Update: An Empirical Analysis of Federal Consumer Fraud Class Action Settlements (2019-2020)

© 2021 Jones Day. All rights reserved. Printed in the U.S.A. Jones Day publications should not be construed as legal advice on any specific facts or circumstances. The contents are intended for general information purposes only and may not be quoted or referred to in any other publication or proceeding without the prior written consent of the Firm, to be given or withheld at our discretion. To request reprint permission for any of our publications, please use our “Contact Us” form, which can be found on our website at www.jonesday.com. The mailing of this publication is not intended to create, and receipt of it does not constitute, an attorney-client relationship. The views set forth herein are the personal views of the authors and do not necessarily reflect those of the Firm. WHITE PAPER Update: An Empirical Analysis of Federal Consumer Fraud Class Action Settlements (2019–2020) Class action settlements in consumer fraud cases have generated significant controversy. Critics opine that these settlements primarily benefit lawyers, and that class members have often suffered little or no injury to begin with. These criticisms frequently turn to calls for legal reform. Our Jones Day White Paper published in April 2020, “An Empirical Analysis of Federal Consumer Fraud Class Action Settlements (2010–2018),” analyzed data showing that lawyers—not class members—frequently are the ones primarily benefitting from monetary settlement awards.   This White Paper updates our 2020 study with data drawn from 31 cases in which federal courts approved consumer class action settlements in 2019 and 2020. We analyzed data regarding class member participation rates and the allocation of monetary benefits among class members, class counsel, and other recipients—all in light of amendments made to Federal Rule of Civil Procedure 23. Those amendments went into effect in December 2018, after the settlements in our previous study were finalized. Based on the number of settlements approved in the two years since, there is sufficient data to meaningfully consider the 2018 amendments’ effects on consumer fraud class action settlements. The new data show that: (i) typically only a small fraction of class members receive any monetary benefits from the settlements; (ii) after the amendments to Rule 23, some courts continue to approve class action settlements without key data about take rates; and (iii) in claims-made settlements, class members as a whole receive on average less than 30% of any monetary award. July 2021 ii Jones Day White Paper TABLE OF CONTENTS EXECUTIVE SUMMARY 1 THE JONES DAY 2010–2018 WHITE PAPER 2 DECEMBER 2018 AMENDMENTS TO FEDERAL RULE OF CIVIL PROCEDURE 23 2 FINDINGS AND ANALYSIS: 2019–2020 CONSUMER FRAUD CLASS ACTION SETTLEMENTS 3 Take Rate Analysis 3 General Public Notice Settlements: 2019–2020 4 General Public Notice Settlements: 2010–2020 4 Direct Notice Settlements: 2019–2020 4 Direct Notice Settlements: 2010–2020 5 Claims Rates Analysis Takeaways 5 Settlement Allocation Analysis 6 Range of Amounts Paid to the Class Members and Class Counsel: 2019–2020 6 Range of Amounts Paid to the Class Members and Class Counsel: 2010–2020 6 Settlement Allocation in Claims-Made Settlements: 2019–2010 7 Settlement Allocation in Claims-Made Settlements: 2010–2020 7 Settlement Allocations in Non-Reversionary Fund Settlements: 2019–2020 7 Settlement Allocations in Non-Reversionary Fund Settlements: 2010–2020 8 Settlement Award Allocation Analysis Takeaways 8 Key Takeaways: The Decade in Review 9 POSSIBLE AREAS FOR FUTURE AMENDMENTS TO THE FEDERAL RULES OF CIVIL PROCEDURE 9 CONCLUSION 10 LAWYER CONTACTS 10 APPENDIX A: METHODOLOGY 11 APPENDIX B: LIST OF CASES IN 2019–2020 DATASET 13 ENDNOTES 15 1 Jones Day White Paper EXECUTIVE SUMMARY In our previous White Paper, we analyzed a set of 110 consumer fraud class action settlements approved by federal courts from 2010 to 2018.1 Our analysis of the data yielded three main takeaways: • Typically only a small fraction of class members receive monetary benefits from the settlements; • Some settlements did not redress class members’ alleged economic harms in a meaningful way at all but awarded class counsel hundreds of thousands or millions of dollars in attorneys’ fees; and • On average, class members receive 30% or less of a monetary award. In that study, we noted that our dataset comprised settlements that were approved before new amendments to Federal Rule of Civil Procedure 23 were implemented on December 1, 2018. We observed that those amendments seek to improve class member participation rates (“claims rates” or “take rates”) by focusing on effective notice strategies, and attempt to improve class settlements by adding a list of factors for courts to consider before approval.2 However, we also pointed out that the amendments did not address the underlying problem of lack of injury or interest in the suit among class members. This White Paper seeks to assess whether the data from consumer fraud class action settlements from 2019–2020 meaningfully differ from the 2010–2018 settlements. The data reflects that they do not. The data suggest that amendments to Rule 23 have not made a measurable difference in take rates or the allocation of settlement funds. Despite the Rule 23 amendments, courts continue to approve settlements with very low take rates and continue to permit significant payouts to plaintiffs’ lawyers even when the class relief is miniscule—in one case, only $3.92 per class member.3 Ultimately, while class actions do play a legitimate role in our legal system, the data suggests the need for reform in consumer fraud class actions. In particular, our analysis of the data from 2019–2020 consumer fraud class action settlements reflects that: • Typically only a small fraction of class members received monetary benefits from the settlements. Across 20 settlements in which class members were required to submit claim forms, the average participation rate was 4.91% and the median participation rate was 3.90%, with only two cases having rates higher than 15%. That range is consistent with what we found in the 2010–2018 data, and it suggests that the claims of economic harm or loss continue to be overstated from the start, with the vast majority of consumers having little or no interest in participating in the settlements regardless of what benefits they stand to receive. Indeed, one