here 2 make the cross-check more salient. See Alexander,

, at *2 (“[I]n 3 megafund cases, the lodestar crosscheck assumes particular importance.”). In applying the 4 lodestar cross-check, this Court first will determine whether to (1) consider only hours 5 worked between rounds two and three of settlement, and compare this to the requested 6 amount (as this Court has done previously), (2) compare all hours worked to all fees 7 awarded and requested, or (3) consider the lodestar ratio with respect to Plaintiffs’ 8 “unreimbursed lodestar.” The Court begins with Plaintiffs’ “unreimbursed lodestar” 9 method. 10 Plaintiffs’ approach calculated a lodestar amount by subtracting from their 11 cumulative lodestar ($45,152,522) the amounts already awarded by the Court 12 ($20,038,071.51) for an “unreimbursed” lodestar of $25,114,450.49. See Joint Decl. ¶ 92. 13 Accordingly, their requested fee, $18,647,081.15, yields a lodestar ratio of 0.74. See

14 Plaintiffs cite authority supporting this method. See Lobatz v. U.S. West. Cellular of Cal., 15

, 1149–50 (9th Cir. 2000) (noting that district court did not determine that its 16 first fee award grant was to compensate counsel for all hours worked up to the first 17 settlement; concluding subsequently that “[b]y later calculating the lodestar value for the 18 entire case and then subtracting the amount class counsel had previously been paid, the 19 district court ensured that the . . . [award after a second settlement] only included those 20 hours that class counsel had not been compensated for by the earlier attorney fee award.”). 21 Nonetheless, this method yields a result similar to simply calculating the cumulative 22 lodestar ratio, and hence does not greatly alter the analysis. The lodestar ratio would be 23 0.86 if the Court simply compared all of the awarded and requested fees to Plaintiffs’ 24 overall lodestar running from the inception of the case.15 The biggest difference would be 25 if the Court calculated a lodestar considering only the hours worked since the second 26 27 1 settlement. This lodestar ratio would be 5.84.16 2 The Court will consider the lodestar ratio with respect to the cumulative lodestar— 3 for simplicity and consistency, and in recognition of counsel’s work as a whole at this 4 stage. See Order Granting Fees, In re Capacitors,

, at *6 (“Because the 5 total work performed by counsel from inception of the case makes each settlement 6 possible, courts typically base fee awards in subsequent settlements on all work performed 7 in the case.”).17 “Indeed, when considering fee awards for subsequent settlements, courts 8 typically calculate the lodestar multiplier by dividing (1) all past and requested fee awards 9 by (2) all of counsel’s time from inception of the case.”

10 So long as the total hours worked by Plaintiffs’ counsel and hourly rates charged are 11 reasonable, then this lodestar percentage, a multiplier of less than 1, is reasonable as well.18 12 4. Attorneys’ Fees—Conclusion 13 Overall, in awarding fees at this stage, the Court balances the following: 14  the reduced financial burden and risk Plaintiffs faced in reaching the third and final 15 settlement (lowering the value); 16  the challenges and risk Plaintiffs faced prior to the first settlement, a period of time 17 that made up the bulk of the work and undoubtedly laid substantial groundwork for 18 a successful settlement later on (raising the value); 19  the risk undertaken in rigorously preparing for trial against the last holdout 20 defendant up to one month before the trial date (raising the value); 21 22 16 This is calculated with respect to the 5,132.25 hours of work done between rounds two and 23 three of settlement (as calculated previously), and a lodestar amount of $3,190,707 (the difference between Plaintiffs’ current total, $45,152,522.00, and their total pursuant to round two of fees, 24 $41,961,815). See Fees Mot. II (dkt. 1227) at 1. 17 The Court in its second Fees Order stated that it considered “most of [Plaintiffs’ work done] in 25 awarding Plaintiffs thirty percent of the net Settlement Fund in the first round of settlements; nonetheless, “the court, in its judgment looks at the overall settlement in determining the 26 appropriate award.” See Fees Order II at 5–6. 18 See Joint Decl. Exs. 3–5. Hourly rates for attorneys at Cotchett, Pitre, and McCarthy, LLP 27 ranged from $250 to $950. Joint Decl. Ex. 3. Hourly rates at Hausfeld, LLP ranged from $290 to 1  the high overall settlement (including the settlement with ANA, the largest of the 2 three), indicating an excellent benefit for the Plaintiff classes and confirming the 3 skill and quality of counsel’s work (raising the value); 4  statistics illustrating the inverse relationship between size of settlement and 5 percentage of recovery, especially in larger settlements (lowering the value); 6  the difficulty of antitrust cases and particularly of this case (raising the value); and 7  the reasonableness of the lodestar crosscheck value (raising the value, since the 8 Court considered the cumulative lodestar figure). 9 As noted, there seemed minimal risk in the last settlement round, given Plaintiffs’ 10 prior successes, cooperation by defendants,19 and reimbursement.20 Nonetheless, the work 11 as a whole reflects a high-risk, high-reward approach contributing to the final settlement. 12 See Pearl Decl. ¶ 22(d) (“Contingent cases that must be tried or prepared for trial are 13 always far riskier than cases that settle earlier in the process.”).21 14 In light of these factors, the Court grants a reduced fee of $14,126,576.64, which 15 equals 25% of the round three net settlement fund and yields total fees to Plaintiffs of 16 $34,164,648.15—this represents 25% of the net award across all three settlements. 22 The 17 reduced award results in an overall (cumulative) lodestar ratio of 0.76. This amount 18

