COMMONWEALTH OF PENNSYLVANIA, ET AL., PETITIONERS V. DELAWARE VALLEY CITIZENS' COUNCIL FOR CLEAN AIR, ET AL. No. 85-5 In the Supreme Court of the United States October Term, 1985 On Writ of Certiorari to the United States Court of Appeals for the Third Circuit Brief for the United States PARTIES TO THE PROCEEDING In addition to the parties identified in the caption of this case, the Secretary of the Pennsylvania Department of Transportation and the Secretary of the Pennsylvania Department of Environmental Resources were named as defendants in the district court and are petitioners in this Court. The United States of America was a plaintiff in the district court and is a respondent in this Court. See Rule 19.6 of the Rules of this Court. TABLE OF CONTENTS Question Presented Parties to the Proceeding Opinions below Jurisdiction Statute involved Statement Summary of argument Argument: I. The risk of failing to prevail in litigation does not justify an upward adjustment of statutorily awarded attorneys' fees A. Contingency multipliers improperly compensate nonprevailing parties B. Contingency multipliers are contrary to basic notions of fairness and lead to anomalous results C. Contingency multipliers provide improper double compensation D. Contingency multipliers are contrary to the intent of Congress E. The award of a contingency multiplier in this case was improper II. The award of multipliers based on the quality of representation in this case was erroneous III. Section 304(d) of the Clean Air Act does not authorize compensation for time spent commenting on state and federal regulatory proposals Conclusion OPINIONS BELOW The opinion of the court of appeals (Pet. App. 46a-70a) is reported at 762 F.2d 272. The opinion of the district court (Pet. App. 1a-45a) is not reported. JURISDICTION The judgment of the court of appeals was entered on May 14, 1985. The petition for a writ of certiorari was filed on July 3, 1985, and was granted on October 7, 1985. The jurisdiction of this Court is invoked pursuant to 28 U.S.C. 1254(1). STATUTE INVOLVED Section 304(d) of the Clean Air Act. 42 U.S.C. 7604(d), provides in pertinent part as follows: The court, in issuing any final order in any action brought pursuant to subsection (a) of this section, may award costs of litigation (including reasonable attorney and expert witness fees) to any party, whenever the court determines such award is appropriate. QUESTIONS PRESENTED 1. Whether a presumptively reasonable attorneys' fee award under Section 304(d) of the Clean Air Act, 42 U.S.C. 7604(d), may be "multiplied" to reflect the risk that plaintiffs might not have prevailed and, therefore, might have obtained no attorneys' fees at all. 2. Whether the lower courts in this case erred by enhancing a presumptively reasonable attorneys' fee to reflect the high quality of representation rendered by respondent's counsel. 3. Whether Section 304(d) of the Clean Air Act authorizes attorneys' fee awards for time spent commenting on state and federal regulatory proposals. STATEMENT 1. Petitioners challenge certain aspects of an award of attorneys' fees granted to respondent Delaware Valley Citizens' Council for Clean Air (Delaware Valley). The attorneys' fees in question were awarded for work performed by Delaware Valley's counsel between 1979 and 1983 in connection with the enforcement of a 1978 federal court consent decree. That decree was in settlement of consolidated lawsuits brought by Delaware Valley and the United States to require petitioners to meet their obligations under the Clean Air Act, 42 U.S.C. 7401 et seq. Pursuant to the Act, the Pennsylvania Department of Environmental Resources (Penn DER) submitted to the United States Environmental Protection Agency a plan for attaining federal air quality standards. As modified and approved by EPA, Penn DER's plan provided for the development of a program to inspect and maintain automobile emission-control systems (the I/M program). The I/M program was designed to meet carbon monoxide and ozone standards in metropolitan Philadelphia and southwestern Pennsylvania, and it was to be implemented by May 1, 1975. Pet. App. 2a. Both Delaware Valley and the United States filed suit against petitioners when they failed to implement the I/M program; the two suits were later consolidated. Ibid. /1/ On August 29, 1978, petitioners -- the Commonwealth of Pennsylvania and two of its executive agencies, the Department of Transportation (Penn DOT) and Penn DER -- signed a consent decree terminating the litigation. Petitioners agreed to implement an I/M program by August 1, 1980, using one of two approaches specified in the consent decree. The decree provided that Penn DOT would seek legislation authorizing that agency to enter into contracts with garage owners to establish inspection stations. If such legislation were not enacted, however, Penn DOT would issue regulations under which privately-owned garages would be certified as inspection stations. /2/ Finally, the decree provided for payment to Delaware Valley of attorneys' fees and costs incurred to that date; the total fee award was $38,700, with $30,000 to be paid by petitioners and $8,700 to be paid by EPA to account for the time during which EPA was a defendant in Delaware Valley's lawsuit (see note 1, supra). Pet. App. 2a-3a. 2. Unfortunately, implementation of the I/M program did not proceed as specified in the consent decree. In March 1980, the district court approved a modification to the decree, as requested by petitioners and agreed to by Delaware Valley and the United States, delaying implementation of the I/M program until May 1981. Pet. App. 3a. In February 1981, petitioners asked Delaware Valley and the United States to agree to a further postponement of the implementation date, to January 1, 1983. Petitioners sought the second delay to enable them to redraft their not-yet-promulgated regulations identifying the type of testing equipment that private garages would be required to obtain in order to become certified inspection stations. Specifically, petitioners wanted to require computerized analysis equipment, as had been recommended by EPA for decentralized inspection programs such as Pennsylvania's. However, no manufacturer had yet marketed such equipment or even developed a prototype. Id. at 3a-4a. After extensive negotiations regarding the second request for an extension, the parties reached an impasse. Accordingly, on April 29, 1981, petitioners filed a motion asking the district court to grant the second extension. In response, Delaware Valley filed a contempt motion, based on petitioners' failure to implement the I/M program by May 1, 1981, as required by the modified consent decree. Pet. App. 4a. On May 20, 1981, the district court denied both Delaware Valley's contempt motion and petitioners' motion for a further extension; instead, the court ordered petitioners to submit a plan for immediate implementation of an I/M program. Such a plan was submitted and approved by the district court. On June 16, 1981, the decree was further modified, and the deadline for implementation of the I/M program was extended to May 1, 1982. Pet. App. 4a. Petitioners appealed the May 20 and June 16 orders, both of which were affirmed by the court of appeals. Delaware Valley Citizens' Council for Clean Air v. Pennsylvania, 674 F.2d 976 (3d Cir.), cert. denied, 459 U.S. 905 (1982). During the pendency of that appeal, the Pennsylvania General Assembly enacted a statute, H.B. 456, over the governor's veto, which prohibited the expenditure of state funds for implementation of the I/Mprogram. Act of Oct. 5, 1981, No. 99, 1981 Pa. Legis. Serv. 31 (Purdon). As directed by the state statute, Pennsylvania's executive agencies promptly ceased virtually all activities related to the I/M program. On October 10, 1981, however, Penn DOT did publish final regulations establishing specifications for the emissions analysis equipment to be used in the inspection program. 11 Pa. Bull. 3519 (Oct. 10, 1981). Pet. App. 5a. On October 12, 1981, petitioners moved to stay implementation of the consent decree in light of H.B. 456. Delaware Valley opposed that motion and again sought contempt penalties. On January 22, 1982, the district court denied petitioners' motion for a stay and held petitioners in civil contempt. As a sanction, the court ordered the United States Secretary of Transportation to refrain from approving any projects or awarding any grants for highways in the two regions covered by the consent decree, except for projects required for purposes of safety, mass transit, or air quality improvement. Pet. App. 5a-6a. The court of appeals upheld the contempt ruling, and this Court declined to grant review. Delaware Valley Citizens' Council for Clean Air v. Pennsylvania, 533 F. Supp. 869 (E.D. Pa.), aff'd, 678 F.2d 470 (3d Cir.), cert. denied, 459 U.S. 969 (1982). Thereafter, the Pennsylvania General Assembly passed legislation authorizing petitioners to proceed with the I/M program; the governor signed that bill into law on May 4, 1983. Subsequently, the parties to the litigation negotiated a new compliance schedule, under which the I/M program would be implemented by June 1, 1984. On May 16, 1983, the district court ratified the new schedule and vacated the contempt sanctions. Pet. App. 6a. 3. Delaware Valley then sought attorneys' fees for the work it had performed between 1979 and 1983 in connection with enforcement of the consent decree. By order dated February 24, 1984, the district court awarded Delaware Valley $216,488.03 in attorneys' fees and costs for this work. Pet. App. 1a-45a. /3/ The district court followed Delaware Valley's suggested approach of dividing the post-decree litigation into nine "phases," and the court separately calculated the fee award for each phase. Although petitioners raised numerous objections to Delaware Valley's fee request, we confine our discussion of the district court's decision to the issues that petitioners raise in this Court. a. For the most part, the hours for which Delaware Valley sought compensation were expended on the post-decree litigation itself. In two instances, however, petitioners objected that Delaware Valley was seeking compensation for its participation in state and federal administrative proceedings. Phase II of the post-decree proceedings consisted of time spent by Delaware Valley "monitor(ing) (petitioners') performance under the consent decree and submitt(ing) comments on the regulations which were published in the Pennsylvania Bulletin" (Pet. App. 17a). The district court rejected petitioners' contention that time spent commenting on the state's proposed regulations was not compensable (ibid.): While it is true that the opportunity to submit comments to the proposed regulations was available to any person, * * * (Delaware Valley) had an interest in the regulations since (it was) a party to the consent decree. (Its) interest was based on a desire to ensure compliance with the consent decree and to protect (its) rights thereunder. The usefulness of (Delaware Valley's) comments was manifested in the revisions that were made to the original proposed regulations. Phase IX of Delaware Valley's post-decree activities consisted of work performed "in regulatory actions, such activities not directly required by (the district) court's Orders" (Pet. App. 34a). Specifically, Delaware Valley "participated in EPA hearings during which: 