concurring.
I join wholeheartedly Judge Garth’s conclusion that the Hague appeal should be allowed notwithstanding its untimely filing. I also agree that each of these appeals must be dismissed for want of appellate jurisdiction and that the circumstances do not warrant relief by way of mandamus. I write separately, however, to highlight certain considerations, though not present here, that I believe would have permitted appellate review. Moreover, because the majority opinion by necessity stops at the jurisdictional gate, the District Court lacks our Court’s comment on the fee award issues. I thus write as but one voice that risks regard as simply a pundit without portfolio.
I. Rule 54(b) Certification
As a threshold matter and as Judge Garth emphasizes, we must be satisfied that we have jurisdiction to hear these appeals. Metro Transp. Co. v. N. Star Reinsurance Co., 912 F.2d 672, 676 (3d Cir.1990). “This Court’s appellate jurisdiction is conferred and limited by Congress’s grant of authority.” Berckeley Inv. Grp. Ltd. v. Colkitt, 259 F.3d 135, 139 (3d Cir.2001) (citing Sheldon v. Sill, 49 U.S. 441, 449, 8 How. 441, 12 L.Ed. 1147 (1850)). Under 28 U.S.C. § 1291, our jurisdiction is limited to “final decisions” of the district courts. Here we are dealing with an award and allocation of counsel fees, embodied in Pretrial Orders Nos. 2622 and 2859, that the District Court designated as “interim.” We have held that a denial of an interim award of attorneys’ fees is not final within the meaning of § 1291. Yakowicz v. Commonwealth of Pennsylvania, 683 F.2d 778, 782 (3d Cir.1982). Accord, e.g., Shipes v. Trinity Indus., Inc., 883 F.2d 339, 341 (5th Cir.1989).
Although an order that disposes of “fewer than all of the claims or the rights and *162liability of fewer than all the parties” is normally not appealable, an exception to the general rule exists when an order is certified as appealable by a district court pursuant to Federal Rule of Civil Procedure 54(b).29 That is, Rule 54(b) “per-mitís] the district court to separate out final decisions from non-final decisions in multiple party and/or multiple claim litigation” in order to allow immediate appeal. Weiss v. York Hosp., 745 F.2d 786, 802 (3d Cir.1984); see also Allis-Chalmers Corp. v. Philadelphia Electric Co., 521 F.2d 360, 363 (3d Cir.1975) (explaining that Rule 54(b) “attempts to strike a balance between the undesirability of piecemeal appeals and the need for making review available at a time that best serves the needs of the parties”); Bendix Aviation Corp. v. Glass, 195 F.2d 267, 269 (3d Cir.1952) (explaining that “in a multiple claims case the judgment which finally adjudicates all the claims is the only judgment having finality unless [the district] court in its discretion enters a final judgment pursuant to [Rule 54(b) ]”).
As Judge Garth points out, in this case no party sought the District Court’s certification and therefore this avenue cannot provide a.basis for our jurisdiction. Nevertheless, the issue of Rule 54(b) certification deserves further discussion in the context of our case. While Judge Garth’s statement that the appeals “charg[e] essentially that the district court abused its discretion in awarding and allocating an interim award of attorneys’ fees” is accurate, what is potentially at stake is both far-reaching and nuanced in the context of this ease- — -an appropriate allocation of compensation to counsel in a cutting-edge class action. Not only is the Nationwide Class Action Settlement Agreement (the “Settlement Agreement”) of record-setting scale and scope, but it also contains numerous innovative features that have potential significance for future class actions. See generally Richard A. Nagareda, Autonomy, Peace, and Put’ Options in the Mass Tort Class Action, 115 Harv. L.Rev. 747 (2002). Additionally, under the Settlement Agreement, payments- can be made to class members who develop serious levels of valvular heart disease at any time for years to come, specifically until December 31, 2015. While it does not appear that the District Court intends to wait that long before entering a final fee award, it remains useful to explore considerations that would allow for appellate review in complex class action litigation in order to work around requirements that might lead to substantial delay.
A.
A district court’s Rule 54(b) certification is necessarily predicated on its affirmative answer to two questions: is the judgment final and is it ready for appeal. Gerardi v. Pelullo, 16 F.3d 1363, 1368 (3d Cir.1994). Thus, in certifying an order for appeal under Rule 54(b), a court must first decide whether it is dealing with a determination that is final for purposes of certification. Curtiss-Wright Corp. v. Gen. Elec. Co., 446 U.S. 1, 7, 100 S.Ct. 1460, 64 L.Ed.2d 1 (1980). “Finality is defined by the requirements of 28 U.S.C. § 1291, which are generally described as ‘ending the litigation on the merits and leaving nothing for the court to do but execute the judg*163ment.’ ” Gerardi, 16 F.3d at 1369 (quoting Gulfstream Aerospace Corp. v. Mayacamas Corp., 485 U.S. 271, 275, 108 S.Ct. 1133, 99 L.Ed.2d 296 (1988) (quoting Catlin v. United States, 324 U.S. 229, 233, 65 S.Ct. 631, 89 L.Ed. 911 (1945))). Rule 54(b) does not alter this definition, but allows a judgment to be entered if the order has the requisite degree of finality as to a specific claim in a multi-claim action or a specific party in a multi-party action. Sussex Drug Prods. v. Kanasco, Ltd., 920 F.2d 1150, 1153-54 (3d Cir.1990). “Although a district court has discretion in certifying a judgment for appeal under Rule 54(b), the district court cannot, in its exercise of its discretion, treat as ‘final’ that which is not ‘final’ within the meaning of [28 U.S.C.] § 1291.” Waldorf v. Shuta, 142 F.3d 601, 611 (3d Cir.1998) (quotation omitted). While the Rule “allows immediate appeal of separate disputes comprised within a larger litigation[,] ... [i]t does not, however, allow appeal ... when the district court will revisit the issues.” Trs. of Chicago Truck Drivers v. Cent. Trans., Inc., 935 F.2d 114, 116 (7th Cir.1991).
