SELYA, Circuit Judge. These appeals call upon us to descend once more into the belly of a beast previously (and accurately) described as a "litigatory monster." In re Recticel Foam Corp., 859 F.2d 1000, 1001 (1st Cir. 1988). On this occasion, we are involved less with the litigants than with their champions. One group of lawyers, appellants here, contend that the procedures followed by the district court in awarding attorneys' fees and directing cost reimbursement were fundamentally unfair, ignored the requirements of procedural due process, and constituted an abuse of discretion. Another group of lawyers, appellees here, vehemently deny these charges and seek to uphold the award. Because we agree that the proceedings in question were profoundly flawed, we vacate the district court's order.
Given the narrow purview of these appeals, we need not dwell upon the tragic facts of the incendiary fire that claimed nearly one hundred lives at the San Juan Dupont Plaza Hotel on December 31, 1986. For our purposes, a bareboned preface serves to place the instant appeals into workable perspective.1
In 1987, the Judicial Panel on Multidistrict Litigation consolidated over two hundred seventy cases, involving approximately twenty-three hundred plaintiffs, and placed them under the aegis of the Honorable Raymond L. Acosta in the District of Puerto Rico. See In re Fire Disaster at Dupont Plaza Hotel, 660 F. Supp. 982 (J.P.M.L. 1987) (per curiam). Shortly thereafter, the Chief Justice appointed the Honorable Louis C. Bechtle as a "settlement Judge." While the litigation moved toward trial in Judge Acosta's court, Judge Bechtle endeavored to advance settlement prospects by determining individual and aggregate values for the cases.
In an effort to organize the plaintiffs' side of the litigation, Judge Acosta appointed a steering committee (the PSC) to act as lead and liaison counsel for the plaintiffs. Because PSC members would exert the lion's share of control over the direction of the litigation and would, therefore, likely lay claim to a larger slice of the fee pie than their non-member colleagues, appointment to the PSC was much coveted; only attorneys who had been retained by individual plaintiffs were eligible to apply, and more than forty of the fifty-six individually retained plaintiffs' attorneys (IRPAs) declared their candicacies. Judge Acosta initially chose nine members and, in June 1988, increased the complement by two.2
By dint of the PSC's role, a rough division of labor emerged. The PSC members looked after the big picture: mapping the overarching discovery, trial, and settlement strategies and coordinating the implementation of those strategies. The IRPAs handled individual client communication and other case-specific tasks such as answering interrogatories addressed to particular plaintiffs, preparing and attending the depositions of their clients, and taking depositions which bore on damages. The IRPAs also worked with Judge Bechtle on a case-by-case basis in his efforts to identify and/or negotiate appropriate settlement values for individual claims. When Judge Acosta determined that the plaintiffs should try twelve representative claims as a means of facilitating settlement, a collaborative composed of three PSC members and four IRPAs bent their backs to the task.
For the most part, Judge Acosta's work on this massive litigation has been a model of judicial craftsmanship and practical ingenuity. The exigencies of these appeals do not demand that we describe the Judge's many innovations. We do note, however, that he wisely segmented the liability inquiry into phases. The first two phases, now concluded, established the liability of the hotel's owners and certain suppliers, respectively; when the results of these phases were integrated with the representative trial outcomes and with Judge Bechtle's handiwork, the combination generated an aggregate settlement fund of approximately $220,000,000.
At the close of the second phase, Judge Acosta opted to disburse the settlement fund because, in his words, "the Phase III litigation will not affect the results of the previous phases in any way and consequently [there is] no reason why distribution of the settlement fund to the victims and their attorneys should not go forward at this time." In re San Juan Dupont Plaza Hotel Fire Litig., 768 F. Supp. 912, 936 (D.P.R. 1991) (hereinafter "Fees Op."). The parties to these appeals -- who agree on little else -- do not dispute this assessment of the situation.3
The IRPAs were originally retained under a variety of contingent-fee agreements, most of which were capped at the legal maxima: twenty-five percent for minors or incompetents; thirty-three and one-third percent for adults.4 In addition, the plaintiffs agreed to pay certain costs. Thus, the amount available for legal fees was a fixed percentage of the overall recovery pool and the only figures which the court needed to compute prior to distributing the settlement fund were (1) the amount of PSC fees to be deducted from the attorneys' fund (a fund comprising the portion of the settlement reserved for legal fees) and (2) the amount of costs to be deducted from the plaintiffs' fund (a fund comprising the portion of the settlement reserved for the victims, excluding counsel fees).
Before discussing the former computation, some introductory comments are in order. Under standard "American rule" practice, each litigant pays his or her own attorneys' fees. See, e.g., Alyeska Pipeline Serv. Co. v. Wilderness Soc'y, 421 U.S. 240, 245, 44 L. Ed. 2d 141, 95 S. Ct. 1612 (1975). Yet, there are times when the rule must give way. For example, when a court consolidates a large number of cases, stony adherence to the American rule invites a serious free-rider problem. See generally Mancur Olson, The Logic of Collective Action (1971). If a court hews woodenly to the American rule under such circumstances, each attorney, rather than toiling for the common good and bearing the cost alone, will have an incentive to rely on others to do the needed work, letting those others bear all the costs of attaining the parties' congruent goals.
A court supervising mass disaster litigation may intervene to prevent or minimize an incipient free-rider problem and, to that end, may employ measures reasonably calculated to avoid "unjust enrichment of persons who benefit from a lawsuit without shouldering its costs." Catullo v. Metzner, 834 F.2d 1075, 1083 (1st Cir. 1987). Such courts will most often address the problem by specially compensating those who work for the collective good, chiefly through invocation of the so-called common fund doctrine. In its paradigmatic formulation, the common fund doctrine permits the trustee of a fund, or a person preserving or recovering a fund for the benefit of others in addition to himself, to recover his costs, including counsel fees, from the fund itself, or alternatively, from the other beneficiaries. See Boeing Co. v. Van Gemert, 444 U.S. 472, 478, 62 L. Ed. 2d 676, 100 S. Ct. 745 (1980); Alyeska, 421 U.S. at 257; Catullo, 834 F.2d at 1083.5 Although common fund cases are sui generis, they typically evince certain characteristics. These include ease in identifying the persons, or classes of persons, benefitted by the recovery; ease in tracking the benefit flow; the ability to trace benefits with enough accuracy that, in the end, the flow chart inspires confidence; and the ability to shift litigation costs with enough precision and reliability that cost and benefit are fairly proportionate to one another. See Boeing, 444 U.S. at 478-79; Alyeska, 421 U.S. at 265 n.39.
