Consolidated Edison Co. of New York, Inc. v. Bodman

445 F.3d 438, 370 U.S. App. D.C. 359, 2006 U.S. App. LEXIS 10066, 2006 WL 1042362

• Court: Court of Appeals for the D.C. Circuit
• Decided: April 21, 2006
• Docket: 05-5089, 05-5090, 05-5223, 05-7009
• Status: Published

Opinion

Opinion for the Court filed by Senior Circuit Judge WILLIAMS.

WILLIAMS, Senior Circuit Judge.

For eight years (from 1973 to 1981) the government imposed price controls on the sale of crude oil. In and since that period, it has collected refunds from the suppliers whose prices exceeded the ceilings and has distributed the proceeds to persons and firms that paid supra-ceiling prices. Indeed, the process of distributing refunds continues to this day. The statutory authority underlying these efforts appears in the Emergency Petroleum Allocation Act of 1973, Pub.L. No. 93-159, 87 Stat. 627 (1973) (“EPAA”), incorporating the Economic Stabilization Act Amendments of 1971, Pub.L. No. 92-10, 85 Stat. 743 (1971) (“ESA”). We deal here with claims for attorneys’ fees for litigation undertaken by Philip P. Kalodner in connection with the distribution.

For roughly two decades, Kalodner has represented a group of six electric utility companies and three paper manufacturers (collectively, the “clients”) in their quests to obtain crude oil pricing refunds. On behalf of himself and his clients, he invokes the “common fund” theory to support claims for legal fees, making claims against both the government and many of the refund beneficiaries who were not his clients. The common funds, he claims, arose out of alleged legal victories in two cases, known here as Con Ed IV (more formally, Consolidated Edison Co. of New York v. Abraham, 271 F.Supp.2d 104 (D.D.C.2003)), and Con Ed V (more for mally, Consolidated Edison Co. of New York v. Abraham, No. 03-1991 (D.D.C. June 30, 2004)). Although the amounts at stake keep shifting for a variety of reasons, Con Ed TV involves about $264 million, Con Ed V about $35 million (which may overlap with the $264 million). 1 We have, then, two sets of claimants (Kalodner and his clients), two sets of possible fee payers (the government and the beneficiaries), and two alleged legal “wins” (Con Ed TV and Con Ed V). Because of the complexity we provide a scorecard:

[[Image here]]

The fee claims under review were asserted in a motion in Con Ed V and in two separate lawsuits. Specifically, the claims against the government took the form of (1) a motion in 2004 in Con Ed V on behalf of Kalodner’s clients for fees in winning the alleged victories in both cases (claims 1 & 2 in Table 1), and (2) a separate lawsuit in 2005 on behalf of Kalodner himself, again for both alleged victories (claims 3 & 4 in Table 1). In each the clients and Kalodner sought a fee of 10% of the final distribution. In a consolidated opinion, the district court rejected Kalodner’s claims (against the government) with respect to both Con Ed TV and Con Ed V, Reasoning primarily that the partial waiver of sovereign immunity provided by the Equal Access to Justice Act, 28 U.S.C. § 2412(b) (“EAJA”), runs in favor only of parties, not their lawyers. See Mem. Op. at 6 (D.D.C. Jan. 26, 2005) (“Consolidated Mem. Op”), filed in both Con Ed V and Kalodner v. Abraham, No. 05-0024, 2005 WL 354483 (D.D.C. Jan.26, 2005) (“Kalodner-Abraham”). Kalodner appeals (Kalodner v. Abraham appears sub nom. Kalodner v. Bodman, No. 05-5090), 2 and we affirm (thus rejecting claims 3 & 4 in Table 1).

In the same opinion the court also rejected the clients ’ claims against the government with respect to Con Ed TV (claims i & 2 in Table 1). Consolidated Mem. Op. at 5-6. In doing so it said it was applying the law-of-the-case doctrine, citing its own prior decision finding the fees claim barred by sovereign immunity. See Order, Con Ed TV (Dec. 4, 2003) (the “Dec. 4, 2003 Order”), aff'd, Consolidated Edison Co. of New York v. Abraham, 101 Fed.Appx. 356 (Fed.Cir.2004). (The district court later characterized the second motion for fees for work in Con Ed IV as “an improper collateral attack on the decision of the Federal Circuit.” Consolidated Mem. Op at 11, thus invoking issue and/or claim preclusion, which appear, given the Federal Circuit’s affirmance of the court’s earlier decision, to be the most apt doctrines.) But the court accepted the clients’ claims for their lawyer’s supposed contribution in Con Ed V, and awarded a fee of 30% of the roughly $35 million there at stake. See Consolidated Mem. Op. at 9-11. (We have discovered no request by Kalodner in these cases for more than 10%.) The clients appeal as to Con Ed IV (our No. 05-5089), and the government cross-appeals as to Con Ed V (our No. 05-5223). 3 We reverse and remand the judgment against the clients as to Con Ed IV, because Kalodner’s efforts in that case may have satisfied the causal requirements for a common fund recovery. (The government having failed to argue issue or claim preclusion, we express no opinion on those defenses.) We reverse the judgment in favor of the clients as to Con Ed V, because it is clear that that lawsuit failed to yield any court-ordered relief and more generally played no material role in the successes claimed.

In the second independent lawsuit, Kalodner brought claims on his own behalf (not for the clients) against refund beneficiaries in 2004 (claims 5 & 6 in Table 1), again seeking 10% of the total recovery. The district court resolved the claims against him, Kalodner v. Public Service Electric & Gas, No. 04-152 (D.D.C. Dec. 20, 2004) (“Kalodner-Public Service ”), and he appeals (our No. 05-7009). 4 We reverse that decision in part and remand, finding that Kalodner may be able to show that his activity in Con Ed IV played a sufficient role to justify a fee recovery; we also affirm in part, as the record makes clear that there was no such victory in Con Ed V.

We note by way of background that the common fund theory conventionally rests on a theory that beneficiaries of the lawsuit would be unjustly enriched if not compelled to pay a share of the fees that made success possible. See, e.g., Swedish Hospital v. Shalala, 1 F.3d 1261, 1265 (D.C.Cir.1993). It may well be that courts have found it sensible to apply the unjust enrichment principle here (after all, human life abounds in windfalls) because doing so answers a potential free-rider problem. See Wal-Mart Stores Health & Welfare Plan v. Wells, 213 F.3d 398, 402 (7th Cir. 2000) (noting that free riding on attorney’s efforts would be “contrary to the equitable concept of ‘common fund’ ”); cf. United States v. Tobias, 935 F.2d 666, 668 (4th Cir.1991) (“Generally, a fund claimant who is represented by counsel ... is deemed not to have taken a ‘free ride’ on the efforts of another’s counsel.”); John P. Dawson, Lawyers and Involuntary Clients: Attorney Fees from Funds, 87 Harv. L. Rev. 1597, 1647-51 (1974) (discussing incentives to free ride on attorneys’ efforts). If lawyers considering representation of some but not all of a cluster of beneficiaries can recover compensation only from beneficiaries who actively retain them, claims will not be brought — even though meritorious — where the expected value of the gains for beneficiaries willing to participate can’t generate adequate compensation for counsel (and thus enable the bringing of suit). Under a rule awarding fees out of litigation proceeds received by passive beneficiaries, lawyers’ anticipation of fee recoveries will provide the requisite incentive. In some cases, of course, a subset of potential beneficiaries will have stakes large enough to call forth ample litigation effort; if so, the free-rider concern declines, possibly to nil. This last point would be pertinent, if at all, in calculation of fees.