It appears that four of the appeals before us would be sufficiently final for purposes of the Rule. The appeals in Benjamin, 30 Fleming,31 Gooch-Kiel,32 and Hague33 challenge the MDL 1203 Fee & Cost Account assessments—assessments that funded, in part, the interim award of counsel fees. Under Pretrial Order No. 467, a percentage of all payments made by the defendant in settlements in cases transferred to MDL 1203 was to be paid into the MDL 1203 Fee & Cost Account out of individual attorneys’ share of their clients’ recoveries. In addition, under Pretrial Order No. 517, state actions could become eligible for state-federal coordination provided that, among other things, an assessment would be sequestered for the MDL 1203 Fee & Cost Account. Initially, the percentage for the assessments in federal actions was set at 9% and the state-coordinated action was set at 6%; these percentages were later reduced to 6% and 4%, respectively, in Pretrial Order No. 2622. The difference was refunded to counsel who had paid the higher assessments, leaving the MDL 1203 Fee & Cost Account nearly spent. The sequestered funds have been paid out to counsel in connection with attorneys’ fees awarded in Pretrial Order No. 2622 and allocated among counsel as specified in Pretrial Order No. 2859.
The Benjamin, Fleming, Gooch-Kiel, and Hague Appellants are contesting the District Court’s allocation of funds that have been paid out to the recipients of the attorneys’ fee award. It appears that these Appellants are no longer involved in the MDL proceedings. If so, and if they had requested certification under Rule 54(b), the assessments levied against the Benjamin, Fleming, Gooch-Kiel, and Hague Appellants appear to be sufficiently final to permit that certification.34
*164B.
Once having found finality, a district court must then determine whether the judgment is ready for appeal, or put differently, whether there is any just reason for delay. Curtiss-Wright Corp. v. Gen. Elec. Co., 446 U.S. 1, 8, 100 S.Ct. 1460, 64 L.Ed.2d 1 (1980). The decision to certify a final judgment under Rule 54(b) is committed to the discretion of the district court, taking into account the interest of sound judicial administration as well as the equities of the case. Id.; Sears, Roebuck & Co. v. Mackey, 351 U.S. 427, 437, 76 S.Ct. 895, 100 L.Ed. 1297 (1956).
In our Circuit, Allis-Chalmers Corp. v. Philadelphia Electric Co., 521 F.2d 360 (3d Cir.1975), instructs district courts to consider the following factors in deciding whether to grant Rule 54(b) certification: (1) the relationship between the adjudicated and non-adjudicated claims; (2) the possibility that the need for review might be mooted by future developments; (3) the possibility that the reviewing court might be obliged to consider the same issue a second time; (4) the presence or absence of a claim or counterclaim that could result in set-off against the judgment sought to be made final; and (5) miscellaneous factors such as delay, economic and solvency considerations, shortening the time of the trial, frivolity of competing claims, expense and the like. Id. at 363. “[Djepending upon the facts of the particular case, all or some of the above factors may bear upon the trial court’s discretion in certifying a judgment as final under Rule 54(b).” Waldorf v. Shuta, 142 F.3d 601, 609 (3d Cir.1998)
Here, the first factor — the relationship between the adjudicated and non-adjudicated claims — favors certification because the parties’ claims on the merits have been resolved by settlement. Similarly, looking ahead to the fourth factor, the absence of any pending counterclaim also favors certification.
The second factor — the possibility that the need for review might be mooted by future developments — is less clear-cut. That is, more than $25 million in additional assessments have been deposited into the MDL 1203 Fee & Cost Account since the entry of Pretrial Order No. 2622, and it is estimated that $60 million will be added to that account. Additionally, in Pretrial Order No. 2622, the District Court declined to perform a Gunter analysis until certain issues surrounding' of the Settlement Agreement are resolved. See Pretrial Order No. 2622 at 24-26 (explaining that “[qjuestions regarding the value of the settlement ... clearly remain” and that the court “will be in a better position to make a final fee award to Class Counsel” after the resolution of “pressing issues” surrounding the administration of the Settlement Agreement). At the time such a final award is made, additional funds may be distributed from the MDL 1203 Fee & Cost Account to counsel that performed work for the benefit of the class. It is also possible that some funds will be returned to the attorneys who have been assessed a percentage of their recoveries.