1) (petitioners) requested that EPA modify the state implementation plan (SIP) so that the I/M program would cover a smaller geographic area; and 2) proposals of cut-offs of EPA funding were discussed" (id. at 36a). The district court reasoned that the first item was compensable but the second was not (ibid.): The activities which occurred due to (petitioners') request for a modification of the I/M program were related to this litigation since an EPA modification of the I/M program would have affected (Delaware Valley's) rights under the decree. For these activities (Delaware Valley) will be compensated. However, participation in hearings concerning the possibility that EPA would cut-off federal grants to Pennsylvania was activity which was not sufficiently related to this litigation since any such EPA decision would not have affected (Delaware Valley's) rights under the decree. Therefore the hours spent on this latter activity will be denied. b. Petitioners' other major objection to the district court's fee award was their opposition to Delaware Valley's request that the "lodestar" fee for certain phases of the post-decree litigation be increased by the use of upward adjustments, or "multipliers." Specifically, Delaware Valley sought a multiplier of two for Phases IV and V and a multiplier of three for Phase VII. Pet. App. 38a. /4/ The district court, which rendered its decision prior to this Court's decision in Blum v. Stenson, No. 81-1374 (Mar. 21, 1984), did enhance Delaware Valley's "lodestar" fee through the use of multipliers, but neither the size of the multipliers nor the phases of the litigation for which they were awarded coincided with Delaware Valley's request. Thus, the district court awarded a multiplier of four for Phase V, even though Delaware Valley had requested only a multiplier of two, and the court awarded a multiplier of two for Phase VII, as to which Delaware Valley had sought a multiplier of three. Pet. App. 39a-40a. The district court did acquiesce in Delaware Valley's request for a multiplier of two for Phase IV. Ibid. The court did not explain the reasons for the two discrepancies between the request and the award. The district court's entire analysis of multipliers consisted of three paragraphs (Pet. App. 39a-40a). The court recited both "the contingent nature of (Delaware Valley's) success" and "(t)he quality of (Delaware Valley's) work" as factors justifying the award of multipliers (id. at 39a). Although the district court's opinion is somewhat ambiguous, it appears that the two-fold multipliers awarded for Phases IV and VII were based solely on the risk of nonsuccess, while the fourfold multiplier awarded for Phase V was based both on the risk of nonsuccess and the quality of Delaware Valley's work. c. The court of appeals affirmed the district court's award of attorneys' fees and costs in its entirety (Pet. App. 46a-70a). The court affirmed the award of fees for time spent commenting on petitioners' proposed regulations for the reasons stated by the district court (id. at 55a). As for the time devoted to EPA's regulatory proceedings, the court of appeals agreed that a fee award was proper "because adoption of the state plan modification would have impaired the rights won by (Delaware Valley) in the consent decree" (id. at 57a). The court took note (id. at 57a n.7) of this Court's decision in Webb v. County Bd. of Educ., No. 83-1360 (Apr. 17, 1985), in which the Court suggested in dicta that time spent on administrative proceedings may be compensable under 42 U.S.C. 1988 if the work was "both useful and of a type ordinarily necessary to advance the civil rights litigation to the stage it reached before settlement" (Webb, slip op. 9). The court of appeals concluded that the administrative work in this case was "useful and necessary for securing full enforcement of the decree" (Pet. App 57a n.7). With respect to the use of multipliers, the court of appeals concluded that "this was 'the rare case where the fee applicant offer(ed) specific evidence to show that the quality of service rendered was superior to that one reasonably should expect in light of the hourly rates charged and that the success was 'exceptional.'" Pet. App. 64a (quoting Blum, slip op. 11-12). The court also approved the use of "contingency" multipliers to compensate Delaware Valley for the risk of nonsuccess. The court stated (Pet. App. 68a): Unlike Blum, (Delaware Valley) specifically identified the risks inherent in this litigation in its brief to the district court and, although the Supreme Court considers it an open question whether contingency of success can properly justify a lodestar increase, we have resolved the question in this court. See Hall v. Borough of Roselle, 747 F.2d 838 (3d Cir. 1984); (Lindy Bros. Builders, Inc. v. American Radiator & Standard Sanitary Corp., 540 F.2d 102, 117 (3d Cir. 1976) (en banc)). Judge Becker dissented from the majority's affirmance of the district court's award of multipliers (Pet. App. 69a-70a n.12). Judge Becker was of the view that the risk of not prevailing in Phases IV and VII was "simply insufficient to justify the very substantial multipliers awarded by the district court," because, as to those two phases, petitioners "had to meet an extemely heavy burden to gain modification of the consent decree" (ibid.). With respect to the four-fold multiplier awarded for Phase V, Judge Becker would have remanded the matter to the district court for reconsideration in light of Blum. Judge Becker's views were expressed as follows (Pet. App. 69a-70a n.12): Judge Becker acknowledges that Blum did not address the question of how high a multiplier the district court may award * * *. Nevertheless, he believes that because the Blum court carefully identified the cases in which fees may be awarded and characterized a 50 percent multiplier as "substantial" (Blum, slip op. 11), Blum suggests that in only the rarest case would a total multiplier in the range of four be permitted. Judge Becker concludes that, even assuming an award of quality and contingency multipliers is appropriate as to phase V, the multipliers must be recalculated because this case was not so very rare as to justify in light of Blum the award of this extraordinary multiplier. SUMMARY OF ARGUMENT I. In Blum v. Stenson, No. 81-1374 (Mar. 21, 1984), slip op. 10, the Court held that a reasonable number of hours multiplied by a reasonable hourly rate "is presumed to be the reasonable (attorneys') fee contemplated by (42 U.S.C.) Section 1988." Although the Court declined to hold that upward adjustments, or "multipliers," are never permissible (Blum, slip op. 10), the Court did severely limit the circumstances under which such adjustments may be made. However, because the record in Blum contained no evidence whatever to support the lower courts' award of a 50% multiplier based on the risk of nonsuccess -- a "contingency" multiplier -- the Court specifically left open the principal question presented in this case -- "whether the risk of not being the prevailing party, * * * and therefore not being entitled to an award of attorney's fees from one's adversary, may ever justify an upward fee adjustment." Id. at 13 n.17. In our submission, at least three separate problems associated with "contingency" multipliers demonstrate that Congress never intended their use in the calculation of "reasonable" attorneys' fee awards under fee-shifting statutes such as Section 1988 or Section 304(d) of the Clean Air Act. First, contingency multipliers improperly compensate nonprevailing parties for their unsuccessful lawsuits, in contravention of Congress's directive, upheld by this Court in cases such as Ruckelshaus v. Sierra Club, 463 U.S. 680 (1983), and Hensley v. Eckerhart, 461 U.S. 424 (1983), that fees be awarded only to prevailing parties. Second, contingency multipliers lead to profoundly anomalous and inequitable results. Because contingency multipliers are set according to the district court's after-the-fact assessment of a plaintiff's chance of success at the outset of the litigation, the net result is that the more questionable a plaintiff's original claim -- and hence the greater the liklihood that his attorney might receive no fee at all -- the greater the attorney's "contingency" bonus. This system unfairly penalizes the defendant who legitimately relied on apparently meritorious defenses, while allowing the defendant who had virtually no defense whatever to escape liability for any "contingency" bonus. Not only is this compensation system unfair to defendants, but it produces strong economic incentives for plaintiffs to pursue marginal cases in hopes of obtaining a fee bonus. In addition, the actual setting of a contingency multiplier more closely resembles the "Wheel of Fortune" or the Irish Sweepstakes than the calculation of "reasonable" attorneys' fees. Under the current system of contingency compensation, a defendant may be saddled with hundreds of thousands of dollars in additional liability that bears no relationship to the harm actually suffered by the plaintiff or the hours worked by the plaintiff's attorneys. Third, contingency multipliers substantially duplicate elements of compensation that are already reflected in the "lodestar" fee. When the underlying facts of a case are convoluted or the legal theories are novel or complex, the attorney will spend correspondingly large amounts of time gathering information, conducting discovery, reviewing documents, researching precedents, investigating the legislative history, drafting pleadings and writing briefs to formulate persuasive legal and factual arguments. Thus, the same obstacles that are likely to make it more difficult for the plaintiff to succeed on the merits also substantially increase the amount of time reasonably expended on a case or the hourly rates that a court may reasonably award. In these circumstances, the "lodestar" fee will produce the fully compensatory fee that Congress intended, and any "enhancement" for the risk of nonsuccess will constitute double counting. These perverse aspects of contingency multipliers cannot be squared with the legislative history of the Civil Rights Attorney's Fee Awards Act of 1976, 42 U.S.C. 1988, in which Congress established the basic rules applicable to nearly all fee-shifting statutes. In enacting Section 1988, Congress clearly identified the touchstone for determining the amount of a "reasonable" award of attorneys' fees: "(F)ees (should be) adequate to attract competent counsel, but (should) not produce windfalls to attorneys." S. Rep. 94-1011, 94th Cong., 2d Sess. 6 (1976). Clearly, paying an attorney for years of work he never performed -- the actual effect of contingency multipliers -- strays far beyond the "reasonable" fee needed to attract competent counsel. With these factors in mind, it is apparent that the multipliers of two and four awarded in this case were improper. Neither of the courts below found that the number of hours for which Delaware Valley's counsel was compensated or the hourly rates they received were insufficient to reflect fully any "risk" they may have assumed. This omission is not surprising, because the presumed "risk" was in fact virtually nonexistent. Once the consent decree was signed in 1978, Delaware Valley was assured of prevailing in this case. The post-decree litigation, while indeed protracted and oftentimes complex, never seriously threatened to take