4s * * * *

The history preceding these cases is a long and tortured one, recounted in bits and pieces elsewhere. See Kalodner v. Abraham, 810 F.3d 767 (D.C.Cir.2002) (“Kalodner I ”); Consolidated Edison Co. of New York v. Abraham, 303 F.3d 1310 (Fed.Cir.2002); Consolidated Edison Co. of Netv York v. Ashcroft, 286 F.3d 600 (D.C.Cir.2002); Consolidated Edison Co. of New York v. Richardson, 233 F.3d 1376 (Fed.Cir.2000) (“Con Ed III”); Kalodner v. Abraham, 309 F.Supp.2d 100 (D.D.C. 2004); Con Ed IV; Consolidated Edison Co. of New York v. O’Leary, 4 Energy Management (CCH) ¶ 26,698 (D.D.C.1996), aff'd, 117 F.3d 538 (Fed.Cir.1997) (“Con Ed II ”); Consolidated Edison Co. v. Herrington, 752 F.Supp. 1082 (D.D.C.1990). For our purposes, it suffices to summarize only the bare background facts.

The statutes mentioned at the outset empowered the Department of Energy (“DOE”) to recover overcharges in violation of the price controls, and it has done so to the tune of several billion dollars. As part of a settlement in a multidistrict litigation, In re Department of Energy Stripper Well Exemption Litigation, 653 F.Supp. 108 (D.Kan.1986) (“Stripper Well”), DOE authorized its Office of Hearings and Appeals (“OHA”) to begin distributions of this money to parties who had paid supra-ceiling prices. DOE placed 20% of the remaining crude oil overcharge funds into escrow for potential distribution to private firms and persons that were not parties to the settlement, leaving the remaining 80% to be split between federal and state governments. See Statement of Modified Restitutionary Policy in Crude Oil Cases, 51 Fed.Reg. 27,899 (Aug. 4, 1986). Since then, OHA has processed over 100,000 claims, and in two rounds of distributions has paid out roughly $610 million to private beneficiaries (as opposed to governments). See Con Ed IV, 271 F.Supp.2d at 106-07. The cases here concern the distribution of roughly $280 million remaining in escrow at the date of oral argument.

Kalodner’s clients will cumulatively receive up to 15% of all crude oil refunds to private parties. The clients have compensated him for his services, but he and the clients now seek a common fund fee of approximately $27 million out of the sums paid or to be paid to many of the roughly 56,000 other refund recipients (past or future). We say “many” because the fee claimants have evidently chosen not to pursue parties receiving relatively small amounts, as well as about 30 beneficiaries with whom Kalodner has fee agreements. (Although in all instances the fees would in economic reality be paid by the beneficiaries, we distinguish (as does the law) between claims made against the government and ones made against beneficiaries.)

Kalodner and his clients assert that his civil litigation in Con Ed IV and Con Ed V preserved and increased the remaining final distribution for the benefit of the whole class. In Con Ed IV, they claim, Kalodner created or preserved a common fund (now amounting to about $280 million) by securing a declaratory judgment that the government distribute all remaining amounts of money in the 20% reserve for private parties. There his clients had moved for partial summary judgment on prayers for relief (1) that the fund be expanded beyond the 20% reserve, (2) that certain proceeds from a settlement agreement be included in the 20% reserve, and (3) that the funds collected be distributed without further delay. The district court denied the motion with respect to the first two requests, and granted it in part with respect to the third, declaring the beneficiaries’ entitlement to have DOE distribute the money “insofar as practicable.” 271 F.Supp.2d at 112. But the court neither ordered an immediate distribution nor issued a timetable. Contrary to the clients’ request, the court found itself “unable either to issue a writ of mandamus ordering DOE to complete distribution of those funds or to declare that further delay in making the final distribution is unjustified.” Id. at 111.

As noted above, the decision under review denying Kalodner’s clients’ request for a fee based on Con Ed TV is the second district court decision to do so. The clients initially moved for fees in Con Ed TV in 2003 (then seeking only 5%), and the court denied the motion on the ground that the money was in the possession of the U.S. government and thereby protected by sovereign immunity. Dec. 4, 2003 Order at 1-2. The clients at the same time moved for joinder of sixteen refund beneficiaries as class representatives of the “respondents” to the fee motion. See Motion to Add Parties as Respondents to Motion for Award of Common Fund Fee at 1, Con Ed TV (Oct. 9, 2003). The court denied the motion for joinder, on the grounds that it would be “inappropriate, particularly given the Court’s ruling on [sovereign immunity].” Id. The clients appealed the Dec. 4, 2003 Order to the Federal Circuit, which has exclusive jurisdiction over ESA issues, see ESA § 211(b)(2), amended by Pub.L. No. 102-572, 106 Stat. 4506 (1992), and that court affirmed without opinion. See Consolidated Edison Co. of New York v. Abraham, 101 Fed.Appx. 356 (Fed.Cir. 2004). In 2004 the clients filed another motion, renewing the claim, which the district court, noting the unsuccessful appeal to the Federal Circuit, denied on grounds of preclusion. The clients appeal.

As to Con Ed V, Kalodner and his clients argue that the litigation increased the beneficiaries’ refunds by roughly $35 million by compelling DOE to modify its “volumetric method” of calculation. See Mem. Op., Con Ed V (June 30, 2004). The clients’ lawsuit sought an order of final distribution and alluded in general terms to the methodology for computing refunds. Of the two modifications that Kalodner and the clients would now attribute to Con Ed V, one (“deferral”) is not mentioned in the complaint; the other (inclusion of the “Citronelle account”) is mentioned but as we shall see (in Part III below on “Causation”) appears never to have been in dispute. After the case was filed, DOE issued its notice of proposed procedures for final distribution. See Notice of Proposed Procedures for Distribution of Remaining Crude Oil Overcharge Refunds and Opportunity for Comment, 68 Fed.Reg. 64,098 (Nov. 12, 2003) (“Proposed Procedures”). Kalodner participated on behalf of his clients in the ensuing administrative proceeding, and in that forum urged inclusion of the Citronelle account and, for the first time, “deferral.” Another participant urged the same points. DOE accepted these suggestions, with the result (we are told) of effectively increasing the total refund amount by $35 million, and published its final order several months later. See Notice of Final Procedures for Distribution of Remaining Crude Oil Overcharge Refunds, 69 Fed.Reg. 29,300 (May 21, 2004) {“Final Procedures"). Because the final distribution appeared to be well under way by the time the court ruled, the court dismissed Con Ed V as moot. Mem. Op., Con Ed V (June 30, 2004).

j}: ‡ ❖

We review the several dispositions de novo. This is obvious for the outright dismissals of claims, see, e.g., Masonry Masters v. Nelson, 105 F.3d 708, 710-11 (D.C.Cir.1997), but also applies to the court’s grant of fees in favor of Kalodner’s clients for his work on Con Ed V, as the issues, with one exception, are ones of law, for which the standard of review is almost invariably de novo. See Edmonds v. FBI, 417 F.3d 1319, 1322 (D.C.Cir.2005). The exception relates to Kalodner’s causal role in generating the beneficiaries’ recovery, an issue of course containing elements of fact. But the district court has so far engaged in no fact-finding, and all we have before us are the movants’ allegations. We assume the correctness of the allegations of specific facts, but not of conclusions. For these we ask whether the record and specific allegations support an inference of causation. See Judicial Watch, Inc. v. U.S. Senate, 432 F.3d 359, 360 (D.C.Cir.2005).

We proceed below in three steps. First, we consider sovereign immunity. We find that three claims (claims 2, 3 & 4) are barred. The fee sought by Kalodner’s clients from the government (claim 2) is barred with respect to his efforts in Con Ed V (allegedly modifying the volumetric method): sovereign immunity applies, so that the clients are barred in the absence of a waiver, and they were not prevailing parties within the meaning of EAJA’s waiver provision. With respect to work in Con Ed IV, however, the clients qualify under EAJA as prevailing parties in the minimal sense of the term; but (as we see in the third step) there is considerable doubt whether their litigation efforts played the causal role needed to qualify for a common fund fee recovery. Kalodner’s suit against the government in his own name (claims 3 & 4) enjoys no EAJA waiver because he was not a party to the underlying suits. And because Kalodner’s claims against the beneficiaries (claims 5 & 6) are not against the government (except with respect to one remedy request, which is severable), sovereign immunity is completely inapplicable. Thus, the only claims surviving sovereign immunity are the clients’ claim against the government for Con Ed IV and both of Kalodner’s claims against beneficiaries, (claims 1, 5 & 6).

