Summary
explaining the Ninth Circuit's frequent "observation that 'in many instances the increase in recovery is merely a factor of the size of the class and has no direct relationship to the efforts of counsel.'" (quoting In re Bluetooth, 654 F.3d at 943)
Summary of this case from In re Yahoo! Inc. Customer Data Sec. Breach Litig.Opinion
Nos. 17-15065 17-17439 Nos. 17-15067 17-17436 No. 17-15143
05-15-2020
IN RE OPTICAL DISK DRIVE PRODUCTS ANTITRUST LITIGATION, Indirect Purchaser Class, Plaintiff-Appellee, v. Conner Erwin, Objector-Appellant, v. Panasonic Corporation; Panasonic Corporation of North America; NEC Corporation; Sony Corporation; Sony Optiarc, Inc. ; Sony Optiarc America, Inc.; Sony NEC Optiarc, Inc. ; Hitachi Ltd.; Hitachi-LG Data Storage, Inc.; Hitachi-LG Data Storage Korea, Inc., Defendants-Appellees. In re Optical Disk Drive Products Antitrust Litigation, Indirect Purchaser Class, Plaintiff-Appellee, v. Christopher Andrews, Objector-Appellant, v. Panasonic Corporation; Panasonic Corporation of North America; NEC Corporation; Sony Corporation; Sony Optiarc, Inc. ; Sony Optiarc America, Inc.; Sony NEC Optiarc, Inc. ; Hitachi Ltd.; Hitachi-LG Data Storage, Inc.; Hitachi-LG Data Storage Korea, Inc., Defendants-Appellees. In re Optical Disk Drive Products Antitrust Litigation, Indirect Purchaser Class, Plaintiff-Appellee, v. Barbara Cochran, Objector-Appellant, v. Panasonic Corporation; Panasonic Corporation of North America; NEC Corporation; Sony Corporation; Sony Optiarc, Inc. ; Sony Optiarc America, Inc.; Sony NEC Optiarc, Inc. ; Hitachi Ltd.; Hitachi-LG Data Storage, Inc.; Hitachi-LG Data Storage Korea, Inc., Defendants-Appellees.
Robert Clore (argued) and Christopher A. Bandas, Bandas Law Firm P.C., Corpus Christi, Texas, for Objector-Appellant Connor Erwin. George W. Cochran, Law Office of George W. Cochran, Streetsboro, Ohio, for Objector-Appellant Barbara Cochran. Christopher Andrews (argued), Livonia, Michigan, pro se Objector-Appellant. Kevin Kamuf Green (argued), Hagens Berman Sobol Shapiro LLP, San Diego, California; Steve W. Berman, Hagens Berman Sobol Shapiro LLP, Seattle, Washington; Jeff D. Friedman and Shana E. Scarlett, Hagens Berman Sobol Shapiro LLP, Berkeley, California; for Plaintiff-Appellee Indirect Purchaser Class. No appearance by Defendants-Appellees.
Robert Clore (argued) and Christopher A. Bandas, Bandas Law Firm P.C., Corpus Christi, Texas, for Objector-Appellant Connor Erwin.
George W. Cochran, Law Office of George W. Cochran, Streetsboro, Ohio, for Objector-Appellant Barbara Cochran.
Christopher Andrews (argued), Livonia, Michigan, pro se Objector-Appellant.
Kevin Kamuf Green (argued), Hagens Berman Sobol Shapiro LLP, San Diego, California; Steve W. Berman, Hagens Berman Sobol Shapiro LLP, Seattle, Washington; Jeff D. Friedman and Shana E. Scarlett, Hagens Berman Sobol Shapiro LLP, Berkeley, California; for Plaintiff-Appellee Indirect Purchaser Class.
No appearance by Defendants-Appellees.
Before: Jerome Farris, Carlos T. Bea, and Morgan Christen, Circuit Judges.
CHRISTEN, Circuit Judge:
Objectors appeal the district court’s approval of two rounds of settlements in this consumer class action litigation, as well as the district court’s award of fees and litigation expenses to class counsel. Objectors also argue that the district court erred by allowing class counsel’s fee bid to remain under seal when it considered the motions to approve the settlements and awards of fees and expenses. We have jurisdiction pursuant to 28 U.S.C. § 1291. In a separately filed memorandum disposition, we affirm the district court’s approval of the first- and second-round settlements. Here, we vacate the awards of fees and litigation expenses and remand for a more complete explanation of the district court’s reasoning.
I.
In 2009, the Department of Justice disclosed that it was conducting a criminal investigation into possible antitrust violations based on price-fixing within the optical disk drive (ODD) industry. As a result of the federal criminal investigation, Hitachi-LG Data Storage (Hitachi) and several of its senior officials pleaded guilty to criminal antitrust violations. The DOJ investigation resulted in significant fines and at least four prison terms.
Optical disk drives are hardware products that can read and write data. They were once commonly included in computers and video game consoles.