Here, Judge Acosta's decision to use a steering committee created an occasion for departure from the American rule. In apparent recognition of the free-rider problem, the Judge served notice from the beginning that he would eventually make what he, relying in part on appellees' counsel, see Fees Op., 768 F. Supp. at 924 n.42, later termed a "common fund fee award" to remunerate PSC members for their efforts on behalf of communal interests. This was a proper exercise of judicial power. See Mills v. Electric Auto-Lite Co., 396 U.S. 375, 392, 24 L. Ed. 2d 593, 90 S. Ct. 616 (1970); see also In re "Agent Orange" Prod. Liab. Litig., 818 F.2d 226, 240 (2d Cir. 1987) (upholding a fee award to a plaintiffs' steering committee under the equitable fund doctrine); Bebchick v. Washington Metro. Area Transit Comm'n, 256 U.S. App. D.C. 296, 805 F.2d 396, 402 (D.C. Cir. 1986) (collecting cases); In re MGM Grand Hotel Fire Litig., 660 F. Supp. 522, 526 (D. Nev. 1987).
After Phase III ended and court-appointed accountants determined that the attorneys' fund was in the approximate amount of $66,000,000, Judge Acosta held a two-day hearing addressed to the method and amount of compensation to be awarded to PSC members for their efforts on behalf of the whole. In the course of the hearing, five of the eleven PSC members broke rank and proposed that PSC fees should be calculated on a lodestar basis, not as a percentage of the settlement fund. A prominent law professor outlined this position on behalf of the five insurgents.6 The majority of the PSC members repudiated the insurgents' proposal and asked the court to award PSC fees as a percentage of the settlement fund, in line with Judge Acosta's stated intent when forming the PSC.
This intramural dispute was not so much motivated by principle as by profit. Although each PSC member was in some sense an IRPA -- each member had been retained by at least one plaintiff -- there was a wide disparity in the number of clients that each lawyer represented. One consequence of this disparity was that a large PSC award stood to benefit PSC members who had few clients and to disadvantage PSC members who had many clients.7 The interests of the PSC members and the non-PSC IRPAs diverged even more starkly; after all, each dollar that went to the former was a full dollar less that would otherwise be divided among the latter.
Despite this adversariness, however, Judge Acosta severely limited the IRPAs' participation throughout the fee-determination process. For instance, the Judge's ground rules barred any non-PSC members from testifying at the hearing. Moreover, the IRPAs were told that they had to file their objections to the insurgents' fee proposal on a one-page form designed by the court. The form (a copy of which appears as an appendix hereto) allotted only seven lines for substantive comment, i.e., "reason(s) for opposition" to the fee proposal. The court stressed that longer submissions would be ignored.
When the fee-determination hearing began, Judge Acosta flatly refused to permit the IRPAs, either individually or through a designated representative, to examine the witnesses. The IRPAs were not allowed to offer oral argument. After the hearings ended, Judge Acosta gave the IRPAs leave to file any written objections which were not foreseeable beforehand, but he restricted these filings to the same one-page form. The court again admonished the IRPAs that, if they deviated from the mandated format, the court would not consider their objections.
On the basis of the insurgents' fee proposal, the testimony at the fee hearing, and those objections filed in accordance with the form, Judge Acosta awarded some $36,000,000 of the $66,000,000 attorneys' fee fund to PSC members for their services in that capacity. The court asserted that the fee calculation was the composite product of the percentage and lodestar methods. See Fees Op., 768 F. Supp. at 926-27 ("The Court finds that the use of both the percentage and the lodestar approach is both logical and practical and will lead to a fair and reasonable result because one method serves to check the other."). The court also awarded costs of approximately $10,000,000 to PSC members, chargeable against the plaintiffs' fund.8
Two members of the PSC majority, the PSC alternate, a number of IRPAs, and several individual plaintiffs then prosecuted the instant appeals. The five insurgent PSC members appear as appellees.
II. APPELLATE JURISDICTION
The threshold question in this case implicates our jurisdiction. In the appellees' view, the fee distribution order does not possess the necessary attributes of finality and, consequently, this court lacks jurisdiction to hear the appeals. We disagree.
The baseline criterion for appellate jurisdiction is con-tained in the statutory directive that "the courts of appeals. shall have jurisdiction of appeals from all final decisions of the district courts of the United States." 28 U.S.C. § 1291 (1988).9 An order is usually considered final when it resolves a contested matter, leaving nothing to be done except execution of the judgment. See United States v. Metropolitan Dist. Comm'n, 847 F.2d 12, 14 (1st Cir. 1988). A ruling may nonetheless be final for purposes of appellate jurisdiction even though it is not "the last order possible to be made in a case." Gillespie v. United States Steel Corp., 379 U.S. 148, 152, 13 L. Ed. 2d 199, 85 S. Ct. 308 (1964). Ultimately, "the requirement of finality is to be given a practical rather than a technical construction." Id. (citation and internal quotation marks omitted); accord Metropolitan Dist. Comm'n, 847 F.2d at 14.
Viewed against this backdrop, it is not surprising that, over the years, judicial decisions have placed a gloss on the finality principle, recognizing that certain kinds of orders, while not marking the end of a case, are sufficiently final to satisfy the requirements of 28 U.S.C. § 1291. See, e.g., Mitchell v. Forsyth, 472 U.S. 511, 524-30, 86 L. Ed. 2d 411, 105 S. Ct. 2806 (1985) (permitting immediate appeal by state actor in connection with denial of qualified immunity); Morales-Feliciano v. Parole Bd. 887 F.2d l, 3-4 (1st Cir. 1989) (permitting immediate appeal of certain civil contempt orders), cert. denied, 494 U.S. 1046, 108 L. Ed. 2d 646, 110 S. Ct. 1511 (1990); Krohn v. United States, 742 F.2d 24, 27 (lst Cir. 1984) (permitting immediate appeal by federal official of district court's refusal to dismiss case on grounds of absolute immunity); see generally Cohen v. Beneficial Indus. Loan Corp., 337 U.S. 541, 545-47, 93 L. Ed. 1528, 69 S. Ct. 1221 (1949) (formulating collateral-order exception to finality principle). The appeals we consider today satisfy the finality requirement through a similar, albeit less familiar, mechanism: the common fund exception.