Second, preclusion issues abound. The beneficiaries fail to make out a case for issue preclusion of the claims against them (claims 5 & 6). The government has failed to press its possible preclusion arguments (claim 1); thus we discuss them only briefly and note that under our case law the omission need not be fatal to the preclusion arguments’ resurrection on remand. See Stanton v. District of Columbia Court of Appeals, 127 F.3d 72, 77 (D.C.Cir.1997).

Finally, as our summary has made clear, for claims not barred by sovereign immunity, the controlling issue is whether Kalodner’s civil litigation played a sufficient role in generating the supposed “common funds” to warrant a fee award. The answer is “maybe” for Con Ed IV (for the claim by the clients against the government (claim 1) and the claim by Kalodner against the beneficiaries (claim 5)), and “no” for Con Ed V (claim 6).

We note that the government distributed the bulk of the remaining refunds on the day of oral argument. See Final Procedures for Distribution of Remaining Crude Oil Overcharge Refunds, 71 Fed. Reg. 2,195 (Jan. 13, 2006). But as the government set aside 10% of the private party refunds pending this litigation (i.e., 10% of the 20% reserved for private parties under the Stripper Well settlement, see discussion in Part III.A. below), id. at 2,195-96, the distribution doesn’t moot the case.

I. Sovereign Immunity

The threshold issue for the fee recovery suits is whether the funds are protected by sovereign immunity. Monetary claims against the government are barred by sovereign immunity unless the government has expressly waived its immunity. See Lane v. Peña, 518 U.S. 187, 192, 116 S.Ct. 2092, 135 L.Ed.2d 486 (1996). To some degree Kalodner and his clients argue that sovereign immunity is simply out of the picture because of the nature of their claims and the status of the refund process. In anticipation of the failure of this theory, they assert a waiver theory under EAJA. We address first the suit by Kalodner’s clients against the government, then Kalodner’s own suit against the government, and finally his suit against the beneficiaries. Lastly, in all claims Kalodner and his clients invoke ESA § 209 as yet another waiver theory. We find that resolution of an ESA issue, which falls under the exclusive appellate jurisdiction of the Federal Circuit, is not likely to be required.

A. Kalodner’s clients’ claims against the government

In Kalodner I, addressing Kalodner’s efforts to recover a common fund fee in other crude oil refund litigation, we held that “the sine qua non of federal sovereign immunity is the federal government’s possession of the money in question.” 310 F.3d at 770. We found sovereign immunity applicable, without more, once we had determined that the government was in possession of the relevant funds. Thus Kalodner I makes clear that government possession of funds is itself sufficient to establish sovereign immunity. Here, except for the money distributed, which the government no longer possesses, the money in dispute is clearly in the government’s possession so that, under Kalodner I, sovereign immunity appears to apply. Unless Kalodner’s clients can point to a waiver, their claims against the government (claims 1 & 2) are barred.

The clients’ first effort to overcome that conclusion rests on a number of inapplicable cases. First they cite the Supreme Court’s decision in Boeing Co. v. Van Gemert, 444 U.S. 472, 100 S.Ct. 745, 62 L.Ed.2d 676 (1980), especially its observation that a common fund fee recovery would be appropriate “when each member of a certified class has an undisputed and mathematically ascertainable claim to part of a lump-sum judgment recovered on his behalf.” Id. at 479, 100 S.Ct. 745. But the decision involves fee recovery in private litigation and has nothing to do with sovereign immunity. Were the clients to establish the inapplicability of sovereign immunity, or a waiver, Boeing might help them meet the ordinary common fund prerequisites, but it does nothing to get them over the initial sovereign immunity hurdle. Commonwealth of Puerto Rico v. Heckler, 745 F.2d 709 (D.C.Cir.1984), and Swedish Hospital Corp. v. Shalala, 1 F.3d 1261 (D.C.Cir.1993), are equally useless in the clients’ effort to finesse the sovereign immunity problem. Although both were suits against the government, in Puerto Rico we found EAJA applicable (thus presupposing a sovereign immunity bar), and in Swedish Hospital the only issue was computation of the fee, the entitlement having evidently been conceded or established. Finally, National Treasury Employees Union v. Nixon, 521 F.2d 317 (D.C.Cir.1975), is of no use to the clients on sovereign immunity; as we said in Kalodner I, the money with respect to which a fee was claimed had already been distributed. See Kalodner I, 310 F.3d at 770.

In a more realistic vein, the clients assert waiver under EAJA. Although much of the clients’ language seems to disclaim any reliance on EAJA, see Cross-Appellees’/Appellants’ Reply Brief at 31 (“they are not” “seeking a fee pursuant to the EAJA”) (emphasis added); see also Appellant’s Initial Brief at 20 (“By his Complaint and his Motion for Preliminary Injunction, Kalodner sought a fee ... not pursuant to the EAJA.”) (emphasis added), the briefs also rather obscurely reserve EAJA as a “back-up,” see Cross-Appellees’/Appellants’ Reply Brief at 54 (“Even if it were necessary for Kalodner to rely on the EAJA, the reliance is solely for the purpose of obtaining a waiver of sovereign immunity.”); Appellant’s Initial Brief at 40 (“if not already so waived, sovereign immunity is waived by the EAJA”). Giving the clients the benefit of the doubt, we proceed to the EAJA analysis.

EAJA provides:

Unless expressly prohibited by statute, a court may award reasonable fees and expenses of attorneys ... to the prevailing party in any civil action brought by or against the United States or any agency or any official of the United States acting in his or her official capacity in any court having jurisdiction of such action. The United States shall be liable for such fees and expenses to the same extent that any other party would be liable under the common law or under the terms of any statute which specifically provides for such an award.

28 U.S.C. § 2412(b) (emphasis added). (This waiver appears quite distinct from the more familiar § 2412(d), which contains additional qualifications. See, e.g., § 2412(d)(1)(B) & (d)(2)(B).)

In Buckhannon Board & Care Home, Inc. v. West Virginia Department of Health & Human Resources, 532 U.S. 598, 603-04, 121 S.Ct. 1835, 149 L.Ed.2d 855 (2001), the Supreme Court interpreted two (non-EAJA) statutes authorizing fee-shifting for prevailing parties, and held that a party has not “prevailed” unless it has secured some form of court-ordered relief. In so ruling, it rejected the “catalyst theory,” under which a party could be found to prevail if a defendant changed its conduct in response to a pending law suit. Id. at 603, 121 S.Ct. 1835. We have held that this understanding of “prevailing party” applies to EAJA’s use of the term. See Select Milk Producers, Inc. v. Johanns, 400 F.3d 939, 945 (D.C.Cir.2005).

, In Con Ed V, the clients obtained relief — but not from the court. It came as a result of the agency’s favorable response to their (and others’) comments in the agency proceeding, suggesting two changes in the “volumetric” computation. Although the complaint alluded vaguely to the method of computation, it never framed a request for a change. The agency accepted the theory — presumably on its merits, there being no detectable judicial pressure to do so, much less a judgment or any other form of court-ordered relief. The clients are therefore not prevailing parties with respect to Con Ed V, and the funds (sought in claim 2) remain protected by sovereign immunity.

We note that the district court mistakenly distinguished Kalodner I, evidently believing that EAJA had not been considered by the court nor raised by the parties in that case. Consolidated Mem. Op. at 8-9. In fact EAJA had been raised in Kalodner I, albeit by the government. See Brief for the Appellees at 28-29, Kalodner I. Our omission of any discussion was plainly because of the ample reasons why EAJA would not have availed Kalodner, the most obvious being that Kalodner was simply not a “party” at all.