Several putative class claims were filed alleging civil antitrust violations arising from the same activity that was the focus of the DOJ investigation, and in April 2010, the panel on multidistrict litigation consolidated the claims. The consolidated case was assigned to Judge Vaughn Walker in the Northern District of California. Two groups of plaintiffs sought class certification—direct purchaser plaintiffs (DPPs) and indirect purchaser plaintiffs (IPPs). The former prospective class comprised individuals and entities who purchased ODDs directly, such as personal computer manufacturers; the latter included plaintiffs who purchased products containing ODDs. The IPPs alleged that eleven groups of associated corporate defendants had conspired to restrain competition for ODDs. Objectors here are members of the IPP class.
As firms jockeyed to represent the putative class of IPPs in the consolidated action, the district court ordered prospective class counsel "to provide information on any subject pertinent to the appointment and to propose terms for attorney fees and costs in representing a prospective class." The court reasoned that it was "appropriate to consider the matter of fees and costs at the outset." Three groups of attorneys, including Hagens Berman, applied for appointment as interim class counsel and the district court granted their motions to seal their respective bids. The court explained that "[b]ecause their applications contain attorney work product, all three groups requested that the precise terms of their applications remain confidential during the pendency of this litigation."
For a discussion of the use of competitive bidding to select lead counsel in class actions, see 3 William B. Rubenstein, Newberg on Class Actions § 10:14 (5th ed. 2012). It appears this type of competitive bidding process is not widely used by trial courts.
Hagens Berman proposed a sliding-scale award for both fees and costs based on the percentage-of-recovery method, specifically citing the district court’s earlier decisions endorsing this approach. See, e.g. , Wenderhold v. Cylink Corp. , 188 F.R.D. 577, 588 (N.D. Cal. 1999) (" Wenderhold I "); Wenderhold v. Cylink Corp ., 189 F.R.D. 570, 572 (N.D. Cal. 1999) (" Wenderhold II "). The firm’s sealed bid included a grid with a sliding-scale fee calculated as a percentage of the class recovery. One axis of the grid comprised four stages of litigation. The grid called for lower fees for earlier-stage recovery scenarios, and higher fees if the case proceeded to trial and/or appeal. The grid’s other axis depicted the size of the class recovery and showed a lower fee percentage as the overall class recovery increased. The grid’s maximum fee award was fourteen percent of the class recovery from each defendant.
In June of 2010, the district court appointed Hagens Berman interim class counsel pursuant to Federal Rule of Civil Procedure 23(g)(3). The appointment order explained that the competing firms’ applications focused primarily on "their respective fee and cost proposals and an analysis of the prospects for a recovery." Without revealing the actual fee percentages in Hagens Berman’s grid, the court’s appointment order concluded that the fee proposals favored Hagens Berman, and explained that the firm proposed declining fees for larger recoveries, and its "proposal entail[ed] one fee that covers both compensation of attorneys [fees] and reimbursement of attorneys’ out-of-pocket expenses." The court’s order specifically observed that Hagens Berman had proposed "folding expense recovery into the fee."
The IPPs’ putative class action was reassigned to Judge Seeborg in October 2010, and the parties began extensive discovery and litigation practice. Four years later, the district court denied the motions for class certification filed by the IPP and DPP putative classes. After our court denied interlocutory review, DPPs obtained certification of a settlement class and entered into settlement agreements resolving their claims. Hagens Berman continued litigating on behalf of the putative IPP class, and eventually prevailed on its renewed motion for class certification in February 2016. Our court denied defendants’ request for interlocutory review in June 2016.
During the period immediately before and after class certification, the IPP class reached tentative settlement agreements with four groups of affiliated corporate entities (Hitachi, NEC, Panasonic, and Sony) that yielded a common fund totaling $124.5 million. In its first motion for fees, Hagens Berman requested twenty-five percent of the total first-round recovery as attorneys’ fees, or $31.125 million, and separately requested $3,704,323.97 for litigation expenses. The motion for fees and litigation expenses was filed six years after the district court’s appointment order. The motion did not mention that the firm’s sealed bid had proposed a different structure for fees and expenses. The motion stated that the request for a twenty-five percent fee equated to a 1.29 lodestar multiplier.
We refer to affiliated defendants by their primary names. For the first-round settlements, Hitachi contributed $73 million, NEC contributed $6.5 million, Panasonic contributed $16.5 million, and Sony contributed $28.5 million.
Class members objected that the requested fee was excessive because the common fund constituted a "megafund," and objector Conner Erwin asked the district court to unseal class counsel’s fee bid so the class could learn what fee arrangement the firm proposed when it sought appointment. In response to the objections, class counsel stated that it would have been awarded twelve to thirteen percent of the first-round settlement fund pursuant to what it called its "fee proposal structure." Class counsel also argued that Erwin and his attorney were professional objectors with suspect motives.