away Delaware Valley's status as a prevailing party. Particularly in light of the presence of the United States as a plaintiff intent on ensuring petitioners' compliance with the requirements of the Clean Air Act, the "risks" faced by Delaware Valley in the post-decree litigation were never sufficiently substantial to justify the award of contingency multipliers. Finally, even if one assumes that the use of multipliers was appropriate in this case, the extraordinarily high multipliers actually employed were grossly excessive. The net effect of the two-fold and four-fold multipliers was to compensate Delaware Valley's counsel for more than $124,000 they never earned -- i.e., for over 1,700 hours never expended on the case. Under any standard, such an award constitutes a "windfall to attorneys" (S. Rep. 94-1011, supra, at 6). II. The lower courts also erred in awarding multipliers based on the superior quality of Delaware Valley's representation and the "excellent" results obtained. Because the district court did not have the benefit of this Court's decision in Blum, it is not surprising that its opinion wholly fails to articulate a sufficient rationale for the use of "quality" multipliers. And the court of appeals' post hoc attempt to apply Blum to the inadequate findings of the district court was similarly defective. Delaware Valley simply failed to offer the "specific evidence" necessary to demonstrate that it obtained "excellent" results and that its counsel's services were superior to the quality of representation to be expected from attorneys being compensated at the high hourly rates employed by the district court. III. The lower courts also erred in awarding Delaware Valley compensation for time spent participating in state and federal regulatory proceedings. In only one case, New York Gaslight Club, Inc. v. Carey, 447 U.S. 54 (1980), has this Court held that a fee-shifting statute authorizes reimbursement for work done at the administrative level. The instant case, however, contains neither of the two factors that supported the result reached in Carey. First, the language of Section 304(d) of the Clean Air Act is significantly different from the language of Section 706(k) of the Civil Rights Act of 1964, 42 U.S.C. 2000e-5(k). That section authorizes fees for work in "any action or proceeding" brought under Title VII. As the Court recognized in Carey, 447 U.S. at 61, a "proceeding" is something different from an "action." Section 304(d) of the Clean Air Act, on the other hand, authorizes fees only in connection with "action(s)." The "actions" contemplated by Section 304(d) are judicial actions, not administrative proceedings. Second, it is undisputed that Delaware Valley was under no obligation to participate in the state and federal administrative proceedings. This situation contrasts sharply with that before the Court in Carey, in which the enforcement structure of Title VII required a potential plaintiff to pursue available state and local remedies before resorting to federal court. Contrary to the court of appeals' conclusion (Pet. App. 57a n.7), nothing in this Court's recent decision in Webb v. County Bd. of Educ., No. 83-1360 (Apr. 17, 1985), calls for a different result. To begin with, the Court in Webb was construing 42 U.S.C. 1988, which, like Section 706(k) of the Civil Rights Act, expressly refers to "proceeding(s)" as well as "action(s)." Furthermore, the most that can be said of Webb, in which the Court rejected a fee applicant's claim for reimbursement for administrative work, is that work is "both useful and of a type ordinarily necessary" to advance the litigation (Webb, slip op. 9) is a necessary but not sufficient condition for a fee award. Here, the work done by Delaware Valley at the administrative level, while perhaps "useful," was not "of a type ordinarily necessary" to protect its rights under the consent decree, because at all times the district court retained jurisdiction to ensure the integrity of the decree. ARGUMENT I. THE RISK OF FAILING TO PREVAIL IN LITIGATION DOES NOT JUSTIFY AN UPWARD ADJUSTMENT OF STATUTORILY AWARDED ATTORNEYS' FEES In Blum v. Stenson, No. 81-1374 (Mar. 21, 1984), slip op. 10, the Court held that a reasonable number of hours multiplied by a reasonable hourly rate "is presumed to be the reasonable (attorneys') fee contemplated by (42 U.S.C.) Section 1988." /5/ Although the Court declined to hold that upward adjustments, or "multipliers," are never permissible (Blum, slip op. 10), the Court did severely limit the circumstances under which such adjustments may be made. Thus, the Court held that "(n)either complexity nor novelty of the issues * * * is an appropriate factor in determining whether to increase the basic fee award," because those factors should be reflected in the reasonable number of billable hours and the hourly rates (id. at 11). Similarly, the Court held that "acknowledgement of the 'results obtained' generally will be subsumed within other factors used to calculate a reasonable fee"; therefore, that factor "normally should not provide an independent basis for increasing the fee award" (id. at 13 (footnote omitted)). Finally, the Court held that the number of persons benefited is not a relevant consideration in setting the fee award, because, "(p)resumably, counsel will spend as much time and will be as diligent in litigating a case that benefits a small class of people, or, indeed, in protecting the civil rights of a single individual." Id. at 13 n.16. With respect to the "quality of representation," the Court held that superior legal work generally is reflected in the reasonable hourly rate. Blum, slip op. 11. Thus, the Court concluded that this factor may be used to justify an upward adjustment to the lodestar "only in the rare case." Ibid. Moreover, the Court limited such an adjustment to instances (id. at 11-12 (emphasis added)) where the fee applicant offers specific evidence to show that the quality of service rendered was superior to that one reasonably should expect in light of the hourly rates charged and that the success was "exceptional." The Court expressly left open the principal issue here presented: "whether the risk of not being the prevailing party, * * * and therefore not being entitled to an award of attorney's fees from one's adversary, may ever justify an upward fee adjustment." Blum, slip op. 13 n.17. /6/ The court of appeals acknowledged that this Court had never held that "contingency of success can properly justify a lodestar increase" (Pet. App. 68a), but it stated that "we have resolved the question in this court" (ibid.). The court then proceeded to uphold the use of contingency multipliers that had the effect of increasing petitioners' liability by more than $124,000. Although the courts below failed to articulate any rationale underlying the use of "contingency" multipliers in this case, courts and commentators have traditionally viewed such multipliers as compensating the attorney for bearing an appreciable risk of nonpayment as a result of lack of success on the merits. Under this view, because the lodestar is calculated on the basis of a reasonable hourly rate for noncontingent litigation (i.e., that in which the attorney is paid regardless of outcome), in some cases additional compensation is thought to be necessary in order to induce attorneys to take on a case in which there is a chance they will receive no fee. See, e.g., Lindy Bros. Builders, Inc. v. American Radiator & Standard Sanitary Corp., 487 F.2d 161, 168 (3d Cir. 1973) (Lindy I) (quoting Cherner v. Transition Electronic Corp., 221 F. Supp. 55, 61 (D. Mass. 1963)) ("'No one expects a lawyer whose compensation is contingent upon his success to charge, when successful, as little as he would charge a client who in advance had agreed to pay for his services, regardless of success.'"); Copeland v. Marshall, 641 F.2d 880, 893, (D.C. Cir. 1980) (en banc) ("It is important to recognize that the contingency adjustment is designed solely to compensate for the possibility at the outset that the litigation would be unsuccessful and that no fee would be obtained."); Jones v. Diamond, 636 F.2d 1364, 1382 (5th Cir.), cert. granted sub nom. Ledbetter v. Jones, 452 U.S. 959, dismissed by agreement, 453 U.S. 950 (1981) ("Lawyers who are to be compensated only in the event of victory expect and are entitled to be paid more when successful than those who are assured of compensation regardless of result."); Berger, Court Awarded Attorneys' Fees: What is "Reasonable"?, 126 U. Pa. L. Rev. 281, 324-326 (1977). This rationale may be appealing in the context of actions such as Lindy I, in which the fee award comes out of a "common fund" generated by the plaintiff's suit, /7/ but it is of dubious validity in the context of fee-shifting statutes that make the defendant directly liable for the plaintiff's fees. In the latter situation, three separate factors demonstrate that Congress could not have intended the use of contingency multipliers in calculating reasonable attorneys' fee awards under fee-shifting statutes such as Section 304(d) of the Clean Air Act: first, the impropriety of awarding fees that indirectly compensate nonprevailing parties for unsuccessful suits; second, the fundamental unfairness of imposing a higher fee on certain defendants merely because the strength of the case against them was initially dubious; and third, the degree to which such multipliers reflect improper double billing. A. Contingency Multipliers Improperly Compensate Nonprevailing Parties. The award of a substantial contingency multiplier in effect compensates a prevailing party for the risk that he might have lost the case and, thus, cross-subsidizes other parties (or, more aptly, their counsel) for unsuccessful litigation. For example, if the chance of success in a particular case in which a party actually prevails were adjudged (with hindsight) to have been 50% at the time the complaint was filed, a risk multiplier of two might be employed. /8/ Assuming the risk assessment were accurate, it could be expected that counsel would succeed in one out of every two similar cases filed. The effect of doubling counsel's fee recovery for prevailing in the first case is indistinguishable from paying for the time counsel devotes to its hypothetical unsuccessful twin. /9/ This compensation for unsuccessful litigation is contrary to the intent of Congress and, in the case of governmental entities, improperly expands Congress's limited waiver of sovereign immunity. See note 17, infra. Several courts have acknowledged that contingency awards compensate attorneys not only for their successful efforts, but for their unsuccessful efforts as well. As the court stated in Stanford Daily v. Zurcher, 64 F.R.D. 680, 685 (N.D. Cal. 1974), aff'd, 550 F.2d 464 (9th Cir. 1977), rev'd on other grounds, 436 U.S. 547 (1978) (emphasis added): From the attorneys' standpoint, the contingent fee insures that counsel are compensated not only for their successful efforts but also for unsuccessful litigation. Its use allows attorneys -- including attorneys who could not otherwise absorb the costs of lost cases -- to take the financial gamble of representing penurious clients, since, over the long run, substantial fees awards in successful cases will provide full and fair compensation for all legal services rendered to all clients. For precisely this reason, however, other courts have disapproved