We are faced once again with the elusive task of deciding whether and under what circumstances attorneys for a successful plaintiff entitled to a statutory fee award may augment the lodestar with a multiplier for risk and/or delay. To date, the Supreme Court has been unable to produce a majority opinion on this issue. See Pennsylvania v. Delaware Valley Citizens' Council for Clean Air, 483 U.S. 711, 107 S. Ct. 3078, 97 L. Ed. 2d 585 (1987) (Delaware Valley II ). This precise case was previously before us, and we remanded the portion of the district court's order dealing with the multiplier with directions to reconsider in light of the Supreme Court's opinion in Delaware Valley II. See Blum v. Witco Chemical Corp., 829 F.2d 367 (3d Cir.1987) (Blum I ). Instead, the district court, without concealing its disapproval of both the Supreme Court's decision and ours, proceeded in accordance with its own views. See Blum v. Witco Chemical Corp., 702 F. Supp. 493, 494 (D.N.J.1988) ("The Supreme Court has sent a Christmas gift to this court delivered via the Third Circuit Court of Appeals. It is called 'How to Make an Attorney Fee Multiplier.' However, the instructions are so confusing and inconsistent that this court has been unable to put the 'gift' together."). On this appeal, we must thus decide both whether the district court's order is consistent with the law in its present state and whether it is faithful to the mandate.

The original plaintiffs, three research chemists discharged by Witco Chemical Corporation during a reduction in force, sued under the Age Discrimination in Employment Act and, after a jury trial, received a judgment for front pay (in the form of lost pension benefits), damages for pain and mental suffering (pursuant to a pendent state law claim), and attorneys' fees. On appeal, this court affirmed the front pay award of $58,217.05, but set aside the award on the pendent state claim tort of wrongful discharge for age discrimination. 829 F.2d at 373-77. We also affirmed the lodestar amount of $113,314.50 for attorneys' fees1 but remanded that portion of the order on attorneys' fees granting a 20 percent multiplier for delay and the contingent nature of the lawsuit for reconsideration in light of Delaware Valley II, which was decided after the district court's orders. In the course of our opinion, we stated that "plaintiffs must develop an evidentiary presentation" to justify a contingency multiplier and that the presentation "will most certainly require expert testimony from someone familiar with the economics of the legal profession." 829 F.2d at 380.

On remand, the district court referred the matter to a magistrate. Plaintiffs moved in April 1988 for restoration of the 20 percent multiplier for pre-judgment delay and risk (seeking $22,662.80 computed as $113,314 X 20%), for $10,872.45 for post-judgment interest, and for an additional counsel fee of $3,575.00 for work in post-judgment matters. In support, plaintiffs' counsel, John M. Esposito, filed an affidavit stating that the firm would not have undertaken the matter on the basis of normal hourly rates to be paid more than two years hence and, "[u]pon information and belief, no other firm using rational management guidelines could have justified accepting the matter on that basis." App. at 12-13. Esposito attached as an exhibit an affidavit of Judith L. Vladeck, a New York attorney, stating that "the contingency multiplier is appropriate and necessary to attract competent counsel to represent plaintiffs in civil rights cases.... Few seasoned and experienced lawyers would be willing, or able, to dedicate themselves to such cases if all that could ultimately be recovered is a base hourly rate." App. at 31-32.

Witco's counsel, Samuel D. Bornstein, filed an affidavit in opposition to a contingency multiplier in which he stated that plaintiffs' attorneys filed an action in a New Jersey state court for four other Witco employees alleging termination as a result of age discrimination and agreed to be paid on a contingency basis notwithstanding that the New Jersey statute under which suit was brought does not authorize contingency multipliers. App. at 41-42. Witco also filed three affidavits from New Jersey attorneys who represent plaintiffs on a contingency fee basis. Each of them, Henry Gurshman, Ronald M. Salzer, and Arnold Shep Cohen, stated that his firm did not refuse meritorious claims because neither a fee-sharing statute nor a multiplier was involved. App. at 36-37, 104-05, 112-13. Esposito filed a reply affidavit which, inter alia, challenged Bornstein's suggestion that the New Jersey Civil Rights Act, which is silent on the issue, does not authorize contingency multipliers. App. at 107.