The history of appellate jurisdiction vis-a-vis fee awards from common funds extends at least as far as Trustees v. Greenough, 105 U.S. 527, 26 L. Ed. 1157 (1881). In Greenough, the defendant sought appellate review of a district court order granting attorneys' fees from a common fund accumulated through the efforts of one particular lawyer. The Supreme Court determined that the appeal was timely, observing that, although the district court was still administering the fund, an order disbursing trust funds to pay litigation expenses incurred by a party in preserving the trust assets was "so far independent as to make the decision substantially a final decree for the purposes of an appeal." Id. at 531. Later Supreme Court opinions have cemented Greenough's place in our jurisprudence. See, e.g., Sprague v. Ticonic Nat'l Bank, 307 U.S. 161, 168-69, 83 L. Ed. 1184, 59 S. Ct. 777 (1939) (holding that the district court could award common fund fees while the merits of the case were being appealed because the two were sufficiently separate) (citing Greenough); see also Boeing, 444 U.S. at 484 (Rehnquist, J., Dissenting) (noting the Supreme Court's "willingness to treat the division of the common fund as a separate piece of litigation for purposes of appeal").
This court has relied on Greenough for the proposition that "an order fixing counsel fees, though incidental and collateral to the main cause, is so far independent of it and so finally dispositive of a particular matter 'as to make the decision [as to counsel fees] substantially a final decree for the purposes of an appeal.'" Angoff v. Goldfine, 270 F.2d 185, 186 (1st Cir. 1959) (quoting Greenough, 105 U.S. at 531). Other circuits have likewise followed Greenough and allowed parties to appeal the award of attorneys' fees from a common fund without waiting for the overall litigation to end. See, e.g., Overseas Dev. Disc Corp. v. Sangamo Constr. Co., 840 F.2d 1319, 1324 (7th Cir. 1988) (declaring it to be "settled that a decision awarding or denying attorney's fees and expenses from a common fund. is severable from the decision on the merits and separately appealable"); Memphis Sheraton Corp. v. Kirkley, 614 F.2d 131, 133 (6th Cir. 1980) (holding that an award of attorneys' fees from a common fund "is collateral and may be appealed independently of the principal action"); cf. Shipes v. Trinity Indus., Inc., 883 F.2d 339, 344 (5th Cir. 1989) ("If an interim award of fees corresponds to an order which is itself appealable, 'and represents compensation for work that is compensable no matter what the course of subsequent events,' it may well be 'conclusive' within the meaning of the collateral order doctrine.") (quoting 15 Charles Wright, Arthur Miller & Edward Cooper, Federal Practice and Procedure § 3915 (Supp. 1989)).10
We conclude, therefore, that an order which definitively resolves claims for attorneys' fees and expenses payable out of a common fund is severable from the decision on the merits and sufficiently final to be separately appealable under 28 U.S.C. § 1291.
We turn next to the related matter of whether Judge Acosta's order meets the criteria for appealability under Greenough and its progeny. We must resolve two questions: Is the order definitive? Does it implicate the common fund doctrine? We think the answers to both of these queries are affirmative.
The definitiveness hurdle is easily cleared. At the Conclusion of Phase II, all the plaintiffs executed general releases. Phase III involves an entirely different group of defendants (mostly insurance companies whose insureds subrogated their rights to the plaintiffs as part of the overall settlement). See supra note 3. Although the plaintiffs might recover more money in the future, they will do so only in proportion to their success against the discrete Phase III defendants. When and if this occurs, the lower court will necessarily have to reassess the relationship between the IRPAs and the PSC members in order to determine counsel fees for the Phase III litigation -- but, with one possible (and immaterial) exception noted below,11 those incremental fees will be in addition to the fees awarded at the close of Phase II. In effect, the Phase II fees are inviolate. Thus, the order appealed from possesses the requisite finality.
By the same token, the facts of this case fit snugly within the contours of the common fund exception. The funds in question -- the attorneys' fund and the plaintiffs' fund -- are similar to stereotypical common funds in that they were generated in substantial part by the disproportionate strivings of a few (the PSC members) to the benefit of a much larger number (the plaintiffs and, derivatively, the IRPAs). Both funds exhibit the three primary characteristics that the Boeing Court thought were significant. See Boeing, 444 U.S. at 478-79. First, the funds' beneficiaries can be determined with complete assurance. Second, while the extent to which each individual plaintiff and each IRPA benefitted from the PSC's efforts cannot be quantified with mathematical precision, it is possible to study the PSC's contribution to the overall success of the litigation and approximate the incremental benefits with some accuracy. Finally, the district court controls both funds and, therefore, possesses the ready ability to prorate the cost of achieving the incremental benefits in an equitable manner.
We will not paint the lily. The order appealed from was essentially the final determination of a common fund attorneys' fees award. Accordingly, we have jurisdiction under Greenough to entertain these appeals.12
III. DUE PROCESS CONSIDERATIONS
The gravamen of appellants' claim is that Judge Acosta's ruling tapped the settlement funds, in which appellants had a constitutionally protected property interest, without affording them the process that was due. We canvass the controlling legal principles and then inspect the particulars of the proceedings below.
The Supreme Court "consistently has held that some form of hearing is required before an individual is finally deprived of a property interest." Mathews v. Eldridge, 424 U.S. 319, 333, 47 L. Ed. 2d 18, 96 S. Ct. 893 (1976); accord Wolff v. McDonnell, 418 U.S. 539, 557-58, 41 L. Ed. 2d 935, 94 S. Ct. 2963 (1974). Perfunctory gestures will not suffice. At bedrock, "the fundamental requirement of due process is the opportunity to be heard at a meaningful time and in a meaningful manner." Mathews, 424 U.S. at 333 (citation and internal quotation marks omitted).
The Court has not forged a boilerplate standard for determining what process is due in all circumstances, but, instead, has articulated three areas of particular importance:
The specific dictates of due process generally require consideration of three distinct factors: First, the private interest that will be affected by the official action; second, the risk of an erroneous deprivation of such interest through the procedures used, and the probable value, if any, of additional or substitute procedural safeguards; and finally, the Government's interest, including the function involved and the fiscal and administrative burdens that the additional or substitute procedural requirement would entail.