of contingency multipliers. See, e.g., McKinnon v. City of Berwyn, 750 F.2d 1383, 1392 (7th Cir. 1984) ("The fundamental problem of a risk bonus is that it compensates attorneys, indirectly but effectively, for bringing unsuccessful * * * suits, even though the attorneys' fee statute is expressly limited to cases where the party seeking the fee prevails."); Laffey v. Northwest Airlines, Inc., 746 F.2d 4, 27 (D.C. Cir. 1984), cert. denied, No. 84-1655 (June 17, 1985) ("To multiply all awards to account for the risk of losing would ultimately generate a pool of legal fees sufficient to compensate all attorneys bringing all civil rights suits; the fund generated would differ from insurance only in the manner in which it was collected and distributed."); Murray v. Weinberger, 741 F.2d 1423, 1431 (D.C. Cir. 1984) ("Awarding an upward adjustment to the lodestar for the risk of losing and the concomitant risk of not obtaining an award of attorney's fees is not unlike compensating an attorney for unsuccessful claims, it hedges the * * * requirement that only prevailing parties may recover attorney's fees."). The rationale for contingency multipliers as expressed in Stanford Daily is thus inconsistent with the Court's decision in Hensley v. Eckerhart, 461 U.S. 424, 435-436 (1983), holding that the extent of a plaintiff's success is the most critical factor in determining the proper amount of an attorney's fee award under 42 U.S.C. 1988. See Laffey, 746 F.2d at 27; Murray, 741 F.2d at 1431. If, as the Court held in Hensley, Congress did not intend to saddle a losing defendant with attorneys' fees incurred by a partially successful plaintiff in the pursuit of discrete claims on which the plaintiff did not prevail (461 U.S. at 434-436), then it makes no sense at all to require a defendant to pay (indirectly) the attorneys' fees of a totally unrelated plaintiff whom neither the defendant nor anyone else has harmed in any way and, indeed, with whom the defendant has no connection whatever. In addition to the conflict with Hensley, the rationale for contingency multipliers cannot be squared with the Court's explicit prohibition of attorneys' fee awards under the Clean Air Act to nonprevailing parties. Ruckelshaus v. Sierra Club, 463 U.S. 680, 691-692 (1983) (construing Section 307 of the Clean Air Act and noting that the same construction governs Section 304). Thus, contingency multipliers of the sort approved by the lower courts in this case violate the will of Congress by indirectly compensating attorneys for unsuccessful actions and losing legal theories. See pages 28-32, infra. B. Contingency Multipliers Are Contrary To Basic Notions Of Fairness And Lead To Anomalous Results. Contingency multipliers are governed by a profoundly perverse rule: the more reasonable a defendant's decision to defend its actions in court, the larger its exposure for fees should the plaintiff finally prevail. /10/ This irrational system unfairly penalizes with a higher payment of fees the defendant who legitimately relies on an apparently meritorious defense or who vigorously defends against a seemingly baseless claim. As one commentator has explained (Leubsdorf, The Contingency Factor in Attorney Fee Awards, 90 Yale L.J. 473, 488-489 (1981) (footnotes omitted)): (T)he defendant must pay more when the balance of precedent and evidence was relatively favorable to him. On the other hand, when the plaintiff was certain of success because the defendant's position was hopeless or frivolous, the defendant pays no contingency bonus. Furthermore, the contingency bonus is extracted from that defendant in order to reward the plaintiff's bar for bringing similar but unsuccessful suits against other defendants. If such litigation is to be subsidized, one may well ask why the subsidy should come from the defendant in another case. See also McKinnon, 750 F.2d at 1392; Laffey, 746 F.2d at 26; Note, Promoting the Vindication of Civil Rights Through the Attorney's Fees Awards Act, 80 Colum. L. Rev. 346, 375 (1980). Moreover, contingency multipliers produce an economic incentive for plaintiffs to pursue nonmeritorious cases in hopes of a fee bonus. /11/ Approval of contingency multipliers thus encourages marginal litigation that further congests the already overcrowded federal court dockets. "If the capacity of the judicial system remained constant, these marginal claims would displace civil rights cases with a greater probability of success. The net effect would be a dilution in the deterrent force of the civil rights statutes." Laffey, 746 F.2d at 27. /12/ The same observation holds true for the meritorious environmental cases that Congress intended to encourage through the enactment of Section 304(d) of the Clean Air Act. Absent strong evidence that Congress authorized contingency bonuses, these deleterious effects counsel strongly against their use. /13/ Finally, the method employed in calculating contingency multipliers is totally arbitrary and standardless. Under the guise of a pseudo-scientific approach that requires a court to estimate, after the plaintiff actually has prevailed, the likelihood of success at the outset of the litigation (Copeland v. Marshall, 641 F.2d at 893), district courts in fact wind up subjecting themselves and the parties to a process that more closely resembles the "Wheel of Fortune" or the Irish Sweepstakes than the calculation of "reasonable" attorneys' fees. Yet the outcome of this irrational and wholly subjective exercise may add hundreds of thousands of dollars to a defendant's fee liability, and the ultimate award may bear no relationship to the harm suffered by the plaintiff or the work performed by his attorney. Cf. City of Riverside v. Santos Rivera, No. A-122 (Aug. 28, 1985) (Rehnquist, Circuit Justice). As one commentator has recognized, district courts face nearly insurmountable obstacles in determining what the chance of success was at the outset of a lawsuit (Leudsdorf, supra, 90 Yale L.J. at 486-487 (footnote omitted)): When we deal with the probability of winning a single litigation * * *, we must speculate about retrials of the same case before different judges and jurors, or imagine well-informed gamblers betting on the outcome of the case. * * * Yet once the result is known, it is hard for judges and lawyers to regain a perspective of ignorance and to treat the result as only one of several that were initially possible. * * * For a court that has already upheld the plaintiff's case to assess how likely it was to do so raises problems of propriety as well as practicality. In effect, the court is being asked to decide how likely it was to reach what is now recognized as the right result. One may question whether a judge should be called upon to look for the dust swept under the same rug on which he is standing. For all these reasons, it is highly doubtful that the risk factor can contribute to determining "reasonable" attorneys' fees with any degree of accuracy, objectivity, or consistency. /14/ Certainly, the result in this case, in which the multiplier accounts for more than half of the total attorneys' fee award, demonstrates that something is seriously wrong with the current system of compensation. C. Contingency Multipliers Provide Improper Double Compensation. Use of a contingency multiplier is particularly objectionable because it substantially duplicates compensation for elements that either are already reflected in the lodestar or are not compensable at all. The contingency multiplier essentially seeks to compensate the plaintiff for assuming the risk of not prevailing -- and of thus receiving no fee award. /15/ That risk basically has two components: first, the risk of bringing an unmeritorious case (i.e., one not actionable on the legal theory presented to the court or one not supported by the facts), and second, the risk of failing to persuade the court in a meritorious case. Given that this Court has determined that Congress did not intend that attorneys' fees be awarded to parties who bring nonmeritorious claims (see pages 20-21, supar), it is difficult to perceive why the first component -- the mere risk of bringing nonmeritorious claims -- should be considered properly compensable. The latter component of the risk -- that arising from failure to persuade the court on the merits -- is the force driving the attorney's work in preparing his presentation of the case. When the underlying facts are convoluted or the legal theories novel or difficult, counsel will spend correspondingly large amounts of time gathering information, conducting discovery, reviewing documents to develop the factual record, researching precedents, investigating legislative history, drafting pleadings, and writing briefs to formulate persuasive legal and factual arguments. Thus, the same unfavorable facts or unfavorable law that undermine a plaintiff's chance of success also substantially increase the number of hours reasonably expended or lead the court to accept as reasonable a higher hourly rate. In either event, the lodestar fully and accurately compensates plaintiff's counsel for any "risk" that he may have assumed. Double billing also occurs in more subtle ways. Some element of risk allowance is built into market rates for attorneys' services, which necessarily cover hours for which the lawyer for one reason or another does not receive payment directly from the client. See Copeland v. Marshall, 641 F.2d at 917-919 (Wilkey, J., dissenting). /16/ Similarly, as the district court observed in Stanford Daily v. Zurcher, 64 F.R.D. at 682, "the contingent nature of success and the quality of the attorneys' work * * * oftentimes will be interrelated. For example, an increase in the attorneys' fees because the chances of success (and for fees) at the beginning of litigation appeared slight is implicitly if not explicitly an increase due to the high caliber of the attorneys' representation." But that quality factor should be fully and fairly reflected in the hourly rate applied to determine the compensatory fee award. Moreover, the Court has limited fee award adjustments for the quality of representation to "the rare case" in which the fee applicant offers specific evidence to demonstrate that the superior service is not reflected in the hourly rate charged and "the success (obtained) was 'exceptional.'" Blum, slip op. 11-12. Accordingly, if the Court finds that contingency multipliers are ever appropriate, they must be limited to the rarest of circumstances. Any other result would permit an attorney to secure indirectly that which was rejected in Blum -- the routine award of a bonus for the high quality of representation, needed only to overcome the apparent weaknesses in his case. D. Contingency Multipliers Are Contrary To The Intent Of Congress. As we have explained in the preceding sections of this brief, contingency multipliers impermissibly compensate nonprevailing parties, produce perverse incentives that generate needless litigation, raise profound issues of fairness, and lead to unearned windfalls for attorneys based on totally arbitrary considerations. In light of these serious problems, the Court should not interpret the phrase "reasonable attorney * * * fees" to incorporate contingency multipliers without a clean indication that Congress so intended. See Ruckelshaus v. Sierra Club, 463 U.S. at 686 (legislative history of the Clean Air Act does not suggest "that Congress meant to abandon