With respect to the work in Con Ed TV, however, the clients appear to meet the minimum qualifications for prevailing parties (claim 1). Although the Con Ed TV court rejected two of the three claims sought in their motion for partial summary judgment, it did grant a declaratory judgment that the clients were entitled “to a distribution of the entire 20% reserve, insofar as practicable.” 271 F.Supp.2d at 112. As the court rejected the clients’ claims with respect to the amount to be distributed, and as it imposed neither deadlines nor even criteria for judging practicability, this was pretty thin gruel, as we shall see when we discuss whether the judgment may have had enough of a causal effect to justify a common fund fee. But it does appear to meet the minimum requirement of constituting court-ordered relief. Insofar as qualification as “prevailing” requires more than the raw Buckhannon minimum, see, e.g., Farrar v. Hobby, 506 U.S. 103, 109, 113 S.Ct. 566, 121 L.Ed.2d 494 (1992) (requiring that plaintiffs “succeed on any significant issue in litigation which achieves some of the benefit the parties sought in bringing suit”), that inquiry is here subsumed in our discussion in Part III of whether Kalodner’s civil litigation played enough of a role in generating the beneficiaries’ recovery to warrant a common fund fee.

B. Kalodner’s claims against the government

We affirm the district court’s dismissal of Kalodner’s suit on his own behalf against the government (claims 3 & 4). Sovereign immunity applies unless waived, for the reasons addressed above. As to any EAJA waiver, Kalodner was counsel in Con Ed TV and Con Ed V, not a party, and EAJA provides attorneys’ fees only for parties. See Consolidated Mem. Op. at 6.

C. Kalodner’s claims against the beneficiaries

The beneficiaries argue that sovereign immunity also bars Kalodner’s attempt to recover fees from them (claims 5 & 6) by virtue of Kalodner I’s holding that sovereign immunity applies if the government is in possession of the relevant funds. With respect to some of the relief sought by Kalodner, this counter-intuitive proposition is correct. He indeed asks for “[a]n Order directing the defendants [i.e., named non-client beneficiaries] on behalf of each of the class members to direct the DOE to withhold the fee awarded to plaintiff [i.e., Kalodner].” Complaint at 18, Kalodner-Public Service (Feb. 3, 2004). Unless the requested communication to DOE were purely precatory (“Would you be so kind as to send some of my money to Mr. Kalodner?”), it would pose the same sovereign immunity issues as a direct court order against the government. But Kalodner appears independently to also ask for an order “awarding to plaintiff [from the beneficiaries] 10% of the distribution to each member of the [beneficiary] class.” Id. Indeed in a later filing, Kalodner clarified that he was requesting a declaratory judgment that beneficiaries have an obligation to pay attorneys’ fees once the mon ey is distributed. See Plaintiffs Motion for Summary Judgment at 1, KalodnerPublic Service (June 1, 2004); Plaintiffs Statement of Material Facts in Support of Plaintiffs Motion for Summary Judgment at 19, 38, Kalodner-Public Service (June 1, 2004). And at oral argument, Kalodner verified that the two requests were independent. See Oral Argument Tape at 18:40-19:28, Kalodner-Public Service; see also Appellant’s Initial Brief at 37-38; Appellant’s Reply Brief at 6-7. Sovereign immunity poses no bar to Kalodner’s fee claims against beneficiaries (claims 5 & 6).

D. ESA ivaiver theory

We come finally to the theory — asserted for all claims — that Congress waived the government’s sovereign immunity in ESA § 209. At the outset, we note that we’re puzzled by the theory of § 209’s relevance. Kalodner and his clients argue that because the underlying suits waived sovereign immunity under ESA, immunity was also waived as to any request for attorneys’ fees. But the ESA jurisdictional basis that they asserted for their suits in Con Ed TV and Con Ed V was § 210, not § 209. See Complaint at 3, Con Ed TV (Mar. 15, 2001); Complaint at 2-3, Con Ed V (Sep. 25, 2003).

As to the merits of the ESA § 209 theory, matters of interpretation of EPAA and ESA generally fall under the exclusive appellate jurisdiction of the Federal Circuit. See ESA § 211(b)(2), amended by Pub.L. No. 102-572, 106 Stat. 4506 (1992) (providing that “[ajppeals from orders or judgments ,. in cases or controversies arising under [the ESA] shall be brought in the ... Federal Circuit”); 28 U.S.C. § 1295(a) (providing that the “Federal Circuit shall have exclusive jurisdiction ... of an appeal under section 211 of the [ESA]”); Consolidated Edison Co. of New York v. Abraham, 303 F.3d 1310, 1313-16 (Fed. Cir.2002); Consolidated Edison Co. of New York v. Ashcroft, 286 F.3d 600, 602-05 (D.C.Cir.2002); Con Ed II, 117 F.3d at 541-42; Texas American Oil Corp. v. United States Department of Energy, 44 F.3d 1557, 1563 (Fed.Cir.1995). For all claims before us, the Federal Circuit has deferred parallel appeals to await our decisions. See supra notes 2-4. In determining the scope of the Federal Circuit’s exclusive jurisdiction, we not surprisingly follow that circuit’s two-fold criteria: “First, resolution of the litigation must require application or interpretation of the ESA or regulations issued thereunder; and second, the ESA issue must have been adjudicated in the district court.” Consolidated Edison Co. of New York v. Ashcroft, 286 F.3d 600, 603 (D.C.Cir.2002) (citing Texas American Oil Corp., 44 F.3d at 1563).

In the end it seems quite likely that no claim will meet the first criterion. The three claims based on the work in Con Ed V (claims 2, 4 & 6) cannot win regardless of any ESA waiver; as we discuss in Part III, the absence of causation is fatal. The claims based on litigation in Con Ed TV (claims 1, 3 & 5) may well also be finally resolved without regard to the ESA. If on remand the district court finds that the Con Ed mitigation had insufficient causal effect, that is the end of the matter. None of these fee claims could succeed. Even if the court finds causation, Kalodner’s claim against the government may be unavailing because any further fee recovery for work on Con Ed TV would duplicate his recovery against the beneficiaries and the clients’ recovery against the government. The same is also true for the special type of relief in Kalodner’s claim against the beneficiaries that we found barred by sovereign immunity, namely the demand for an order directing them to direct DOE to pay a portion of their entitlements to Kalodner. While there may be scenarios under which the application of preclusion would give a potentially broad ESA waiver significance, it is premature to evaluate such possibilities at this stage.

In closing, we note that the clients try to make something of our statement in Kalodner I that “Congress has waived sovereign immunity for Subpart V claimants.” 310 F.3d at 770. See Motion for Award of a Common Fund Fee at 24, Con Ed V (July 12, 2004). But the sentence does them no good. Our sole concern in that passage was to rebut Kalodner’s claim under ESA § 210, and doing so required us only to observe that Kalodner was not a “Subpart V claimant,” 310 F.3d at 770; he was their lawyer.

To recap, sovereign immunity bars Kalodner’s claims against the government (claims 3 & 4) and the clients’ claim against the government for Con Ed V (claim 2). Three claims survive: Kalodner’s claims against beneficiaries (claims 5 & 6) and the clients’ claim against the government for Con Ed IV (claim 1).

II. Preclusion

Before considering the merits of the surviving common fund claims (Kalodner against the beneficiaries for both cases, and the clients against the government for Con Ed IV), we must note the issue of possible preclusion from the district court’s December 4, 2003 rejection of the clients’ claims for a fee for the work in Con Ed IV and that decision’s later affirmance by the Federal Circuit.