of the lowest take rates we observed (0.10%) was in a case where class members could receive up to $400.4 Low take rates also may suggest that there are superior means of compensating genuinely dissatisfied consumers, such as through money-back guarantees or other customer satisfaction programs where consumers can receive a full refund of their purchase price rather than the more likely outcome of a delayed recovery of only a fraction of the price through a class action settlement. • Some courts continued to approve class action settlements without key data about take rates. Key data, such as an estimated class size or claims rate, continues to be missing from many of the cases in our 2019–2020 dataset. That is so even though Rule 23 added criteria that district courts must consider before determining a settlement is “fair, reasonable, and adequate.” • On average, particularly in claims-made settlements, class members received less than 30% of any monetary award. Across seven claims-made settlements (where the settlement award paid to class members was based on the number of claims submitted rather than a set fund) where the amount paid to class members is compared to the amount paid for attorneys’ fees, expenses, or other non-class distributions, class members received on average only 23.89% of the settlement amount. Across 16 settlements of all types, more than half of the settlement on average went to attorneys or others who were not class members. The first section of this report briefly discusses the previous White Paper, analyzing data from consumer fraud class action settlements from 2010 to 2018. Section two reviews the December 2018 amendments to Federal Rule of Civil Procedure 23. Section three presents our empirical findings regarding take rates and settlement award allocations, as well 2 Jones Day White Paper as the implications of those findings for the 2019–2020 data. Section four presents potential areas for additional changes to Rule 23. Section five summarizes conclusions drawn from the data. THE JONES DAY 2010–2018 WHITE PAPER Our first study analyzed eight years of consumer fraud class action settlements.5 Our data came from 110 cases in which federal courts approved class settlements between 2010 and 2018.6 Those settlements were finalized before the 2018 amendments to Rule 23 went into effect.7 Based on that dataset, a number of conclusions may be drawn about consumer fraud class action settlements. First, the data showed that only a small fraction of class members received monetary benefits from the settlements. “Across 40 settlements in which class members were required to submit claim forms, the average participation rate was 6.99% and the median participation rate was 3.40%, with only four cases having a rate higher than 15%.”8 The low rates suggested that the severity or extent of economic harms may be overstated. Second, the data showed that class counsel took away hundreds of thousands or millions of dollars even when settlements failed to redress class members’ alleged economic harms. In eight injunctive relief cases, for example, “class counsel received an average amount of $491,717, while class members received no monetary relief.”9 Finally and relatedly, the data showed that the bulk of cash settlements went to paying attorneys’ fees, expenses, or cy pres awards rather than to class members. The dataset included 44 cases between 2010 and 2018 in which there was had sufficient information to compare the amount received by counsel with that paid to class members.10 Although the average total amount paid to the class was approximately 1.7 times the average amount paid to class counsel, there were 16 cases in which the amount paid to class counsel exceeded that paid to class members.11 Overall, across the 44 cases, “an average of 38.42% of the settlement award was paid to class members” and an “average of 33.20% of the settlement award was paid to class counsel.”12 In other words, the data showed that class members received less than half of the settlement awards in consumer fraud class action settlements during the years under review. DECEMBER 2018 AMENDMENTS TO FEDERAL RULE OF CIVIL PROCEDURE 23 Our previous study was based on settlements finalized before December 1, 2018. On December 1, 2018, amendments to Federal Rule of Civil Procedure 23 went into effect.13 Those amendments, while important, have been characterized by some as “modest.”14 In this section, we explain those changes. We limit this discussion to the changes to Rule 23(e) and Rule 23(c), because we identified those changes as most likely to affect the trends observed in the data on 2010–2018 consumer fraud class action settlements: low take rates and a settlement allocation skewed toward plaintiffs’ lawyers rather than class members. First, before 2018, Rule 23 did not provide district courts with a process for deciding motions for preliminary approval of class action settlements.15 District courts’ treatment of preliminary approval therefore varied,16 and some described this level of review as “just enough to ensure that sending notice to the class is not a complete waste of time.”17 As a result of the lack of guidance at this stage, courts “rarely denied” preliminary approval.18 The amendments to Rule 23(e) now require that, before directing notice, the court must conclude “that the prospect of class certification and approval of the proposed settlement justifies giving notice.”19 The Rule requires the parties to submit a “solid record,” and the Advisory Committee’s commentary suggests that parties should submit “all available materials they intend to submit to support approval under Rule 23(e)(2).”20 The commentary provides a non-exhaustive list of such materials, which should include the anticipated claims rate, the claims process, and a plan for distributing unclaimed funds (for example, cy pres or pro rata redistribution).21 Further, the parties should address how attorneys’ fees will be handled, and the commentary stresses that “it will be important to relate the amount of an award of attorneys’ fees to the expected benefits to the class.”22 The commentary suggests—but the Rule does not require—“defer[ring] some or all of the award of attorney’s fees until the court is advised of the actual claims rate and results.”23 3 Jones Day White Paper Second, before 2018, Rule 23(e)(2) required only that a proposed class action settlement be “fair, reasonable, and adequate” without providing