The district court held a fairness hearing in December 2016 to consider the first-round settlements. With minor modifications, the court adopted class counsel’s proposed order granting final approval of the $124.5 million settlement, as well as the requested fees and separate litigation expenses. The district court did not require that counsel unseal its original bid, and it overruled all objections in an order that stated the "original fee structure does not apply." The order suggested the firm’s proposal had not contemplated that it would have to litigate a second motion for class certification and multiple appeals to the Ninth Circuit. One objector appealed the approval of the first settlement: Christopher Andrews. Three objectors appealed the order awarding fees and expenses for the first settlement: Erwin, Andrews, and Barbara Cochran.
While the appeal of the first-round settlements was pending, the underlying ODD litigation continued, and in 2017, the class entered into a second round of tentative settlement agreements with three additional groups of affiliated defendants (PLDS, Pioneer, and Teac). The second-round settlements resulted in an additional $55.5 million common fund. Class counsel moved for a fee award of twenty-one percent of the common fund, or $11.655 million. The motion stated this was equivalent to a lodestar multiplier of 1.53 for the cumulative fee award (a $42,905,000 fee for the 76,772 attorney and non-attorney hours that generated the cumulative, $180 million, common fund). The motion also requested $1,368,718.95 for litigation expenses incurred after the first-round settlements. As before, the firm’s fee motion did not state that its request diverged from its original fee and expense proposal. Several objections were filed, including one by Erwin, who complained that the class still did not have access to class counsel’s original bid. Erwin also objected that the second lodestar cross-check should be limited to the hours invested since the last fee award, or the additional $3.7 million that class counsel said it invested after its first motion for fees. Erwin measured the requested fee against the $55.5 million common fund and calculated the second lodestar multiplier at 3.08. In response to the objection that twenty-one percent was too high, the firm represented that "[u]nder the interim lead counsel application, the fee guide was 12 percent."
For the second-round settlements, PLDS contributed $40 million, Pioneer contributed $10.5 million, and Teac contributed $5 million.
A second fairness hearing was held in September 2017, and the district court largely adopted class counsel’s proposed order approving its second motion for fees and expenses. The district court again denied Erwin’s request to unseal class counsel’s fee bid, along with all other objections. The court relied on class counsel’s lodestar cross-check, which used the cumulative lodestar (all time invested) measured against the total recovery ($180 million), but the court acknowledged that if the second settlement were viewed in isolation, the multiplier would be in line with Erwin’s calculation.
Erwin timely appealed the district court’s order awarding fees and expenses for the second-round settlements. Andrews objected to the second-round settlement agreements, but did not assert specific objections to Hagens Berman’s motion for fees and expenses arising from those settlements. Cochran did not object to the second-round fee award.
Class counsel filed a third motion for fees in 2018 after the IPP class reached settlement agreements with two additional defendant groups (Samsung and Toshiba). At the third fairness hearing, the district court stated that it had been unable to find Hagens Berman’s bid in the court files and directed the firm to submit a copy of the bid for in camera review. The order approving Hagens Berman’s third fee request was issued two weeks later, and it acknowledged that the previous awards had been "substantially higher than ... the terms of [the firm’s] proposal." The order partially granted Erwin’s request to unseal class counsel’s bid, requiring counsel to file the single page of the bid showing the sliding-scale fee grid. The third order observed that Hagens Berman’s original bid "offered to accept representation on the terms that no separate expense award would be made on top of any percentage-based fee award," and denied class counsel’s request for expenses "in light of the total fee and expense recovery to date, and given that Hagens Berman obtained the right to represent IPPs in part as a result of the payment structure it proposed." The order approved the requested twenty percent fee, an additional $5 million.
Six months later, Hagens Berman filed on the public docket the single page of its bid showing the fee grid, but other portions of its fee application remain under seal. The objectors who appealed the firm’s fee and expense awards arising from the first- and second-round settlements did not have access to the sliding-scale fee grid when they briefed the issues on appeal, but they did have the grid by the time oral argument was held before our court.
Class counsel represented to us in a related appeal that the firm’s bid was filed within twenty-four hours after the third fairness hearing. It may be that, within twenty-four hours, the firm submitted a copy of the bid in response to the district court’s request for in camera review, but the single page of the bid that the court ordered unsealed did not appear on the public docket until six months after the third fairness hearing.
In a separately filed memorandum disposition, we affirm the district court’s approval of the first- and second-round settlements. This opinion addresses the appeals filed by Erwin, Andrews, and Cochran arising from the district court’s awards of fees and litigation expenses for these settlements. Objections to the third-round fee award are also addressed in a separate memorandum disposition.
All arguments raised by Cochran that we do not address here are deemed waived because they were either inadequately briefed or raised for the first time on appeal. See Yamada v. Nobel Biocare Holding AG , 825 F.3d 536, 543 (9th Cir. 2016).
II.