Id. at 335; see also Connecticut v. Doehr, 115 L. Ed. 2d 1, 111 S. Ct. 2105, 2112 (1991); Reardon v. United States, 947 F.2d 1509, 1518 (1st Cir. 1991) (en banc). The Court's language in Mathews explicates what is essentially a constitutional balancing test. See Amsden v. Moran, 904 F.2d 748, 753 (1st Cir. 1990), cert. denied, 112 L. Ed. 2d 702, 111 S. Ct. 713 (1991); Needleman v. Bohlen, 602 F.2d 1, 5 (1st Cir. 1979). We do not approach the Mathews factors in a vacuum. Rather, we sally forth mindful that, in many, if not most, instances, due process does not require a full-scale trial, or even a hearing strictly conforming to the rules of evidence. See Mathews, 424 U.S. at 348; Doyle v. Secretary of HHS, 848 F.2d 296, 301 (1st Cir. 1988). Due process is malleable, calling "for such procedural protections as the particular situation demands." Morrissey v. Brewer, 408 U.S. 471, 481, 33 L. Ed. 2d 484, 92 S. Ct. 2593 (1972). A court may hand-tailor procedures to account for the nature of the affected interests and the circumstances of the threatened deprivation. See Neron v. Tierney, 841 F.2d 1197, 1201 (1st Cir. 1988), cert. denied, 488 U.S. 832, 102 L. Ed. 2d 66, 109 S. Ct. 90 (1988)."As the rubric itself implies, 'procedural due process' is simply 'a guarantee of fair procedure.'" Amsden, 904 F.2d at 753 (quoting Zinermon v. Burch, 494 U.S. 113, 125, 108 L. Ed. 2d 100, 110 S. Ct. 975 (1990)); see also Newman v. Burgin, 930 F.2d 955, 961 (1st Cir. 1991). Hence, we review cases involving adversarial hearings to determine whether, under the specific facts and circumstances of a given situation, the affected individual has had a fundamentally fair chance to present his or her side of the story. Gorman v. University of R.I., 837 F.2d 7, 14 (1st Cir. 1988). Throughout, we resist bright line rules and mechanical tests, mindful that a nisi prius court may use its sound discretion to assemble an appropriate process by choosing from among a multitude of available methods, so long as the end product stays within the four corners of that discretion and satisfies the rigors of the Due Process Clause.BWith this flexible standard in mind, we focus the lens of our inquiry on whether the procedures employed below satisfied the constitutional imperative. In so doing, we follow the Mathews framework.
1. The Private Interest. In this instance, the affected private interest is the monetary share of the funds due each IRPA.13 The district court necessarily deprived the appellants of money when it invaded the attorneys' fund to pay PSC fees. Not only do the appellants have a property interest in escrowed funds, see Reed v. Commissioner, 723 F.2d 138, 146 (1st Cir. 1983), but the amounts involved are substantial and the deprivation is complete. For these reasons, we find the affected interest to be an important one.
2. Risk of Error.
During the four years preceding the filing of the insurgents' lodestar proposal in February 1991, the district court engendered the assumption that it would calculate PSC fees as a percentage of the total fee award -- "probably less than 10%." Order 2 (March 18, 1987).14 Judge Acosta reiterated his intention a year and one-half later in Order 127, In re Dupont Plaza Hotel Fire Litig., MDL 721, at 40 (D.P.R. Dec 2, 1988). Although these orders did not bind the district court to a ten percent cap, they did induce the IRPAs to rely on the percentage mechanism. This reliance manifested itself in two ways. Because the court led the IRPAs to believe that it would calculate the PSC's fees as a percentage of the settlement fund, the IRPAs had virtually no incentive to monitor the PSC members' time sheets (which were regularly filed with the court). Similarly, because the IRPAs were working under contingent-fee agreements and had every reason to assume that time records would not be material in deciding what portion of the attorneys' fund would be diverted to remunerate the PSC, they were ill prepared to supply the type of documentation necessary to counter the late-blooming lodestar proposal advanced by the PSC minority.15
The initial indication that PSC fees might be determined by anything other than a percentage approach surfaced on January 31, 1991, when the PSC submitted a petition for a fee award together with a motion for partial distribution of funds. In the petition, the PSC discussed the need for a hearing to determine the hours spent by PSC members and consider appropriate multipliers. The petition's import was clouded, however, by the contemporaneous motion for partial distribution, in which the PSC suggested a hearing to establish a percentage allotment to be withheld from the attorneys' fund for PSC fees and a percentage allotment to be withheld from the plaintiffs' fund for PSC costs. On February 5, 1991, the district court notified the parties in interest that it would convene a fee-determination hearing on February 20. Although the order, Order 305 (February 5, 1991), indicated that the court would entertain proposals anent the amount of fees and method of fee calculation, it was not until the deadline date for applications (February 15, 1991) that a filing by an insurgent member of the PSC gave appellants the first clear notice that any PSC member would actively campaign for the lodestar method.In a nutshell, after four years of justifiable reliance on the district court's stated intent to calculate fees for PSC members on a percentage basis, probably not to exceed ten percent, the IRPAs had only a few days to consider the ramifications of a possible switch to the lodestar method, reconstruct their own hours in order to protect their interest in the funds,16 and sift the hours claimed by PSC members. To compound the problem, the court required the IRPAs to confine their objections to a one-page form.
In the ordinary course, due process might have been expected to mitigate the IRPAs' concerns, because "when a person has an opportunity to speak up in his own defense, and when the [decisionmaker] must listen to what he has to say, substantively unfair and simply mistaken deprivations of property interests can be prevented." Fuentes v. Shevin, 407 U.S. 67, 81, 32 L. Ed. 2d 556, 92 S. Ct. 1983 (1972). But here, that expectation never blossomed into actuality. Cf. John McGahern, The Leavetaking, pt. II (1974) ("we have to learn never to expect anything"). The hearing procedures outlined by the district court in Order 310 became part of the problem, not part of the solution.To be blunt, the protocol that governed the hearing hogtied the IRPAs, severely restricting their ability to participate in the fee-determination process. The court summarily abrogated the PSC's "majority decision" rule without prior notice, allowing the insurgents' proposal to go forward. The court then stated, in Order 310, that "the role of [the IRPAs] at the hearing will be limited to that of spectators, that is, they may not interrogate the witnesses, raise objections or otherwise address the Court." Even the PSC members were admonished to testify only as to their participation as PSC members and not to recount their services in their dual capacity as IRPAs. As a final indignity, the court limited the IRPAs' posthearing objections to matters which could not have been foreseen prior to the hearing and confined them to the same tourniqueted form used for prehearing objections. Such miserly process is certainly not sufficient to meet constitutional minima. While "notice and an opportunity to be heard have traditionally and consistently been held to be the essential requisites of procedural due process," Gorman, 837 F.2d at 12, a last-minute warning of a fundamental change in the fee structure is not constitutionally adequate notice17 and a rigid limitation of one affected group's input into the hearing while giving members of the other affected group a much broader array of participatory rights does not comprise a constitutionally adequate chance to be heard. Appellants were entitled to reasonable forewarning and an opportunity to present their side of the controversy in a meaningful manner. They received neither. Hence, the risk that appellants would be erroneously deprived of their property was intolerably great.18
