Testimony of Brian Wolfman
June 1, 2001
Statement Before the Third Circuit Task Force on Appointment of Class Counsel
Good afternoon. I am Brian Wolfman, a staff lawyer at Public Citizen Litigation Group, a Washington-D.C. based public interest law firm. I want to thank Chief Judge Becker and the members of the Task Force for inviting me to testify.
My interest in class actions is longstanding. Early in my legal career, as a legal services lawyer in rural Arkansas in the 1980's, I filed class actions, mainly against state and federal governmental agencies to enforce statutory rights, but also against private parties. Usually these cases involved injunctive relief, but some involved damages as well, including difficult questions of how to distribute those damages to a widely dispersed class. Since 1990, at Public Citizen Litigation Group, I have continued to work from time to time representing classes, as do other lawyers in my office. Mainly, however, I represent absent class members who object to settlements, because, in their view, the class representatives and class counsel did not adequately protect the interests of the absentees. I have submitted to the Task Force my resume, which highlights some of this class action work. In addition, a comprehensive discussion of Public Citizen Litigation Group's efforts regarding class action settlements can be found on our website at http://www.citizen.org/litigation/briefs/class_act.html. I note that my colleague, Alan B. Morrison, who was also invited to testify before the Task Force, has reviewed my testimony and concurs in it.
II. Preliminary Observations.
At first blush, it appears that there is a wide divergence of views on the principal question presented by the Task Force: whether, and in what circumstances, a pre-litigation auction should be used to appoint class counsel. Some of the witnesses are opposed to auctions in any circumstance; other witnesses are skeptical, but have an open mind; and other witnesses endorse their use. However, almost all agree that an auction will work only in a fairly narrow category of cases -- antitrust, securities matters, perhaps a few others -- and even there, probably only in cases where liability and the amount of potential damages are fairly clear. But, in the relatively large category of securities fraud cases, it is fair to say that Congress, in enacting the PSLRA, preferred a different method for controlling lawyer conduct and legal fees. I mention this not because the question whether auctions should be used is unimportant -- it clearly is not. Nor do I believe that the auction concept is fundamentally flawed -- what I have read in preparing for this hearing indicates that, in some instances, it has worked well, resulting in quality legal services at a lower price than might have otherwise been achieved, to the benefit of the class. The testimony of those opposed to any use of competitive bidding is unconvincing, for the simple reason that it fails to address cases where auctions have been successful or to explain why the prevailing model of post-litigation fee determination is a better method in all circumstances.
I raise the issue of the rather narrow range in which this Task Force appears to be operating for another reason. If we are to achieve the purpose of the auction -- netting more for the class members by aligning their interests with those of the lawyers -- this Task Force, or perhaps its successor, should look more comprehensively at the question of class counsel compensation, in effect updating the work of the 1985 Task Force. I'm not alone, it appears, in this view, because many of the witnesses have chosen to deal not so much with the appointment of class counsel (through auctions or otherwise) but with an array of related issues, such as the use of the percentage-of-the-fund fee method, whether the percentage fee should rise, decline, or stay constant as the size of the fund increases, the composition of plaintiffs' steering committees, and the like.
Therefore, in the pages that follow, I first address some issues regarding appointment of attorneys via auction, but then move on to related issues that I believe the Task Force should address.
III. Use of Auctions.
A. General Observations. What I have read and learned, both prior to, and in preparation for, this testimony suggests that appointing class counsel via auction can benefit the class in some circumstances. In money damages actions where the alleged cause of action is well established and the amount of potential damages is well understood at the outset, an auction may produce a better deal for the class than the prevailing method of retrospective fee determination. Not only can such a system produce a fee well below the traditional fee based on a "benchmark" percentage of the common fund, but it ought to do away with large plaintiffs' steering committees, which, in some cases, are highly inefficient (more on that later).
My chief concern is that in many, and probably most, cases, the putative plaintiffs' lawyer simply does not have the information to participate sensibly in a pre-litigation auction. Drawing from my experience, both in litigating on behalf of classes and in representing absent class member-objectors, many keys points of law and fact are in substantial doubt at the outset of the case (or even half way into it), and only gradually come into focus as the litigation matures. Some of the unknowns are:
- the legal viability of one or more of the key claims for relief;
- facts that might prove necessary or important to establish a viable legal claim (indeed, in some cases, the "historical" facts themselves are still developing);
- whether the case can be maintained as a class action, or whether instead the case cannot be maintained at all or can be maintained only individually or in some other form of aggregated litigation;
- key facts bearing on the amount of damages suffered by the class (assuming liability).