"[W]e review for abuse of discretion the district court’s award of attorney’s fees and costs to class counsel as well as its method of calculating the fees." In re Hyundai & Kia Fuel Econ. Litig. , 926 F.3d 539, 556 (9th Cir. 2019) (en banc). The factual findings underlying these decisions are reviewed for clear error. Id. "In order for this [c]ourt to conduct a meaningful review of the fee award’s reasonableness, ... the district court must ‘provide a concise but clear explanation of its reasons for the fee award.’ " Stanger v. China Elec. Motor, Inc. , 812 F.3d 734, 739 (9th Cir. 2016) (per curiam) (quoting Hensley v. Eckerhart , 461 U.S. 424, 437, 103 S.Ct. 1933, 76 L.Ed.2d 40 (1983) ).
III.
Courts must ensure that attorneys’ fees awarded pursuant to Federal Rule of Civil Procedure 23(h) are reasonable. In re Online DVD-Rental Antitrust Litig. , 779 F.3d 934, 949 (9th Cir. 2015). This duty exists independent of any objection from a member of the class. Zucker v. Occidental Petroleum Corp. , 192 F.3d 1323, 1328–29 (9th Cir. 1999). District courts have discretion to choose which method they use to calculate fees, but their discretion must be exercised to reach a reasonable result. In re Bluetooth Headset Prods. Liab. Litig. , 654 F.3d 935, 942 (9th Cir. 2011). We have approved fee awards in class litigation using either the lodestar method or the percentage-of-recovery method. Hyundai , 926 F.3d at 570. And we have encouraged courts using the percentage-of-recovery method to perform a cross-check by applying the lodestar method to confirm that the percentage-of-recovery amount is reasonable. Online DVD-Rental , 779 F.3d at 949.
Because the relationship between class counsel and class members turns adversarial at the fee-setting stage, district courts assume a fiduciary role that requires close scrutiny of class counsel’s requests for fees and expenses from the common fund. Vizcaino v. Microsoft Corp. , 290 F.3d 1043, 1052 (9th Cir. 2002) ; see also In re Mercury Interactive Corp. Sec. Litig ., 618 F.3d 988, 994 (9th Cir. 2010) ("As a fiduciary for the class, the district court must ‘act with a jealous regard to the rights of those who are interested in the fund in determining what a proper fee award is.’ " (internal quotation marks omitted) (quoting In re Wash. Pub. Power Supply Sys. Sec. Litig. , 19 F.3d 1291, 1302 (9th Cir. 1994) (" WPPSS "))); 4 William B. Rubenstein, Newberg on Class Actions § 13:40 (5th ed. 2012).
In Vizcaino , we identified several factors courts may consider when assessing requests for attorneys’ fees calculated pursuant to the percentage-of-recovery method: (1) the extent to which class counsel achieved exceptional results for the class; (2) whether the case was risky for class counsel; (3) whether counsel’s performance generated benefits beyond the cash settlement fund; (4) the market rate for the particular field of law; (5) the burdens class counsel experienced while litigating the case; (6) and whether the case was handled on a contingency basis. 290 F.3d at 1048–50 ; see also Online DVD-Rental , 779 F.3d at 954–55.
Vizcaino did not establish an exhaustive list of factors for assessing fee requests calculated using the percentage-of-recovery method, but district courts have frequently referred to the factors it identified when considering fee awards for class counsel. See Online DVD-Rental , 779 F.3d at 955. Ultimately, district courts must ensure their fee awards are "supported by findings that take into account all of the circumstances of the case." Vizcaino , 290 F.3d at 1048 ; see also id. at 1050 (affirming fee award where "the district court considered the relevant circumstances"). Vizcaino recognized that in some class action litigation, "lawyers compete for lead counsel status and may even bid in a court-supervised auction," but because counsel in Vizcaino had not submitted that type of fee proposal, our court did not have occasion to consider how or whether such a bid should factor into the analysis when requests for fees and expenses are considered. Id. at 1049. Nor has our court had such an occasion since Vizcaino .
Counsel and the named plaintiffs in Vizcaino had entered into retainer agreements "promising to pay class counsel 30% of any recovery." 290 F.3d at 1049. The court noted that "[w]here evidence exists ... about the percentage fee to which some plaintiffs agreed ex ante , that evidence may be probative of the fee award’s reasonableness." Id . at 1050. The circumstances here are different because counsel made representations about its proposed fee structure directly to the district court in a competitive bidding process. Its fee bid remained under seal at the time objectors had an opportunity to weigh in on the motion for fees.
IV.
Class counsel’s motion for fees and expenses was submitted over six years after Hagens Berman was appointed interim class counsel, and nearly 2,000 district court docket entries later, but it included just one reference to counsel’s bid: "Had this case been successfully certified originally according to typical proceedings projected at the outset of the case, [Hagens Berman]’s lodestar and total attorneys’ fees for resolution at that stage would have been less." A twenty-page attorney declaration accompanied the fee motion, but the closest it came to mentioning the fee bid was the acknowledgment that "Judge Walker appointed Hagens Berman as sole lead counsel on behalf of the indirect purchaser class in a contested leadership fight."