3. The Public Interest.
We do not gainsay the substantial governmental interest in conserving scarce judicial resources. Particularly in mass disaster litigation, courts must run tight ships to ensure that litigation stays on course. We recognize that Judge Acosta imposed the restrictions about which appellants complain in a well-intentioned effort to keep the case velivolant. His renitency was understandable in light of the number of lawyers involved, the four-year life span of the litigation, and the Supreme Court's admonition that "[a] request for attorney's fees should not result in a second major litigation." Hensley v. Eckerhart, 461 U.S. 424, 437, 76 L. Ed. 2d 40, 103 S. Ct. 1933 (1983).Nonetheless, the court was not faced with an all-or-nothing choice between allowing the IRPAs free rein and unfairly hobbling them. There were a wide range of procedures available which, in any one of several possible combinations, could have brought a sense of fundamental fairness to the fee-determination hearing while at the same time husbanding the court's resources. The court could, for example, have granted the IRPAs more latitude in making written submissions, heard some testimony on their behalf, allowed them to designate one or more representatives for purposes of cross-examination, and/or given them a right to participate in oral argument. To be sure, the court was free to pick and choose from among these (and other) alternatives or to impose reasonable restrictions -- but the court was not free to construct a set of ground rules that largely ignored the IRPAs' substantial stake in the controversy.19 While we do not suggest that a full-scale evidentiary hearing was essential, the court, having opted for one (wisely, we think, considering the stakes), should at the very least have permitted the IRPAs to make submissions of a reasonable length -- certainly, more than seven lines -- and should have allowed the IRPAs to appear or to designate a prolocutor to question witnesses and address the court. After all, the root purpose of the fee-determination proceeding was to decide upon an equitable allocation of fees between the IRPAs and the PSC -- a purpose which could scarcely be achieved by hearing only one side of a long, complicated story.20
Let us be perfectly clear. We do not hold today that the Due Process Clause requires the district court to convene an evidentiary hearing for all attorneys' fee disputes. See Weinberger v. Great N. Nekoosa Corp., 925 F.2d 518, 528 (1st Cir. 1991) (stating that hearings are not always obligatory in connection with fee awards). Nor do we hold that a district court must always take live testimony or afford adversarial discovery in such matters. We do rule, however, that, once Judge Acosta elected to convene a fee-determination hearing, the hearing format itself had to be fair. In other words, when a Judge constructs a process for setting fees, the process must contain at least the procedural minima that the Due Process Clause requires; and, moreover, those procedures must apply in a fair and evenhanded manner to the parties in interest, without preferring one group of disputants over another.In this instance, we cannot see how the court could fairly adjudicate this dispute unless it afforded the IRPAs a viable means (comparable to the means afforded the PSC) of describing their contribution to the litigation, contrasting their contribution with the PSC's contribution, and questioning the PSC members regarding their work. See Briscoe v. LaHue, 460 U.S. 325, 333-34, 75 L. Ed. 2d 96, 103 S. Ct. 1108 (1983); California v. Green, 399 U.S. 149, 158, 26 L. Ed. 2d 489, 90 S. Ct. 1930 (1970); Kennerly v. United States, 721 F.2d 1252, 1257-58 (9th Cir. 1983) (reversing district court award of funds where interested party did not receive a hearing); see also Weinberger, 925 F.2d at 525 ("The court's role as the guarantor of fairness obligates it not to accept uncritically what lawyers self-servingly suggest is reasonable compensation for their services."); In re Air Crash Disaster at Fla. Everglades, 549 F.2d 1006, 1021 (5th Cir. 1977) (requiring plaintiffs' committee and counsel to offer evidence and submit to cross-examination at fee-determination hearing). Although using such a model would have prolonged the hearing, it would not have created an unreasonable or intolerable burden on the district court. Given the panoply of available alternatives, the public interest in the integrity of court proceedings demanded considerably more. 4. Recapitulation. After assessing the relevant criteria and giving proper weight to the litigation's great complexity and the size of the pecuniary stakes, we find the process afforded appellants as unbalanced as a two-legged stool. The IRPAs' important property interests and the inversely limited means that the lower court afforded for representing and protecting those interests were not counterbalanced by any sufficient governmental interest. Contrary to our Concurring brother, we do not see how a district court's "equitable judgment," post at 32, no matter how familiar the Judge may be with the underlying litigation or how laudable his intentions, can displace the Due Process Clause. We are thus constrained to conclude that the IRPAs received too short shrift. In sum, the court below employed constitutionally inadequate process. C Although our Discussion to this point focuses on the attorneys' fund, we have other concerns as well. Certain individual plaintiffs also appealed and appellants' consolidated brief limns a fully developed due process claim on their behalf. In that wise, appellants contend that the district court taxed the plaintiffs' fund for PSC costs, amounting to more than $10,000,000, see supra note 8 and accompanying text, without providing a meaningful process through which individual plaintiffs could air their concerns.It would serve no useful purpose to repastinate ground that has already been thoroughly ploughed. Suffice to say that the same analysis we have used in connection with the IRPAs' claims applies ex proprio vigore to the plaintiffs' claims. First, the affected private interest is significant: each individual plaintiff has an independent pecuniary interest in the monies contained in the plaintiffs' fund; because the district court decreed that various expenses should be recouped by deductions from the plaintiffs' fund, the plaintiffs' loss of their property interest is complete; and the amount of the loss is substantial. Second, the risk of an erroneous deprivation is fairly steep: because the court relegated plaintiffs' lawyers (the IRPAs) to the sidelines, the plaintiffs' interests were, for all practical purposes, unrepresented in the hearing which determined cost reimbursement; the plaintiffs were given virtually no chance to be heard on the subject at a meaningful time and in a meaningful manner; and, moreover, some of the costs for which reimbursement was sought (and granted) were at least arguably unreimbursable. Third, muting the plaintiffs' voices served no countervailing public interest. The short of it is that the hearing procedures devised by the district court deprived the individual plaintiffs of meaningful input into the decisionmaking process by which the court decided what costs were properly to be charged against the plaintiffs' fund. Hence, because the plaintiffs, like the IRPAs, did not receive due process of law, the costs, as well as the fees, must be determined afresh.