While the matter was pending before the magistrate, plaintiffs' attorneys filed a supplemental affidavit on June 17, 1988 requesting a fee enhancement of 100%, App. at 114-17, referring to a similar award by an Alabama district court, App. at 154, and also requesting compensation for delay of 16.77%, based on the percentage increase in the cost of living. App. at 116-17. In support, plaintiffs filed seven additional affidavits of New Jersey attorneys. Nancy Erica Smith stated that "[i]f fees are subject to the contingency of prevailing in the proceeding, my firm will normally enter an arrangement where the contingent payment will be substantially greater than what our estimated compensation would be at our normal hourly rates." App. at 120. She further stated that in cases where plaintiffs have meritorious claims but limited damages, the only viable method of compensating counsel for the hours, the risk, and the delay would be an award that "would have to be two or three times the normal hourly rate." App. at 121. The only other affidavit to refer to a specific multiplier was that of Peter Vanschaick who attached a New Jersey state court opinion which adjusted the lodestar by 140 percent. App. at 137, 140-42.

Paul Schachter stated that his regular retainer agreement for employment discrimination cases provides "an enhancement of [his] normal rates from the recovery in cases successfully concluded." App. at 123. He stated that because his office does not have the ability to take on every new case which it is requested to take on, "were we to take on cases solely on an unenhanced fee basis, we would be substantially lowering the overall amount of fees we earn on employment cases" and that "[w]ithout the prospect of being awarded an additional enhancement, we would refuse low recovery cases." App. at 122, 125-26. Steven Edelstein and Stephen Blader both stated that without the potential for fees with a multiplier effect, there is less incentive for competent counsel to undertake employment cases dealing with discrimination. App. at 127-28, 129-30. The affidavit of Charles R. Church was similar. App. at 133-35. Ronald S. Levitt, a personal injury lawyer, stated that he was not interested in taking wrongful discharge cases because of the risk involved. App. at 131-32.

The magistrate, at the hearing on June 27, 1988, stated that the affidavits produced by plaintiffs fell "far short" of the expert evidence on the economic analysis of the legal profession that this court required. Report of Magistrate Hedges at 17 (filed Mar. 23, 1989). On July 5, 1988, on the basis of the affidavit evidence and the hearing, the magistrate recommended that "[p]laintiffs' motion for an upward adjustment of the lodestar be denied" but that they be awarded post-judgment interest from April 30, 1986 to November 6, 1987 at the interest rate of 6.31 percent, for a total of $10,872.45, and that they be awarded $1,887.50 for supplemental attorneys' fees for the proceedings on remand. App. at 171. By order of July 28, 1988, the district court, apparently overlooking the extension it had granted to plaintiffs for filing exceptions, adopted and affirmed the magistrate's report. See App. at 174-75.

On July 28, 1988, within the extended period, plaintiffs filed objections to the proposed findings and recommendations of the magistrate. Plaintiffs argued that the magistrate, having rejected their proffer of a delay multiplier based on the cost-of-living multiplier, should have permitted them to produce information based on market rates. They included a chart demonstrating the cost of delay using market interest rates which calculated delay interest as $14,818.63. App. at 181. In addition, plaintiffs now claimed that they were entitled to a 200 percent contingency multiplier because this was the figure awarded in Black Grievance Committee v. Philadelphia Electric Co., 690 F. Supp. 1393 (E.D.Pa.1988), which had been filed on June 20, 1988. In their objections to the magistrate's proposed findings, plaintiffs argued, "[t]his figure is plainly established as a sufficient enhancement which will attract competent counsel to the representation of plaintiffs in these matters." App. at 180.