The motion for attorneys’ fees and litigation expenses arising from the first-round settlements requested twenty-five percent of the common fund, approximately double the amount called for by the firm’s fee bid. Perhaps mindful that we have said a twenty-five percent benchmark may be "of little assistance" in megafund cases, Vizcaino , 290 F.3d at 1048, the fee motion acknowledged that the request was "not per se valid," Online DVD-Rental , 779 F.3d at 955. But the motion argued that each of Vizcaino ’s enumerated factors supported the requested award.
Erwin, Andrews, and Cochran objected to the first-round request for fees and expenses on various grounds, and Erwin sought disclosure of the sealed bid. Erwin argued that the class had no way of knowing whether a fee less than twenty-five percent had been proposed at the competitive bidding stage. In response, class counsel disclosed that the firm would have been awarded "12 to 13 percent" of the recovery pursuant to its proposed fee structure, and also identified the four stages of litigation on one axis of its fee grid: "(1) From Pleading Through Decision on Motion to Dismiss; (2) After Motion to Dismiss Through Adjudication of Class Certification; (3) After Adjudication of Summary Judgment; and (4) Through Trial Verdict and Final Appellate Determination." Counsel also argued that its original fee structure did not apply, invoking Judge Walker’s statement at the outset of the case that there were many "imponderables" that could impact the litigation. The firm argued that its bid had not contemplated the possibility that its first motion for class certification might be denied, that it devoted an additional two years to successfully litigate a second class certification motion, and that the period following the denial of the first certification motion was a period of heightened risk. Class counsel argued that it invested $5.6 million in hourly fees during that period, and it drew a favorable contrast with the outcomes achieved by counsel for DPPs, who settled after the district court declined to certify the proposed DPP class.
Counsel’s reference to "imponderables" quotes Judge Walker’s appointment order, which used "imponderables" to refer to the fact that the ultimate size of class recovery was unknowable. It was in this context that the district court’s appointment order explained that, in holding a competitive bidding process for prospective counsel, the court "understood fully that counsel, at this early stage of litigation, have limited information concerning the probability of a recovery and the amount or range of such recovery," and that unlike other types of cases, "potential recovery ... in this litigation is subject to a greater variety of imponderables." The order observed that Hagens Berman’s bid presented "a very impressive array of possible recovery scenarios that suggest a high level of analysis of potential recoveries."
The firm opposed objectors’ request that the bid be unsealed, arguing that disclosure of the bid was unnecessary because the court had a copy of the bid and the class had the court’s original appointment order. The district court overruled all objections to the first-round settlements, approved the request for fees and litigation expenses, and denied objectors’ request to unseal the bid. This order did not discuss the variance between the actual amount the court awarded and the amount class counsel would have received pursuant to its proposal.
In fact, it appears the district court did not have a copy of the firm’s bid. The bids submitted by the firms competing for appointment were filed under seal, and although there were docket entries showing they were filed, the bids were not available on the public docket. It appears they were stored separately.
The proposed order class counsel lodged approving its second-round fee and expense request included the statement that the district court "did not request the original proposed fee guidelines." The district court struck that line, but the record does not show whether the district court requested a copy of the bid at that time. As explained infra , the district court stated during the third fairness hearing that the sealed bid was unavailable in the court file, and asked class counsel to file a copy.
A similar process followed the request for fees and expenses arising from the second-round settlements. The motion sought twenty-one percent of the $55.5 million settlement fund and an additional $1,368,718.95 in litigation expenses. It discussed why the enumerated Vizcaino factors supported counsel’s request, but it did not address the bid or the variance from its proposed fee award, nor identify the amount that would have been called for had the grid been followed. Erwin again objected that Hagens Berman’s bid should be unsealed, and class counsel again responded that unsealing the bid would benefit the remaining defendants without providing useful information to the class. The firm’s response to the objectors represented that it would have been entitled to a twelve percent award pursuant to its bid.
The district court’s order approving the second fee motion adopted the firm’s analysis of the Vizcaino factors and counsel’s arguments in favor of keeping the bid sealed. Regarding the firm’s bid, the court’s written order stated that "[t]he guidelines submitted at the outset of the case ... have remained relevant" because the court awarded less than counsel’s twenty-five percent benchmark, "even though absent those guidelines, an argument could be made that the circumstances here might warrant an upward deviation from the benchmark." The order concluded that a lodestar cross-check "confirms that an upward variance from the original proposed fee guidelines is amply justified."
With both motions for fees, counsel submitted summaries showing the total hours worked and hourly rate for each timekeeper to support its suggested lodestar cross-check. The firm offered to provide the district court with time records, and one of the objectors argued that billing records should be filed, but the district court determined that this was unnecessary. Thus, the court’s references to lodestar cross-checks refer to class counsel’s summary calculations.
V.