We need go no further. We admire the district court for its unflagging efforts, effective stewardship, and mastery of the unwieldy dimensions of this massive piece of litigation. We appreciate that the court struggled doggedly to confront an increasingly contentious and fragmented dispute over the allocation of sizable attorneys' fees and substantial costs. Even so, challenging circumstances cannot excuse fundamental flaws. Because the district court, in an excess of zeal, impermissibly abridged appellants' due process rights as to the distribution of both the attorneys' and the plaintiffs' funds, we must vacate the district court's order.The order appealed from is vacated and the cause is remanded to the district court for further proceedings consistent herewith. Costs in favor of appellants.
LAY, Senior Circuit Judge, Concurring.
Although I concur in the vacation and remand to the trial court on the issue of the division of attorney fees, I do so for very different reasons. I find that the attorney fees and costs granted to the PSC are excessive as a matter of law. I express considerable reservation concerning a procedural remand for a renewed evidentiary hearing.
First, I remain fearful that a literal reading of the majority's opinion could be perceived as requiring all attorney fee applications to become "second major litigation."21 Such collateral litigation is not welcome or needed in federal courts. Permitting the IRPA to take discovery (as urged by the IRPA) and have an extensive evidentiary hearing in which they explore the hours and costs expended by PSC attorneys could result in a collateral trial of mind-boggling proportions. Judge Acosta, a worthy and experienced trial Judge, was well aware of the limitless litigation a collateral trial might produce. Obviously, he was mindful of the "property interests" of all attorneys involved. He was a close hand observer of four and one-half years of litigation and the respective roles of the some fifty-seven lawyers involved. This panel of Judges cannot begin to discern the division of labor by the lawyers from the paper briefs filed. In my judgment, the trial Judge deserves greater deference in the exercise of his management role in the division of fees than the majority opinion has afforded. His equitable judgment as to the proper procedure to be followed in the division of the attorney fees should be given controlling weight. This court must therefore balance appellants' due process arguments against the recognized procedures followed by Judge Acosta, who was intimately familiar with the underlying litigation and the division of work in that litigation.
There is no question that the hours spent by the IRPA in processing their individual clients' claims before Judge Bechtle and pursuing "boiler-plate" discovery on individual damages were infinitesimal compared to the burden carried out by the PSC's. The record clearly substantiates this fact. Judge Acosta did not need an evidentiary hearing to acknowledge what he had lived through for four and one-half years.
The IRPAs complain that Judge Acosta failed to provide an adequate hearing. The PSC members presented two full days of testimony in support of their petition for fees, informing the court on such matters as each PSC member's hourly billing rate, academic background, litigation experience and a summary of tasks performed. IRPA members were allowed to be present, but could not interrogate the witnesses or raise objections. Although a very few individual plaintiffs' attorneys requested a full-fledged hearing,22 Judge Acosta denied these requests for credible reasons. Such a hearing would have been unmanageable due to the large number of attorneys involved, tasks performed and hours logged by the PSC members and their staff.
Moreover, Judge Acosta did provide various procedural safeguards to IRPA members in order to protect their due process rights. For example, although the hearing was closed, individual IRPA members were allowed to attend, and sealed transcripts were made available to these parties upon request. The reports containing the total PSC attorney hours and expenses incurred were available to IRPA members for review throughout the case.23 Furthermore, in the event fees would be computed using a different formula than the contingency calculation, Judge Acosta requested individual plaintiffs' counsel to submit fees and expenses incurred during their preparation of the damage cases for the representative plaintiffs and also for the damage cases scheduled for the Phase II trial on damages. Finally, IRPA members were given an opportunity to file objections to the PSC's petition for fees both before the hearing and again subsequent to the hearing.
Under these circumstances, I respectfully submit that Judge Acosta's procedural approach was fair and provided the necessary process that was due to all counsel.24
I do find, however, that the fee award must be vacated on the ground that the fee is excessive and not in compliance with existing law.
Judge Acosta's fee award was made June 21, 1991. On June 24, 1992, the Supreme Court decided City of Burlington v. Dague, 120 L. Ed. 2d 449, 112 S. Ct. 2638 (1992). Although there were storm warnings along the way,25 Judge Acosta did not have the benefit of the Dague opinion. In Dague, the Court held that the lodestar amount in cases involved in federal fee-shifting statutes may not be enhanced to reflect that the prevailing party's attorneys were retained on a contingency fee basis. Id. at 2643-44. Dague involved a class action under the Solid Waste Disposal Act, 42 U.S.C. § 6972(e), and the Federal Water Pollution Control Act, 33 U.S.C. § 1365(d). Both of these statutes authorize the court to award attorney fees to a "prevailing party." The district court enhanced the lodestar amount by twenty-five percent, reasoning that the "risk of not prevailing was substantial" and that but for the opportunity for an enhancement, the class "would have faced substantial difficulty in obtaining counsel of reasonable skill and competence in this complicated field of law." Id. at 2640. The Court of Appeals for the Second Circuit affirmed the fee award, id., but the Supreme Court reversed, holding that enhancement for contingency is not permitted under the typical fee-shifting statute.
In so holding, the Court reasoned that an attorney's risk in taking a contingency case is the product of two factors: (1) the legal and factual merits of the claim, and (2) the difficulty of establishing those merits. Id. at 2641. The second factor is already included in the lodestar calculation -- a difficult case will necessarily result in a higher fee for attorneys, either because more hours are needed to perform the more difficult case, or because the hourly rate of an attorney with the requisite skill and experience will necessarily be higher. Id. Therefore, to again take this factor into account in the form of a lodestar enhancement would amount to "double-counting." Id.
The first factor (relative merits) is not considered in calculating the lodestar, but the Court noted strong policy reasons supporting its exclusion. For example, this factor exists in every claim to some extent, so courts would in effect always be adding a multiplier to a lodestar calculation in contingent-fee cases. Moreover, to reward counsel for taking actions of questionable merit would encourage attorneys to bring nonmeritous claims. Id. at 2641-42.
Finally, the Dague Court expressed its opinion that contingency enhancement is inconsistent with its general rejection of the contingent-fee model for fee awards, noting that "to engraft this feature onto the lodestar model would be to concoct a hybrid scheme that resorts to the contingent-fee model to increase a fee award but not to reduce it." Id. at 2643.