With respect to the contingency multiplier, the district court, expressing its view "that enhancers should be the rule and not the exception," was concerned that the proof now required to sustain multipliers or enhancers would discourage counsel who otherwise would undertake such matters. 702 F. Supp. at 496. The court noted that plaintiff had consulted an expert in financial modeling, econometrics and computers who had estimated that a model to determine the relationship between hourly rates and contingency compensation rates would cost $17,600. Because the model would not explain to what extent the multiplier would be necessary to attract competent counsel, which is concededly the single most important consideration, the court concluded that an econometric study would not be practicable in this case. Id. at 498.

The district court next applied the two-step inquiry it understood was required by Delaware Valley II. First, looking to the relevant market of "all civil cases", the court concluded that plaintiffs' affidavits demonstrated "that lawyers generally expect and receive a premium for contingency cases." Id. Second, the court concluded from the affidavits submitted on behalf of plaintiffs "that contingency multipliers are necessary to ensure sufficient legal representation in [civil rights and employment litigation]." Id. at 499. The court stated, however, that the affidavits were not specific enough to "provide the court with a basis to make a market-based, quantitative finding. Unlike the record [in the Black Grievance Committee case], the record herein simply does not include any substantiated amount by which fees need be enhanced. Thus, the court has insufficient 'market' evidence to quantify the amount of enhancement necessary to ensure an adequate supply of competent counsel in employment discrimination cases." Id. at 500.

The court then looked to the "facts of the particular case," to determine the amount for enhancement, stated that this case was "difficult, lengthy, and risky" and concluded that therefore "the lodestar fee does not adequately compensate plaintiffs' counsel for the contingency inherent in this case." Id.

The court stated that based on its review of the affidavits "as to the expectations of counsel in undertaking these matters generally, and [its reconsideration of] the specific risks confronted by counsel in this particular matter," its original multiplier of 20 percent was inadequate and that a 50 percent enhancement was necessary "to attract competent counsel and adequately compensate counsel here." Id. The court offered no explanation why the specific figure of 50 percent was chosen. In this connection, we note that in its original order entered in 1986, awarding a 20 percent increase in the fee award "to reflect the contingent nature of the litigation and the delay in payment," App. at 30, the district court rejected plaintiffs' request for a 50 percent increase due to the contingent nature of the litigation and the delay in payment of fees, stating that plaintiffs' counsel failed to provide documentation of the cost to them caused by the delay in payment and that the court "hesitat[ed] to allow an award that is tremendously disproportionate to the amount of the monetary recovery." App. at 29.

Finally, the court reviewed the material submitted by plaintiffs with respect to market interest rates during the period of the litigation and the certification of a member of the law firm that represented plaintiffs that the delay of payment of the fee was significant and that the firm borrowed money to cover the shortfall, and concluded that plaintiffs' attorney is entitled to the $14,818.63 requested as compensation for delay in payment. 702 F. Supp. at 501.

The district court vacated its earlier order adopting the magistrate's report and thereafter adopted and affirmed the magistrate's report with regard to post-judgment interest and supplemented attorneys' fees. On December 27, 1988, the district court filed its opinion on the issue of contingency and delay enhancements.

The court thereafter entered the order of February 6, 1989 which granted the plaintiffs: (1) a 50 percent contingency multiplier; (2) interest on the contingency multiplier from April 30, 1986; (3) $14,818.63 as an enhancement for delay; and (4) interest on the delay enhancement from April 30, 1986. It is this order that represents the subject of this appeal. We will consider first the contingency multiplier and then the enhancement for delay.