Nevertheless, the fact that additional funds have been deposited into the MDL 1203 Fee & Cost Account since the entry of Pretrial Order No. 2622 does not preclude Rule 54(b) certification. Even in the event that the assessments are reduced to a lower percentage — an event that is by no means certain and, indeed, is not even expressly contemplated by the District Court in Pretrial Order Nos. 2622 or 2859 — the Benjamin, Fleming, Gooch-Kiel, and Hague Appellants are objecting to the fact that they were subject to any assessment whatsoever. There is no plausible scenario under which their liability would reduce to zero or an amount ap*165proaching zero. In this respect, there is not a great possibility that future developments will moot the issues raised by the four Appellants.
With respect to the third factor, there does not appear to be a significant possibility that the same issues would be presented for review a second time. There is some overlap among the issues raised in the Benjamin, Fleming, Gooch-Kiel, and Hague appeals. They all challenge the District Court’s finding that they benefit-ted from the work of the Plaintiffs’ Management Committee (“PMC”).35 Given that this issue has been raised for appellate review in only a relatively small number of the total cases in which funds were sequestered for the MDL Fee & Cost Account, there is not a high probability that this argument would be raised again.
Lastly, the miscellaneous factor of delay argues (albeit slightly) in favor of review. For the reasons discussed above and in Judge Garth’s opinion, it is not yet known when a final fee award will be rendered. It has already been more than two years since the District Court entered Pretrial Order No. 2622. Even though it appears that the final fee award will be made well before 2016, the four Appellants face delay that is not insignificant.
Viewing all of the Allis-Chalmers factors together, this case is one in which a court could find Rule 54(b) certification is warranted. Assuming that the settlement remains sound — a matter that the District Court is much better-situated to assess— the resolution of these issues by our Court would be consistent with Rule 54(b)’s policy of striking a “balance between the undesirability of piecemeal appeals and the need for making review available at a time that best serves the needs of the parties.” Allis-Chalmers Corp., 521 F.2d at 363.
II. Collateral Order Doctrine
As an avenue apart from Rule 54(b) certification, interlocutory orders may be immediately appealed if the order falls within the narrow confines of the collateral order doctrine.36 First brought into play by the Supreme Court in Cohen v. Beneficial Industrial Loan Corp., 337 U.S. 541, 69 S.Ct. 1221, 93 L.Ed. 1528 (1949), the collateral order doctrine provides a narrow exception to the general rule permitting appellate review only of final orders. In re Ford Motor Co., 110 F.3d 954, 958 (3d Cir.1997). Under this doctrine, an appeal of a nonfinal order is appropriate if: (1) the order from which the appellant appeals conclusively determines the disputed question; (2) the order resolves an important issue that is completely separate from the merits of the dispute; and (3) the order is effectively unreviewable on appeal from a final judgment. Id.; see also United *166States v. Bertoli, 994 F.2d 1002, 1010 (3d Cir.1993) (“The flexibility given by Cohen, commonly called the collateral order doctrine, permits appeal of some district court orders that do not terminate the entire case, or even a discrete part of it.”)- If the order at issue fails to satisfy any one of the three prongs, it is not an appealable collateral order. Gulfstream Aerospace Corp. v. Mayacamas Corp., 485 U.S. 271, 276, 108 S.Ct. 1133, 99 L.Ed.2d 296 (1988). Courts have reached mixed conclusions with respect to whether interim fee awards are appealable under the collateral order doctrine. See Dardar v. Lafourche Realty Co., 849 F.2d 955, 957 n. 8 (5th Cir.1988) (collecting cases).
A.
Under the first prong of the collateral order doctrine test — labeled the “conclusiveness prong” — the order appealed must “finally resolve a disputed question.” Praxis Properties, Inc. v. Colonial Sav. Bank, S.L.A., 947 F.2d 49, 54 (3d Cir.1991). In determining this, the Supreme Court has contrasted two types of orders: those that are “inherently tentative” and those that are “technically amendable, but made with the expectation that they will be the final word on the subject addressed.” Christy v. Horn, 115 F.3d 201, 204 (3d Cir.1997) (citing Gulfstream Aerospace Corp. v. Mayacamas Corp., 485 U.S. 271, 277, 108 S.Ct. 1133, 99 L.Ed.2d 296 (1988)).
The conclusion that the conclusiveness prong is not met in our case would be far less certain, however, if the District Court had not indicated a willingness to make a final fee award when the “entire picture is less clouded,” Pretrial Order No. 2622 at 25, which suggests that the fees will be awarded when practical — presumably well before 2016. Though there is scant indication it intends to do so, the Court is free to revisit its allocation of fees through June 30, 2001. Moreover, it has expressly declined to make a final fee award so that the Gunter factors may be applied. In this context, while I do not share my colleagues’ view that the conclusiveness prong is clearly not satisfied, the current state-of-play nonetheless tilts to the conclusion that conclusiveness does not yet exist.37
B.