Although Dague 's holding technically only applies to contingency enhancements involving fee-shifting statutes,26 the Court's reasoning applies with equal force to multipliers in the context of equitable fund cases. Rather than relying on statutory interpretation to reach its holding, the Dague Court examined the policy reasons behind contingency enhancements and concluded that these reasons did not support their use. Courts in equitable fund cases have employed the same reasons, now discredited by the Dague Court, in deciding whether to add contingency multipliers. See, e.g., In re "Agent Orange" Product Liability Litig., 611 F. Supp. 1296, 1310-14 (E.D.N.Y. 1985) (considerations in determining whether to add a multiplier include the risk of failure, the complexity of the issues, and the demonstrated skill of the attorneys), aff'd in part, rev'd in part on other grounds, 818 F.2d 226 (2d Cir. 1987); Purdy v. Security Savings & Loan Ass'n, 727 F. Supp. 1266, 1276 (E.D. Wis. 1989) (factors included difficulty of plaintiff's case, considering the legal and factual complexity and probability of defendant's liability; and risk of loss, considering the potential for uncompensated hours and out-of-pocket expenses). For example, among the reasons given by Judge Acosta to justify a multiplier in this case was the risk of non-recovery and the quality of representation. I can discern no logical distinction between an equitable fund case and a fee-shifting case that would cause these reasons to be worthwhile in an equitable fund case, but not in a fee-shifting case.27
I note that some pre-Dague cases have found multipliers to be more appropriate in the common fund case than in the fee shifting case. See, e.g., Skelton v. General Motors Corp., 860 F.2d 250, 252-54 (7th Cir. 1988), cert. denied, 493 U.S. 810, 110 S. Ct. 53, 107 L. Ed. 2d 22 (1989). However, the fund in this case is somewhat hybrid, and the concerns in dividing the fund are more akin to the fee-shifting statutory concerns dealt with in Dague.28
First, the traditional definition of a common fund is monies recovered by a litigant or lawyer for persons other than the lawyer or his client. Boeing Co. v. Van Gemert, 444 U.S. 472, 478, 62 L. Ed. 2d 676, 100 S. Ct. 745 (1979). The "fund" thus consists of the pool of money to which the plaintiffs are entitled, less reasonable attorney fees. Here, the fund consists not of monies to which the plaintiffs are entitled, but of a sum that is to be split between the different plaintiffs' attorneys.
Moreover, the common fund doctrine rests on a theory of unjust enrichment -- persons who obtain the benefit of a lawsuit without contributing to its cost are unjustly enriched at the successful litigant's expense. See id. Here, regardless of the outcome of this case, the individual recovery of each plaintiff has been fixed and adjudicated. Furthermore, in the present case, the entire class of attorneys to which the monies are owed contributed some efforts. The question therefore is to what extent the PSC members' additional efforts should be compensated. As in a fee-shifting case, therefore, the fundamental issue in determining the reasonable attorney fee for the PSC is "the amount of attorney time reasonably expended on the litigation." See Blum v. Stenson, 465 U.S. 886, 900, 79 L. Ed. 2d 891, 104 S. Ct. 1541 n.16 (1984). I respectfully submit the lodestar method provides the fairer calculus for this determination. As Dague makes clear, use of the lodestar method in such computation should not include a multiplier for contingency reasons.29
Thus, I believe that the Dague case controls and that no multiplier may properly be added in this case. I respectfully submit that the fee is excessive as a matter of law because it includes a contingent multiplier. Accordingly, I would vacate and remand with instructions that no multiplier may be used.
I have one other reservation concerning the award of fees and costs. I respectfully submit that ordinary overhead costs which are distributed under the categories of "hard and soft" costs are ordinarily subsumed within an attorney's fees. To award overhead costs as well as an attorney fee would appear to me to double count and to allow PSC attorneys to make a profit from the award of costs. I am certain this was not intended. I refrain from being more specific since Judge Acosta will have to appraise anew this entire award. Reasonable fees for law clerks and paralegals should not include a profit margin. The original order contemplated the time of the PSC's only. Thus, I submit the court erred in allowing fees for law clerks and paralegals in excess of actual costs.
In sum, I respectfully submit that fee procedures can be equitably determined without collateral litigation requiring a more extensive evidentiary hearing than what has already been afforded here. In this respect, I do not feel Judge Acosta erred. I would, however, vacate the judgment of division of fees for the PSC's for reconsideration under Dague and the other concerns discussed.
* Of the Eighth Circuit, sitting by designation.
** Of the Ninth Circuit, sitting by designation.
1. We refer readers interested in the arson and the course of the ensuing litigation to the plethora of opinions which grace the federal reports. See, e.g., In re San Juan Dupont Plaza Hotel Fire Litig., 888 F.2d 940 (1st Cir. 1989); S.C., 789 F. Supp. 1212 (D.P.R. 1992); S.C., 768 F. Supp. 912 (D.P.R. 1991); S.C., 745 F. Supp. 79 (D.P.R. 1990); see also Recticel, supra, 859 F.2d 1000.
2. To prevent down time during the selection process, Judge Acosta appointed an interim plaintiffs' steering committee (the IPIC). Certain IPIC members were not appointed to the PSC but were ultimately awarded fees and costs for IPIC-related work. In addition, Judge Acosta named an alternate member of the PSC.
3. The final phase of the liability inquiry, popularly called "Phase III," is aimed at determining the contractual liability of various insurers. Phase III was underway when these appeals were argued before us.
4. A few of the contingency agreements appear to be for less than the maximum amounts allowed by Puerto Rico law. We need not concern ourselves with this phenomenon.
5. The district court's power to award attorneys' fees through an application of the common fund doctrine is derived from its traditional power to do equity. See Boeing, 444 U.S. at 478; Weinberger v. Great N. Nekoosa Corp., 925 F.2d 518, 523 (1st Cir. 1991).
6. The district court characterized the professor's presentation as akin to the testimony of a friend of the court. We find this characterization to be wholly insupportable. The hearing transcript makes it clear that the professor appeared as a paid advocate of the PSC minority's parochial interests.
7. The former would anticipate receiving the bulk of their compensation from PSC fees while the latter would anticipate receiving most of their compensation from their clients' cases.
8. Insofar as costs are concerned, the total charge-back to the plaintiffs' fund was in excess of $16,000,000. Of this amount, however, $3,563,575 represented a contingency set-aside reserved for future allocation; slightly over $1,500,000 represented trustees' fees; $766,700 represented reimbursement of cost assessments previously levied by the court to fund ongoing expenses of the litigation; and close to $900,000 represented miscellaneous charges, such as PSC office expenses and local taxes.