- the financial ability of the defendant to withstand full liability, or, rather, whether recovery is possible only on a "limited fund" basis, in bankruptcy, or must be sought, in whole or in part, from other parties not initially believed to be principally responsible.
- the form of relief (for instance, cases in which money damages initially are sought, but, as the litigation develops, only injunctive relief, ADR, or some other form of relief is viable).
- the size of the class, and perhaps more important, the approximate number of class members likely to be eligible for relief on account of the defendant's alleged misconduct.
- the potential for intra-class conflict, and the resulting need for sub-classes requiring separate sub-class counsel. See Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 626-27 (1997).
I'd like to explore some of these categories in a bit more detail. One of them -- assessing the viability of the plaintiffs' legal theories -- involves the kind of judgment that lawyers make all the time, and perhaps enough lawyers would be willing make a pre-litigation bid even where the legal viability of the case is in substantial doubt (with the bids reflecting that doubt). And the legal system could insist on a bidding process where the lawyers are in the dark about many of the foregoing issues, but that system would not likely attract many qualified bidders, and, more fundamentally, given the risks, would be unlikely to benefit class members any more than the post-hoc fee system generally employed today. Indeed, in this regard, it is important to ask why courts have generally not employed one of the key methods for simulating market conditions at the outset of the litigation that the 1985 Task Force recommended -- appointment of an independent person to negotiate a fee, as soon as practicable, on behalf of the absent class. See 108 F.R.D. 237, 256 (1985). I think it is reasonable to assume that many courts and counsel have felt that such negotiations would not be meaningful early in the case, because neither the fee negotiator nor class counsel would have sufficient information to negotiate a reasonable fee at that juncture.
The above-listed categories of doubt are not mere abstractions. In fact, in almost all of the class actions on which I have worked, even the lawyers who filed the case lacked information about some key factual and legal issues when the case was filed. It goes almost without saying, therefore, that potential rival bidders would also be in the dark. The facts also change in complex class litigation. In cases concerning motor vehicle defects, for instance, the actions of government regulators sometimes alter the prospects for recovery, or at least the type of recovery that might be available. In a couple personal-injury class actions on which I have worked, it was impossible to determine whether the defendant could afford full relief until the litigation had proceeded for a number of years. Moreover, in those and other cases, recovery may depend on the availability of insurance, which may, in turn, depend on success in ancillary litigation, involving questions that cannot be posed, let alone answered, by an array of potential bidders before the class action is underway. Indeed, in such cases, discovery sometimes reveals potentially responsible parties that could not possibly have been known when the case was filed.
At the outset of some securities fraud class actions, it will be fairly easy to know both the approximate number of class members and the subset of the class that may have suffered harm. Indeed, in Judge Walker's impressive opinion in In re Quinta Securities Litigation, attached to his testimony before this Task Force, the court conducted a sort of mini-auction under the PSLRA rubric in which the known damages suffered by the proposed lead plaintiffs were key to Judge Walker's determination of who could best represent the class. And it was clear, inferentially, that knowledge about the total damages suffered by the plaintiff class as a whole was important to the bidding lawyers in structuring their bids. Put otherwise, had that information not been available in Quinta, it is difficult to see how either the putative class counsel or Judge Walker could have properly evaluated the issues bearing on appointment of counsel.