The case law addressing whether orders respecting interim fee awards fit within the collateral order doctrine places considerable weight on the fact that the fee awards can generally be reviewed at the conclusion of the litigation in the district court. However, an interim fee order may be reviewable when the “mere payment of the fees would make them unrecoverable.” Ruiz v. Estelle, 609 F.2d 118, 119 (5th Cir.1980); see also, e.g., Palmer v. City of *167Chicago, 806 F.2d 1316, 1319-20 (7th Cir.1986).
The Seventh Circuit Court of Appeals has addressed the unreviewability prong of the collateral order doctrine in the context of interim fee awards in a series of cases beginning with Palmer, in which the Court held that an interim fee award was appeal-able when fees were to be paid not to attorneys but to a “revolving fund” for prisoner-plaintiffs. 806 F.2d at 1319-20; see also, e.g., People Who Care v. Rockford Bd. of Educ., 171 F.3d 1083, 1086 (7th Cir.1999); Constr. Indus. Ret. Fund v. Kasper Trucking, Inc., 10 F.3d 465, 468 (7th Cir.1993); People Who Care v. Rockford Bd. of Educ., 921 F.2d 132, 134-35 (7th Cir.1991); Richardson v. Penfold, 900 F.2d 116, 117-18 (7th Cir.1990). In a subsequent case, the Seventh Circuit indicated that although interim awards “are not final in the traditional sense ... [,] they are appealable under the collateral order doctrine when the defendant may have difficulty getting the money back.” People Who Care, 921 F.2d at 134. In this respect, the crucial consideration in determining that an order is immediately ap-pealable is whether “postponing appellate review till the end of the case would cause substantial irreparable harm to the party against whom the order was directed.” Palmer, 806 F.2d at 1319.
My colleagues emphasize that the fee distributions at issue in this case have been made to counsel instead of to the parties. Cf. id. at 1319 (explaining that if the “fees would have been disbursed to the lawyers rather than retained by the prisoners and defendants, the problem would be less serious.... [W]e assume that the district court has an inherent power to order attorneys to whom fees were paid over by their clients pursuant to court order to repay the fees should the order be reversed.”). Although that consideration is relevant in the short-term, in the long-run this fact becomes a less compelling basis for finding that the unreviewability prong is not met. Whether the funds are distributed to an attorney or to a private litigant, it is difficult to maintain a high degree of confidence that either will remain available to be returned for re-distribution if the final tabulation were not to be made for more than a decade. For this reason, if it did not appear that a final distribution will occur in the relatively near future, this case, as a practical matter, would fall much closer to the situation in Palmer, where review was necessary to avoid irreparable harm.
III. Fee Allocation Procedures
The District Court described the task of allocating $160 million in counsel fees as “herculean.” Pretrial Order No. 2859 at 5. This description was apt. By the time the Court entered the order allocating interim fees, the litigation had spanned more than five years, produced more than 2800 orders, and resulted in a complex Settlement Agreement that had been amended multiple times. See id. In light of the size of this task and our Court’s prior lack of exploration of the issues involved in such an allocation, discussion of the merits follows. The discussion in this section is not intended to express a view on the correctness of the actual allocation among counsel (save my comment in B.l below), but rather addresses the procedure by which the allocation was rendered.
A.
As noted, our Court has offered little guidance on how fees should be allocated among counsel in MDL class actions. The most direct guidance came in a footnote in which we posited that the approach of allowing lead counsel to allocate and distribute counsel fees among various law *168firms frees district courts from “under* tak[ing] the difficult task of assessing counsels’ relative contributions.” In re Prudential Ins. Co. of Am. Sales Litig., 148 F.3d 283, 329 n. 96 (3d Cir.1998).
Perhaps implicitly acknowledging the lack of detailed guidance from our Court, Appellees cite a number of decisions in which courts have delegated the task of allocating fees among counsel to lead counsel or have relied on an agreement reached by counsel. See, e.g., In re Linerboard Antitrust Litig., 333 F.Supp.2d 343, 351 (E.D.Pa.2004); In re Copley, 50 F.Supp.2d 1141, 1147-50 (D.Wyo.1999); In re Indigo Sec. Litig., 995 F.Supp. 233, 234 (D.Mass.1998); In re Magic Marker Sec. Litig., 1979 WL 1248 (E.D.Pa. Sept.16, 1979), 1979 U.S. Dist. LEXIS 9777; In re Ampicillin Antitrust Litig., 81 F.R.D. 395, 400 (D.D.C.1978); DelNoce v. Delyar Corp., 457 F.Supp. 1051, 1055 (S.D.N.Y.1978). In one of the earlier decisions to address these issues, the judge went as far as to say that “it is virtually impossible for the Court to determine as accurately as can the attorneys themselves the. internal distribution of work, responsibility and risk.” In re Ampicillin Antitrust Litig., 81 F.R.D. at 400. He then accepted the “unanimous position of [the] attorneys ... that the Court should take no part in the ultimate division of any fee awarded ... [and] defer[ed] to the attorneys’ request that the fee award be made to” the committee of counsel for the settling class. Id. More recently, a court has justified delegating the task of formulating a proposed fee allocation as follows:
Attorney fee allocation is an unenviable task for any court. It is a difficult matter that, frankly, even the trial court is often not in the best position to decide. This is especially true in complex class ■ actions, like the one at bar. In such a circumstance, ideally, allocation is a private matter to be handled among class counsel. The rationale for this policy is both logical and practical. Class counsel are better able to decide the weight and merit of each other’s contributions.