A. Kalodner’s claims against the beneficiaries

In its December 4, 2003 fee decision in Con Ed IV, the district court dismissed the clients’ motion to join refund beneficiaries. They appealed to the Federal Circuit, which denied the appeal by order. The refund beneficiaries argue that the dismissal (and loss of the appeal) should preclude Kalodner’s fee claim against beneficiaries for Con Ed IV’s declaratory judgment (claim 5). Leaping over the issue of whether Kalodner should be bound by his clients ’ loss, we address the nature of the district court’s order of dismissal, a more obvious obstacle to the beneficiaries’ theory. The order appears not to have been based on the merits or on any other substantive theory. The court said simply that “joinder of [beneficiaries] at this stage of the litigation and for this purpose would be inappropriate, particularly given the Court’s ruling on the previous motion [denying a fee claim against the government on sovereign immunity grounds].” Dec. 4, 2003 Order at 3. Sovereign immunity, of course, would be no bar to a claim directed to the beneficiaries, so the court’s entire substantive discussion would have been, as to them, beside the point. Indeed, the court seemed affirmatively to contemplate the clients’ future pursuit of fees, suggesting that “[a] more suitable option would be ... to initiate a separate lawsuit against applicable claimants ... once [the government] distribute[s] the monies from the 20% reserve.” Id. As the court was evidently ruling only that the clients’ fee claim against beneficiaries should be addressed in some other context, it clearly did not resolve the issue before us — the merits of that claim (or Kalodner’s). See Yamaha Corp. of America v. United States, 961 F.2d 245, 254 (D.C.Cir.1992).

We do not understand the beneficiaries to be arguing claim preclusion — really a rule against claim splitting. See, e.g., Gener-Villar v. Adcom Group, Inc., 417 F.3d 201, 205 (1st Cir.2005) (noting that claim preclusion “generally binds parties from litigating or relitigating any [claim] that was or could have been litigated in a prior adjudication and prevents claim split ting”) (internal quotation omitted, brackets in original). The case is unusual in that the district court created the split by declining to reach the merits of the claim against the beneficiaries. But there might be an argument that the claims against the government and against the beneficiaries were properly viewed as a single claim, so that the clients’ failure to get that aspect of the district court’s judgment reversed would bind the claimants (and even Kalodner, if the beneficiaries’ theory of privity is correct). We express no opinion on such a theory.

The beneficiaries also make a distinctly confusing argument that certain of the decisions under review here bar Kalodner’s claims against them by virtue of issue preclusion. In one respect the claim has merit, though the beneficiaries’ labeling is wrong. In so far as they argue that Kalodner cannot double dip, recovering both through his clients and/or against the government, and independently against themselves, the beneficiaries are right, as Kalodner forthrightly conceded at oral argument. See Oral Argument Tape at 0:37-1:12 (in appeal No. 05-5089). That is not a matter of issue preclusion, but of double recovery. See, e.g., Commissioners Court of Medina County v. United States, 719 F.2d 1179, 1182 n. 6 (D.C.Cir.1983). So far as issue preclusion is concerned, the dispositive issue in the one case not on review before us (i.e., the December 4, 2003 decision and the failed appeal) related to sovereign immunity, which provides the beneficiaries no defense. Thus, issue preclusion cannot bar Kalodner’s claims against the beneficiaries (claims 5 & 6).

B. Kalodner’s clients’ claim against the government

With respect to the clients’ surviving fee claim against the government (claim 1), DOE’s brief proclaimed it unnecessary to delve into the preclusive effect of the December 4, 2003 Order, instead relying on sovereign immunity alone. This tactical choice is especially perplexing because the district court invoked preclusion in finding in the government’s favor as to Con Ed IV. As to those fees, we’ve just ruled, the clients formally qualify as “prevailing parties” under EAJA’s waiver provision. As the government failed to brief the preclusion issue for fee claims against it, and as we are remanding the issue, for prudential reasons we do not address it here. We note for the benefit of the parties and the district court, however, that because interests of judicial economy are at stake in preclusion doctrines, courts retain the power to consider such doctrines sua sponte. See Stanton v. District of Columbia Court of Appeals, 127 F.3d 72, 77 (D.C.Cir.1997).

III. Common Fund Causation

Our circuit law permits “a party who creates, preserves, or increases the value of a fund in which others have an ownership interest to be reimbursed from that fund for litigation expenses incurred.” Swedish Hospital, 1 F.3d at 1265 (emphasis added). All three variants express the necessity that the claiming parties’ litigation have played a causal role in achieving the benefits for which they seek fee reimbursement. Similarly, the Supreme Court has demanded that “[t]he benefits could be traced with some accuracy.” Alyeska Pipeline Service Co. v. Wilderness Society, 421 U.S. 240, 265 n. 39, 95 S.Ct. 1612, 44 L.Ed.2d 141 (1975). See also In re Holocaust Victim Assets Litigation, 424 F.3d 150, 157 (2d Cir.2005) (“The actions of the party seeking to recover costs must ... be a substantial cause of the benefit obtained.”) (citation and internal quotation omitted); Knight v. United States, 982 F.2d 1573, 1579-80 (Fed.Cir.1993) (de scribing typical cases to involve third party beneficiaries of enhancement or preservation of assets or trust); Vincent v. Hughes Air West, 557 F.2d 759, 771 n. 10 (9th Cir.1977) (“[T]he common fund doctrine requires that the work of the attorney seeking an extra fee be a cause-in-fact of any claimed benefit to the fund and its beneficiaries.”); see generally Federal Judicial Center, Awarding Attorneys’ Fees And Managing Fee Litigation 62-64 (2005) (“the plaintiff must ... establish that its suit was a ‘but for’ cause of the fund (or at least ensured access to the fund).”).

The question hence becomes whether Kalodner and the clients have pleaded facts supporting an inference of the requisite causation on the three potentially viable claims (claims 1, 5 & 6). We remand the two claims that ride on the alleged success in Con Ed IV so as to give the fee claimants a chance to make their case (claims 1 & 5). As to Kalodner’s claim against the beneficiaries based on Con Ed V (claim 6, the only claim based on Con Ed V not already found barred by sovereign immunity), the record shows the absence of causation as a matter of law.

A. The declaratory judgment in Con Ed IV

Kalodner argues that Con Ed TV’s declaratory judgment created or preserved the fund by requiring the government to distribute the previously undistributed portion of the private parties’ 20% of collections set aside pursuant to the Stripper Well settlement, overcoming DOE’s alleged reservation of a right not to do so. Yet DOE’s expression of a reservation does not mean in itself that Con Ed IV was a cause (much less a substantial cause) of the final distribution. Reluctance is not refusal. We find the record inconclusive.

We note a few basic points at the outset. First, the decision to grant private crude oil purchasers 20% of certain overcharge collections dates back to the 1986 settlement. See Stripper Well, 653 F.Supp. at 114 (noting that DOE “will ... establish an initial reserve for [private beneficiaries not party to the settlement agreement] amounting to twenty percent of the funds received by the DOE”); Statement of Modified Restitutionary Policy in Crude Oil Cases, 51 Fed.Reg. 27,899, 27,900 (Aug. 4, 1986) (providing that “OHA will establish an initial reserve fund for these claims of twenty percent” of crude oil overcharges). Although DOE undoubtedly hemmed and hawed a good deal on the follow-through, Kalodner and his clients have never pointed to any statement indicating an affirmative intent to renege on the planned distribution of the 20% reserve.

Second, although some of the language in the Con Ed IV decision seems directed to getting DOE moving, there is no claim that Kalodner’s civil litigation helped the beneficiaries by accelerating pay-out. Nor does it appear that there could be. First, it will be recalled that the Con Ed IV expressly declined to impose any deadline. More pertinently, interest has been accruing on the funds (evidently from the outset, and certainly during the period relevant to Kalodner’s litigation activities in Con Ed IV and Con Ed V), see, e.g., Citronelle-Mobile Gathering, Inc. v. Edwards, 669 F.2d 717, 723 (Temp.Emer.Ct.App.1982) (noting “that the Government has a duty to try to ascertain those overcharged, and refund them, with interest, from the restitution funds”) (second emphasis added); Final Procedures, 69 Fed.Reg. at 29,301 (noting that “interest will continue to accrue ... until the refund process is completed”), so the beneficiaries have been and are being held harmless from the effects of delay.