When this court in Lindy Brothers Builders, Inc. v. American Radiator & Standard Sanitary Corp., 487 F.2d 161 (3d Cir.1973), initiated the material change in the manner in which court-awarded attorneys fees were calculated, it envisioned the possibility of increasing, or decreasing, the lodestar based on the contingent nature or risk involved and the quality of the attorney's work. The Lindy lodestar approach thereafter received the approval of the Supreme Court. See Hensley v. Eckerhart, 461 U.S. 424, 433, 103 S. Ct. 1933, 1939, 76 L. Ed. 2d 40 (1983); Blum v. Stensen, 465 U.S. 886, 888, 104 S. Ct. 1541, 1543, 79 L. Ed. 2d 891 (1984). Blum v. Stenson held that although the lodestar product is "presumed to be the reasonable fee," 465 U.S. at 897, 104 S.Ct. at 1548, there may be circumstances in which it must be adjusted. However, the Court rejected an upward adjustment based on the novelty and complexity of the issues because those factors are fully reflected in the number of billable hours and the reasonableness of the hourly rates. Id. at 898, 104 S.Ct. at 1548. The Court left open the possibility of an upward adjustment in the "rare case" where there is specific evidence of "exceptional" success and of the superior quality of service rendered but not reflected in the hourly rate. Id. at 899, 104 S.Ct. at 1549.

Thereafter, in Pennsylvania v. Delaware Valley Citizens' Council for Clean Air, 478 U.S. 546, 106 S. Ct. 3088, 92 L. Ed. 2d 439 (1986) (Delaware Valley I ), the Court pulled back even from this formulation for a quality multiplier and instead held that "the overall quality of performance ordinarily should not be used to adjust the lodestar, thus removing any danger of 'double counting.' " Id. at 566, 106 S.Ct. at 3099. The Court stated that in the absence of "detailed findings as to why the lodestar amount was unreasonable, and in particular, as to why the quality of representation was not reflected in the product of the reasonable number of hours times the reasonable hourly rate," there was no reason to increase the fee award. Id. at 568, 106 S.Ct. at 3099. It left open for subsequent argument the issue of enhancement of the fee award for contingency of success.

In Delaware Valley II, the Supreme Court announced three opinions. Four justices, in an opinion written by Justice White, would have held that "multipliers or other enhancement of a reasonable lodestar fee to compensate for assuming the risk of loss is impermissible under the usual fee-shifting statutes." 483 U.S. at 727, 107 S.Ct. at 3087. Four justices, in an opinion authored by Justice Blackmun, believed that "[a]n adjustment for contingency is necessary if statutory fees are to be competitive with the private market and if competent lawyers are to be attracted in their private practice to prosecute statutory violations." Id. at 740, 107 S.Ct. at 3094.

Justice O'Connor, writing separately, formed the majority. She agreed with the plurality of four reflected in the opinion of Justice White that no contingency was appropriate in that particular case, but agreed with Justice Blackmun that "Congress did not intend to foreclose consideration of contingency in setting a reasonable fee under [statutory] fee-shifting provisions." Id. at 731, 107 S.Ct. at 3089. Although there is some awkwardness in attributing precedential value to an opinion of one Supreme Court justice to which no other justice adhered, it is the usual practice when that is the determinative opinion and, in any event, is the way we analyzed the legal principles emerging from Delaware Valley II in Blum I. See 829 F.2d at 379.

Thus, from Justice O'Connor's opinion can be gleaned several factors that must be met before a contingency multiplier can be awarded. First, "compensation for contingency must be based on the difference in market treatment of contingent fee cases as a class, rather than on an assessment of the 'riskiness' of any particular case." 483 U.S. at 731, 107 S. Ct. at 3089 (emphasis in original). Second, the risk enhancement must be necessary to attract competent counsel. Id. at 733, 107 S.Ct. at 3090. Justice O'Connor also noted that "the fee applicant bears the burden of proving the degree to which the relevant market compensates for contingency." Id. Finally, courts "should treat a determination of how a particular market compensates for contingency as controlling future cases involving the same market." Id.