additional factors for that evaluation.24 As a result, the courts of appeals developed “an intricate jurisprudence on their own.”25 The changes to Rule 23(e) largely codified that jurisprudence, and so are “unlikely to generate a significant change in the settlement process or outcome.”26 The Rule now requires a focus on “core concerns” for approval, including the effectiveness of distributing relief and the timing of payment of an attorneys’ fee award.27 As noted, however, the amendments do not require that the parties present data on the claims rate to the court before it determines the fee award. Finally, after the Supreme Court decided Eisen v. Carlisle & Jacquelin in 1974 and signaled a preference for mailed notice to members of 23(b)(3) settlement classes, “many courts . . . read [Rule 23(c)] to require notice by first class mail in every case.”28 Amended Rule 23(c)(2) now requires district courts to consider “appropriate” notice, not only “best practicable” notice, and the Rule now authorizes class notice to be made by “electronic means.”29 However, the focus remains on “the means or combination of means most likely to be effective,” and the Advisory Committee notes that in some cases, “a significant portion of class members … may have limited or no access to email or the Internet.”30 Indeed, a 2019 Federal Trade Commission (“FTC”) study suggests that email may be less effective at notifying class members, due to recipient skepticism.31 In sum, these changes largely codify tests or criteria that federal courts had already developed on their own. Below, we assess whether these changes to the rules had an impact on the trends observed in our previous White Paper by analyzing consumer fraud class action settlements finalized over the last two years. FINDINGS AND ANALYSIS: 2019–2020 CONSUMER FRAUD CLASS ACTION SETTLEMENTS As in our previous study, we reviewed consumer fraud class action settlements in which class members allege only economic loss due to alleged misrepresentations or false advertising.32 We analyzed: (i) the take rates and (ii) the settlement allocations, to the extent that the cases we collected included sufficient information about those points. For each of these analyses, we also provide the data drawn from the full 2010– 2020 dataset. Notably, despite Rule 23’s new preliminary approval procedures and analysis of the “core factors” for approval, it is still frequently the case that basic information about class size, take rates, and relief to the settlement class is missing from public dockets. Thus, there were some cases upon which we could rely in our claims rate analysis but not in our allocation analysis, and others that enabled an assessment of the relative allocations between class members and class counsel while not providing sufficient detail to estimate the take rates in those cases.33 Take Rate Analysis We found 20 consumer fraud class action settlements for which there was sufficient data to assess the participation rate of class members. We excluded cases in which the class size was undisclosed or only vaguely estimated (such as: “potentially millions”).34 For this preliminary review, we used the take rate for valid claims where that data was available. Across those 20 cases, the majority featured take rates below 10%. Despite robust notice campaigns by class administrators, the vast majority of class members continued to decline to participate in settlements. The average take rate was 4.91%, and the median take rate was 3.90%. Only three settlements had a rate higher than 10%, and only two had a rate higher than 15%.35 Figure 1 presents this data. Figure 1: Take Rates Across All 20 Settlements, 2019–2020 Additionally, when combined with our 2010–2018 data to analyze the entire 2010–2020 dataset, the average take rate was only 6.30%, and the median was 3.74%. 4 Jones Day White Paper Table 1: Take Rates Across All Settlements, 2010–2020 TAKE RATE Average 6.30% Median 3.74% Minimum 0.01% Maximum 48.99% And when we removed three outliers36 from the dataset, the average and median of our 2010–2020 dataset were even lower, 4.55% and 3.22% respectively. Table 2: Take Rates Across All Settlement Excluding Three Outliers, 2010–2020 TAKE RATE Average 4.55% Median 3.22% Minimum 0.01% Maximum 25.53% General Public Notice Settlements: 2019–2020. Twelve of the 20 take-rate cases involved general public notice. Class members in those cases received notice of the settlement by general publication in magazines, newspapers, and internet advertisements rather than by direct mail or email. This general public notice subcategory also included cases in which the defendants sent direct notice to some members, as long as the percentage of class members receiving direct notice was less than 50% of the entire estimated class. Across the general public notice cases, take rates ranged from 0.83% to 11.50%. The average was 3.89%, and the median was 3.61%. These percentages are presented in Figure 2. Figure 2: Take Rates Across General Public Notice Settlements, 2019–2020 General Public Notice Settlements: 2010–2020. When combined with our 2010–2018 data to analyze the entire 2010– 2020 dataset, the percentages were similar. The average take rate was only 4.27%, and the median was 2.80%. Table 3: Take Rates Across General Public Notice Settlements, 2010–2020 TAKE RATE Average 4.27% Median 2.80% Minimum 0.01% Maximum 32.45% When we excluded one outlier,37 the take rates were even lower: The average take rate was only 3.30%, and the median was 2.60%. Table 4: Take Rates Across General Public Notice Settlements Excluding One Outlier, 2010–2020 TAKE RATE Average 3.30% Median 2.60% Minimum 0.01% Maximum 11.50% Direct Notice Settlements: 2019–2020. Our dataset included eight cases in which at least 50% of the estimated settlement class members received direct notice of the settlement, by mail or email. The average take rate was 6.45%, and the median take rate was 5.09%. These percentages are presented in Figure 3. Figure 3: Take Rates Across Direct Notice Settlements, 2019–2020 5 Jones Day White Paper Direct Notice Settlements: 2010–2020. When combined with our 2010–2018 data to analyze the entire 2010–2020 dataset for direct notice settlements, the average take rate was 8.32%, and the median take rate was 4.45%. Table 5: Take Rates Across Direct Notice Settlements, 2010–2020 TAKE RATE Average 8.32% Median 4.45% Minimum 0.10% Maximum 48.99% But when we removed two outliers,38 the take rates