historic fee-shifting principles and intuitive notions of fairness"). Congressional approval of contingency multipliers is, however, conspicuously absent. /17/ Fee-shifting statutes reflect Congress's express determination that, in specific limited circumstances, a departure from the traditional "American Rule" that each party bear its own attorneys' fees regardless of the outcome of the litigation is warranted to encourage citizens to act as "private attorneys general" to vindicate important public rights. See, e.g., S. Rep. 91-1196, 91st Cong., 2d Sess. 38 (1970). Although Congress sought to ensure that attorneys would be fairly compensated for certain "public interest" litigation, nowhere has Congress evinced a desire to replicate every aspect of private fee agreements. Such agreements typically provide either that the attorney is entitled to payment at an hourly rate regardless of outcome or that the attorney will receive some portion of the monetary judgment awarded to the client. Neither scenario remotely parallels the workings of the fee awards made pursuant to fee-shifting statutes, under which defendants directly pay their opponents' attorneys' fees. Nothing in the Clean Air Act or its legislative history speaks to the size of a "reasonable" attorneys' fee award. Congress has, however, briefly addressed that issue in enacting the attorneys' fee provision of the Civil Rights Attorney's Fee Awards Act of 1976, 42 U.S.C. 1988. /18/ There, Congress clearly identified the touchstone for determining the amount of a "reasonable" award of attorneys' fees: "(F)ees (should be) adequate to attract competent counsel, but (should) not produce windfalls to attorneys." S. Rep. 94-1011, 94th Cong., 2d Sess. 6 (1976). See also H.R. Rep. 94-1558, 94th Cong., 2d Sess. 9 (1976) (fee awards should be sufficient "to attract competent counsel in cases involving civil and constitutional rights, while avoiding windfalls to attorneys"). The committee reports cited Johnson v. Georgia Highway Express, Inc., 488 F.2d 714, 717-719 (5th Cir. 1974), as listing the factors to be considered in fixing a fee award. /19/ H.R. Rep. 94-1558, supra, at 8: S. Rep. 94-1011, supra, at 6. In addition, the Senate Report cited three district court decisions said to "correctly appl(y)" the Johnson standards in fixing a fee award. Ibid.; see also Hensley v. Eckerhart, 461 U.S. at 430 & n.4. A fair reading of Johnson and the three cases cited in the Senate Report does not support the conclusion that Congress authorized contingency multipliers as an element of "reasonable" attorneys' fees. Admittedly, one of the factors listed in Johnson was "(w)hether the fee is fixed or contingent" (488 F.2d at 718). But the Fifth Circuit's "original elucidation of this factor in 1974 made clear that it referred to the fee contract between the plaintiff and his lawyer." Leubsdorf, supra, 90 Yale L.J. at 479 n.38. At most, therefore, the Fifth Circuit meant only that contingency arrangements should be taken into account in determining the reasonableness of a particular fee award, not that the contingent nature of a fee warrants enhancement of a fully compensatory award by application of a multiplier. See Bonner v. Coughlin, 657 F.2d 931, 936 (7th Cir. 1981) ("The contingent nature of the fee is an appropriate factor to weigh in determining the overall reasonableness of the fee, but it alone does not justify the use of a multiplier."). An analysis of the three district court cases cited with approval in the Senate Report demonstrates that Congress merely viewed the risk of an attorney's obtaining no compensation as a factor to be considered in assessing the reasonableness of the amount of time expended and the hourly rate requested. In Davis v. County of Los Angeles, 8 Empl. Prac. Dec. (CCH) Paragraph 9444 (C.D. Cal. 1974), the district court added a "Result Charge" of $7,193.32 to the basic award of $39,868 for "Attorneys' Time" at the "normal hourly rates." The bonus awarded in Davis was in no way intended to compensate the lawyers for assuming the risk of nonpayment that might arise out of failure on the merits. Instead, as the label used suggests, the fee was enhanced because the court believed that counsel had "achieved excellent results" and that "(t)he nature of the case made it difficult to litigate" (8 Empl. Prac. Dec. Paragraph 9444, at 5048). /20/ In Swann v. Charlotte-Mecklenburg Bd. of Educ., 66 F.R.D. 483 (W.D.N.C. 1975), the district court did not award any bonus or multiplier. In fact, the court did not explicitly compute the fee award based upon hours reasonably expended. Instead, the court simply reviewed nine factors, similar to those listed in Johnson, and reduced the prevailing party's fee request by roughly 15%. The court stated that the amount requested "would be reasonable," but that it "would prefer to err on the conservative side in dealing with any fee question" rather than "contribute unnecessarily to the overpricing of litigation in this or any other court" (66 F.R.D. at 486). Finally, in Stanford Daily v. Zurcher, 64 F.R.D. at 688, the district court ruled that "the contingent nature of compensation, the quality of the attorneys' work, and the results obtained by the litigation warrant(ed) increasing the base fees figure (hours worked times average billing rate)" of $37,500 by $10,000. /21/ Thus, in approving the approach of the Stanford Daily court, Congress did not authorize the mechanical multiplication of the lodestar to compensate attorneys for the risk of receiving no payment. Rather, Congress ratified a process -- consistent with the one used in Johnson -- that evaluates whether the "hours worked" and the "billing rate" accurately reflect not only "the contingent nature of compensation," but also the "quality of the attorneys' work" and the "results obtained." In Blum, the Court recognized that the latter two factors are subsumed within a properly calculated "lodestar" fee, and nothing in Stanford Daily suggests that the "contingency" factor should be treated any differently. /22/ E. The Award Of A Contingency Multiplier In This Case Was Improper. For the reasons explained in the preceding sections of this brief, we submit that contingency multipliers should be entirely excluded from the calculation of "reasonable" attorneys' fees. If contingency multipliers are ever appropriate, however, the Court should require at least as strong a showing in favor of their use in a particular case as it requires for the use of "quality" multipliers. As we noted at the outset, Blum prohibits the award of "quality" multipliers except in "the rare case" in which the fee applicant offers "specific evidence" to demonstrate that the high caliber of the representation is not reflected in the hourly rates charged. Blum, slip op. 11-12. In light of the serious problems associated with contingency multipliers, no lesser showing should be allowed to sustain the award of a risk bonus. Free of a requirement of "specific evidence" to support a contingency multiplier, the district courts would be vested with unbridled discretion to confer a large financial "bounty" on subjectively selected plaintiffs. Here, for example, the district court made no finding that the number of hours compensated or the hourly rates awarded failed to reflect fully the "risk" allegedly assumed by Delaware Valley's counsel. Indeed, the district court's three-paragraph analysis of the multiplier issue does not even clearly identify which portions of the multipliers reflected contingency and which reflected quality of representation. For this reason alone, the district court's award of multipliers was erroneous. In any event, no contingency multiplier should have been awarded under the facts of this case, because the "risk" of nonpayment was not significant. Throughout these proceedings, Delaware Valley has sought to enforce a judicially-approved consent decree. Whatever "risk" accompanied the initial filing of this lawsuit was effectively terminated with the signing of the consent decree in 1978. Although petitioners made strenuous efforts to avoid their obligations under that decree, they never claimed that those responsibilities -- or the underlying requirements of the Clean Air Act -- had been satisfied. While there may, at times, have been doubt that the consent decree as written would be implemented, some form of relief was a virtual certainty in light of petitioners' failure to fulfill any portion of their consent decree obligations and their undisputed noncompliance with the Clean Air Act. This conclusion is all the more inescapable in light of the United States' participation as a plaintiff in the litigation; whatever "risks" were associated with the case for Delaware Valley were effectively offset by the federal government's insistence that petitioners bring themselves into compliance with federal air quality standards. /23/ Thus, the danger of failing to prevail in the post-decree proceedings simply was not significant enough to require compensation at a level higher than the lodestar fee. See Ramos v. Lamm, 713 F.2d 546, 558 (10th Cir. 1983) ("The real controversies and uncertainties lay in litigating the remedy * * *. Thus, some measure of success on the merits was fairly certain, and no bonus for contingency appears warranted."). See also Beazer v. New York City Transit Authority, 558 F.2d 97, 100-101 (2d Cir. 1977), rev'd on other grounds, 440 U.S. 568 (1979) (despite "complex factual issues," the court eliminated $50,710 risk "premium," because danger of losing case was low due to "relatively simple" legal issues). Finally, even if some contingency multiplier were appropriate, the district court's award of multipliers of two and four was grossly excessive. As noted above, Congress has cautioned the courts not to award fees that "produce windfalls to attorneys." S. Rep. 94-1011, supra, at 6; see also H.R. Rep. 94-1558, supra, at 9. Consistent with this direction, if any bonus above the basic fee was to be awarded, it should have been far more modest in size. Instead, the district court utilized extraordinarily high multipliers, see note 8, supra, that resulted in payment to Delaware Valley's counsel of more than $124,000 for over 1,700 hours that they never worked on the case. /24/ In Johnson v. Georgia Highway Express, Inc., 488 F.2d at 719, the Fifth Circuit warned against the corrosive effects of excessive attorneys' fee awards: (C)ourts must remember that they do not have a mandate under (a fee-shifting statute) to make the prevailing counsel rich. Concomitantly, the (statute) should not be implemented in a manner to make the private attorney general's position so lucrative as to ridicule the public attorney general. The statute was not passed for the benefit of attorneys * * *. To appreciate the significance of this reminder, it should be noted that, at about the time the legal services in this case were rendered, the average cost to the taxpayer of the legal services of the United States Department of Justice (including attorneys' salaries, support staff salaries, and other overhead) ranged from $27.48 per hour for a newly hired attorney to $48.28 per hour for the most experienced litigators. See Copeland v. Marshall, 641 F.2d at 912 n.10 (Wilkey, J., dissenting). The multipliers employed below -- in effect compensating Delaware Valley's counsel at rates up to $400 per hour -- unquestionably "ridicule the public attorney general" and are