In most cases, however, it is difficult to know either the aggregate potential damages or the subset of the class that has suffered injury. In one mass-tort class action on which I worked, even the defendant had only a very rough estimate of the number of people who had used the allegedly defective product, and the key fact -- how many class members were harmed by the product in a manner sufficient to justify relief -- was not even estimable until many years into the litigation (after the applicable law and facts were refined, and the class members were surveyed and medical records were obtained). In two public benefit class actions in which I was lead counsel, getting a handle on the number of injured plaintiffs proved even more difficult. In one, defendants conceded that the class met Rule 23's numerosity requirement, but depositions and document discovery showed that the subset of class members who had been harmed could not be identified through the defendant's records, and thus the size of the recovery could not measured until liability had been established and a post-judgment notice sought to locate injured plaintiffs. Williams v. Patterson, No. LR-C-87-380 (E.D. Ark.). In the other case, through discovery we learned that one of the defendant's computer "fields" did, in fact, capture the class, albeit overinclusively, but the number of class members eligible for relief, and the number who actually obtained relief, was not known until two post-settlement notices had been sent to the class. Hannah et al. v. Glickman, No. 94-3004 (D.S.D.); see South Dakota v. Madigan, 824 F. Supp. 1469 (D.S.D. 1993). The point here is that, if fees are to be gauged according to the potential harm to the class (i.e. the aggregate potential recovery), or better yet, as a reasonable percentage of what the class actually receives (discussed further in Part IV.C. below), an auction when the case is filed makes sense only when the harm to the class can be closely approximated at that time.
B. The Cendant Issue. I agree with the observations of several other witnesses that, if an auction is conducted and the bidding process itself is proper, the resulting fee should be treated as if it were a contractual obligation of the class. If, by conducting an auction, and forcing the participating attorneys to compete with one another, the court avoids an allegedly deficient, court-driven retrospective fee determination, it makes little sense to impose that type of fee determination on top of the auction. In a world where only a minority of cases are subject to auction, plaintiffs' lawyers are less likely to participate in the bidding process (and instead spend their time working on other cases), if the winning bid does not really represent the fee, but rather a ceiling on the fee. See generally In re Cendant Corp. Prides Litig., 243 F.3d 722 (3d Cir. 2001).
It is true, as Cendant noted, that the court must review fees in all class actions, id. at 730 (citing In re General Motors Corp. Pick-up Truck Fuel Tank Litig., 55 F.3d 768, 819 (3d Cir. 1995), but the scope of review of a bidding process should be different from that performed with respect to a traditional retrospective fee award. In the district court, when appointing class counsel, the court should of course exercise tight control over the bidding process itself. But at the end of the case, when approving the fee award to the winning bidder, both the district court and the court of appeals should not disturb the results of the auction absent unusual factual circumstances that could not have been reasonably anticipated at the onset of the litigation or a showing of bid-rigging or other improprieties in the bidding process, such as a violation of auction rules imposed by the district court or the deliberate withholding of information that the district court requested to conduct the auction. With respect to a change in factual circumstances, its invocation should be rare and, if invoked, it should be a two-way street, leaving open the possibility that the auction-generated fee could be increased as well as reduced.
C. Use of Auctions In Non-Damages Cases. The Court asked whether appointment of class counsel through an auction makes sense in cases involving injunctive relief or relief other than cash damages. As in all class actions, in most injunctive relief cases, the class benefits through settlement rather than through a litigated judgment. In non-cash settlements, the fee must still be paid in cash, but it must come from the defendant rather than out of a cash fund also used to pay the class members. In those cases, the court should closely scrutinize the fee request and ask whether a large fee was a quid pro quo for the plaintiffs' lawyers agreement to accept a less-than-optimal injunctive settlement for the class. General Motors, 55 F.3d at 819-20; see generally Weinberger v. Great N. Nekoosa Corp., 925 F.2d 518 (1st Cir. 1991). However, because there is no direct relationship between the fee and the class recovery, the use of an auction to set the fee for class counsel is unlikely to benefit the class.
On a related matter, the Court asked whether an auction could be used in cases where the anticipated relief is coupons or an alternative dispute resolution ("ADR") mechanism through which class members' claims are resolved by impartial decision makers. This question simply underscores the problem, discussed in Part III.A. above, of conducting an auction at the outset of the litigation where the imponderables are significant. No substantive law authorizes coupon or ADR relief; put differently, a jury or judge cannot award coupons or order classwide ADR in a litigated case alleging violations of a state consumer protection statute, the UCC, common law duties, or any other law. Rather, the complaint in such a case must seek money damages and/or an injunction prohibiting future unlawful practices. Thus, generally, it would be an inadequate plaintiffs' counsel who "anticipated" coupon relief when filing a class action, and thus it would certainly be inappropriate to conduct an auction on the assumption that the class will be getting coupons.