for direct settlements across the 2010–2020 dataset were only slightly higher than the general public notice settlements, with an average 5.85% take rate. Table 6: Take Rates Across Direct Notice Settlements Excluding Two Outliers, 2010–2020 TAKE RATE Average 5.85% Median 4.04% Minimum 0.10% Maximum 25.53% Claims Rates Analysis Takeaways. As in our previous study, our data from consumer fraud class action settlements in 2019 and 2020 suggest that settlement awards reach only a small share of class members. The vast majority of class members receive no benefits from settlements. The low take rate persists even when Direct Notice is given to a majority of potential class members. Although slightly higher than in general public notice cases, the average take rate for direct notice settlements was still just 6.5% across the eight cases in our 2019–2020 dataset. In our 2010–2018 report, we proposed several possible explanations for the low take rates in consumer fraud class actions. Those proposals continue to appear valid based on the 2019–2020 data.39 First, many class members may not consider themselves to have been injured by the alleged fraud. It is possible—perhaps even likely—that most potential class members did not base their purchasing decisions on the misrepresentations alleged in the class actions, and that they therefore do not feel aggrieved enough to participate in the class. That would explain why, as the FTC recently observed, in consumer class actions, “[t]here does not appear to be a statistically significant relationship between median compensation and claims rates.”40 Indeed, even in a case where some settlement members who had incurred out-of-pocket expenses for repairs to their truck doors were eligible for a $400 award, the take rate was only 0.10%.41 Second, perhaps potential class members are simply uninterested in participating in settlements that promise only miniscule awards. When potential awards are as low as $0.60 per product purchased (as was the case in Pettit v. Procter & Gamble Company),42 the opportunity costs of participating may be too high. Where potential class members must locate proof of purchase, even where proof (such as receipts) may be available, the time required to locate that proof of purchase may be seen as far outweighing the sometimes-paltry awards. What is more, some manufacturers may already offer a moneyback guarantee program, providing a full refund to dissatisfied customers. Many consumers may see this as a superior means of addressing their concerns, as they prefer to receive a refund by contacting the manufacturer directly rather than participate in a class action where relief may be delayed or less than a full refund.43 Finally, it is possible that existing class notice methods could be improved upon. The FTC reports that “[t]here are marked difference in the claims rates across notice methods,” with regular mail notice resulting in “a median claims rate of 16%” and email notice resulting in a median claims rate of just 3%.44 It is entirely possible that a majority of class members are simply unaware of ongoing class action settlements in which they could participate. Whatever the reason, take rates remained low in 2019 and 2020. As a result, relatively few potential class members received any compensation from settlement awards. If class relief is the main objective of consumer fraud cases, class actions are apparently an inefficient tool for obtaining that relief. 6 Jones Day White Paper Settlement Allocation Analysis Our analysis of settlement allocations yields a similar conclusion. The data establishes that the entire class receives, on average, less than half of a settlement award. The majority—and in some cases the vast majority—of settlement awards are used to pay for attorneys’ fees, litigation costs, and other administrative expenses. As in our previous study, we examined how consumer fraud class action settlements allocated money, especially between class members and class counsel.45 Range of Amounts Paid to the Class Members and Class Counsel: 2019–2020. Our dataset included 16 settlements in which we had sufficient information to compare the amounts allocated to the settlement class and class counsel in attorneys’ fees and expenses. The calculations for all 16 cases are presented in Table 7.46 Table 7: Comparison of Settlement Amount Paid to Settlement Class and Class Counsel, 2019–2020 AMOUNT PAID TO SETTLEMENT CLASS AMOUNT PAID TO CLASS COUNSEL Average $1,802,891.03 $1,984,645.72 Median $1,221,758.80 $1,268,039.06 Lowest Value $164,060.00 $386,321.56 Highest Value $7,867,518.00 $7,065,940.93 The amount paid to class counsel exceeded that paid to the settlement class in nine of 16 cases, a fact that is also reflected in the relative averages above. The average amount paid to class counsel was approximately 10% higher than the amount paid to the class. Our dataset included a range of cases with sometimes wildly different amounts allocated between class members and class counsel. For example, in Brickman v. Fitbit (a case in which the defendant allegedly misrepresented the ability of Fitbit watches to accurately track users’ sleep), class counsel received $7,065,940.93 in attorneys’ fees and litigation costs while only $1,768,625 was allocated to the settlement class.47 In other words, class counsel received almost four times as much as class members in that case. Across our cases, an average of 43.08% of the settlement award was paid to class members (see Figure 4), and the median amount was 46.51%. An average of 44.42% of the settlement award was paid to class counsel (see Figure 4), and the median amount was 38.03%. Our dataset included seven cases in which the amounts awarded to class counsel exceeded 35% of the total settlement award. Figure 4: Average Percentages Paid to Class Members and Class Counsel Overall, 2019–202048 Range of Amounts Paid to the Class Members and Class Counsel: 2010–2020. When combined with our 2010–2018 data to analyze the entire 2010–2020 dataset for all cases, the amount paid to the class was slightly higher than the amount paid to class counsel. However, when we removed one outlier,49 the average amount paid to class counsel was almost as high as the amount paid to the class. Table 8: Allocation Analysis Across All Cases, 2010–2020 AMOUNT PAID TO SETTLEMENT CLASS AMOUNT PAID TO CLASS COUNSEL Average $2,966,358.93 $1,963,024.25 Median $1,004,256.61 $990,837.22 Lowest Value $11,913.00 $54,194.00 Highest Value $68,000,000.00 $10,200,000.00 7 Jones Day White Paper Table 9: Allocation Analysis Across All Cases Excluding One Outlier, 2010–2020 AMOUNTS PAID TO SETTLEMENT CLASS AMOUNT PAID TO CLASS COUNSEL Average $1,864,093.82 $1,823,414.49 Median $937,718.21 $982,500.00 Lowest Value $11,913.00 $54,194.00 Highest Value $16,739,712.00 $10,000,000.00 Settlement Allocation in Claims-Made Settlements: 2019– 2010. Within the 16 cases in our settlement allocation dataset, seven settlements were “claims-made” settlements. In these cases, the amount paid to class members was based on the number of valid claims submitted by class members. Our calculations for these settlements are presented in Table 10. Table 10: Comparison of Settlement Amounts Paid to Settlement Class and Class Counsel in Claims-Made Settlements, 2019–2020 AMOUNT PAID TO SETTLEMENT CLASS AMOUNT PAID TO CLASS COUNSEL Average $1,024,606.05 $2,895,674.76 Median $537,879.00 $3,221,468.26 Lowest Value $164,060.00 $3,415,761.91 Highest Value $2,472,940.07 $7,065,940.93 The average, median, lowest, and highest values paid to class counsel far exceeded that paid to the class in claimsmade settlements. The average amount paid to class counsel was approximately 2.8 times that paid to class members in these cases. In claims-made settlements, the average amount paid to class counsel was 62.96% of the total settlement award (the median was 63.90%), and the average amount paid to class members was 23.89% (the median was 18.84%). Figure 5 presents this comparison. Figure 5: Average Percentage Paid to Class Members and Class Counsel in Claims-Made Cases, 2019–202050 Settlement Allocation in Claims-Made Settlements: 2010– 2020. When combined with our 2010–2018 data to analyze the entire 2010–2020 dataset for all claims-made cases, the average amount paid to class counsel was more than $1,000,000 higher than the amount paid to the class. Table 11: Allocation Analysis Across Claims-Made Settlements, 2010–2020 AMOUNT PAID TO SETTLEMENT CLASS AMOUNT PAID TO CLASS COUNSEL Average $1,406,990.48 $2,512,593.52 Median $522,387.66 $1,500,000.00 Lowest Value $11,913.00 $54,194.00 Highest Value $9,202,862.67 $7,065,940.93 Settlement Allocations in Non-Reversionary Fund Settlements: 2019–2020. The remaining nine settlements in our dataset featured “non-reversionary funds.” The settlements in these cases established set funds from which the class members—and occasionally attorneys’ fees, expenses, and settlement administration costs—are paid. Any funds remaining after the initial payouts to class members would either be directed to cy pres recipients or be reallocated among class members that submitted valid claims. Our calculations for the nine non-reversionary fund cases are presented in Table 12. 8 Jones Day White Paper Table 12: Comparison of Settlement Amounts Paid to Settlement Class and Class Counsel in Non-Reversionary Fund Settlements, 2019–2020 AMOUNT PAID TO SETTLEMENT CLASS AMOUNT PAID TO CLASS COUNSEL Average $2,408,223.78 $1,276,067.58 Median $1,300,000.00 $808,821.15 Lowest Value $490,420.00 $386,321.56 Highest Value $7,867,518.00 $4,149,585.00 The average percentage of the total settlement award paid to the settlement class was 55.07% (the median was 57.79%), and the average percentage paid in attorneys’ fees and expenses was 32.83% (the median was 32.68%). Figure 6 presents this comparison. Figure 6: Average Percentage Paid to Class Members and Class Counsel in Non-Reversionary Settlement Fund Cases, 2019–202051 Settlement Allocations in Non-Reversionary Fund Settlements: 2010–2020. When combined with our 2010–2018 data to analyze the entire 2010– 2020 dataset for non-reversionary fund settlements, the average amount paid to the settlement class was higher than class counsel. However, that gap shrank when we removed one outlier.52 From 2010–2020, excluding one outlier, the amount paid to the entire class was only slightly higher than the amount paid to class counsel. Table 13: Allocation Analysis Across Non-Reversionary Fund Settlements, 2010–2020 AMOUNT PAID TO SETTLEMENT CLASS AMOUNT PAID TO CLASS COUNSEL Average $3,582,853.43 $1,745,752.68 Median $1,143,517.60 $950,000.00 Lowest Value $24,682.00 $69,563.46 Highest Value $68,000,000.00 $10,200,000.00 Table 14: Allocation Analysis Across Non-Reversionary Fund Settlements Excluding One Outlier, 2010–2020 AMOUNT PAID TO SETTLEMENT CLASS AMOUNT PAID TO CLASS COUNSEL Average $2,049,111.84 $1,544,461.07 Median $1,109,363.37 $936,000.00 Lowest Value $24,682.00 $69,563.46 Highest Value $16,739,712.00 $10,000,000.00 Settlement Award Allocation Analysis Takeaways. Across the dataset of cases from 2019 to 2020 for which we had sufficient information to compare the amounts allocated to the settlement class to those allocated to other costs including attorneys’ fees, class members received on average less than half of total settlement amounts. Even in non-reversionary fund settlements—in which we would expect a greater amount of settlement funds to be allocated to class members—just 55.07% of settlement funds (on average) was paid to the class. In claims-made settlements, almost two thirds of the average settlement went to attorneys’ fees and litigation costs. As in the previous study, the data demonstrate that the lion’s share of consumer fraud class action awards do not go to class members. As with the take rate analysis, this fact suggests that class actions are an inefficient tool for redressing alleged consumer fraud. 9 Jones Day White Paper Key Takeaways: The Decade in Review Across 10 years of class action settlements, our analysis yields a clear finding: Consumer fraud class actions frequently yield relatively little for a small percentage of the settlement class but promise substantial awards for the lawyers who take on those cases. When we merged the data from 2019 and 2020 with that from 2010–2018, our analyses produced similar results. In an aggregate 57 cases for which we had data about class participation,53 the average take rate was 4.55%, and the median was 3.22%. Eleven settlements had take rates at or below 1%, and 42—the vast majority of settlements—had take rates below 5%. Figure 7: Take Rates, 2010–2020 Likewise, across 59 cases,54 the data reflects that class members received an average of $1,864,093.82 (with a median of $937,718.21) and class counsel received $1,823,414.49 (with a median of $982,500.00). Class counsel received an average of 36.53% of settlement awards