We begin by considering objectors’ arguments that the district court abused its discretion by approving Hagens Berman’s fee requests. On appeal, objectors first argue the district court erred by not taking into account the size of the common fund, which they refer to as a "megafund," when it determined the appropriateness of the fee request. We have not identified a bright-line definition for "megafund," but the first-round settlements here yielded a $124.5 million common fund, and there is no question that a common fund of this size qualifies. See Vizcaino , 290 F.3d at 1047 ; 5 William B. Rubenstein, Newberg on Class Actions § 15:81 (5th ed. 2012) (noting that "[m]ost courts define mega-funds as those in excess of $100 million"). It is also clear that where a megafund recovery is achieved, "fund size is one relevant circumstance to which courts must refer" in determining the fee award. Vizcaino , 290 F.3d at 1047. We have cautioned that the twenty-five percent benchmark referred to in other cases—and upon which Hagens Berman anchored its fee requests—"is of little assistance" in megafund cases. Id. at 1047–48 (quoting WPPSS , 19 F.3d at 1297 ).
Objectors argue that a megafund settlement warrants a lower fee percentage because "it isn’t ten times as hard to try a $100 million case as it a $10 million case." This argument is consistent with our observation that "in many instances the increase in recovery is merely a factor of the size of the class and has no direct relationship to the efforts of counsel." Bluetooth , 654 F.3d at 943 (quoting In re Prudential Ins. Co. Am. Sales Practice Litig. Agent Actions , 148 F.3d 283, 339 (3d Cir. 1998) ). Other circuits have made the same general observation. See, e.g. , Wal-Mart Stores, Inc. v. Visa U.S.A., Inc. , 396 F.3d 96, 122 (2d Cir. 2005) (reasoning that because "economies of scale could cause windfalls in common fund cases, courts have traditionally awarded fees for common fund cases in the lower range of what is reasonable"). Consistent with this case law, counsel’s bid for appointment recognized and employed the same sliding-scale fee grid that Judge Walker had approved in earlier cases. The bid also quoted Judge Walker’s observation in Wenderhold II that "increasing amounts of recovery do not require correspondingly increased levels of attorney effort." 189 F.R.D. at 572.
We agree with objectors’ contention that the district court was required to consider the size of the first- and second-round settlements, but the record does not support the objectors’ assertion that the district court overlooked this factor. During the first fairness hearing, the court recognized that the percentage-of-recovery awarded as fees typically declines as the size of the common fund increases. And when the second-round settlements generated another $55.5 million common fund, the district court correctly recognized that the overall recovery had grown to $180 million. We are persuaded by our review of the record that the district court considered the size of the two settlements when it assessed fees and litigation expenses for the first- and second-round settlements.
Next, Erwin and Cochran urge this court to adopt a new rule requiring a sliding-scale fee award. Specifically, they advocate a rule requiring that percentage-based fee awards must decline as the size of class recovery increases to account for economies of scale in megafund settlements. But we have already declined to adopt a bright-line rule requiring the use of sliding-scale fee awards for class counsel in megafund cases, and we are bound by circuit precedent. See Vizcaino , 290 F.3d at 1047.
Objectors separately argue that class counsel’s fee grid was a critical factor in securing appointment as interim class counsel, and they argue the court erred by failing to consider it adequately. We agree that the proposed fee grid was a relevant circumstance the district court was required to consider, see id. at 1048–1050 ; Stanger , 812 F.3d at 739, but the objectors do not persuasively argue that the district court was bound by it. Objectors’ stronger argument is that the district court erred by failing to explain adequately the variance between counsel’s fee grid and the actual awards. The variance was significant. Based on class counsel’s approximation of the percentage that would have been called for had the grid been applied, Erwin argues that the first fee award was double what it should have been, and approximately $21 million too high. It is difficult to tell what role the fee bid played in the district court’s approval of the motions for fees and expenses. The record makes clear that Hagens Berman’s proposed fee structure, while not the lowest-cost proposal, was an important factor in Judge Walker’s decision to appoint the firm as interim class counsel. But despite the bid’s importance in the appointment process, the bid was likely not available to the court during the first and second fairness hearings. The district court’s order awarding $31.125 million in fees from the $124.5 million in first-round settlements stated that class counsel’s bid "does not apply," but the court’s second order awarding $11,655,000 in fees from the $55.5 million in second-round settlements included the statement that the firm’s bid "remained relevant." Finally, after the court requested a copy of the fee bid at the third fairness hearing and had an opportunity to review it, the court declined to award any expenses for the third-round settlements because class counsel had "obtained the right to represent IPPs in part as a result of the payment structure it proposed."
In the district court, part of class counsel’s rationale in support of its fee award was its suggestion that it encountered unanticipated litigation challenges during the course of the litigation. Counsel cited the initial denial of its first class certification motion and the need to litigate a discovery dispute in an interlocutory appeal to our court. The latter was undertaken jointly with DPPs to affirm the district court’s order allowing plaintiffs to subpoena records from DOJ’s criminal investigation. The firm cited the need for multiple interim appeals, but only the discovery dispute concerning the DOJ investigation was litigated on an interlocutory basis—we denied interlocutory review of the district court’s orders regarding class certification. See In re Optical Disk Drive Antitrust Litig ., 801 F.3d 1072 (9th Cir. 2015). It appears that both of these circumstances were contemplated by the stages of litigation incorporated into the fee grid, which called for a somewhat increased fee as the case advanced through the various stages of litigation. If other unexpected difficulties accounted for the district court’s fee awards, they are not evident in the orders from which these appeals were taken.