In Blum I, this court attempted to give some "guidance" to the district courts in the wake of Delaware Valley II for the "daunting task" of calculating any contingency multiplier. 829 F.2d at 380. We stated, inter alia: first, "the plaintiff has a significant burden to carry to justify a contingency multiplier," id., and second, the evidentiary presentation required of the plaintiff "will most certainly require expert testimony from someone familiar with the economics of the legal profession." Id. We added, "[i]t may also be that an expert economist will be required, even one able to develop some kind of econometric model." Id. However, Judge Becker, writing for the court, recognized that a study into the mathematical relationship between hourly rate and contingency would be extremely expensive and would not address the question of whether a multiplier is necessary to attract competent counsel. Id. at 381. Accordingly, our opinion stated that "[t]he district court may wish to consider whether such studies are feasible and whether there are alternatives that will meet the [Delaware Valley II ] test." Id.

We also drew from Justice O'Connor's opinion the conclusions that the class of cases studied should be nothing less than "all contingency cases in a given geographic market," "that the district court's determination of how the market compensates for contingency in this case will control future cases involving the same market," and that there must be a "focus on the class of cases, as opposed to the risks of the individual case at issue." Id.

We must review the district court's order granting a 50 percent contingency multiplier and the rationale given to support that order in light of the applicable precedent. We conclude that in at least four respects essential to its decision, the district court applied the incorrect legal standard.

First, although the district court recognized that "the Supreme Court has held that determination of [the necessity and quantification of a contingency enhancement] requires a marketwide analysis of the legal community," 702 F. Supp. at 495, and this court reinforced that interpretation when we stated in Blum I that "[i]t does not appear that Justice O'Connor ... contemplated that the class of cases to be studied be anything less than all contingency cases in a given geographic market, including personal injury cases, which constitute the biggest group," 829 F.2d at 381, the district court believed that "evidence of the practices and expectations in non-statutory fee cases is not relevant." 702 F. Supp. at 495. It appears that the court proceeded to follow its own view of the relevant market in ascertaining the availability of adequate legal representation, because it stated that, "[t]he relevant market cannot be deemed to include all civil practitioners, for a healthy abundance of personal injury lawyers willing to take cases on a contingency is not meaningful to a person who has been wrongfully discharged from employment." Id. at 498-99.

Second, the district court did not consider the risk associated with all contingency cases stating, "the amount for enhancement must be based on the facts of the particular case, although the need arises from a market-wide analysis." Id. at 500. In making its determination on the risk associated with this individual case, the court failed to follow the clear direction of this court quoting from Delaware Valley II that " 'compensation for contingency must be based on the difference in market treatment of contingent fee cases as a class, rather than on an assessment of the "riskiness" of any particular case.' " 829 F.2d at 379 (quoting 483 U.S. at 731, 107 S. Ct. at 3089) (O'Connor, J., concurring). The district court made no secret of its disagreement with the instruction it received on this issue, stating, "[t]he risk for which counsel should be compensated is the risk of no recovery in the particular case for which she seeks fees. Even if lawyers do not need, seek or expect such enhancement in most other cases, this conclusion would be irrelevant if no one can be found to undertake the particular matter in issue because of the substantial risk of no recovery." 702 F. Supp. at 495. Accordingly, in making its determination as to a multiplier, the district court used its view of the risk associated with this particular case rather than the risk involved in contingency cases as a whole. Id. at 500. Such an approach is also at variance with our decision in Student Public Interest Research Group v. AT & T Bell Laboratories, 842 F.2d 1436 (3d Cir.1988) (SPIRG ), where we stated, "we believe that because Justice O'Connor opposed delving into the individualized risk of the case, she would agree with Justice Blackmun that 'it is the fact of contingency, not the likelihood of success in any particular case, that mandates an increase in an attorney's fee.' " 842 F.2d at 1451.