ADR presents a somewhat different situation because, even if class counsel seeks to impose classwide liability and damages through verdict, counsel could appropriately anticipate ADR as a means for distributing relief. Cf., e.g., In re Prudential Ins. Co. of America Sales Practices Litig., 148 F.3d 283 (3d Cir. 1998). Thus, it is not inconceivable that, in a case in which ADR is the expected means for apportioning classwide damages, class counsel could be picked by auction (with the fee denominator being the overall value of the relief actually distributed via ADR). However, even there, it is only after much is learned about the case that plaintiffs' counsel generally would know, for instance, whether all class members have been injured in much the same way (making claims administration rather simple) or whether the existence and magnitude of the injuries vary dramatically across the class (making ADR the preferred remedial tool). In sum, in most cases, plaintiffs' counsel cannot (and should not) anticipate ADR relief in connection with a pre-litigation auction, because it is only after discovery has been taken, the matters in dispute have been refined, and settlement is in the offing, that the possibility for such relief can be rationally considered.
D. Joint Auction Bids. The Court asked whether joint bids submitted by multiple law firms should be permitted. As some witnesses have stated, such bids should not be automatically foreclosed because, in some instances, competition might be increased if joint bids are received from small law firms that would be unwilling to participate without pooling their resources. However, joint bidding ought to be viewed with skepticism. Obviously, joint bidding could be an attempt to decrease competition. It could also re-create gross inefficiencies that arise in the large steering committees and other multiple-firm practices typically employed in complex class actions. I have reviewed fee applications in many nationwide class actions on behalf of objecting class members. As a rule, those applications reveal a large amount of duplication of effort, with multiple law firms doing the same or similar research and document review. These large lawyer committees often involve a number of law firms whose roles in the litigation, and in negotiating the settlement, cannot be discerned. Sometimes it appears that the lead law firms do the work, and the hangers-on bloat the fee to the detriment of the class. In my experience, some participants can rightly be described as bottom feeders, who file related litigation and sign up numerous clients, standing as potential objectors unless placed on the committee. To gather data on the "bloating" problem, in one fee challenge on behalf of absent class members, my office attempted to obtain discovery of fee-splitting deals among counsel, only to be rebuffed. Bowling v. Pfizer, Inc., 102 F.3d 777, 780-81 (6th Cir. 1996). However, in one case we did obtain informal discovery and learned that a lawyer whose role was to file a 20-page brief and appear at the fairness hearing was cut in for a quarter million dollars, while another lawyer, whose role was somewhat more extensive, received nearly half a million. Thus, in structuring a bidding process, courts ought to think long and hard before they allow law firms to work together rather than bid against each other.
IV. Improving Retrospective Fee Determinations.
For the reasons stated above, I believe that fee setting at the beginning of the litigation, whether via auction or otherwise, is not likely to be practicable in most cases. Instead, I believe that the traditional retrospective fee determination can be considerably improved. I discuss some ideas for improvement below.
A. Better Approximating the Market. In my view, the current system for awarding a reasonable percentage-of-the-fund attorney's fee in class actions is completely inadequate. Here's what usually happens: The plaintiffs' lawyer files a fee application requesting a percentage, rather than a lodestar, fee. The lawyer begins by reminding the court that the typical contingent fee in ordinary bi-polar litigation is 30 to 40 percent, and that therefore the 25 percent fee that she is requesting is a great deal for the class. She reminds the court that the case was taken at great risk (no recovery, no fee), but usually does little to explain the real risks of non-recovery in this or similar litigation (other than saying that the defendant litigated the case vigorously and was a tough, arms-length negotiator). The lawyer then tells the court that case law exists approving class fees both below and above the 25 percent rate, but that 25 percent is reasonable under the circumstances and consistent with the "benchmark" established in that circuit. More often than not, the court awards the requested fee or something like it, although it sometimes reduces the percentage when the size of the fund is very large, consistent with the case law in those circumstances. See Empirical Study of Class Actions in Four Federal District Courts ("FJC Study") 69-74 (Federal Judicial Center 1996) (discussing prevailing fee award methodologies).