In re Copley, 50 F.Supp.2d at 1148 (quotations and citations omitted).
Against this background and in view of the slowly emerging consensus (or, at least, trend) that it is difficult for courts to assess the contribution of various counsel to the litigation, the District Court here decided to create a five-member Fee & Cost Allocation Committee (the “Allocation Committee”). Three of the five attorneys on the Allocation Committee are members of the PMC, the body of attorneys that oversaw the coordinated and consolidated pretrial proceedings and conducted discovery of widespread applicability on behalf of plaintiffs in this multidistrict litigation. Two of the five members of the Committee are partners in the same law firm — Levin, Fishbein, Sedran and Berman (“Levin, Fishbein”).
Under Pretrial Order No. 2622, the Allocation Committee had forty-five days to propose an allocation of the $160 million. It held meetings during that period in secret and in late November 2002 issued a report with its proposed allocation. That proposal set aside approximately $28.7 million from the interim class fee award for counsel in certified class actions in various state courts and certain other attorneys who had performed services that contributed to class recovery and were entitled only to recover from the Fund A Legal Fee Escrow Account (“Fund A”) or the Fund B Legal Fee Escrow Account (“Fund B”).
The Allocation Committee then combined the remaining fees available from Fund A with those funds approved for the payment of fees from Fund B and the MDL 1203 Fee & Cost Account. These *169funds totaled approximately $131 million. The Allocation Committee formulated a plan to allocate this amount to more than two dozen law firms entitled to share in both the interim award of class fees and the interim award from the MDL 1203 Fee & Cost Account. In arriving at its proposed allocation, the Allocation Committee considered the relative contribution of each MDL Firm to various stages of the litigation. Within each stage of the litigation, it considered certain factors to determine the contribution of each firm, including, among other things, the: (1) quality of the work performed and relative skill and efficiency of the attorneys involved; (2) duration and intensity of the firm’s commitment to the litigation; (3) level at which firm partners participated in the litigation; and (4) extent to which the firm was engaged in the litigation for the common benefit of the class members independent of any case specific recoveries. See Pretrial Order No. 2859 at 6-7. The Allocation Committee also examined each firm’s reported lodestar38 as determined by the court-appointed auditor. Generally, it considered the lodestar for each firm through June 30, 2001, but added over $6.3 million to the lodestar for the Levin, Fishbein firm to reflect time expended on matters up to September 30, 2002. After weighing the above criteria, and reviewing the various lodestars, the Allocation Committee measured the relative contribution of each firm and quantified that contribution by assigning the firm a percentage of the interim award.
A number of firms — including the Bloom, Lopez-Hodes, and Nisen & Elliott Appellants — filed objections to the Allocation Committee’s proposed allocation with the District Court. In May 2003, in Pretrial Order No. 2859 the District Court approved the Committee’s fee allocation with the exception of modifying the award to exclude the Levin, Fishbein fee award for time incurred after June 30, 2001. See Pretrial Order No. 2859 at 22-23. As a result of the order, 52% of the MDL portion of the fee award was allocated to three law firms to which four out of five Allocation Committee members belong, and one firm (with two of its attorneys on the Allocation Committee) received nearly $58 million of the $131 million allocated to MDL firms. The Bloom, Lopez-Hodes, and Nisen & Elliott Appellants essentially challenge, inter alia, the District Court’s almost complete approval of the Committee’s allocation.
B.
A review of the various decisions addressing the allocation of attorneys’ fees among counsel reveals two competing lines of analysis — the delegation approach and the reexamination approach. The former rests generally on practical considerations and stems from decisions which, for the most part, have not involved numerous serious objections to both the outcome of the fee allocation and the procedure from which the fee allocation was set.
Although the delegation approach has gained acceptance, it is not beyond criticism. Significantly, along the line of analysis of the reexamination approach, the Second Circuit Court of Appeals has acknowledged that there is “authority for a court, under certain circumstances, to award a lump sum fee to class counsel in an equitable fund action under the lodestar approach and then to permit counsel to divide this lodestar-based fee among themselves under the terms of a private fee sharing agreement....” In re “Agent *170Orange” Prod. Liab. Litig., 818 F.2d 216, 223 (2d Cir.1987). The Second Circuit rejected this authority, however,
to the extent it allows counsel to divide the award among themselves in any manner they deem satisfactory under a private fee sharing agreement. Such a division overlooks the district court’s role as protector of class interests under [Federal Rule of Civil Procedure] 23(e) and its role of assuring reasonableness in the awarding of fees in equitable fund cases.