inconsistent with the court's obligation to limit fees to the level needed to attract competent counsel and to avoid windfalls to attorneys. In sum, an award of attorneys' fees computed by applying a reasonable hourly rate to the number of hours reasonably expended upon a successful claim should, by definition, be sufficient to attract competent counsel. Accordingly, absent truly exceptional circumstances, further enhancement of such an award -- particularly enhancement by arbitrary, large multipliers such as the district court chose in this case -- results in an improper windfall to the prevailing party's counsel. II. THE AWARD OF MULTIPLIERS BASED ON THE QUALITY OF REPRESENTATION IN THIS CASE WAS ERRONEOUS The Court held in Blum, slip op. 11-12, that enhancement of a presumptively reasonable fee to account for the high quality of representation is prohibited except in "the rare case" in which "the fee applicant offers specific evidence to show that the quality of service was superior to that one reasonably should expect in light of the hourly rate charged and that the success was 'exceptional.'" To the extent that the multipliers awarded by the district court were intended to compensate for the high quality of Delaware Valley's representation and "excellent" results (but see page 8, supra), this case falls far short of Blum's stringent requirements. We start with the "specific evidence" offered by Delaware Valley. Significantly, in its initial fee application, Delaware Valley expressly disclaimed any entitlement to a "quality" multiplier (see J.A. 80a-81a). This disclaimer clearly encompassed Phase IV of the litigation. Later, Delaware Valley submitted a supplemental fee application, in which it requested quality multipliers for Phases IV, V, and VII (J.A. 162a-168a). Although the work performed in Phase IV obviously had not changed between the time of the initial and supplemental fee applications, Delaware Valley did not even attempt to explain, in its supplemental application, why a quality multiplier for that phase had suddenly become appropriate. As for Phases V and VII, Delaware Valley merely contended that it had borne the laboring oar in the litigation and that the issues "concern(ed) the nature of our federal form of government in the context of cooperative federalism" (J.A. 166a-167a). Clearly, this submission did not satisfy Blum's requirement for "specific evidence" of the superior quality of Delaware Valley's work, nor did it support a finding of "excellent" results. The district court then took this inadequate submission and based its award of multipliers on a hodge-podge of undifferentiated, conclusory statements that included factors clearly prohibited by Blum. For example, the district court stated (Pet. App. 39a) that the case involved "new and novel issues, the resolution of which had little or no precedent." Compare Blum, slip op. 11. The court also noted that the "quality of (Delaware Valley's) work was superior" in Phase V (Pet. App. 39a), but it did not explain how it reached that conclusion, other than to note that Delaware Valley had "helped not only this court but also the court of appeals" (ibid.). This was an insufficient justification for a quality increase in a fee already calculated on the basis of "high hourly rate(s)" (id. at 25a). Finally, the district court, without any explanation whatever, increased the multiplier requested by Delaware Valley for Phase V from three to four. Clearly, neither Delaware Valley nor the district court came forward with the requisite "specific evidence" to support this award of an additional $27,372.50. The court of appeals, which had the benefit of this Court's decision in Blum, made a more extensive attempt to justify the multipliers than did the district court. Pet. App. 65a-67a. The court of appeals described at some length petitioners' repeated attempts to avoid compliance with the consent decree and explained that these unusual circumstances required Delaware Valley to "walk() the tight, and sometimes imperceptible, line that divides the competencies of two governmental sovereignties" (id. at 66a-67a). These considerations, however, do not demonstrate that the work of Delaware Valley's counsel was superior to what one would expect in light of the high hourly rates awarded or that Delaware Valley's success was "exceptional." It was the Supremacy Clause of the Constitution that accounted for the results obtained, rather than any extraordinary efforts of counsel. Cf. Washington v. Washington State Commercial Passenger Fishing Vessel Ass'n, 443 U.S. 658, 695-696 (1979). The courts below did not find that the hourly rates awarded to Delaware Valley's counsel failed adequately to reflect the quality of the services rendered or the results obtained. It is worth emphasizing in this regard that Delaware Valley's lead counsel was, by the court of appeals' own acknowledgement, an "inexperienced attorney( )" (Pet. App. 62a n.10), yet the district court chose to compensate him "at a high hourly rate * * * of $100 per hour" (id. at 25a; see also id. at 29a-30a, 33a). This determination provided ample reward for the quality of the services rendered and made the award of quality multipliers improper under Blum. III. SECTION 304(d) OF THE CLEAN AIR ACT DOES NOT AUTHORIZE COMPENSATION FOR TIME SPENT COMMENTING ON STATE AND FEDERAL REGULATORY PROPOSALS As previously noted (see pages 6-7, 8-9, supra), the courts below approved compensation for the time Delaware Valley spent commenting on Pennsylvania's proposed I/M regulations and participating in EPA administrative proceedings concerning proposed revisions to Pennsylvania's SIP. In our submission, such time is not properly compensable under Section 304(d) of the Clean Air Act. In only one instance has this Court held that a fee-shifting statute authorizes reimbursement for work performed at the administrative level. New York Gaslight Club, Inc. v. Carey, 447 U.S. 54 (1980). That case, however, arose under the attorneys' fee provision of Title VII of the Civil Rights Act of 1964, 42 U.S.C. 2000e-5(k). The result reached by the Court turned both on the express language of Section 706(k) and on the special structure of Title VII's enforcement scheme. First, Section 706(k) expressly authorizes fees for work in "any action or proceeding" under Title VII. The Court held that "Congress' use of the broadly inclusive disjunctive phrase 'action or proceeding' indicates an intent to subject the losing party to an award of attorney's fees and costs that includes expenses incurred for administrative proceedings" (447 U.S. at 61). Second, the Court interpreted Title VII as requiring resort to state and local remedies before the filing of suit in federal court. 447 U.S. at 65. More recently, in Webb v. County Bd. of Educ., No. 83-1360 (Apr. 17, 1985), the Court affirmed the denial of attorneys' fees under 42 U.S.C. 1988 for work performed before a local school board in advance of the filing of a suit under 42 U.S.C. 1983. Although Section 1988 contains the same "action or proceeding" language as Section 706(k), the Court distinguished Carey by noting that nothing in Section 1983 requires resort to administrative proceedings prior to the filing of suit (Webb, slip op. 6). The Court further explained that the time that is compensable under Section 1988 is that "'reasonably expended on the litigation.'" Webb, slip op. 8 (quoting Hensley v. Eckerhart, 461 U.S. at 433). Although the Court noted that some pre-litigation work, such as drafting the initial pleadings and developing a theory of the case, is performed "on the litigation," it stressed that there was no difficulty in the case before it in identifying the dividing line between the administrative proceeding and the judicial action. Webb, slip op. 8. For the same reason, the Court had no difficulty in determining that the work performed before the school board was not work performed "on the litigation." Id. at 8-9. In the instant case, the language of Section 304(d) and analogous fee-shifting statutes strongly suggests that time spent on administrative proceedings is noncompensable. First, Section 304(d) is limited by its terms to "action(s)" and does not include "proceeding(s)." Several lower courts have recognized that the omission of the word "proceeding" from statutes that are virtually identical to Section 304(d) is dispositive of the issue. Roosevelt Campobello International Park Comm'n v. United States Environmental Protection Agency, 711 F.2d 431, 438-439 (1st Cir. 1983) (interpreting Section 505(d) of the Clean Water Act of 1977, 33 U.S.C. 1365(d), as precluding fees for work performed in EPA administrative proceedings); Sierra Club v. Gorsuch, 672 F.2d 33, 42 (D.C. Cir. 1982), rev'd on other grounds sub nom. Ruckelshaus v. Sierra Club, 463 U.S. 680 (1983) (fee awards under Section 307(f) of the Clean Air Act, 42 U.S.C. 7607(f), limited to work performed in "judicial proceedings"); Florida Power & Light Co. v. Costle, 683 F.2d 941, 943 (5th Cir. 1982) (same). A "proceeding" is clearly something different from an "action." The "actions" contemplated by Section 304(d) are judicial actions, not administrative proceedings. In the instant case, neither of the courts below gave any consideration to the express language of limitation in Section 304(d). Second, the particular administrative proceedings in which Delaware Valley participated bore little resemblance to "judicial actions." In Webb, for example, the Court approved the denial of fees for work performed before a local school board even though "witnesses were examined and opposing arguments considered and refuted in those proceedings," leading the plaintiff to characterize the work as "analogous to discovery, investigation and research that is part of any litigated proceeding" (Webb, slip op. 7-8). Here, by contrast, Delaware Valley, both in its comments on the proposed Pennsylvania regulations and in its participation in the EPA proceedings, did not participate in anything remotely "judicial." Moreover, participation in the activities for which Delaware Valley was awarded compensation is equally available to all citizens, whether or not they have brought, or intend to bring, litigation against an agency, and does not necessarily call for legal expertise warranting the expenditure of "attorneys' fees." See, e.g., J.A. 65a (emphasis added), in which Delaware Valley characterized its comments on petitioners' proposed regulations as time spent "investigating technical aspects of Commonwealth proposals respecting I/M regulations." Third, it is undisputed that Delaware Valley was under no obligation either to comment on the proposed Pennsylvania rules or to participate in the EPA proceedings; indeed, the district court expressly recognized that participation in the EPA proceedings was not required by any of the court's orders (Pet. App. 34a). Thus, this case does not fall within the rationale of Carey. Nevertheless, the courts below were of the view that Delaware Valley's participation in the state and federal administrative proceedings was useful because either or both of those proceedings could have resulted in impairment of the rights secured by Delaware Valley under the consent decree. Pet. App. 17a, 36a, 55a, 57a & n.7. This rationale provides insufficient justification for ignoring the language of Section 304(d). While it may well have been reasonable, and indeed useful, for Delaware