In the above scenario, the lawyer has honestly stated the facts and the relevant case law. So what's wrong with this picture? Starting from the market rate for contingency fee lawyers in individual cases (assuming for present purposes that the market for individual cases operates rationally) and tweaking the numbers a bit, while doing nothing to evaluate the actual risks of class litigation, does not reflect the market in which the class action lawyer is actually operating. As a result, the benchmark "market" rates for class actions, accepted over and over again by the courts, were established tautologically, with courts simply looking to other courts that have engaged in the identical exercise.
Indisputably, plaintiffs' lawyers are at risk because they invest money and time in their cases without assurance that they will recover. But if a lawyer must invest $10,000 to represent one plaintiff in a case against a particular defendant, the lawyer almost surely will not have to invest 1,000 times that much money to represent a class of 1,000 people bringing the same class action claim against that defendant. So, too, with the investment of the lawyer's time. No class lawyer would take a class action on behalf of 1,000 people if it was anticipated that litigating the case would take 1,000 times more effort than it would to litigate the matter individually. After all, class actions exist because they create economies of scale, so that the lawyer need not expend anywhere near the amount of time and expense it would take to represent the same number of people in individual, bi-polar litigation. In sum, it is improper for courts to set class fees by reference to fees in individual contingent litigation, because doing so may permit the class lawyer to achieve an economy of scale without sharing its benefits with the clients.
The same problem can be viewed from a different perspective. Contingency fees of 30 to 40 percent in bi-polar litigation are defended, quite appropriately in many instances, on the ground that the risk of non-recovery or an insignificant recovery is large, and therefore many cases result in a net loss to the lawyer. Is that true in class litigation in the securities, antitrust, consumer, and/or personal-injury cases? We need an answer to that question before we can look to the 33 percent contingency fee in ordinary litigation, lop off a few percentage points in a class action, declare ourselves satisfied, and go home. In my experience, the majority of class actions result in some recovery for the class, and, since the fee almost always meets or exceed the lodestar, a significant recovery for the lawyers. Of course, there are total losses and poor recoveries, although in many such cases, particularly where class certification is denied, losses are cut quickly. The problem is that there is little hard information and courts generally have not sought to create it.
But why can't it be done? For instance, the court could inquire about success and failure in securities fraud class actions generally, or in those securities cases raising the particular issues before the court, or of those filed by class counsel's law firm over the past five years, and so forth. The Federal Judicial Center could expand the inquiry that it did several years back concerning class actions in four judicial districts. That study showed that median fee rates were 27 to 30 percent, FJC Study at 69, and that, once the hurdle of class certification was overcome, the risk of non-recovery was fairly small. Id. at 60. (The study also debunked the notion that many class actions served only the interests of lawyers, finding that, in almost all cases, the class recovery substantially exceeded the lawyers' fee. Id. at 68-69). In short, more information is needed, and courts can obtain some of that information by getting more from the parties about their experiences in similar litigation. Of course, in any particular case, class counsel may argue that the risks of that case were greater than the norm, but that argument, by definition, could only work so often.
Having answers to these questions would not, of course, yield perfect fee determinations. Economic expertise would have to be brought to bear to determine a reasonable range of fees given the economies of scale and the true risks of class litigation. If, as Judge Walker believes, unleashing true market forces in a court-conducted auction yield benefits for the class, more data on these topics ought to yield similar benefits in the retrospective fee determination process.
B. Fee Holdbacks. One technique for aligning the interests of the class members and their lawyers is to hold back a significant amount of the fee until the lawyer's work has been successfully completed. In relatively simple cases, where the class settles for a sum certain that is, in turn, promptly distributed to the class, the fee can be paid simultaneously with the class recovery. However, in many cases, important work, upon which the class's fate depends, remains to be done after a litigated judgment or settlement. In cases where the class members' recoveries depend on success in ADR, for instance, the court should hold back a significant portion of the fee, with the full payment forthcoming only if the predicted level of ADR recovery is met. See Duhaime v. John Hancock Mut. Life Ins. Co., 989 F. Supp. 375 (D. Mass. 1997). In this manner, class counsel are strongly encouraged to ensure that their clients take advantage of the ADR and are well represented in it.
