Critics of our society’s penchant for litigation could make a strong argument on the basis of attorney’s fees litigation alone. This ease is perhaps paradigmatic. In an action that was settled before it even arrived in the courtroom, this fight over attorney’s fees has lasted five years (and counting), has required two decisions by both the district court and this court and has itself generated a request by the plaintiff alone for fees of $45,000. And all of this over a ease that we estimated to be worth just $25,000. Considered judgment has been lost in the clamor of combat. With two relatively minor exceptions, we affirm the decision of the district court with respect to both parties’ appeals.
I.
We assume a familiarity with our first opinion in this case,’
Nanetti v. University of Illinois at Chicago,
The district court addressed the apportionment of fees on the merits first. Using calculations we will describe in greater detail later, the district court found that Nan-etti’s lawyers were due $8,290 for their role in securing the salary settlement. It arrived at that award by multiplying a reasonable hourly rate for each of plaintiffs counsel by the number of hours that attorney spent working on Nanetti’s salary claim. Mem.Op. at 5-10, 11-12 (April 11, 1990). The district court agreed to the rate requested by each of plaintiff’s counsel, with the exception of her lead counsel, whose requested rate it adjusted from $175 to $125 per hour. Id. at 2-3.
The district court then attempted to untangle the fees-on-fees issue. It first decided, in accord with an agreement between the parties, that because the plaintiff’s arguments to the court of appeals had been only partly successful, not all of the fees generated during that litigation would be compensated. The court accepted Nanet-ti’s suggested 15% reduction in her lodestar for the fee litigation. 1 Id. at 11. The district court went on to apply an additional 35% reduction because plaintiff had been only partly successful in her fee litigation, recovering just 58% of her original request. Implementing these calculations, the district court awarded Nanetti $17,982 in fees-on-fees.
II.
We review a district court’s award of attorney’s fees for abuse of discretion,
*1418
unless the error alleged is one of law.
Brooms v. Regal Tube Co.,
We first reach Nanetti’s argument that the district court abused its discretion in reducing the hourly rate of her lead counsel, Martin Oberman, through most of this litigation. Our deferential posture in reviewing fee awards notwithstanding, we have previously admonished district courts that would grant less than the hourly rate requested “to give reasons justifying the particular cut that [it] want[s] to impose.”
Henry v. Webermeier,
With all due deference to the district court’s knowledge of the rates for Chicago attorneys, we believe this reduction constituted an abuse of discretion. As an initial matter, the lack of a challenge to the rate requested by Oberman makes it more difficult to apply any reduction to that rate. Oberman has had no opportunity to build a record on his usual billing rate, other counsel’s assessment of his talents or any other relevant, reliable information concerning his request. Particularly in the absence of a challenge to counsel’s requested rate, it is imperative that the district court set out with specificity its reasons for reducing that rate. Those reasons were not adequately spelled out in the opinion below. Ordinarily, we would remand this matter for further consideration by the district court, but, in light of the history of this litigation, we think it more appropriate to end the matter here and now. The judgment is therefore modified to award Oberman fees at the rate of $175 per hour.
Unlike the only case cited to us by the University in which a district court reduced counsel’s requested rate,
see Chrapliwy v. Uniroyal, Inc.,
We turn next to the errors claimed in the district court’s calculations of the hours recoverable in settling the merits (periods one and two). The University argues that no hours in period one should have been counted while Nanetti argues that sixteen hours were wrongly excluded from the two
*1419
periods. We reject both arguments. With only minor exceptions, the district court allowed the hours requested by Nanetti’s counsel for the period prior to filing the EEOC charge. The University contends that our earlier opinion precludes any award for fees expended during this first period, because that period involved only efforts expended on Nanetti’s tenure dispute, not her salary claim. The defendant claims that, under
Hensley,
the tenure issue and the salary issue are not sufficiently related that work on the former (for which fees are not recoverable under our first opinion in this case) can be counted as work on the latter.