19 19 For instance, the eight settling parties between 2010 and 2014 executed settlement agreements with Plaintiffs providing for cooperation and payments. See Joint Decl. ¶¶ 64–71. The 20 settlements were premised upon “each settling Defendant’s agreement to provide cooperation to the Class[.]” See id. at ¶ 72. Later settlements were similar. See id. at ¶¶ 77–80. 21 20 Just how much Plaintiffs had to lose depends on the current unreimbursed litigation fund expenses. If the current unreimbursed litigation fund expenses include a twice-rejected Nathan 22 Associates, Inc., invoice, then in fact Plaintiffs spent only $442,160.55 in litigation fund expenses since the last round. See Joint Decl. Ex. 6. If so, then the expenses are covered by the $1.25 23 million settlement that Plaintiffs received in a vendor dispute. See Joint Decl. ¶ 97. 21 Declarant Pearl framed the issue of risk by considering the litigation as a whole, instead of 24 considering risk attendant to the last settlement round—this exercise is not as helpful in deciding how much risk Plaintiffs’ counsel actually faced. See Pearl Decl. ¶ 22(a) (“If this Action had not 25 been successful—if interim settlements had not been obtained—and Plaintiffs had been required to try the Action against all the Defendants and lost, Class Counsel would have lost far more than 26 their total $45,152,522.00 lodestar . . .”). 22 The total net settlement funds equal $136,658,592.58, and 25% of this value is $34,164,648.15. 27 The Court has awarded $20,038,071.51 in fees in prior rounds. See Fees Mot. III at vii–2. The 1 reflects the Vizcaino factors and the Ninth Circuit’s benchmark, strikes a balance among 2 competing empirical data, and reflects the value of Plaintiffs’ work across 12 years of 3 complex and challenging antitrust litigation. 4 B. Expenses 5 Total expenses incurred by the litigation fund throughout the case were 6 $6,341,702.95, of which the Court has reimbursed $4,984,604.31 to date.23 Plaintiffs’ 7 counsel seeks $1,357,098.64 in unreimbursed litigation fund expenses. See Fees Mot. III 8 at 15. Plaintiffs also seek individual firm expenses: $7,537.12 in out-of-pocket expenses 9 incurred by Cotchett, Pitre, & McCarthy, LLP, between May 17, 2018 and July 31, 2019, 10 and $43,262.72 in out-of-pocket expenses incurred by Hausfeld LLP in the same time. See 11 Joint Decl. ¶ 95. The total unreimbursed litigation fund and individual firm expenses total 12 $1,407.898.48. Id. at ¶ 96. These are offset by $1.25 million that a litigation vendor 13 provided to Plaintiffs following settlement of a dispute. Id. at ¶ 97. 14 Overall, Plaintiffs seek $157,898.48 in unreimbursed expenses. See Joint Decl. 15 ¶¶ 94–7. Plaintiffs properly filed categorized and itemized lists of litigation fund expenses 16 and out-of-pocket individual firm expenses. See Joint Decl. Exs. 6-7. However, it is 17 unclear whether, in their litigation fund expenses, Plaintiffs include a February 27, 2015 18 invoice from Nathan Associates, Inc., which the Court has twice rejected, since the 19 summary of expenses is cumulative from March 28, 2008 to July 31, 2019. 20 Regarding individual firms’ out-of-pocket costs covering May 17, 2018 to July 31, 21 2019 for Cotchett and Hausfeld: the expense categories are reasonable, but travel 22 23 The Court awarded $1,877,660.12 in litigation fund expenses pursuant to the first round of 23 settlements, $3,000,000 in future litigation fund expenses, also pursuant to the first round of settlements, and $106,944.19 in litigation fund expenses pursuant to the second round of 24 settlements (the $106,944.19 was for an Epiq invoice that had been added to the first motion for fees via supplemental declaration, which the Court earlier rejected; the Court again rejected, 25 however, a February 27, 2015 invoice of $914,938.09 from Nathan Associates). See Fees Order II at 3. Now, Plaintiffs’ counsel have excluded, in the calculation of the requested litigation fund 26 expense amount, $930,039.61 and $38,426.02 that the Court awarded as individual firm expenses pursuant to the first and second round of settlement, respectively. See Joint Decl. Ex. 6 n.2. 27 Accordingly, their current request divides up the requested expenses into $1,357,098.64 for 1 comprises the vast majority of expenses incurred for each firm and may be excessive as to 5 Hausfeld. See Joint Decl. Exs. 7-8 (Cotchett’s travel expenses totaled $2,760.87 (airfare 3 and ground travel) and $2,269.04 (meals and lodging) while Hausfeld’s cost $11,746.00 (airfare and ground travel) and $17,309.31 (meals and lodging)). The Hausfeld travel ‘ expenses might indeed be reasonable, but the expense lists do not take a sufficiently ° detailed approach for the Court to know. The Court requires further itemization of the 6 Hausfeld travel expenses to ensure that these expenses are not excessive. IV. CONCLUSION For the foregoing reasons, the Court GRANTS attorneys’ fees in the reduced ° amount of $14,126,576.635 (25% of the round three net settlement fund), and GRANTS the requested expenses, pending provision of the following information: (1) what the noted litigation vendor dispute and settlement was for, and its effect on the litigation fund (2) whether Plaintiffs’ list of litigation fund expenses (in Exhibit 6 to the Plaintiffs’ Joint 3 Declaration) includes a previously-rejected February 27, 2015 invoice from Nathan 3 Associates, Inc., and (3) a detailed itemization of travel expenses from Hausfeld. 2 16 IT IS SO ORDERED. = Dated: November 26, 2019 ; Y CHARLES R. BREYER UNITED STATES DISTRICT JUDGE

19 20 21 22 23 24 25 26 27 28