across over the last decade— more than a third of the average award.55 Figure 8: Settlement Allocations, 2010–2020 POSSIBLE AREAS FOR FUTURE AMENDMENTS TO THE FEDERAL RULES OF CIVIL PROCEDURE As our analysis confirms, the 2018 amendments to the Federal Rules of Civil Procedure are an important start but largely codified what many federal courts were already doing or merely acknowledged the realities of technology’s effect on notice. As reflected in the data on 2019–2020 settlements, it appears the amendments have had little effect on the trends observed in 2010–2018. As one district court put it, the “changes are mostly form over substance.”56 There are additional possible changes that could have a significant impact on the trends observed from 2010–2020, including the low take rates and the allocation of relief to the class as compared to the attorneys’ fee award. First, in light of low take rates for consumer fraud class action settlements, the Rules could provide further guidance for how courts determine “that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy” in cases of consumer complaints or dissatisfaction with a product.57 In particular, if a defendant offers a moneyback guarantee for customer satisfaction, that program may offer more meaningful relief to potential class members than protracted litigation where class members likely stand to recover some fraction of the price they paid for the product or services.58 Second, the Advisory Committee’s suggestion that the attorneys’ fee award cannot be determined until all class data is submitted could be made part of Rule 23(e)(2)’s criteria, rather than a suggestion for implementing Rule 23(e)(2)(C)(iii). Some courts are already using this approach.59 Additionally, approval of class relief should be untethered from approval of attorneys’ fees, to ensure that appeals of attorneys’ fee awards do not jeopardize or delay relief for the class members. Again, provisions like this one are already included in many class action settlements. Including these two procedural mechanisms in the Rule will benefit consumers and ensure uniformity. 10 Jones Day White Paper Third, the Rules could also provide more specific guidance for how courts should calculate attorneys’ fee awards. Currently, the Rules do not impose any “rigid limits . . . for such awards.”60 Additional changes to the Rules could provide criteria linking attorneys’ fee awards to the overall claims rates, for instance, or guidelines for an allocation of settlement funds that would avoid some of the extreme examples observed in the data. Any evaluation of the potential efficacy of such proposals is beyond the scope of this paper; however, the trends observed show that consumer fraud class action settlements merit further study and review. LAWYER CONTACTS Darren K. Cottriel Irvine +1.949.553.7548 [email protected] Robert N. Stander Washington +1.202.879.7628 [email protected] Craig E. Stewart San Francisco +1.415.875.5714 [email protected] Amelia A. DeGory and Justin D. Rattey, associates in the Washington Office, contributed to this White Paper. CONCLUSION The findings in this study align with the view of many that consumer fraud class action settlements often do not provide meaningful relief to consumers and instead primarily benefit class counsel. While it is generally true that the average percentage of settlement awards paid to the class exceeds the average percentage paid to class counsel, that is not always the case, and typically only a small fraction of class members actually receive monetary relief due to low take rates. Although the 2018 amendments to Federal Rule of Civil Procedure 23 are designed to increase take rates and improve transparency in the settlement process, the 2019–2020 data does not reflect significant changes post-amendment. That accords with the observation that low take rates may actually reflect that the claims of economic harm were overstated to begin with, or that some companies already make full refunds available to dissatisfied customers. This may be why consumers have little interest in participating in settlements regardless of the benefits they stand to receive. 11 Jones Day White Paper APPENDIX A: METHODOLOGY In building our dataset, we replicated many of the collection procedures we used for our previous study. Our objective was to collect consumer fraud or false advertising cases in which a class action settlement was approved by a federal court in 2019 and 2020. We utilized two strategies to select our pool of settlements: 1. We reviewed issues of the BNA Class Action Litigation Reporter, the Mealey’s Class Action Litigation Reporter, and the Westlaw Journal Class Action Litigation Reporter from 2019 and 2020 to identify settlements approved by a federal court during the past two years. These reporters allowed us to skim the descriptions of settlements receiving final approval to determine whether we could include them. 2. We ran term searches on Bloomberg, Westlaw, Lexis, and Docket Alarm. Our searches included various combinations of search terms such as: false advertising, consumer fraud, settlement, class action, final approval, preliminary approval, deceptive, unfair, and misrepresentation. Through these two research approaches, we collected a total of 31 cases. In empirically analyzing the 31 cases, we used the public case dockets available on Docket Alarm and Westlaw to examine case filings and court orders. In particular, we looked to the complaints, the parties’ motions for preliminary and final settlement approval, court orders granting preliminary and final settlement approval, settlement administrator declarations, motions for attorneys’ fees and costs, and orders granting motions for attorneys’ fees and costs. Often, exhibits to these documents included the settlement agreements themselves, as well as Class Action Fairness Act notices (some of which contained the estimated settlement class sizes). We collected the following information for each case, to the extent it was available: • Date filed • Case number • Jurisdiction • Defendant(s) • Whether the settlement class was a national or state class • Description of consumer fraud claims • Products or services involved • Total monetary relief under settlement • Non-monetary or injunctive relief under settlement • Whether there were any coupon payments • Whether there were any charitable contributions up front (rather than cy pres) • Whether the settlement involved a non-reversionary