Hensley
held that hours spent on an unsuccessful claim might be unrecoverable where that “claim ... is distinct in all respects from [plaintiff’s] successful claims_”
The University ignores our precedents, which clearly state that when time is spent jointly preparing two distinct claims, the fact that one claim produces no recovery will not deprive the plaintiff of every hour spent in joint preparation. Thus, in
Ustrak v. Fairman,
Before submitting his hours for this period, Nanetti’s attorney deleted 13.25 of his 50.25 hours worked, recognizing that some of his efforts involved legwork specific to the tenure claim. The other hours, he contended, involved legal and empirical research that would eventually support both the tenure and salary case against the University. The district court’s analysis of this point is lucid and convincing:
The legal theories in each of Nanetti’s claims were the same. Nanetti based her claims on Title VII and its prohibitions against discrimination on the basis of gender. Title VII does not distinguish between discrimination in denial of tenure and discrimination in salary. Both can be unlawful employment practices.
The facts underlying each of Nanetti’s claims too were similar. The University was a defendant for both claims. In both instances Nanetti would have had to prove at trial that she was a female and that the University intentionally treated her differently from similarly situated male professors. As a result, Nanetti would have had to investigate who at the University made these decisions, how they were made, and their historical results .... In short, legal work on the University’s tenure decisions easily could have duplicated legal work on the University’s salary decisions.
Mem.Op. at 5-6. We find no abuse of discretion in this careful consideration.
The University argues that suits over tenure and salary are unrelated as a matter of law, citing
King v. McCord,
Nanetti on the other hand contends that the court erred in denying sixteen hours of her attorney’s time during the first period. The district court carefully explained each of its decisions. Two hours were denied because they were specifically recorded as “tenure process” activities. The district court also looked askance at Judson Miner’s claim for ten and one-half hours during settlement negotiations. Among plaintiff’s
*1420
counsel who billed during these negotiations, Miner was the only one not to offer any reduction in hours billed to account for the University’s concession of the tenure claim. Coupled with his failure to document these six and one-half hours adequately, Miner’s decision not to reduce led the district court to do it for him, bringing his total during the period down to four hours. Finally, the court reduced Ober-man’s hours by one-third during period two, because of questionable billing practices and an indication in the records submitted for that period that the attorney was pursuing other, ultimately unsuccessful relief from the University. The district court’s opinion provides us a “concise but clear explanation” of its reasons for reducing the hours requested,
Hensley,
The last issue before us in both the appeal and cross-appeal is the district court’s determination of the fees-on-fees award. The plaintiff complains that the district court abused its discretion in twice reducing the hours she requested for arguing the initial fee petition. The defendant argues conversely that the district court did not reduce enough, specifically because it failed to consider the defendant’s settlement offer following the settlement of the merits. Our starting point in this analysis is
In re Burlington Northern, Inc., Employment Practices Litigation,
Applied to the case before us, therefore, Hensley and Burlington Northern offered the district court two options. It could find the time spent unsuccessfully researching and arguing the fees for tenure distinct from time spent on the salary issue — and therefore count none of the former toward the lodestar — or it could find the two fees claims interrelated, but nevertheless reduce the lodestar due to its view of the plaintiff’s ultimate success in achieving redress.
The 15% reduction offered by the parties themselves fits neatly within the first class of reductions under Hensley. Apparently conceding that she could not recover for her fees expended pursuing only tenure-related fees, Nanetti offered to reduce her attorney’s fees during the fee litigation by 15% across the board. The appellee itself understood the import of this concession: “Since Nanetti did not prevail on her claim that she was a prevailing party on the tenure issue, on remand, Nanetti sought only 85% of the fees incurred on her initial fee petition [and] appeal.” Appellee’s Br. at 19. Although the district court accepted that reduction, it did not stop there. “Still, Nanetti’s success in her fee litigation was only 58%.... [I]t is [close] to the 42% success of the plaintiffs in Burlington Northern, who witnessed a 35% reduction in the lodestar for their lead counsel.” Mem.Op. at 13 (citations omitted). Apparently referring to the second of Hensley’s two types of reduction (examining the overall success of the litigation in obtaining the relief sought), the court therefore discounted the remaining hours spent on the fee litigation by 35%. Nanetti complains that she was thus subjected to a double reduction, and we must agree.
We do not find error in the methodology of employing both
Hensley
analyses to determine a plaintiff’s ultimate fee award. Other opinions from this circuit have recognized situations that might straddle the two types of reduction.