Third, the government and beneficiaries assert that the court should not award a common fund fee because it was DOE’s own pursuit of the overcharging crude oil sellers that led to the accumulation of the funds to be distributed. This is, of course, true, but in significant part it misses the point. It may come as a surprise to counsel, but in all lawsuits producing only money judgments or fund pay-outs, it is not counsel who have created the wealth to be distributed. Mandatory payments do not create wealth (except indirectly, in so far as they enforce rules that provide incentives for wealth-creating behavior); they simply redistribute it. This is true whether the funds distributed originate with taxpayers or, as here, with sellers of crude oil and the government’s refund mavens. But the common fund theory provides a potential basis for payment nonetheless; to the extent that the litigation secured for the beneficiaries sums that otherwise would have flowed to other parties or would have been retained by the government, a common fund fee would be in order. See, e.g., United States v. American Society of Composers, Authors and Publishers, 466 F.2d 917 (2d Cir.1972).

Fourth, Kalodner and the clients mistakenly argue that the district court made a factual finding of causation deserving of deference. Such a finding would of course be surprising, given the court’s dismissal of the claims with respect to Con Ed TV on preclusion grounds. Kalodner points to the district court’s statement that Con Ed V “insured the implementation of the Court’s declaratory judgment in Con Ed IV that DOE should disburse approximately $275 million in funds.” Consolidated Mem. Op. at 10. But this adds up to very little. The language appears more aimed at describing the effect of Con Ed V (which we address below) on ensuring the implementation of the prior declaratory judgment than the effect of the declaratory judgment itself. And the district court’s opinion in Con Ed TV seems in fact (1) to have recognized that its word was by no means the last and (2) to have believed that the government’s primary concern was to be assured that all refund claims should be properly resolved. See 271 F.Supp.2d at 110 (noting that “OHA has advised plaintiffs that it will not be in a position to determine whether any further direct payments to plaintiffs is [sic] warranted until all remaining refund claims are processed”) (internal quotation omitted). Worst for the argument advanced by Kalodner and his clients is that this language runs straight into the earlier order dismissing Con Ed V as moot and saying that the court “decline[d] [plaintiffs’] invitation to declare them ‘victor’ just because the contemporaneous administrative process adopted many of their distribution criteria.” Mem. Op. at 2-3, Con Ed V (June 30, 2004). And the court’s statement that “Kalodner was successful in two adverse civil suits against DOE,” Consolidated Mem. Op. at 5, fails to make a finding as to any substantive consequence of Kalodner’s “success.” While formal success may be minimally sufficient to qualify Kalodner’s clients as prevailing parties under Buckhannon, see 532 U.S. at 604, 121 S.Ct. 1835, it doesn’t establish common fund causation. And the district court simply did not come close to making such a factual finding.

We now turn to the main substantive question of what DOE was likely to have done independent of the litigation. Its communications leave us uncertain how to classify its intent, as between serious contemplation of an ultimate decision not to make the roughly $280 million final distribution and merely a plan to go slow in light of continuing uncertainties. OHA said, for instance, in reply to one of Kalodner’s letters requesting distribution (amid many calling for distribution and also asserting various computational claims), that it “should continue to devote all available resources to the completion of pending original and supplemental applications, before addressing the issue of whether to make a final payment to applicants that have already received refunds [among them, Kalodner’s clients].” Letter from George B. Breznay, Director, OHA, to Philip P. Kalodner (July 11, 2000) (filed as Exhibit D of Plaintiffs’ Motion for Partial Summary Judgment, Con Ed IV (Sep. 4, 2001)). Though the “whether” suggests uncertainty about making any final payment to parties situated as were Kalodner’s clients (and, evidently, the beneficiaries here), the letter also appears to reflect a straightforward matter of priorities— putting work on pending applications first.

Indeed, it isn’t altogether clear that even the clients saw OHA’s position as seriously considering non-payment. Like many communications to the agency, the Con Ed IV complaint seems driven more by plaintiffs’ unsuccessful efforts to get beyond the 20% limitation. Thus the complaint asserted that “Defendant Breznay continues (in decisions issued with regard to claimants being approved for refunds) to refuse to commit DOE to any distribution ... he has indicated that any such subsequent distribution will in any event be limited by employing in all distributions only 20% of the funds.” Complaint at 11, Con Ed IV (Mar. 15, 2001).

On the fee claimants’ side we note that the outstanding potential claims against the fund seem modest in relation to the sums available. In other words, there was no risk that the remaining money in the 20% reserve would be fully or even largely depleted; the lack of money doesn’t seem to have warranted a determination as to “whether any further direct payments ... [are] warranted.” In the government’s motion to dismiss, it described “several hundred refund cases pending” and then noted pending litigation that seemed to put at risk about $11.5 million (DOE noted four pending cases, indicating the amounts at stake in each of three suits, namely $930,063, $3,591,485, and $6,977,635). See Defendants’ Memorandum of Points and Authorities in Support of Their Motion to Dismiss or in the Alternative for Summary Judgment at 15 & n. 4, Con Ed IV (Aug. 1, 2001). In addition, there seem to have been about $1 million outstanding in small claims. See 271 F.Supp.2d at 107, 106 & n. 7, 111 n. 7.

Other statements of the government also seem to reflect an idea that plaintiffs may have been due no more than what they had already received. At one point, for instance, DOE made a rather sweeping statement implying that it thought that a final distribution was entirely discretionary:

[T]he fact that OHA has determined that plaintiffs were eligible to receive an initial distribution does not compel the conclusion that an additional payment is now required. Plaintiffs point to nothing in the record to support such a contention or to show that the funds already paid [to] plaintiffs may not be sufficient to compensate them for any actual injuries suffered.

Defendants’ Reply in Support of Their Motion to Dismiss or in the Alternative for Summary Judgment at 2-3, Con Ed IV (Oct. 1, 2001) (emphasis added). The district court flatly rejected this, finding that plaintiffs “are entitled to the complete distribution of the 20% reserve funds that the DOE created” and that “[t]he DOE cannot now suddenly change that commitment and the implementing regulations unless and until the 20% reserve proves to be more money than needed, which is clearly not the case.” 271 F.Supp.2d at 110.

But other statements cut against the fee claimants’ interpretation. DOE gave strong signs of moving independently towards making a final distribution. For example, DOE spoke of “a determination by DOE as to the distribution of the more than $262 million now in escrow in the U.S. Treasury” and that “Breznay ... has submitted a memorandum [in December 2001] containing his recommendation as to such distribution to the Office of General Counsel of DOE, the contents of which are unknown to plaintiffs, but no action has been taken on such recommendation by the defendant Secretary of Energy.” Joint Memorandum of Status at 6-7, Con Ed TV (Mar. 14, 2002). Indeed, the Joint Memorandum’s summary of DOE’s positions seems to focus on (1) legalistic claims that plaintiffs lack a cause of action to compel immediate distribution and (2) computational issues on which plaintiffs ultimately lost. Id. at 4-6. See, e.g., Letter from Philip P. Kalodner to George B. Breznay, Director, OHA (Mar. 8, 1999) (filed as Exhibit D of Plaintiffs’ Motion for Partial Summary Judgment, Con Ed IV (Sep. 4, 2001)) (stating that “unless you advise me prior to March 31, 1999 that you will recognize my clients’ immediate right to receive the balance of the $2800 per million gallons not yet paid them ... I will institute a mandamus action to require OHA and DOE to make such a supplemental distribution to my clients” and threatening to request an “order[] to make an immediate distribution”). As the plaintiffs in Con Ed IV loudly proclaimed, they had peppered OHA with letters demanding attention and complaining of OHA’s failure to reply promptly. See Plaintiffs’ Statement of Points and Authorities in Opposition to Defendants’ Motion to Dismiss or in the Alternative for Summary Judgment and in Support of Plaintiffs’ Motion for Partial Summary Judgment at 29-31, Con Ed IV (Sep. 4, 2001). Conceivably even a very dutiful official might have come to perceive Kalodner as a nuisance, and such a perception might have colored his reactions and provoked a use of legalistic defenses, even if, as a substantive matter, the government fully intended to distribute the money in any case.