In coupon settlements, the little empirical evidence that exists demonstrates that most class members get nothing because redemption rates are very low. See, e.g., Buchet v. ITT Consumer Financial Corp., 845 F. Supp. 684, 695-96 (D. Minn. 1984) (minuscule coupon redemption rates), amended, 858 F. Supp. 944, 944-45 (D. Minn. 1984) (citing additional information to same effect); "In Camera," 16 Class Action Reports 369, 485-87 nn.2-8 (July-Aug. 1993) (survey of coupon settlements, showing that settling parties generally vastly overstate expected redemption rates and that, without transferability, settlement coupons are generally worthless); B. Meier, "Fistful of Coupons--Millions for Class Action Lawyers, Scrip for Plaintiffs," New York Times, pp. D1, D5 (May 26, 1995) (only one percent redemption rate where coupons could be used toward purchase of new vehicle). Such low rates can result from indifference, lack of proper notice, a lack of desire to use the coupon to purchase the defendant's product, or, in cases involving big-ticket items, an inability to afford the defendant's product. Nonetheless, in some cases, class counsel have simply multiplied the number of certificates issued by the certificate's face value and sought a "reasonable" percentage of the resulting figure. In other cases, fees have been awarded as a percentage of the plaintiffs' expert's prediction regarding the level of coupon redemption, In re Domestic Air Transp. Antitrust Litig., 148 F.R.D. 297, 322 (N.D. Ga. 1993); see General Motors, 55 F.3d at 807-10, predictions that, as noted above, are at odds with the little we know about actual coupon redemption in comparable cases.
However, there is no need for guesswork. In the Domestic Air case -- where a coupon expert was employed by fee-seeking counsel -- the court awarded a $14.3 million fee, or about 5.25% of the predicted redemption value of the settlement coupons. Id. at 357. But why not award the fee based on a percentage of the coupon value that is actually redeemed? Whatever one thinks of coupon settlements -- and there are arguments against their use in any case -- this holdback method will surely eliminate the worst excesses. With the prospect of a paltry fee, no longer would class counsel agree to a settlement in which coupons are non-transferable, Dunk v. Ford Motor Co., 48 Cal. App. 4th 1794, 1805 (1996), or the impediments to redemption so great as to render the coupons valueless to most class members. General Motors, 55 F.3d at 808-10. In fact, by tying counsel's fate to that of their clients, the typical coupon settlement would become a thing of the past, and only settlements in which the coupon has a cash redemption value or the settlement includes the participation of a secondary market-maker -- in other words, a settlement that actually broadly benefits the class -- would be worth counsel's efforts. See National Ass'n of Consumer Advocates--Standards and Guidelines for Litigating and Settling Consumer Class Actions, 176 F.R.D. 375, 382-84 (1998); General Motors, 55 F.3d at 809.
Finally, in some complex cases where the contours of the relief are fairly well understood when the settlement is approved, much work is left to be done to assure that the class members are located, the distribution of the settlement proceeds is widespread, and, in cases where the funds are not distributed directly to the class, the money is efficiently spent on other settlement components, such as medical monitoring or research for the potential benefit of the class. See, e.g., Bowling v. Pfizer, Inc., 143 F.R.D. 141, 149 (S.D. Ohio 1992). Similarly, sometimes the settlement proceeds are paid over an extended period of time, even though ultimately paid in cash to the class members. In all such cases, a significant fee should be held back to assure that the settlement is properly implemented by class counsel. And for post-approval work, class counsel's efforts cannot be classified as contingent, and therefore courts should seriously consider awarding such "future fees" on a straight lodestar basis for work as it is performed, with payment forthcoming after the competent completion of the necessary tasks. Bowling v. Pfizer, Inc., 132 F.3d 1147 (6th Cir. 1998).
C. Tying Fees to Actual Recovery. Although Task Force witnesses have given considerable attention to the "lead plaintiff" provision of the PSLRA, the PSLRA's requirement that fees be tied to actual recovery has generated relatively little discussion. However, that provision is quite important because, like the fee holdback techniques discussed immediately above, it provides an incentive for counsel to obtain actual benefits for the class members. Unfortunately, many courts have permitted fees to be calculated against the stated settlement fund, even where only a small proportion of the fund is claimed and the remainder has reverted to the defendant. In extreme situations, that approach has led to situations where the class recovered almost nothing, but the lawyers obtained a large recovery. See Williams v. MGM-Pathe Communications Co., 129 F.3d 1026 (9th Cir. 1997) (33 percent fee on $4.5 million "fund" resulting in $1.5 million fee, $9,900 class recovery, and $2,990,100 reverter to defendant). Those results are intolerable and can be cured simply by tying fees to actual recovery. Obviously, if such a fee mechanism is in place, class counsel would work diligently to assure clear and effective class notice and simple, user-friendly claiming procedures.