Valley to have participated in the administrative proceedings, its rights under the consent decree were at all times protected by the continuing jurisdiction retained by the district court. Had either or both of the administrative proceedings resulted in any significant impairment of the consent decree, Delaware Valley clearly could have sought relief from the district court. /25/ Finally, it is especially significant that when Congress has authorized funding for the mere presentation of views in agency proceedings, it has spoken clearly. For example, Congress has expressly authorized EPA to award attorneys' fees for participation in rulemakings under the Toxic Substances Control Act, 15 U.S.C. 2605(c)(4)(A). That statute plainly demonstrates that Congress knows how to draft language authorizing fee awards for participants in agency proceedings open to the public at large. Such funding is rare, however, and it has never been authorized for the proceedings in which Delaware Valley participated. Indeed, it is noteworthy that, contemporaneously with the passage of the 1977 amendments to the Clean Air Act, Congress considered and rejected proposals to fund public participation in agency proceedings through grants of attorneys' fees and other costs. See S. 270 and H.R. 3361, 95th Cong., 1st Sess. (1977). See generally Public Participation in Agency Proceedings: Hearings on H.R. 3361 and Related Bills Before the House Subcomm. on Administrative Law and Governmental Relations of the Comm. on the Judiciary, 95th Cong., 1st Sess. (1977). Thus, the court of appeals clearly erred in reading into Section 304(d) authorization to award fees for time that was not spent "on the litigation." Despite the clear differences between Section 304(d) and the statutes considered in Carey and Webb, the court of appeals concluded (Pet. App. 55a, 57a & n.7) that the award of fees for Delaware Valley's administrative work was appropriate under Webb. In our view, the court read far too much into Webb. There, in rejecting Webb's claim to fees for administrative work, the Court noted in passing that Webb had failed to demonstrate "that any discrete portion of the work product from the administrative proceedings was work that was both useful and of a type ordinarily necessary to advance the civil rights litigation to the stage it reached before settlement." Webb, slip op. 9. The court of appeals turned this statement on its head and concluded that fees may be awarded when the applicant does show that the administrative work was "'both useful and of a type ordinarily necessary to advance the . . . litigation' to the point where the party prevailed" (Pet. App. 57a n.7 (quoting Webb, slip op. 9)). Nothing in Webb suggests that this is the only relevant inquiry or even that proof of the two elements noted in Webb will suffice to support a fee award for administrative work. /26/ And even if the Court intended its statement in Webb to establish a test for determining the appropriateness of fee awards for work done in optional administrative proceedings, that test would have no applicability in the case of a statute, such as Section 304(d), that limits fee awards to the "costs of litigation" incurred in judicial "action(s)." CONCLUSION The judgment of the court of appeals should be reversed. Respectfully submitted. CHARLES FRIED Solicitor General F. HENRY HABICHT II Assistant Attorney General KENNETH S. GELLER Deputy Solicitor General KATHRYN A. OBERLY Assistant to the Solicitor General KAREN L. FLORINI Attorney /*/ DECEMBER 1985 /1/ Originally, Delaware Valley sued both petitioners and EPA, the latter for failing to enforce petitioners' obligations under the Clean Air Act. Thereafter, the United States initiated its own action against petitioners. EPA was then dismissed as a defendant in Delaware Valley's suit, and the two actions against petitioners were consolidated. Pet. App. 2a. /2/ Because the legislation authorizing the contracts system was not passed, Penn DOT later promulgated regulations to establish the alternate system. Pet. App. 3a. /3/ Delaware Valley did not seek fees against the United States for any portion of the post-decree proceedings. The United States therefore has not pecuniary interest in the outcome of this particular case. Because the United States frequently finds itself in the position of a fee-paying defendant, however, the government has a substantial programmatic interest in the legal issues presented by petitioners. /4/ Phase IV covered Delaware Valley's motion for contempt based on petitioners' failure to meet the May 1, 1981 implementation deadline and Delaware Valley's concomitant opposition to petitioners' request for a further extension. Pet. App. 20a-22a & n.6. Phase V involved Delaware Valley's motion to have petitioners held in contempt following enactment by the state legislature of H.B. 456, the statute that prohibited petitioners from spending any funds to implement the I/M program. Pet. App. 25a-27a & n.7. Phase VII involved Delaware Valley's "monitoring of the contempt sanction" and its opposition to petitioners' attempt to have certain highway projects certified as exempt from the district court's freeze on federal highway funding. Id. at 32a. We note that the district court's opinion contains a clerical error, stating (Pet. App. 38a) that Delaware Valley sought a multiplier for Phase VI (which it did not) and suggesting, by negative implication, that no multiplier was requested for Phase IV, when in fact Delaware Valley sought a multiplier of two for that phase. /5/ Although the Court in Blum was addressing an attorneys' fee award made in a civil rights action pursuant to 42 U.S.C. 1988, similar attorneys' fee provisions, including Section 304(d) of the Clean Air Act, that authorize the award of "reasonable" fees "should be interpreted pari passu." Ruckelshaus v. Sierra Club, 463 U.S. 680, 691 (1983). /6/ Such multipliers have come to be known as "contingency" multipliers. As the District of Columbia Circuit acknowledged in Copeland v. Marshall, 641 F.2d 880, 893 (1980) (en banc), however, contingency adjustments designed solely to compensate for the possibility at the outset that litigation will be unsuccessful and that no fee will be obtained "are entirely unrelated to the 'contingent fee' arrangements that are typical in plaintiffs' tort representation." /7/ The "common fund" doctrine is a judge-made exception to the American Rule against fee-shifting that authorizes a court to permit a party who "preserv(es) or recover(s) a fund for the benefit of others in addition to himself, to recover his costs, including his attorneys' fees, from the fund or property itself or directly from the other parties enjoying the benefit." Alyeska Pipeline Service Co. v. Wilderness Society, 421 U.S. 240, 257 (1975) (footnote omitted); see also Boeing Co. v. Van Gemert, 444 U.S. 472, 478 (1980); Trustees v. Greenough, 105 U.S. 527, 533 (1882). /8/ As one commentator has noted, "(i)n theory, there should be no upper limit to this process: when the chance of success was one in fifty, the fee should be multiplied by fifty, and so forth." Leubsdorf, The Contingency Factor in Attorney Fee Awards, 90 Yale L.J. 473, 481-482 (1981). The courts, however, have balked at carrying the theory to its logical extremes. Professor Leubsdorf notes that "(t)he highest any court appears to have gone in setting a contingency multiplier is four." Id. at 482 n.48. At least as to one phase of the instant litigation, the lower courts tied this record (Pet. App. 39a-40a). /9/ Laffey v. Northwest Airlines, Inc., 572 F. Supp. 354 (D.D.C. 1983), aff'd in part and remanded in part, 746 F.2d 4 (D.C. Cir. 1984), cert. denied, No. 84-1655 (June 17, 1985), offers a concrete illustration. In that case, the district court found that plaintiffs' counsel had reasonably expended in excess of 11,000 hours pursuing a sex discrimination claim. 572 F. Supp. at 371. This led to a "lodestar" attorneys' fee of approximately $1.48 million. Id. at 388. The district court then approved a "risk premium" of 100%, after determining that plaintiffs' chances of prevailing were approximately 50% when the case was filed. Id. at 378-380, 388. By doubling the attorneys' fee award to account for this contingency factor, the district court compensated plaintiffs' counsel for the equivalent of more than five years of attorneys' time spent on unsuccessful cases. The District of Columbia Circuit reversed this award of a "crude multiplier * * * as contrary to the congressional scheme" of compensating only prevailing parties. Laffey, 746 F.2d at 27-28. The court of appeals also held that plaintiffs were not entitled to a more "sophisticated" contingency "enhancement" to reflect the uncertainty of prevailing (and hence receiving payment). Noting that Blum cast some doubt on the propriety of any contingency adjustments, the court of appeals concluded that, in any event, the case before it was not so exceptional as to warrant such an adjustment to the lodestar. Laffey, 746 F.2d at 28-29. /10/ For example, in Laffey v. Northwest Airlines, supra, the district court's estimate that the plaintiffs' chances of prevailing were 50% could be viewed as a determination that, at the outset, the defendant's case appeared to be just as strong as the plaintiff's case. Yet, by doubling the fee award, the district court effectively penalized the defendant $1.48 million for failing to concede the case. This anomaly did not escape the attention of the court of appeals. See Laffey, 746 F.2d at 26. /11/ Not surprisingly, there is evidence to suggest that Congress never intended fee-shifting statutes to produce such counter-intuitive incentives. During the House hearings on the Civil Rights Attorney's Fee Awards Act of 1976, 42 U.S.C. 1988, Representative Seiberling remarked: Mr. Chairman, since these hearings do concern circumstances under which it would be appropriate to make one side in a lawsuit pay for the legal expenses of the other side, some people may conclude that the hearings are intended to promote the special interests of lawyers. Perhaps that is so to the extent it promotes the interests of successful lawyers who make the right judgments or who handle a case properly so that they win. But it certainly is not calculated to promote the interests of lawyers who make the wrong judgment * * *. Awarding of Attorneys' Fees: Hearings Before the Subcomm. on Courts, Civil Liberties, and the Administration of Justice of the House Comm. on the Judiciary, 94th Cong., 1st Sess. 8 (1975) (emphasis added). See also Ruckelshaus v. Sierra Club, 463 U.S. at 692-693 & n.13 (citation omitted) ("central purpose" of Section 304(d) was to "check the 'multiplicity of (potentially meritless) suits'"). /12/ In this regard, the total irrationality of the contingency multiplier was graphically explained by the Seventh Circuit in McKinnon, 750 F.2d at 1392: (I)f the logic of the risk multiplier were applied consistently, the attorney's fee would be larger the riskier the case, even though this would mean rewarding lawyers for flooding the courts with unmeritorious litigation, something we very much do not need. Imagine a class of cases where only one in 50 plaintiffs prevails. Then the risk multiplier would be 50, and a lawyer who brought all 50 cases and lost 49 would receive the same compensation that he would have received had he been certain to win all 50 cases (in which event there would