Lastly, we are unpersuaded by the fee claimants’ suggestion that DOE’s Proposed Procedures decision itself establishes that Con Ed IV caused the final distribution. In the summary of the order DOE said that Con Ed IV “rendered a declaratory judgment that successful claimants are entitled to a distribution of the entire remaining amount of crude oil overcharges reserved for direct restitution, ‘insofar as practicable.’ OHA will therefore make a final distribution in the longstanding crude oil refund proceeding.” 68 Fed.Reg. at 64,098 (emphasis added). But in this very passage DOE refers to the amount as already “reserved for direct restitution,” id., arguably implying that it would have been distributed even without the declaratory judgment.

In the end, our efforts to draw an inference of causation face major informational deficits. Notably, the record contains allusions to the December 2001 OHA memorandum to DOE counsel proposing a disposition of the $270 million then on hand, see Reply Memorandum in Support of Motion for Award of Common Fund Fee at 16-17, Con Ed TV (June 12, 2003), but not the memorandum itself. Clients assert that DOE has refused to release it. Id. The OHA memorandum itself, of course, may not be dispositive, as the ultimate decision may have lain with others, such as DOE counsel or perhaps the Secretary of Energy. Given the ambiguities in OHA’s formal public position, we are unable to reach a conclusion about causation and we agree with the clients and Kalodner that limited discovery may be useful to bring OHA’s position to light and thus afford them an adequate opportunity to establish that DOE would not have paid out the refunds had Con Ed IV never been brought.

B. The volumetric adjustment of Con EdV

While sovereign immunity bars the clients’ claim to a fee for the legal efforts involved in Con Ed V (for want of court-ordered relief), that doctrine has no effect on Kalodner’s claim against the beneficiaries based on that case (claim 6). We thus must assess whether those efforts increased the common fund by certain adjustments in the “volumetric” amount, or simply “volumetric,” used to calculate refunds. The “volumetric” amount represents the total dollar amount remaining in the reserve (the numerator) divided by the total number of gallons purchased by all eligible claimants (the denominator). See Proposed Procedures, 68 Fed.Reg. at 64,-100. Each claimant would then receive a refund of the volumetric times the number of gallons purchased by that claimant (effectively a weighted average of the reserve). Kalodner claims that Con Ed V increased the common fund via two adjustments of the volumetric adopted in the Final Procedures. First, the Final Procedures included $9.5 million in escrow in the “Citronelle” account in the numerator, thereby increasing the total payout to all beneficiaries; the Proposed Procedures hadn’t mentioned this one way or the other.

Second, the Final Procedures deferred calculation until verification of all claims (other than time-barred ones) was complete, thus excluding ineligible claims from the denominator (and thereby increasing the pay-out). Claims might ultimately be found ineligible for a number of reasons, including: (a) forfeiture by large refund recipients of future claims due to failure to request supplement refunds, (b) failure by small refund recipients to apply for a final distribution due in large part because no notice would be provided, (c) a finding that claimants are unqualified suceessors-in-interest, or (d) reduction of a prior award. See Philip P. Kalodner, Comments of Utilities, Transporters and Manufacturers at 7-13 (Jan. 8, 2004); Douglas B. Mitchell, Comments Regarding the Proposed Procedures for Distribution of Remaining Crude Oil Overchange [sic] Refunds at 3 (Jan. 12, 2004). Here the difference between the Proposed and the Final Procedures appears sharper than for the Citronelle account, as the Proposed Procedures seemed to include such claims in the denominator, whereas Kalodner’s and Mitchell’s proposed use of the (already planned) 180-day notice period could be expected to weed out the ineligibles.

Kalodner claims that the two adjustments increased the common fund by $35 million. As both the Proposed and the Final Procedures made this round of distributions truly final, allocating any leftover sums to state and federal governments, see Proposed Procedures, 68 Fed. Reg. at 64,100; Final Procedures, 69 Fed. Reg. at '29,301-02, the adjustments came at the expense of those governments. (It appears that Kalodner is including interest payments. These of course did increase the gross sum paid out — but only by an amount needed to compensate recipients for the delay that Kalodner himself sought.)

The district court attributed this $35 million increase to Kalodner’s litigating efforts in Con Ed V and awarded a fee calculated as 30% of that supposed increase. We find nothing in the record supporting the idea that Con Ed V played any such role.

The fact that the Con Ed V suit was dismissed for mootness is not in itself dis- positive. It is true that common fund cases typically hinge on some form of court-ordered relief. See, e.g., Mills v. Electric Auto-Lite Co., 396 U.S. 375, 393-94, 90 S.Ct. 616, 24 L.Ed.2d 593 (1970); National Treasury Employees Union v. Nixon, 521 F.2d 317, 320-21 (D.C.Cir. 1975); see generally Alba Conte, I Attorney Fee Awards § 2.1 at 41 (3d ed.2005) (noting that “the unarticulated threshold requirement for application of the common-benefit doctrine is that the claimant must enjoy some form of success on the merits of the litigation”). But in this area some version of the catalyst theory applies, illustrated by decisions awarding common fund fees even where the claimants’ action was dismissed as moot. In Koppel v. Wien, 743 F.2d 129 (2d Cir.1984), for instance, the Second Circuit reversed the district court’s denial of a common fund fee where the defendants, after the suit was filed, had voluntarily abandoned the project plaintiffs had sued to enjoin. Id. at 131-32, 135. See also Savoie v. Merchants Bank, 84 F.3d 52, 56-57 (2d Cir. 1996). Although in Koppel and Savoie the Second Circuit held that where a case is mooted the burden shifts to defendants to prove the absence of causation, see 743 F.2d at 135, 84 F.3d at 57, in both cases the record appeared to offer no plausible explanation for the defendants’ action other than the lawsuit itself. In contrast, the Con Ed V litigation occurred in parallel with an entirely separate administrative proceeding conducted by DOE, in which Kalodner and others participated actively. As the district court said in finding the Con Ed V suit moot, “[t]he appropriate venue for consideration of the plaintiffs’ proposed distribution methodology was the administrative comment process, which they successfully utilized.” Mem. Op. at 3, Con Ed V (June 30, 2004). Kalodner doesn’t even appear to claim that his persuasive efforts before DOE, independent of some supposed judicial pressure induced by his civil litigation, could entitle him to fees. That implied concession appears in full accord with the law. See Knight, 982 F.2d at 1576, 1581 (denying common fund fee for results of administrative action taken before any court order or indeed any filing of suit).

Of the two changes supposedly wrought by Con Ed V, we consider first the idea of deferring the calculation until the end of a 180-day period (already provided for in the Proposed Procedures), so as to exclude unresolved claims from the denominator of the fraction governing the beneficiaries’ entitlements. The complaint in Con Ed V never requests any such deferral. It merely requests “an Order directing the defendants to distribute to plaintiffs and the members of the class an amount per million gallons of qualified product purchases determined pursuant to the formula set forth in paragraph 33 [of the complaint], some $650 to more than $700 per million gallons.” The $650-700 per million gallons is close to the range that DOE itself proposed in its Proposed Procedures. See 68 Fed.Reg. at 64,100 (proposing volumetric amount of $670 per million gallons). Given the numbers in the complaint, we are baffled by Kalodner’s assertion on brief that Con Ed V increased the amount from $670 to $750-800 per million gallons.

Worse for Kalodner, the complaint appears to demand a denominator consisting of “the sum of the volume of purchases by applicant end user claimants already found qualified for recovery and the volume of purchases by claimants whose claims have not as yet been processed.” Complaint at 11, Con Ed V (Sep. 25, 2003) (emphasis added). It thus implicitly urged inclusion of those very claims for which Kalodner, in his administrative comment, successfully advocated exclusion. The relationship completely contradicts Kalodner’s claims for Con Ed V.