It is also important that class recoveries be widely distributed among the class members. Thus, it is no answer to the problem in Williams simply to eliminate the reverter, and allow windfall distributions to the few claiming class members and the 33 percent fee to the lawyers. Thus, in cases where there is no reverter, courts must take steps to maximize widespread relief to the class. The court should make preliminary determinations of expected claiming rates, based on the settling parties' potential ability to effectively locate and notify the class and other demographic factors, and make fee adjustments (either up or down) if counsel exceeds or falls short of the distributional goal (such as a 50% claiming rate). See generally In re Orthopedic Bone Screws Prods. Liab. Litig., 246 F.3d 315 (3d Cir. 2001) (highlighting problems with inadequate notice that suppressed class member participation in common fund). In this manner, the court can assure that counsel will provide informative and widespread notice to the class, eliminate make-work claiming procedures when simply cutting a check will do, and otherwise assist class members who want to partake of the common fund.
D. Injecting An Adversarial Relationship Into The Fee Determination. Courts and commentators have noted the lack of adversariness in most retrospective fee determinations. Outside of the statutory fee-shifting context, where the defendant in some cases has an incentive to contest counsel's fee request, the settling defendant usually has little or no concern about what part of the recovery is paid to the class and what part is paid as a fee. Prandini v. National Tea Co., 557 F.2d 1015, 1020 (3d Cir. 1977). Indeed, class counsel often negotiate "clear sailing" arrangements, where the defendant agrees not to contest a fee up to a certain amount, the very purpose of which is to eliminate an adversary contest regarding fees. Although courts have indicated that such arrangements demand added judicial scrutiny of fees, Great N. Nekoosa, 925 F.2d at 524, the court is often unable to conduct a searching inquiry because no one is there to present the other side of the story.
Two methods exist to provide the missing adversary element. First, courts should assure that objecting class members or organizations have standing to oppose fee requests, regardless of whether the fee is paid from a common fund or paid separately by the defendant apart from the class benefit. See, e.g., General Motors, 55 F.3d at 819-20; cf. Cendant, 243 F.3d at 728-29. Such objectors should be allowed discovery into counsel's time and expense records and fee-sharing agreements, and given a full opportunity to brief all relevant issues. Fee objectors should be eligible for fees if they are successful in reducing class counsel fees, a basic incentive to keep class counsel's fees in line. See Cendant, 243 F.3d at 743-44 (directing that fees be awarded to objector who helped this Court scrutinize excessive district court fee award); Duhaime v. John Hancock Mutual Life Ins. Co., 2 F. Supp. 2d 175 (D. Mass. 1998); cf. Deborah R. Hensler, et al., Class Action Dilemmas -- Pursuing Public Goals For Private Gain 495 (RAND Institute for Civil Justice 2000) (noting importance of awarding fees to successful objectors because they play constructive role in monitoring unlawful or unfair class action settlements).
Moreover, courts should not permit the parties to agree that reduced fees (i.e., any court-awarded fee smaller than that requested by class counsel) be returned to the defendant. Rather, courts should insist that the money be directed to the common fund for distribution to the class or, if that is not practicable, dedicated to some other purpose related to the litigation (such as class-related research, see Bowling, 143 F.R.D. at 149, or use by an organization dedicated to advancing the plaintiffs' interests). In that fashion, the court can assure that the total amount that the defendant was willing to pay in settlement is properly allocated between the lawyers and the clients.
Second, the court can appoint an independent advocate, beholden to no one but the class, whose job is to carefully scrutinize the fee request and oppose it whenever appropriate. The advocate would be compensated on a lodestar basis from the class fund, and given the same access to relevant materials as would an objecting class member. Any "recovery" obtained by the advocate would, of course, be paid to the common fund.
Thank you again for inviting me to appear before the Task Force. If the Task Force has questions or would like further information, please let me know.