TSST-Korea, a joint venture between Toshiba and Samsung, and an anonymous individual were included in telephonic conversations recorded during DOJ’s criminal investigation. They sought to quash the subpoena of these records.
Given this record and the fiduciary duty the court owes to the class at the fee award stage, we remand the first- and second-round fee awards for further consideration and findings. Both parties now agree that the bid is a factor that must be considered and that the bid may now be unsealed. We agree with Hagens Berman that the district court was not bound by the original bid, but counsel’s continued reliance on the twenty-five percent benchmark—the firm referred to the benchmark as "granite" in its briefing on the third-round fee award—misses the mark. We have expressly stated that this benchmark is of little assistance in megafund cases. See Vizcaino , 290 F.3d at 1047–48 ; WPPSS , 19 F.3d at 1297.
The memorandum disposition in Case Number 19-15538 directs the district court to unseal the bid on remand.
We now hold that when class counsel secures appointment as interim lead counsel by proposing a fee structure in a competitive bidding process, that bid becomes the starting point for determining a reasonable fee. The district court may adjust fees upward or downward depending on circumstances not contemplated at the time of the bid, but the court must provide an adequate explanation for any variance. See Stanger , 812 F.3d at 739. We do not endeavor to create a comprehensive list of circumstances that may warrant departure from a fee bid, but we note that a district court would likely not abuse its discretion by departing from a bid based on circumstances the bid did not contemplate. Conversely, departure from a bid based on circumstances that were known at the time the bid was filed may be an abuse of discretion given the court’s fiduciary duty to members of the class. Here, class counsel argues that an upward departure from its bid was warranted in part because it did not anticipate the need to litigate a second class certification motion or interlocutory appeals. Without more, these factors are insufficient to justify a variance of the magnitude approved in the first- and second-round fee awards. The bid to become interim class counsel clearly contemplated that Hagens Berman would move to certify the plaintiff class and it is not unusual for interim class counsel to have to take more than one run at class certification. Finally, the proposed fee structure in this case explicitly contemplated appellate litigation.
VI.
We briefly address other arguments raised by objectors because the district court is likely to encounter them on remand.
First, the district court’s orders awarding fees adopted class counsel’s contention that the DOJ criminal investigation increased the litigation risk for the IPP class. Cochran and Andrews contend that it was error to weigh the DOJ investigation as a factor that increased the risk of this litigation, because absent the discovery generated by the criminal investigation, pursuing this consumer class action would have required significantly more effort and risk. We recognize the non-indicted defendants attempted to argue that they were not civilly liable because they were not subjects of the criminal investigation, but Hagens Berman did not offer a plausible explanation in support of its contention that this circumstance increased the litigation risk. That said, the court cited other factors that amply supported its determination that this litigation was very risky. Indeed, the last two non-settling defendants ultimately prevailed over the class at the summary judgment stage. See In re Optical Disk Drive Prods. Antitrust Litig. , 785 F. App'x 406 (9th Cir. 2019) (concluding that the class did not advance sufficient evidence of "pass-through," and therefore failed to create a genuine dispute of material fact as to injury, causation, and damages). Overall, the district court did not err in assessing the Vizcaino risk factor.
Second, class counsel’s descriptions of the sliding-scale fee grid were not entirely consistent. Objectors noted some of these inconsistencies, but they were not well positioned to assess whether the discrepancies were significant because the grid remained under seal. Some of the variables may seem minor, but they can make a significant difference when applied to a megafund case. Equally important, given that the bulk of the fee proposal remains under seal, these inconsistencies make it difficult for class members to have confidence that a fair outcome is reached. For example, in its response to objections raised to the first fee motion, class counsel stated that its fee grid would have yielded an award based on twelve to thirteen percent of the recovery. The firm seems to have reached that estimate by treating each of the four settlements comprising the first-round recovery as separate awards, rather than inputting a lump-sum $124.5 million recovery. The text above the grid on the single unsealed page of the firm’s proposal permits this approach, but the firm applied the grid differently for the second fairness hearing. There, it treated the second recovery as a $55.5 million lump sum rather than plugging in the individual components that made up the second settlement. Separately, the firm asserted in a related appeal before our court, No. 19-15538, that the award for the first-round settlements "would have been in the range of 12% to 14% ... had the grid been dispositive," rather than twelve to thirteen percent, as it argued at the first fairness hearing. These discrepancies may be addressed on remand.
Third, Erwin contends the district court erred by not considering the role of future settlements when it approved Hagens Berman’s first fee award. In particular, Erwin argues that the first award should have been adjusted downward because additional defendants remained in the case and there were likely to be additional recoveries. We disagree. No future settlements were guaranteed when the first settlement was approved, and the district court was mindful that any future settlement should take into account the firm’s cumulative award. The court correctly reasoned that it could address any potential double-counting if and when future settlements materialized.