Id.
With these considerations in mind, there are at least three ways in which the fee allocation here may be cause for concern. These concerns, however, can be addressed in a way that serves the court’s role as protector of class interests without abandoning the approach of looking to the views of counsel for assistance.
1. Exclusion of Unaudited Time
The District Court’s sole disagreement with the Allocation Committee’s proposed apportioning to the MDL firms was the latter’s inclusion of approximately $6.3 million in unaudited time in the proposed award to Levin, Fishbein. After indicating that it had reviewed the award to Levin, Fishbein with “special care” because of the “potential for unfairness,” the District Court excluded the unaudited time out of an “abundance of caution” so that the “allotments [would] all involve approximately the same time period.” Pretrial Order No. 2859 at 22-23.
The decision simply to reduce the amount of Levin, Fishbein’s award (and the total amount of the interim fee award) by the lodestar amount of $6.3 million is inconsistent with the manner in which the Allocation Committee arrived at its proposed fee allocation. That is, it did not recommend an award of fees simply on the basis of the lodestar. Instead, after weighing the various factors, the Committee allocated a percentage of the total fee award to each firm. Levin, Fishbein received 44% of the fees, which was approximately 2.38 times the amount of its lodestar. Arguably, then, the District Court should have reduced Levin, Fishbein’s award by 2.38 times the lodestar sum of $6.3 million (that is, by approximately $15 million) or reassessed Levin, Fishbein’s contribution or directed the Committee (perhaps absent its Levin, Fishbein members) to reconsider its allocation recommendation.
2. Consultation with the State Judges’ Committee
Evidently in view of certain legal requirements under Illinois law limiting the recovery of medical monitoring costs, the District Court excluded users of diet drugs residing in Illinois from the nationwide certified class. See Pretrial Order No. 865. A state-wide class was certified in Illinois state court in December 1998, several months prior to the initiation of “global” settlement discussions and prior to the certification of the nationwide federal class in August 1999. The Settlement Agreement, entered into in November 1999, provided, inter alia, that class members were entitled to: (1) reimbursement if they obtained private echocardiograms prior to the implementation of the settlement; and (2) refunds for the diet drugs they purchased. These remedies had been pursued by the Nisen & Elliott Appellants with respect to the class certified in Illinois state court.
As a means of assisting the District Court in matters pertaining to the settlement and the award of counsel fees, the Settlement Agreement called for the creation of a judges’ committee:
*171A State Court Judicial Advisory Committee ... will consist of the judges from the State Courts which, as of October 7, 1999, had issued any order certifying state-wide class actions in relation to the effects of Pondimin and/or Redux. The State Court Judicial Advisory Committee shall provide advice and counsel on all matters pertinent to the Settlement. ... In addition, prior to making any award of counsel fees and reimbursement of litigation expenses, the Federal District Court shall consult with and give substantial deference to the views of the State Court Judicial Advisory Committee■ concerning the actual contribution which was made to the overall resolution of the litigation by the attorneys with whom the members of the committee are familiar.
(emphasis added). The District Court established the State Court Judicial Advisory Committee (the “State Judges’ Committee”), and it met on several occasions with the District Court prior to June 2001. Following the District Court’s approval of the Settlement Agreement, attorneys from the MDL firms and attorneys representing the class' actions pending in various state courts filed a joint petition for an award of attorneys’ fees, collectively requesting more than $400 million. Before the joint petition was filed, a private fee-sharing agreement had been negotiated by attorneys representing class actions in Texas, New York, New Jersey, Pennsylvania, and West Virginia. The Nisen & Elliott Appellants in Illinois39 were not parties to that agreement. Under it, the attorneys representing the class actions in those five states would receive approximately 97% of fees potentially available for allocation to state court counsel. The Allocation Committee recommended an award to the Ni-sen & Elliott Appellants that they objected to as significantly understating their contribution • to the nationwide settlement, which included the remedies noted above.
The District Court rejected the Nisen & Elliott Appellants’ argument:
The real issue -is whether the allocation proposed by the [Allocation] Committee for Illinois counsel ... is fair and reasonable. Ultimately, the Committee analyzed Illinois counsel’s participation in this litigation as. it had the other firms, by considering their relative contribution to the overall outcome of the litigation. In the context of the entire litigation, the efforts of the Illinois firms, though valuable and inuring to the common benefit of the Class, were limited. Outside of the state class certification in Illinois, these firms 'performed little, if any, work on this matter.
Pretrial Order No. 2859 at 31-32 (footnote omitted). Thus the District Court found that the Allocation' Committee’s recommended award was fair arid reasonable without mentioning the State Judges’ Committee, notwithstanding that the Ni-sen & Elliott Appellants had requested that the District Court consult with the State Judges’ Committee.