Third, in another departure from the task set for it, the district court established a contingency multiplier for this individual case rather than setting a standard which would be applicable to future litigation within the same market. In Blum I, we stated that "Justice O'Connor's opinion makes clear that the district court's determination of how the market compensates for contingency in this case will control future cases involving the same market." 829 F.2d at 381 (citing 483 U.S. at 373, 107 S.Ct. at 3090). Also, in SPIRG we stated, "Justice O'Connor meant that courts should calculate how the relevant community compensates for contingency and then apply that multiplier in every case". 842 F.2d at 1452. Nothing in the district court's opinion suggests that it made any effort to do more than establish a contingency multiplier applicable to this single case. It made that explicit when it stated, "[t]he court has ... reconsidered the specific risks confronted by counsel in this particular matter. Based upon such consideration, the court is satisfied that its original award was inadequate.... Although the court is unable to establish a general standard, it is satisfied that greater compensation is warranted in this matter than originally determined.... Accordingly, a 50 percent multiplier will be allowed." 702 F. Supp. at 500 (emphasis added).

Finally, and perhaps most importantly, although the district court concluded that the plaintiffs had failed to meet their burden of proof by not quantifying the contingency premium, the court nonetheless relieved the plaintiffs of their burden of proof. The court noted that the magistrate found that the affidavits offered by various lawyers were conclusory and failed to document the relevant economic factors and that the magistrate recommended that "no multiplier be allowed on the issue of risk because plaintiffs had failed to meet the heavy burden of proof required by the Third Circuit in Blum." See 702 F. Supp. at 498. The court never stated the magistrate erred in characterizing plaintiffs' affidavits as "conclusory." Indeed, the court conceded that only one of affiants, other than plaintiff's counsel himself, attempted to quantify the necessary enhancement.

The district court recognized that this evidence was insufficient to meet the plaintiffs' burden with regard to the specific figure required to be employed as a multiplier in the applicable market:

The affidavits submitted on plaintiff's behalf do not provide the court with a basis to make a market-based, quantitative finding. Unlike the record [in Black Grievance Committee ], the record herein simply does not include any substantiated amount by which fees need be enhanced. Thus, the court has insufficient 'market' evidence to quantify the amount of enhancement necessary to ensure an adequate supply of competent counsel in employment discrimination cases.

Id. at 500. Nonetheless, the court allowed a 50 percent multiplier on the ground that "the granting of any multiplier less than 50 per cent might serve to discourage counsel from undertaking representation in other similar matters." Id.

We are not unsympathetic to the syllogism which underlay the district court's opinion, i.e., "congressionally authorized fee awards ... provide[ ] the means to the underprivileged and encourage[ ] the enforcement of certain rights in instances in which it would not otherwise be feasible or affordable to do so," id. at 494; "without an expectation of enhancement for risk, persons who have suffered discrimination in the workplace will face substantial difficulties in finding competent counsel," id. at 500; therefore, because "[t]he rights being vindicated are far more important than concern that from time to time a lawyer may be overcompensated," id. at 496, "enhancers should be the rule and not the exception." Id. Even if we agreed with the court's conclusion, we would not be free to uphold its judgment in this case.

The deficiency is not that there was no econometric study. Our opinion in Blum I did not mandate such a study in every case, and the affidavit of plaintiffs' attorney that a consultant reported it would cost $17,600 for a mathematical model of the relationship between hourly rate and contingency compensation rates supports the district court's conclusion that "an econometric study would not be practicable in this case."2 Instead, the error with the district court's judgment was that the 50 percent multiplier it arrived at was supported only by the court's own intuition. This is precisely what the Supreme Court and this court held is impermissible. Neither the district court nor this court is free to superimpose its own view of what the law should be in the face of the Supreme Court's contrary precedent. Unless and until that Court revises its view or promulgates an opinion of the majority that clarifies the determination that must be made to support a contingency multiplier, the district court and we are bound to the exposition of the prevailing law set out in Blum I. In light of the absence of plaintiffs' submission of any evidence that the district court could use to make a reasonable "market-based quantitative finding," we cannot sustain the district court's award of a 50 percent multiplier.