settlement fund or was based on claims made by class members • Date of preliminary approval • Date of final approval • Estimated settlement class size • Whether relief was automatic or required claim form submissions • Number of claims submitted • Claims rate • Amount paid or available to each class member • Total amount allocated to the class • Total amount and percentage of the settlement actually paid to the class • Total amount and percentage of the settlement paid to cy pres • Total amount and percentage of the settlement paid to class counsel • Total amount and percentage of the settlement paid to settlement administration costs • Where the court records lacked information • Whether the settlement involved general public notice or direct notice to class members Because our data was based on the information available in public court records, it was limited in scope. First, not every case docket contained all the information we sought. As a result, we could not use all 31 cases to analyze both take rates and settlement award allocations. We found that 20 case dockets provided information sufficient to calculate take rates, and 16 case dockets provided information sufficient to calculate settlement award allocations. 12 Jones Day White Paper Further, the dockets often did not contain the most updated information. For example, if the most recent claims rate information available was a settlement administrator’s declaration accompanying a Motion for Final Settlement Approval, and the settlement administrator had not yet determined which of the claims received were valid and nonduplicative, then the claims rate for that particular case likely was lower than reflected in the declaration and in this report. Additionally, even if court records specified the cash amount to class members, they rarely indicated how many class members actually received and cashed their settlement checks once the checks were distributed. Therefore, the settlement award percentages paid to class members may be lower than reported here. 13 Jones Day White Paper APPENDIX B: LIST OF CASES IN 2019–2020 DATASET CASE NAME CASE NUMBER JURISDICTION DATE OF FINAL APPROVAL INCLUDED IN TAKE RATE ANALYSIS INCLUDED IN SETTLEMENT ALLOCATION ANALYSIS INCLUDED IN NEITHER Pettit v. Procter & Gamble Company 3:15-cv-02150 N.D. Cal. 3/29/2019 X X Jackie Fitzhenry-Russell et al. v. Keurig Dr. Pepper Inc. 5:17-cv-00564 N.D. Cal. 4/10/2019 X X Lori Cowen et al. v. Lenny & Larry’s, Inc. 1:17-cv-01530 N.D. Ill. 5/2/2019 X Mednick v. Precor, Inc. 1:14-cv-03624 N.D. Ill. 6/12/2019 X Littlejohn v. Ferrara Candy Co. 3:18-cv-00658 S.D. Cal. 6/17/2019 X Dashnaw et al. v. New Balance Athletics, Inc. et al. 3:17-cv-00159 S.D Cal. 7/29/2019 X X McKnight v. Uber Techs., Inc. 3:14-cv-05615 N.D. Cal. 8/13/2019 X Grant McKee v. Audible, Inc. et al. 2:17-cv-01941 C.D. Cal. 8/16/2019 X Jackie Fitzhenry-Russell v. The Coca-Cola Company 5:17-cv-00603 N.D. Cal. 10/3/2019 X X Bayol v. Health-Ade LLC 3:18-cv-01462 N.D. Cal. 10/11/2019 X Woodard et al. v. Labrada et al. 5:16-cv-00189 C.D. Cal. 10/17/2019 X Miller v. O’Reilly Automotive, Inc. 4:18-cv-00687 W.D. Mo. 12/16/2019 X Carter v. Gen. Nutrition Ctrs., Inc. 2:16-cv-00633 W.D. Pa. 12/19/2019 X Hunter v. Nature’s Way Prods. 3:16-cv-532 S.D. Cal. 1/6/2020 X Shin v. Plantronics, Inc. 5:18-cv-05626 N.D. Cal. 1/31/2020 X Theodore Broomfield, et al. v. Craft Brew Alliance, Inc. 5:17-cv-01027 N.D. Cal. 2/5/2020 X X Miller et al. v. Wise Company, Inc. 5:17-cv-00616 C.D. Cal. 2/20/2020 X X Dickey v. Advanced Micro Devices, Inc. 4:15-cv-04922 N.D. Cal. 2/21/2020 X Rodriguez v. It’s Just Lunch Int’l 1:07-cv-09227 S.D.N.Y. 3/2/2020 X Brickman v. Fitbit, Inc. 3:15-cv-2077 N.D. Cal. 3/20/2020 X X continued on next page 14 Jones Day White Paper CASE NAME CASE NUMBER JURISDICTION DATE OF FINAL APPROVAL INCLUDED IN TAKE RATE ANALYSIS INCLUDED IN SETTLEMENT ALLOCATION ANALYSIS INCLUDED IN NEITHER Wolf v. Hewlett Packard Co. 5:15-cv-01221 C.D. Cal. 3/29/2020 X X Megan Schmitt v. Younique LLC et al. 8:17-cv-01397 C.D. Cal. 4/9/2020 X X Crane v. Sexy Hair Concepts, LLC 1:17-cv-10300 D. Mass. 5/14/2020 X X Cicciarella, et al. v. Califia Farms LLC 7:19-cv-08785 S.D.N.Y. 7/9/2020 X Belfiore v. Procter & Gamble Company 2:14-cv-04090 E.D.N.Y. 7/27/2020 X Hilsley v. Ocean Spray Cranberries, Inc. 3:17-cv-2335 S.D. Cal. 8/3/2020 X Hart v. BHH, LLC 1:15-cv-04804 S.D.N.Y. 9/22/2020 X Ang et al. v. Bimbo Bakeries USA, Inc. 4:13-cv-01196 N.D. Cal. 9/29/2020 Clay v. Cytosport, Inc. 3:15-cv-00165 S.D. Cal. 10/29/2020 X X Schneider v. Chipotle Mexican Grill, Inc. 3:16-cv-02200 N.D. Cal. 11/4/2020 X Kommer v. Ford Motor Co. 1:17-cv-00296 N.D.N.Y. 12/15/2020 X X 15 Jones Day White Paper ENDNOTES 1 Jones Day, An Empirical Analysis of Federal Consumer Fraud Class Action Settlements (2010–2018) (April 2020) (hereinafter “2010–2018 White Paper”). 2 Id. at 13. 3 See Pettit v. Procter & Gamble Co., 2:14-cv-04090 (E.D.N.Y. 2020). 4 See Kommer v. Ford Motor Co., 1:17-cv-00296 (N.D.N.Y. 2020). 5 See 2010–2018 White Paper, supra n. 1. 6 See id. at 1 & Appendix B. 7 See id. at 13. 8 Id. at 1. 9 Id. 10 Id. 11 Id. at 8. 12 Id. at 9. 13 See Fed. R. Civ. P. 23 (2018). 14 Rhonda Wasserman, The New, Improved Class Action Rule: The December 2018 Amendments to Rule 23, 90 Pa. B. Ass’n Q. 182, 182 (Oct. 2019). 15 See William B. Rubenstein, Newberg on Class Actions § 13:10 (5th ed., Dec. 2020). 16 Id. 17 Nieberding v. Barrette Outdoor Living, Inc., No. 12-CV-2353-DDC-TJJ, 2015 WL 1645798, at *4 (D. Kan. Apr. 14, 2015) (quoting Newberg, supra n. 15). 18 Dan Donovan et al., Class Action Settlement Approval in an Era of Heightened Judicial Scrutiny, N.Y.L.J. Online (Feb. 27, 2019). 19 Fed. R. Civ. P. 23(e)(c)(2), advisory committee’s note to 2018 amendment. 20 Fed. R. Civ. P. 23(e)(1), advisory committee’s note to 2018 amendment. 21 See id. 22 Id. 23 Id. 24 See Fed. R. Civ. P. 23(e)(2) (amended 2012). 25 See Newberg, supra n. 15, § 13.13. 26 Id. 27 See Fed. R. Civ. P. 23(e)(2). 28 Fed. R. Civ. P. 23(c)(2), advisory committee’s note to 2018 amendment. 29 See Fed. R. Civ. P. 23(c)(2). 30 See Fed. R. Civ. P. 23(c)(2)(B), advisory committee’s note to 2018 amendment. 31 FTC Staff Report, Consumer and Class Actions: A Retrospective and Analysis of Settlement Campaigns 2 (Sept. 2019). 32 For a description of our methodology in selecting settlements, see Appendix A. 33 See Appendix B. 34 See Hilsley v. Ocean Spray Cranberries, Inc., No. 3:17-cv-2335 (Dkt. 260) (final approval hearing) (S.D. Cal. July 31, 2020). 