In fact, the first time that Kalodner’s clients ever appeared to raise the deferral issue in Con Ed V was in their motion for summary judgment, filed with the court nearly four months after the complaint and eight days after Kalodner filed comments in the administrative proceeding. See Plaintiffs’ Memorandum in Opposition to Defendants’ Motion to Dismiss and Statement of Points and Authorities in Support of Plaintiffs’ Motion for Summary Judgment at 13-14, Con Ed V (Jan. 16, 2004). See also Plaintiffs’ Memorandum in Reply to Defendants’ Opposition to Plaintiffs’ Motion for Summary Judgment at 10, Con Ed V (Mar. 18, 2004) (noting that deferral was raised in the “Initial Memorandum,” i.e., the motion for summary judgment, see id. at 3-4, but not noting the complaint); Defendants’ Memorandum of Points and Authorities in Reply to Plaintiffs’ Opposition to Defendants’ Motion to Dismiss and in Opposition to Plaintiffs’ Motion for Summary Judgment at 2, Con Ed V (Feb. 20, 2004) (correctly noting that “[njone of these allegations [about deferral] are raised in plaintiffs’ complaint in this matter which simply sought the distribution the agency has stated it will undertake.”).

Further weakening the causal link is the fact that not only Kalodner, but another lawyer, Douglas B. Mitchell, acting on behalf of 104 individual claimants and two filing services, filed a comment suggesting deferral. See Douglas B. Mitchell, Comments Regarding the Proposed Procedures for Distribution of Remaining Crude Oil Overchange [sic] Refunds at 2-3 (Jan. 12, 2004) (“Mitchell Comments”) (suggesting deferral until verification is complete, after a 180-day last-chance notice period); see also Declaration of George B.- Breznay, Con Ed V (Feb. 9, 2005). Even where court action is the source of the relief sought, the fact that parties with interests in the common fund were separately represented may militate against the award of a common fund fee. See, e.g., United States v. Tobias, 935 F.2d 666, 668 (4th Cir.1991); Vincent v. Hughes Air West, Inc., 557 F.2d 759, 771 (9th Cir.1977); see generally 20 Am. Jur. 2D Costs § 66. In any event, as we noted earlier, Kalodner appears to concede that triumphs at the agency level, unless shown to have been caused by some sort of actual or realistically threatened judicial action, give rise to no common fund entitlement.

Unlike its treatment of deferral, the complaint at least took the same position on the $9.5 million Citronelle account that Kalodner did in the administrative proceeding. But there is no evidence that the Con Ed V filing caused its inclusion in the numerator in the Final Procedures. As with deferral, Mitchell’s comment also advocated the inclusion of the Citronelle account. See Mitchell Comments at 2. More important, inclusion of the Citronelle refund appears to have already been contemplated by DOE. In its reply to the comments, DOE expressly stated, “It is already DOE’s practice that ‘returned funds’ ... are deposited.” 69 Fed.Reg. at 29,301. Further, the settlement under which the Citronelle funds were recouped (to which Kalodner was a signatory) itself required that those funds be paid to the other crude oil end users. See Declaration of George B. Breznay at ¶ 16 (February 9, 2005). Kalodner offers nothing-other than a conclusory assertion to contradict the reasoning behind Breznay’s explanation of why “those funds would have been included in the final crude oil distribution, regardless of any comment by Mr. Kalodner.” Id. See also Brief for the Appellees/Cross-Appellants at 33 n. 6.

Lastly, we also find that there is no evidence that Con Ed V contributed to the probability of the final distribution vel non. It is undisputed that in a series of telephone conversations with Kalodner from August 25 to September 22, 2003, before the September 25, 2003 filing of the Con Ed V complaint, DOE Assistant General Counsel Skubel indicated that OHA was proceeding with plans for a final distribution. Nonetheless, Kalodner’s clients proceeded to file their complaint. Moreover, that filing occurred only some 20 weeks after Con Ed TV’s declaratory judgment, which itself specifically left timing to OHA’s discretion. See 271 F.Supp.2d at 111. The timing and circumstances suggest that Con Ed V did nothing more than exhibit once again Kalodner’s trigger-happiness (and perhaps that of his clients). Notwithstanding the district court’s language, there is simply no evidence in the record that Con Ed V in any way “caused” the distribution itself.

* * * * *

To recap by reference to the claims as enumerated in Table 1: the December 4, 2003 order may preclude the clients’ fee claim against the government for Con Ed TV (claim 1). If not, the claim turns on the causal effect (if any) of Con Ed TV. The clients’ claim against the government for Con Ed V (claim 2) and each of the claims by Kalodner against the government (claims 3 & 4) are barred by sovereign immunity. Lastly, Kalodner’s claim against the beneficiaries for Con Ed TV (claim 5), if not precluded, turns on the causal effect of Con Ed TV, while that for Con Ed V (claim 6) fails for lack of causation.

We repeat that on remand the preclusion arguments are not themselves precluded. To the extent that claims (not already defeated by sovereign immunity) survive any reconsideration of preclusion, the court should conduct a limited hearing or discovery for purposes of determining the causal effect of Con Ed IV. See Copeland v. Marshall, 641 F.2d 880, 905 n. 57 (D.C.Cir.1980).

If Kalodner or the clients get past the basic causation hurdle, fee computation may be quite complex. In Democratic Central Committee of D.C. v. WMATC, 38 F.3d 603, 606 (D.C.Cir.1994), we noted that payment should be allowed “only as a reasonable proportion of the amount actually collected ... for which petitioners’ attorneys were responsible,” i.e., proportional to the degree to which the civil litigation enhanced the probability of pay-out to the beneficiaries in question and the amount distributed. Presumably the aim should be to assure that Kalodner’s total recovery would approximate what a single claimant to the common fund would have negotiated with him absent the transactions costs due to free-rider temptations and sheer numbers. Thus, if the probable effect of the litigation was to raise the chances of recovery from, say, 95% to 100%, we suppose that this relationship would be reflected in the fee. We note also that Kalodner originally sought a 5% common fund fee in Con Ed TV, whereas he asks for 10% here. Compare Motion for Award of Common Fund Fee at 3, Con Ed TV (May 23, 2003) (requesting “5% of the amount to be distributed to all claimants”); with Motion of Award of Common Fund Fee at 1, Con Ed V (July 12, 2004) (requesting 10% fee); Complaint at 2, Kalodner-Abraham (Jan. 7, 2005) (requesting 10% fee); First Amended Complaint at 18, Kalodner-Public Service (Mar. 8, 2004) (requesting 10% fee). If the 5% represents Kalodner’s own theory of the marginal value of his contribution in Con Ed TV, whereas the 10% request here represents his guess for efforts both in Con Ed TV and Con Ed V, then our denial of his fee request in Con Ed V would have implications for Kalodner’s maximum claim.

We affirm the dismissal of Kalodner’s claims against the government for his efforts in both Con Ed TV and Con Ed V and of his claim against the beneficiaries for his work in Con Ed V (claims 3, 4 & 6). We reverse the grant of the clients’ claim against the government relating to Con Ed V (claim 2). This leaves two claims — the clients’ claim against the government, and Kalodner’s claim against the beneficiaries — both for Kalodner’s efforts in Con Ed TV (claims 1 & 5). As to these, we reverse the judgments denying recovery and remand for further proceedings consistent with this opinion.

So ordered.

1 . See Notice of Final Procedures for Distribution of Remaining Crude Oil Overcharge Refunds, 69 Fed.Reg. 29,300, 29,301 (May 21, 2004). We use these numbers only to give an idea of the magnitudes; in the event that any fees are awarded, sorting out the amounts, and the degree to which they are attributable to the one court decision left standing as conceivably justifying a fee (Con Ed IV), will not be simple.

2 . Kalodner simultaneously appealed to the Federal Circuit as No. 05-1310, which that circuit deferred pending decisions here. See Order (Fed.Cir. May 31, 2005).

3 . The clients simultaneously appealed to the Federal Circuit as No. 05-1309, and DOE cross-appealed there as No. 05-1450. The Federal Circuit entered an order deferring consideration of these parallel appeals. See Order (Fed. Cir. June 17, 2005).

4 . Kalodner simultaneously appealed to the Federal Circuit as No. 05-1214, which the circuit deferred pending decisions here. See Order (Fed.Cir. Mar. 25, 2005).