Moreover, in a letter that was dated just two days before the District Court’s hearing on the recommended fee allocation, a member of the State Judges’ Committee— Justice Ellis Reid of the Appellate Court of Illinois, First District (who had presided over the Illinois diet drug class action when he was the Circuit Court Judge for the Circuit Court of Cook County, Illinois) — wrote to the District Court. The letter pointedly questioned the District Court’s procedures with respect to fee allocations:
*172As a member of the State Court Judicial Advisory Committee it was my understanding, based upon the provisions of the Nationwide Settlement Agreement, that my views concerning the contributions made by Illinois [counsel] would be given substantial deference in any award of counsel fees to the state court attorneys who represented the six statewide class actions certified prior to the Nationwide Settlement. I am writing to express my concerns that this has not happened and, as a result, certain Illinois class counsel are being prejudiced.
Justice Reid further questioned the appropriateness of the Allocation Committee’s recommended portion to the Nisen & Elliott Appellants.
The Nisen & Elliott Appellants argue that the District Court’s failure to consult with the State Judges’ Committee ignored the terms of the Settlement Agreement. They are correct. Under the express terms of the Settlement Agreement approved by the District Court, the State Judges’ Committee should have been consulted. This is significant from a procedural standpoint, notwithstanding the fact that the state court counsel had entered into the fee-sharing agreement. The members of the State Judges’ Committee, unlike those of the Allocation Committee, did not have a financial interest in the outcome of the fee allocation. In such a complex case, soliciting and taking into account the views of disinterested jurists familiar with the proceedings should have occurred. Those views would provide a valuable procedural check on at least some of the recommendations of the Allocation Committee.
3. Degree of Deference to the Committee’s Recommendations
There is yet another aspect of the fee allocation process that raises serious questions about how the Allocation Committee’s proposed “cutting-up-the-pie” should be reviewed by the District Court. The Court afforded a high degree of deference to the Allocation Committee’s recommendation, stating: “Although the ultimate decision with respect to the award and allocation of counsel fees is reserved for the court, we will give substantial deference to the recommendation of the Committee as long as we conclude the recommendations are fair and reasonable.” Pretrial Order No. 2859 at 15-16. Further, a comparison between the District Court’s opinion and the Allocation Committee’s explanation of its recommended fee apportioning through June 30, 2001 reveals that the District Court tracked the Committee’s recommendation to the dollar.
The District Court correctly points out that other courts have afforded deference to the views of lead counsel in allocating awards of fees. In re Copley, for example, explained:
In the case at bar, when the Court became aware that class counsel could not reach a unanimous stipulation, it necessarily gave substantial deference to Lead Counsel’s proposed allocation. In a case of this magnitude, the assistance of Lead Counsel on matters such as this is especially invaluable. Accordingly, this Court, after reviewing the previous submissions of class counsel as to hours and expenses, relying on previous discussions with Lead Counsel as well as other members of class counsel, and weighing the relative responsibilities of class counsel members and their contribution to this litigation, as well as when respective attorneys became involved in this litigation, found Lead Counsel’s allocation to be fair and reasonable.
50 F.Supp.2d 1141, 1147-50 (D.Wyo.1999) (citations omitted); see also, e.g., In re *173Indigo Sec. Litig., 995 F.Supp. 233, 234 (D.Mass.1998). Likely because of perceived practical necessity, courts- .have shown an eagerness to defer to counsel’s views in allocating attorneys’ fees.
But counsel have inherent conflicts. They make recommendations on their own fees and thus have a financial interest in the outcome. How much deference is due the fox who recommends how to divvy up the chickens? A template I suggest for consideration is our deference-determination scale in Employee Retirement Income Security Act (“ERISA”) cases. In Pinto v. Reliance Standard Life Insurance Co., 214 F.3d 377 (3d Cir.2000), we addressed the standards to be employed in reviewing the denial of a request for benefits under an ERISA plan by an insurance company that both determines eligibility for benefits and pays those benefits out of its own funds. That is, we considered what standard of review is appropriate when “the nature of the relationship between the funds, the decisions, and the beneficiary invites self-dealing. [A]n inherent conflict [exists] between the roles assumed by an insurance company that administers claims under a policy it issued.... Because an insurance company pays out to beneficiaries from its own assets rather than the assets of a trust, its fiduciary role lies in perpetual conflict with its profit-maldng role as a business.” Id. at 384-85 (quoting Brown v. Blue Cross & Blue Shield of Ala., 898 F.2d 1556, 1561 (11th Cir.1990)). After surveying the law in other circuits, we rejected the requirement that bias must be specifically demonstrated, and adopted a sliding scale approach to the standard of review. Pinto, 214 F.3d at 389-93. That approach allows each case to be examined on its facts. The court may take into account the sophistication of the parties, the information accessible to them, and the exact financial arrangement between the insurer and the company. Id. at 392.
While the analogy between Pinto and our case is imprecise (Pinto considered the fiduciary duties of insurers), ■ Pinto’s teachings remain relevant for several reasons. First, it illustrates a willingness to examine critically decisions of non-judicial bodies that may have a financial interest in the outcome of their decisions or recommendations. Second, Pinto supports the view that, when a conflict of interest is present, the reviewing court should consider on a fact-specific basis how much deference should be' afforded to the views of a group potentially affected by self-dealing.