See Jackson v. Illinois Prisoner Review Bd.,
Thus, two different reductions from the lodestar can arise as the result of separate and distinct shortfalls of the litigation. The trouble with the reductions allowed in this case is that they were apparently invoked for essentially the same failure. The district court accepted the parties’ agreed 15% discount in plaintiff’s hours spent litigating the original fee petition. Nanetti offered that cut because her petition for tenure fees had been unsuccessful, and she believed 15% represented the portion of hours spent on that issue. Mem.Op. at 11. That reduction, which the district court accepted, ensued directly from our prior decision in this case that the fees expended for tenure were unrecoverable: because Nanetti had failed to convince this court that she should be awarded fees for her tenure battle, she considered those fees a distinct and unsuccessful claim, and reduced her fees-on-fees request accordingly.
The district court’s second reduction stemmed from its observation that “Nanet-ti’s success in her fee litigation was only 58%.” Mem.Op. at 13. It computed that success rate by comparing the amount of fees she sought for fighting the merits ($14,365) to the amount recovered ($8,290). Id. at 12-13. Of course, plaintiff conceded much of that $6,000 reduction as hours spent seeking tenure, a concession that followed immediately from our decision to disallow those hours. Even the marginal cuts to the request for fees on the merits implemented in the district court’s order flowed almost entirely from its findings that Nan-etti was wrongly including tenure hours in her fee request. Therefore the $8,290 figure used by the district court for purposes of computing the fee litigation’s “success rate” was overwhelmingly the result of this court’s decision to deny fees for the tenure dispute.
The district court thus apparently exacted a double reduction for what amounts to the same reason — Nanetti’s failure to convince this court of her right to recover fees for winning tenure. While a remand on such issues would ordinarily be appropriate, we believe none is required in this situation. Taking the plaintiff’s partial defeat into account, the district court believed that 35% represented an adequate reduction in her fees-on-fees request. Having so decided, the court should not have accepted the parties’ offer of a further 15% reduction for the same reasons. Given the amount of Nanetti’s eventual recovery for fees on the merits, we do not find the 35% reduction in itself an abuse of discretion. We will therefore apply a 35% reduction to reach a revised calculation of fees-on-fees.
Defendant’s cross-appeal on this issue requires only brief discussion. The University argues that it was error for the district court to award
any
attorney’s fees for litigating the fee issue because the fees actually recovered for settling the merits of this case only slightly exceeded the amount defendant offered to settle the matter. In several cases discussing exactly this point, this court has exhibited great deference to the district court’s decision whether a rejected settlement offer should affect the amount of fees awarded. Thus, in
Coop v. City of South Bend,
Nothing about this case convinces us to abandon this deferential posture. The district court adequately considered the defendant’s settlement argument and denied it any probative weight. Considering the apparent reluctance of both parties to bring this litigation to a close, we do not find that the district court abused its discretion.
III.
To bring to a close this spiraling litigation, we believe it appropriate to calculate ourselves the fee award as modified by our decision and to resolve the inevitable issue of fees for this appeal.
See Jackson,
Plaintiff appealed only related issues under
Hensley,
and was two-thirds successful in its arguments (and possibly more so when one considers the dollar value sought and recovered).
Cf. Ustrak,
Notes
. The district court factored this reduction into the hours requested by each attorney. Mem.Op. at 12 n. 6. Some of Martin Oberman’s hours and all of Thomas Johnson’s hours were counted at 100% because theirs were hours spent after remand from the court of appeals, at which point all arguments centered on salary-related fees.
. Oberman accounted for 52.5 hours during this time period. Thus the district court's award of $8,290 for fees on the merits must be increased by $2,625, bringing the fee award on the merits to $10,915.
. The increase in fees for this period results from both our decision to award Oberman a rate of $175 and from our decision to reduce the fees-on-fees award only by 35%, rather than by the sum of 15% and 35%. Multiplying the period three hours by the appropriate hourly rate, we reach the following result:
Attorney Hours Rate Total
Oberman 88.35 $175 $15,461.25
Mueller 10.50 125 1,312.50
Johnson 58.38 125 7,297.50
Hyman 142.50 80 11,400.00
$35,471.25
This figure must then be reduced by 35% to account for the plaintiffs limited success in her fee litigation, leaving Nanetti $23,056.31 in fees for this period.