VII.
We next turn to the district court’s award of litigation-related expenses. The district court awarded $5,073,042.92 in litigation expenses for the first- and second-round settlements. By the time this appeal was argued, the district court had conducted a third fairness hearing, requested that a copy of the bid be filed, and decided not to award expenses for the third settlement. The court explained that the firm "offered to accept representation on the terms that no separate expense award would be made on top of any percentage-based fee award." Before our panel, class counsel argued that the original bid’s reference to "costs" did not encompass all litigation expenses. Class counsel more specifically pressed this position when it briefed the district court’s third-round fee award. There, counsel argued that its original bid proposed absorbing only the taxable costs, not the much larger category of nontaxable costs, or litigation expenses. Hagens Berman does not identify any language in its original fee bid that supports this interpretation, and the record squarely refutes it.
See 5 William B. Rubenstein, Newberg on Class Actions § 16:5 (5th ed. 2012) (observing that the terms "nontaxable costs" and "nontaxable expenses," as used in Rule 23(h)(1) and Rule 54(d)(2), are interchangeable).
--------
Hagens Berman’s sealed bid proposed a single award for attorneys’ fees and costs without specifying taxable or nontaxable costs. Notably, the proposal cited Judge Walker’s orders in Wenderhold I , 188 F.R.D. 577, and Wenderhold II , 189 F.R.D. 570. In Wenderhold I , Judge Walker presided over seven consolidated securities fraud class actions and issued an order announcing that the court would select lead counsel by sealed-bid auction. 188 F.R.D. at 578, 587. The order specified that firms’ bids should describe their experience and qualifications, as well as provide evidence that the applicant had evaluated the securities case at bar, including the range and probability of recovery. Id. at 587–88. The district court’s order required that the firms’ bids identify the percentage of recovery the firm would charge in the event of a recovery "as fees and costs for all work performed in connection with the case set forth on the Fee Schedule Grid[ ] affixed as Appendix A" to the order. Id. at 588. One axis of the grid that appeared on the court’s Appendix A exactly mirrors the four stages of litigation reflected in the fee grid Hagens Berman submitted in this case; the other axis identifies potential common fund recoveries, from $500,000 up to any amount over $20,000,000. Id. The sample grid in Judge Walker’s Appendix A left the fee percentages blank. Id.
In Wenderhold II , Judge Walker issued a second order explaining that just one bid had been received in response to the court’s request for proposals, and the court rejected it. 189 F.R.D. at 571. The defect that rendered the bid unacceptable was that it called for separate awards of fees and expenses. Id. at 573. In particular, the court explained that its first order solicited bids setting forth a percentage to be charged as fees and costs for all work performed in connection with the case. Id . at 572. Contrary to the court’s invitation for bids, the rejected proposal stated that the bidding firm would seek separate reimbursement for expenses out of any fund created as a result of the litigation. Id . at 572–73. The court declined to allow fees and costs to be "divorce[d]," because it reasoned that such an arrangement could encourage counsel to categorize as costs "anything that could conceivably be so considered." Id . at 573. The district court viewed computer database and internet connection charges for research and investigation to be, in varying degrees, substitutes for attorney or paralegal library work, document review, and witness interviews. Id . The court sought to incentivize the least costly mix of inputs, and concluded "[c]ompensation which covers attorney effort and all other expenses affords that incentive, while a percentage award that omits non-attorney expenses does not." Id .
We express no opinion on the appointment order’s characterization of reasonable litigation expenses, but recite the details of Wenderhold I and II to explain that the district court correctly interpreted Hagens Berman’s bid. The proposal clearly offered to absorb litigation expenses. The orders entered in Wenderhold I and II , and the bid’s citations to those orders, is important context supporting Judge Walker’s statement that the firm’s bid proposed "one fee that covers both compensation of attorneys [fees] and reimbursement of attorneys’ out-of-pocket expenses." Judge Seeborg interpreted counsel’s bid the same way when he described the firm’s bid as an offer "to accept representation on the terms that no separate expense award would be made on top of any percentage-based fee award."
Though we agree with this part of the court’s interpretation of the bid, the record is unclear about whether the district court had the opportunity to take this into account when it reviewed the expense requests associated with the first two settlement rounds. Objectors did not have the information necessary to squarely raise this issue before the district court, but the fiduciary duty we owe to the class requires that this case be remanded so the class may be assured the issue is considered.
VIII.
District courts enjoy broad discretion to determine reasonable fee awards, but the size of the variance between the bid and the awards in this case requires more explanation. More detailed findings are particularly important because the bid remained under seal, putting objectors at a disadvantage in trying to assess the reasonableness of the fees. Separately, we reject class counsel’s assertion that it did not offer to absorb litigation expenses. Therefore, we vacate the district court’s award of fees and litigation expenses arising from the first- and second-round settlements, and remand to the district court for further proceedings consistent with this opinion.