In addition to awarding the 50 percent contingency multiplier, the district court awarded plaintiffs' counsel $14,818.63 as compensation for delay, covering the time between initiation of the suit and the judgment. In Delaware Valley II, a majority of the Court (Justice White for the plurality joined by Justice O'Connor in this part), stated that adjustment for delay is a distinct issue from adjustment for the risk of non-payment. 483 U.S. at 716, 107 S.Ct. at 3081.3 The Court noted the general practice of the courts in recognizing the delay factor in setting fees for prevailing counsel. It referred to the use by the courts of two methods in adjusting for delay, basing the fee award on current rates or adjusting the fee based on historical rates to reflect its present value. Id. This court's opinions have followed the latter approach, although we do not rule out the former in an appropriate case.4

In Institutionalized Juveniles v. Secretary of Public Welfare, 758 F.2d 897, 923 (3d Cir.1985), we upheld the district court's decision to award a 25 percent enhancement for delay, and expressed our view that the relevant factors in setting the delay enhancement are the market interest rate and the time value of money, rather than the inflation rates. We also stated that the district court was not "required" to award a delay-in-payment multiplier that was equal to the prevailing rate of interest and noted that the court must take into account different rates because an attorney's work is performed over a period of time. Id. See also Black Grievance Comm. v. Philadelphia Elec. Co., 802 F.2d 648, 655-56 (3d Cir.1986), vacated and remanded, 483 U.S. 1015, 107 S. Ct. 3255, 97 L. Ed. 2d 754 (1987).

Thereafter, we held in SPIRG that because the plaintiff had failed to meet its burden of demonstrating to the district court the evidence required under Institutionalized Juveniles, the district court did not abuse its discretion in denying plaintiff's counsel an increase for delay of payment of the fees. 842 F.2d at 1453-54. Although we reiterated "the important policy of recompensing plaintiff for delay," plaintiff had failed to meet its burden to establish a record sufficient to make an award for delay. Id. at 1454.

In this case, the magistrate based his recommendation against any multiplier for delay on the ground that plaintiffs had not made an adequate showing. See 702 F. Supp. at 501. The evidence before the magistrate was limited to the increased cost of living during the relevant period, id., and thus the magistrate's conclusion was consistent with our legal analysis in Institutionalized Juveniles. In contrast, the district court had before it evidence of the market interest rates throughout the period of the litigation. It also had the certification of counsel with respect to loans taken by the law firm and the approximate amount of interest paid thereon during the period while it awaited payment of the fee in this case. We review the district court's decision awarding plaintiffs the sum of $14,818.63 as enhancement for delay under an abuse of discretion standard. See Institutionalized Juveniles, 758 F.2d at 922.

Witco argues that it was inconsistent for the district court to award to plaintiffs' counsel an enhancement for delay in the payment of counsel fees when it had earlier denied plaintiffs' own request for pre-judgment interest. We see no inconsistency. The damages awarded to the plaintiffs were based on loss of pension in the future, and thus could not have logically supported pre-judgment interest. An adjustment for delay in receipt of statutory fees, however, is "designed to compensate the attorney for the time gap between the actual expenditure of services and the fee award," Black Grievance Comm., 802 F.2d at 656, and serves to put the attorney willing to take such cases on par with attorneys who get paid before or during the course of their representation of clients.

We conclude that the district court did not abuse its discretion in awarding an enhancement for delay in the amount of $14,818.63, and we will affirm its order in that respect.

For the reasons set forth above, we conclude that no contingency multiplier shall be allowed plaintiffs in this case, and we will reverse that portion of the district court's order of February 6, 1989 that granted the plaintiffs a 50 percent contingency multiplier. It follows that we will also reverse that portion of the order granting interest on the contingency multiplier. We will, however, affirm that portion of the order that awarded plaintiffs $14,818.63 as an enhancement for delay and the portion granting interest on the delay enhancement from April 30, 1986. Costs on appeal are to be divided equally between the plaintiffs' attorney on one hand and the defendant.