9. The other statutory bases upon which appellate jurisdiction may sometimes rest, see, e.g., 28 U.S.C. § 1292 (1988), are inapposite here.
10. We find added support for the exertion of appellate jurisdiction in the line of cases holding that a merits appeal may be taken (and indeed, must be taken in some instances) despite the fact that applications for counsel fees are unresolved. See, e.g., Budinich v. Becton Dickinson & Co., 486 U.S. 196, 202, 100 L. Ed. 2d 178, 108 S. Ct. 1717 (1988); White v. New Hampshire Dep't of Employment Sec., 455 U.S. 445, 451-53, 102 S. Ct. 1162, 71 L. Ed. 2d 325 & n.14 (1982); Crossman v. Maccoccio, 792 F.2d 1, 3 (1st Cir. 1986) (per curiam). Because this rule rests on the premise that orders setting attorneys' fees are separate from determinations on the merits in a jurisdictionally significant sense, it is a logical corollary that, where a dollar-specific order for attorneys' fees has been entered and further action on the main case will not require revisiting that order, an appeal of the fee award may be prosecuted prior to the very end of the litigation.
11. The district court prudently reserved a contingency set-aside of $3,563,575 from the Phase II funds. See supra note 8. To the extent the reserve is not consumed by the expenses of Phase III, the district court will presumably allocate it among the plaintiffs -- but the court could conceivably award some part of the unexpended reserve to counsel. Although we mention this fact for the sake of completeness, we do not think it detracts in the slightest from the definitiveness of the order at issue here.
12. Appellees argue that two of the appeals were filed out of time. It is, however, undisputed that all nineteen appeals raise the same substantive questions and that seventeen of them were timely filed. Since we believe that the order below must be rescinded and a new fee-determination proceeding conducted, see infra, we need not spin wheels mulling the particulars of the last two notices of appeal. When, as now, the resolution of appeals that are properly before us produces remediation necessarily affecting all putative appellants, whether or not they are parties to those appeals, the procedural regularity of any essentially duplicative appeals is of no moment. See, e.g., Marquis v. FDIC, 965 F.2d 1148, 1151 n.2 (1st Cir. 1992) (where all appeals raise the same substantive point, the court of appeals, after satisfying itself that some are properly before it, need not trace the provenance of the remaining appeals).
13. For simplicity's sake, we refer here only to the IRPAs' stake and the differences of opinion concerning fees (as opposed to costs). Nevertheless, the same analysis applies to the individual plaintiffs and the lower court's subtraction of costs from the plaintiffs' fund. See infra Part III(C).
14. Judge Acosta's initial estimate of a ten percent ceiling for PSC compensation did not itself appear unusual. See, e.g., In re MGM Grand Hotel Fire Litig., 660 F. Supp. at 529 (awarding seven percent of the recovery pool as steering committee compensation).
15. The proposal suggested that the court use the lodestar method, a time-and-rate-based framework for setting counsel fees. The specifics of the lodestar method have been frequently recounted elsewhere, see, e.g., Blum v. Stenson, 465 U.S. 886, 896-902, 79 L. Ed. 2d 891, 104 S. Ct. 1541 (1988); Lipsett v. Blanco, 975 F.2d 934, 937 (1st Cir. 1992); Metropolitan Dist. Comm'n, 847 F.2d at 19, and need not be rehearsed here.
16. The standard form promulgated by the district court, reproduced in the appendix, required each objecting IRPA to specify the total number of hours he had worked on his clients' cases.
17. Appellees argue that any failure of notice was inconsequential because Judge Acosta eventually couched the PSC fee award in terms of a percentage. We agree that, after calculating the award by using lodestar techniques, Judge Acosta then translated it into a percentage. Notwithstanding this effort to touch all bases, it is pellucid that the court relied on the lodestar method to reach a dollar figure, see Fees Op., 768 F. Supp. at 929 n.49 (rejecting percentage suggestions on the ground that they did not encompass "the number of hours, value for various tasks performed, etc."), and then paid lip service to the percentage method. It beggars credulity to argue that this bit of adroit wordplay fulfilled the court's original promise to set the PSC's fees on a percentage basis. Form cannot be allowed to triumph over substance. Because the court posted a different menu when the PSC was formed, due process precludes the court from effectively scrapping the menu by the simple expedient of preparing a lodestar stew and serving it up under another name.
18. The strictures on IRPA participation also likely contributed to Judge Acosta's failure to compensate those IRPAs who conducted the representative trials for their work in that capacity. Although Judge Acosta did ask for, and receive, an accounting from these IRPAs, he did not allow them to participate in the fee hearing. In the end, despite the evident contributions made by these IRPAs in undertaking the representative trials, the Judge "tacitly denied" them special fees, splitting the attorneys' fund between the PSC and the IRPAs generally. Order 402 (Oct. 16, 1991). Moreover, Judge Acosta awarded those counsels' costs to the PSC as part of the PSC cost award. Fees Op., 768 F. Supp. at 934 n.74. The fate that befell these IRPAs is illustrative of the hazards that an adjudicator faces in listening to only one side of a multi-sided story.
19. Although we do not reach the question of whether the district court abused its discretion in the particulars of its fee award, we note that the court's belated shift from a percentage approach to a lodestar approach raises serious questions in this respect. In justifiable reliance on the court's original declaration that it would calculate PSC fees on a percentage basis, see Order 2 (March 18, 1987), the majority of the IRPAs did not maintain time records. Thus, despite our brother's contrary view, post at 34, we cannot see a practical method by which to compare PSC members' time to IRPAs' time on the paper record alone. In view of this seemingly unavoidable impasse, we think that the lower court, at the very least, should have heard the IRPAs' views on the advisability of so basic a change in the method of PSC compensation.
20. There are indications in the record below that the district court viewed its task solely in terms of fixing suitable compensation for the PSC. Judge Lay seemingly falls into the same trap. See post at 33 n.22. In our view, such a focus distorts the reality of the situation. Because the total amount available for counsel fees was limited, the court necessarily determined the amount due to the IRPAs when it determined the amount due to the PSC members. Indeed, a trier who attempted punctiliously to follow the classic lodestar formula, to the exclusion of all else, could theoretically wind up awarding the entire fee pool to the PSC, leaving nothing for the IRPAs. Surely, that result is completely indefensible.