35 In Dickey v. Advanced Micro Devices, Inc., 4:15-cv-04922 (N.D. Cal. 2020), the court announced in its final approval of the settlement that the claims rate was 27.4%. See id. (Dkt. 165). But the court had erroneously calculated that claims rate using only the class members who received direct notice. When calculated using the total class of “6,713,000 potential purchasers,” id. (Dkt. 154), the actual claims rate in that case was 4.09%. 36 Because there were no outliers in the 2019–2020 settlements, this analysis reflects the 2010–2020 data excluding the three outliers from the 2010–2018 dataset. 37 Because there were no outliers in the 2019–2020 settlements, this analysis reflects the 2010–2020 data excluding the one outlier from the 2010–2018 dataset. 38 Because there were no outliers in the 2019–2020 settlements, this analysis reflects the 2010–2020 data excluding the two outliers from the 2010–2018 dataset. 39 See 2010–2018 White Paper, supra n. 1, at 7. 40 FTC Staff Report, Consumer and Class Actions: A Retrospective and Analysis of Settlement Campaigns 11-12 (Sept. 2019). 41 See Kommer v. Ford Motor Co., No. 1:17-cv-00296 (Dkt. 70) (N.D.N.Y. 2020). 42 Pettit v. Procter & Gamble Co., 2:14-cv-04090 (Dkt. 129) (E.D.N.Y. 2020). 43 See, e.g., Conrad v. Boiron, Inc., 869 F.3d 536, 540-41 (7th Cir. 2017) (affirming district court’s denial of class certification because refund option for disappointed consumers was already in place and no “significant extra value” could be attained for class members). 44 FTC Staff Report, Consumer and Class Actions: A Retrospective and Analysis of Settlement Campaigns 25 (Sept. 2019). 45 Unlike our previous study, we did not identify enough cases allocating amounts to cy pres recipients to meaningfully analyze that allocation separately. It is possible that, for several settlements, data about cy pres distributions will be available at some point in the future, once the claims period has run. It is also possible that fewer recent settlements were structured to include cy pres due to the Supreme Court’s signal, by granting certiorari in Frank v. Gaos (No. 17-961), that such settlements may face increased scrutiny. 46 In Schmitt v. Younique, 8:17-cv-01397 (C.D. Cal. 2020), we assume that the amount paid to cover settlement administration costs was exactly $175,000 (the maximum amount set by the court (Dkt. 255)) and calculated the amount paid to the settlement class based on that assumption. In Brickman v. Fitbit, Inc., 3:15-cv-2077-JD (N.D. Cal. 2020), we assume the amount paid to cover settlement administration costs was $548,292.35, the estimated total (Dkt. 292). In Pettit v. Procter & Gamble Co., 2:14-cv-04090 (E.D.N.Y. 2020), we assume the amount paid to cover settlement administration costs was $677,122.48 (Dkt. 139). 47 Brickman v. Fitbit, Inc., 3:15-cv-2077-JD (Dkt. 292) (N.D. Cal. 2020). 48 The “Other” section consists of settlement administration and notice costs, incentive awards for class representatives, and cy pres distribution, if applicable. Because of missing information related to some of these other factors, the dataset for this table includes only 13 settlements. 49 Because there were no outliers in the 2019–2020 settlements, this analysis reflects the 2010–2020 data excluding the outlier from the 2010–2018 dataset. For a brief description of the outlier case, see infra note 54. 50 The “Other” section consists of settlement administration and notice costs, incentive awards for class representatives, and cy pres distribution, if applicable. Because of missing information related to some of these other factors, the dataset for this table includes only five settlements. 51 The “Other” section consists of settlement administration and notice costs, incentive awards for class representatives, and cy pres distribution, if applicable. Because of missing information related to some of these other factors, the dataset for this table includes only eight settlements. 52 Because there were no outliers in the 2019–2020 settlements, this analysis reflects the 2010–2020 data excluding the outlier from the 2010–2018 dataset. For a brief description of the outlier case, see infra note 54. 53 We excluded three outlier cases with take rates above 30%. All three outliers were settlements finalized between 2010 and 2018 and were treated as outliers in our previous study. 54 We excluded Ebarle v. LifeLock, Inc., 4:15-cv-00258 (N.D. Cal. 2016) in our analysis, as we had in the previous study, because that case is apparently an outlier. In that case, the court awarded the class $68,000,000 (84.15% of the total settlement award) and awarded class counsel $10,200,000 (12.62% of the total settlement award). 55 Because of missing information about other expenses—such as administration costs—in several cases, our analysis of the allocation percentages draws on information from 50 settlements. 56 Swinton v. SquareTrade, Inc., No. 418CV00144SMRSBJ, 2019 WL 617791, at *5 (S.D. Iowa Feb. 14, 2019). © 2021 Jones Day. All rights reserved. Printed in the U.S.A. Jones Day publications should not be construed as legal advice on any specific facts or circumstances. The contents are intended for general information purposes only and may not be quoted or referred to in any other publication or proceeding without the prior written consent of the Firm, to be given or withheld at our discretion. To request reprint permission for any of our publications, please use our “Contact Us” form, which can be found on our website at www.jonesday.com. The mailing of this publication is not intended to create, and receipt of it does not constitute, an attorney-client relationship. The views set forth herein are the personal views of the authors and do not necessarily reflect those of the Firm. 57 Fed. R. Civ. P. 23(b)(3). 58 See, e.g., Turcios v. Carma Labs., Inc., 296 F.R.D. 638, 648 (C.D. Cal. 2014) (holding that a class action mechanism was not superior when the defendant manufacturer already provided a full refund, which is greater than the relief the plaintiff sought); In re Aqua Dots Prod. Liab. Litig., 654 F.3d 748, 752 (7th Cir. 2011) (affirming the district court’s denial of class certification where the requested relief was duplicative of the refund option that was already offered). 59 See, e.g., N.D. Cal., Procedural Guidance for Class Action Settlements, (updated Dec. 5, 2018). 60 See Rule 23(e)(2), advisory committee’s note to 2018 amendment.