Though the insurance companies discussed in Pinto may have been affected by a “structural conflict of interest [that] unconsciously encouragefd] even a principled fiduciary to make decisions that are not solely in the interest of the beneficiary,” id. at 384 (discussing Brown, 898 F.2d at 1561), the members of the Allocation Committee had a direct conflict of interest: they were suggesting to the District Court how to proceed on matters near and dear — dividing a limited fund among themselves and other firms. Such a direct conflict of interest strongly suggests that affording substantial deference is inappropriate.
While the District Court’s allocation may ultimately be fair, careful attention must be paid to the procedures by which the allocation is set. In this regard, there is room for flexibility. To the extent the District Court chooses to rely on the recommendations of a committee of interested attorneys,. it then becomes necessary to scrutinize more closely those recommendations. By soliciting the views of less interested individuals or a disinterested body like the State Judges’ Committee, and al*174lowing objections, the conflict of interested counsel becomes less of a factor.
* * * * * X
I join my colleagues’ conclusion that appellate jurisdiction does not exist in our case. However, I am not as sanguine as they that plausible arguments do not exist in certain circumstances for appellate jurisdiction under Rule 54(b) or the collateral order doctrine. The rest — attending to attorneys’ fee allocations — is but dicta once removed. It is, however, an attempt to forestall claims that courts that follow recommendations of fee allocation committees controlled by counsel with conflicts exercise scrutiny-lite.
. The Rule provides in part:
When more than one claim for relief is presented in an action, whether as a claim, counterclaim, cross-claim, or third-party claim, or when multiple parties are involved, the court may direct the entry of a final judgment as to one or more but fewer than all of the claims or parties only upon an express determination that there is no just reason for delay and upon an express direction for the entry of judgment.
. Ronald R. Benjamin (02-4021).
. Fleming & Associates, L.L.P (02-4074).
. Lois Gooch-Kiel and Linda L. Marull (02-4020).
. Randy Hague, et al. (03-4830). .
. The other appeals—Carol Bloom, et al. (No. 03-2695) ("Bloom”); Lopez, Hodes, Restaino, Milman & Skikos, et al. (No. 03-2627) ("Lopez-Hodes"); and Nisen & Elliott, et al. (No. 03-2866) ("Nisen & Elliot”)—present different considerations, as they challenge awards of counsel fees that the District Court expressly indicated were subject to future revision following the application of the factors outlined in Gunter v. Ridgewood Energy Corp., 223 F.3d 190 (3d Cir.2000). See Pretrial Order No. 2622 at 24-26.
. To repeat what is already noted in the majority opinion, the PMC performed work (or assigned work to other attorneys) for the common benefit of plaintiffs in MDL 1203 and in any coordinated state-court proceedings. Among other things, the PMC oversaw pretrial proceedings on behalf of plaintiffs, conducted discovery of widespread applicability, and compiled a widely applicable trial preparation package.
. Rule 54(b) and the collateral order doctrine are conceptually distinct exceptions to the finality rule. Prior to the adoption of Rule 54(b), the entire case was typically treated as a single judicial unit that could give rise to only one appeal, even if that case consisted of numerous discrete claims or numerous parties. Rule 54(b) was therefore designed to relax the “judicial unit” aspect of finality in response to the increasing demands and frequency of complex litigation. See Shipes v. Trinity Indus., Inc., 883 F.2d 339, 342 (5th Cir.1989). The collateral order doctrine is, in contrast, a judicially created exception to the statutory finality requirements that permits appeals from orders that would otherwise be considered interlocutory. See id.
. Regarding the second prong of the collateral order test, this is a case in which the issues surrounding the attorneys' fee allocation are separate from the merits of the underlying litigation. See White v. New Hampshire Dep’t of Employment Sec., 455 U.S. 445, 452, 102 S.Ct. 1162, 71 L.Ed.2d 325 (1982) (explaining that an award of attorney's fees "is uniquely separable from the cause of action to be proved at trial”). Furthermore, the second prong of Cohen contemplates orders that are important in a jurisprudential sense. See Praxis Properties, 947 F.2d at 56 (citing Nemours Found. v. Manganaro Corp., 878 F.2d 98, 100 (3d Cir.1989)); see also Nixon v. Fitzgerald, 457 U.S. 731, 742, 102 S.Ct. 2690, 73 L.Ed.2d 349 (1982) (“Cohen established that a collateral appeal of an interlocutory order must 'present a serious and unsettled question.'") (citation omitted). Given the novelty of the issues implicated in the underlying class action and the lack of guidance from our Court on the allocation of attorneys’ fees in the context of our case, I believe this aspect of the second prong is easily met.
. Under the "lodestar” method, the number of hours reasonably expended by an attorney is multiplied by a reasonable hourly rate to calculate attorneys’ fees.
. The Nisen & Elliott Appellants include four of the five law firms that represented the state-wide certified class in the Illinois diet drugs litigation. The fifth firm that represented the Illinois class did not object to the fee allocation.