have been no multiplier), rather than (virtually) certain to lose 98 percent of them. /13/ In fact, Congress was quite concerned that Section 304(d) of the Clean Air Act not generate needless litigation (S. Rep. 91-1196, 91st Cong., 2d Sess. 38 (1970) (emphasis added)): Concern was expressed that some lawyers would use section 304 to bring frivolous and harassing actions. The Committee has added a key element in providing that the courts may award costs of litigation, including reasonable attorney and expert witness fees, whenever the court determines that such action is in the public interest. The court could thus award costs of litigation to defendants where the litigation was obviously frivolous or harassing. This should have the effect of discouraging abuse of this provision, while at the time time encouraging the quality of the actions that will be brought. /14/ Beazer v. New York City Transit Authority, 558 F.2d 97 (2d Cir. 1977), rev'd, 440 U.S. 568 (1979), nicely illustrates the difficulties district courts face in assessing the plaintiff's initial chances of success. A unanimous panel of the Second Circuit was so confident that the legal issues presented in Beazer were "relatively simple" that it held the district court's award of a risk bonus to be "an abuse of discretion" (558 F.2d at 100). This Court, however, found the issues somewhat more complex: In three separate opinions covering over 50 pages of the official reporter, a divided Court reversed on the merits. /15/ It goes without saying, of course, that a "contingency" multiplier certainly is inappropriate when the fee agreement between the attorney and client in fact guarantees payment. See Ramos v. Lamm, 713 F.2d 546, 558 (10th Cir. 1983) (footnote omitted) ("Lawyers who are to be paid whether they win or lose have a weak claim to a multiplier based on the risk of loss."). /16/ As Judge Wilkey observed in Copeland, 641 F.2d at 918-919 (emphasis in original): What our colleagues (in the majority) fail to realize is that the "market value" fee they have taken as a "lodestar," the starting point to be adjusted for contingencies, already has a substantial contingency factor built into the fee. * * * * * Since the "market value" fee in commercial private practice already includes the contingency of failure and receipt of a lower fee than otherwise obtainable, our colleagues (in the majority) create a danger of duplication when they add this contingency factor to the already generous market value regular hourly rate. Judge Wilkey further pointed out (641 F.wd at 919) that the majority in Copeland recognized the same point (641 F.2d at 893): "To the extent, of course, that an hourly rate underlying the 'lodestar' fee itself comprehends an allowance for the contingent nature of the availability of fees * * *, no further adjustment duplicating that allowance will be made." Thus, Judge Wilkey parted company with the majority only to the extent that he was of the view that market billing practices almost always include an allowance for "contingency," whereas the majority proposed to leave that determination to case-by-case evaluation. Compare 641 F.2d at 893 with 641 F.2d at 919 (Wilkey, J., dissenting). There was no dispute, however, over the proposition that the contingency factor already built into market rates may not be further enhanced by a separate "contingency multiplier." /17/ The absence of express congressional authorization for contingency multipliers is particularly important where, as here, the fee-paying defendant is a governmental entity. As this Court recently reaffirmed in the case of the federal government, fee-shifting statutes represent a limited waiver of sovereign immunity. Ruckelshaus v. Sierra Club, 463 U.S. at 685-686. Waivers of immunity must be strictly construed and may not be "enlarge(d) * * * beyond what the language requires." Eastern Transportation Co. v. United States, 272 U.S. 675, 686 (1927). Accordingly, in determining whether a "reasonable" attorneys' fee award may properly include a contingency multiplier, the Court should be guided by the language of the statute and by what Congress "clearly and unequivocally" intended. See Lehman v. Nakshian, 453 U.S. 156, 162 (1981). By virtue of the Eleventh Amendment, the same rule of strict construction holds true in cases such as this one, where a state is the fee-paying defendant. Cf. Atascadero State Hospital v. Scanlon, No. 84-351 (June 28, 1985), slip op. 7-11. /18/ We agree with the court of appeals' statement that "the jurisprudence regarding the calculation of reasonable attorneys' fees developed in connection with other attorneys' fee statutes -- particularly Section 1988 -- is applicable to cases brought pursuant to Section 304(d)" of the Clean Air Act. Pet. App. 52a. Because Section 1988 and Section 304(d) of the Clean Air Act share a common purpose -- to enable private citizens to aid in the enforcement of important public rights -- we also believe that Section 1988's legislative history is relevant to a proper determination of what constitutes "reasonable" attorneys' fees under Section 304(d). It should be noted, however, that neither the case law interpreting Section 1988 nor its legislative history is tranferable to the Equal Access to Justice Act, 28 U.S.C. 2412(d). Among other reasons, EAJA, unlike Section 1988, contains no express fee cap of $75 per hour (28 U.S.C. 2412(d)(2)(A)(ii)), thus making the use of multipliers -- whether to reflect contingency or any other factor -- incompatible with Congress's limited waiver of sovereign immunity. See Underwood v. Pierce, 761 F.2d 1342, 1348 (9th Cir. 1985), petition for reh'g pending. /19/ The factors discussed in Johnson are (488 F.2d at 717-719): (1) The time and labor required; (2) The novelty and difficulty of the questions; (3) The skill requisite to perform the legal service properly; (4) The preclusion of other employment by the attorney due to acceptance of the case; (5) The customary fee; (6) Whether the fee is fixed or contingent; (7) Time limitations imposed by the client or the circumstances; (8) The amount involved and the results obtained; (9) The experience, reputation, and ability of the attorneys; (10) The "undesirability" of the case; (11) The nature and length of the professional relationship with the client; and (12) Awards in similar cases. /20/ Under Blum, of course, neither of these factors is sufficient to justify a multiplier; rather, both are subsumed within the "lodestar" calculation. Blum, slip op. 11, 13. /21/ Although the district court did not explain how it had translated these factors into the stated dollar amount, it emphasized that the various Johnson factors overlap and that "recognition of their overlap (is needed to avoid) unnecessary inflation of the attorneys' fees award" (64 F.R.D. at 682). /22/ It is, of course, perilous to attribute too much weight to the particular circumstances of district court cases mentioned in passing in a legislative report. But it is significant to note that, in the three cases cited by the Senate Report, one lowered the fee award requested by counsel and the other two added bonuses of only 20% or 25%. Accordingly, if the Senate Report is read as approval for the use of contingency multipliers, it clearly conveys Congress's desire that the use of such multipliers not be automatic and that any multiplier component be modest in amount so as to prevent "windfalls to attorneys" (S. Rep. 94-1011, supra, at 6). /23/ Delaware Valley contended in its fee application (J.A. 160a) that the United States "was not always a staunch defender of the Decree, nor of the contempt sanction imposed by the (district court) to coerce the (petitioners) into compliance," thereby suggesting that Delaware Valley along bore the risks of the litigation. While it is true that the United States at times took a more moderate position than Delaware Valley, it is simply not credible to assert that Delaware Valley did not benefit substantially from the presence of the United States as a plaintiff in this case. It also bears noting that the United States significantly aided Delaware Valley's cause in other ways. Petitioners submitted the consent decree, the I/M regulations spawned by the decree, and other control measures to EPA for approval as revisions to Pennsylvania's state implementation plan (SIP) for attainment and maintenance of the federal standards for ozone and carbon monoxide. Petitioners also requested an extension of the statutory deadline for attainment of those standards to December 31, 1987. In May 1980, EPA approved the measures and the requested extension. 45 Fed. Reg. 33607. This approval automatically triggered the requirement added to the Clean Air Act in 1977 that SIPs for extension areas contain an I/M program (42 U.S.C. 7502(b)(11)(B)). Thus, petitioners are subject to I/M requirements by statute, wholly apart from the terms of the consent decree. In addition, the Clean Air Act Amendments of 1977, 42 U.S.C. 7401 et seq., added stringent new provisions to deal with the failure of many areas, such as Pittsburgh and Philadelphia, to attain national ambient air quality standards on time. See 42 U.S.C. 7407(d), 7410(a)(2)(I), 7501-7506. Pursuant to these provisions, EPA designated the Philadelphia and Pittsburgh areas as "nonattainment areas" for carbon monoxide and ozone. 42 Fed. Reg. 8962 (1978). That designation remains in effect. 40 C.F.R. 81.339. /24/ In our submission, the award of any multipliers in this case was erroneous. The use of the four-fold multiplier in particular demonstrates that something is seriously amiss. As one commentator has remarked (Leubsdorf, supra, 90 Yale L.J. at 487): In such a case (of multipliers greater than two), the obvious question is why the plaintiff won at all -- and the most obvious answer is that he won because the case was wrongly decided. No one, presumably, would argue that the court should increase the lawyer's fee because he succeeded in winning a case that he should have lost. /25/ We do not mean to suggest that Delaware Valley should have forgone participation in the administrative proceedings simply because fees would not be available for that work. As the Court noted in Webb, slip op. 7 n.15: Of course, competent counsel will be motivated by the interests of the client to pursue state administrative remedies when they are available and counsel believes that they may prove successful. We cannot assume that an attorney would advise the client to forego an available avenue of relief solely because Section 1988 does not provide for attorney's fees for work performed in the state administrative forum. /26/ This Court did not clearly define work "of a type ordinarily necessary" to advance the litigation (Webb, slip op. 9). The immediately preceding passage in Webb suggests, however, that the Court had in mind no more than the routine services performed in connection with virtually all litigation, such as research, drafting the initial pleadings, and developing legal theories. See Webb, slip op. 8. If this is what the Court intended, then it is clear that participation in optional administrative proceedings is not work "of a type ordinarily necessary" to advance the litigation. /*/ We acknowledge the assistance in the preparation of this brief of Raymond B. Ludwiszewski, a Department of Justice employee who has passed the Massachusetts bar examination but has not yet been admitted to practice.