CUDAHY, Circuit Judge.
Members of a pension fund needed a named plaintiff and legal representation to combat fund mismanagement. Archie Cook, the named plaintiff, needed an incentive to bring the suit, and the attorneys he ultimately retained needed a reasonable fee for their services. One might have thought that members of the pension fund and Cook’s attorneys would be satisfied with the result in the district court. The pension fund recovered over $14 million and underwent substantial structural reforms, Cook received a $25,000 incentive award, and the attorneys obtained more than $2 million for approximately 5,900 hours of work. But apparently no one feels that he, she, or it received the benefit of the bargain, since the pension fund challenges the propriety of the incentive award and both the fund and Cook’s attorneys object to the amount of attorney’s fees. On appeal we affirm the district judge in all respects.
I. Background
In 1992, Archie Cook, a truck driver and participant in the Teamsters Local 705 Health and Welfare Fund, filed a class action [1009]*1009against the Fund, the International Brotherhood of Teamsters Local 705, and the trustees of both entities. He alleged that the Fund had been mismanaged in violation of the Employee Retirement Income Security Act of 1974 (ERISA), and sought damages for the Fund of more than $20 million and structural reforms. In September 1995, Cook’s attorneys advised the district judge of the possibility of settlement with the defendants’ liability insurer. At Cook’s request and in an effort to facilitate settlement, Judge Manning referred the case to Special Master Frank McGarr. The parties entered into a settlement agreement in January 1996, which provided that the Fund would undergo extensive structural reforms and receive a cash payment of over $13 million. Judge Manning preliminarily approved the settlement and again referred the case to Special Master McGarr, instructing him, in relevant part, to make recommended findings regarding petitions for attorney’s fees and incentive awards.
In accordance with these instructions, Special Master McGarr held hearings, reviewed documents and then prepared a report for the district judge. In his report, the master concluded that Archie Cook was entitled to a $25,000 incentive award because of the time he devoted to the case and the risk he faced in bringing the suit. With respect to attorney’s fees, Cook’s counsel had requested $4,785,000. In evaluating this request, the special master found that contingent fee agreements in the Chicago area typically provide that attorneys will be paid between 25 and 30 percent of the amount they recover for their client. The master determined that, in light of the substantial relief obtained by Cook’s counsel and the riskiness of the litigation, an appropriate fee was $4,400,000, or 30 percent of the amount recovered for the Fund.1 Although the number of horn’s logged by Cook’s attorneys and their hourly rates were not directly relevant to the percentage-of-fund analysis, the special master noted the reasonable hourly rates for the attorneys and paralegals involved in the litigation: between $115 and $325 for the attorneys and $80 for paralegals. Attorneys had logged 4,741 hours and paralegals 1,166 hours, for a total of 5,907 hours. The master acknowledged that a “very small part” of the total hours expended may be excessive, but did not pursue the issue further. The attorney’s fees and Cook’s incentive award were to be paid out of the settlement escrow created for the benefit of the Fund.
When presented with the special master’s report, Judge Manning agreed that Cook should receive a $25,000 incentive award. But she declined to follow the master’s recommendation with respect to attorney’s fees. Instead she employed the lodestar method, multiplying each attorney’s and paralegal’s hourly rate by the total number of hours spent on the case. To reflect the riskiness of the litigation, Judge Manning enhanced the lodestar’ by a multiplier of 1.5. In making these calculations, the district judge used the rates and hours that were reported by the special master. The final result was an attorney’s fee of $2,121,045.40.
Both Cook’s attorneys and the Fund are dissatisfied with this result. Cook’s attorneys assert that Judge Manning erred in rejecting the recommendation of the special master and failed to award a reasonable fee. The Fund challenges the district judge’s use of a multiplier, the hours she included in her computation, and her approval of Archie Cook’s incentive award.
II. Rejection of the Special Master’s Recommendation
We begin by addressing whether Judge Manning erred when she declined to follow the special master’s recommendation to award a percentage of the cash recovery to Cook’s attorneys. Before reaching the merits of the question, we must focus on the applicable standards of review. Because this case involves a special master and a district judge, two standards of review are relevant: the standard the district court applies to the special master’s report and the standard we apply to the district court’s opinion. With [1010]*1010respect to the former, the general rule is that the district court steps into the shoes of an appellate court and employs the same standards that an appellate court uses to review a lower court opinion. See Williams v. Lane, 851 F.2d 867, 884 (7th Cir.1988). Hence a district court reviews a special master’s legal conclusions de novo, see id. at 885, and accepts findings of fact unless they are clearly erroneous, see Fed.R.Civ.P. 53(e)(2). But the choice between the lodestar and percentage-of-fund methods is neither a question of law nor of fact; rather, a district judge has discretion to use either method, depending on the particular circumstances of the case. See Harman v. Lyphomed, Inc., 945 F.2d 969, 975 (7th Cir.1991). What is not clear is how much deference the district court, in exercising this discretion, should afford the special master’s recommendation.
Cook’s attorneys argue that the general rule should apply—that the district court should review the master’s recommendation for an abuse of discretion, the same standard an appellate court would use to review a district court’s choice between the lodestar and percentage-of-fund methods. In support of this position, Cook’s attorneys cite Gottlieb v. Barry, 43 F.3d 474 (10th Cir.1994). We agree that Gottlieb provides the proper standard of review, but do not understand the case as holding that abuse of discretion is the appropriate standard. While the Tenth Circuit analogized the master’s recommendation of a method to a credibility finding, it stated only that the district court should “give[] some deference to the master’s assessment [of which] method was more likely to result in a reasonable fee in this particular case.” Id. at 487 (emphasis added). Although “some deference” is neither precise nor easily quantifiable, we believe it adequately captures the weight that a district court should assign to the master’s recommendation of the lodestar or percentage-of-fund method. It also comports with the language of In re Continental Ill. Sec. Litig., 962 F.2d 566 (7th Cir.1992), where we encouraged a district judge to appoint a special master who would “advise the court” with respect to attorney’s fees. Id. at 573. While In re Continental recognized the value of a master’s opinion, it noted that “[t]he judge may have insights to add by virtue of having observed the lawyers in action.” Id. at 573-74. A district court that extends some deference to the master will consider his recommendation and the factors influencing it, but will not regard the recommendation as the alpha and omega of the fee analysis.
While not as familiar as, for example, “abuse of discretion” and “de novo,” “some deference” is not a novel standard of review. See, e.g., Goluba v. School Dist. of Ripon, 45 F.3d 1035, 1037-38 (7th Cir.1995) (when district court oversees and approves consent decree, its interpretation of the decree is entitled to some deference); Brown & Williamson Tobacco Corp. v. Jacobson, 827 F.2d 1119, 1141 (7th Cir.1987) (appellate court gives some deference to jury’s determination of presumed damages in libel suit while also considering whether the award is excessive); United States v. Ceballos, 812 F.2d 42, 47 (2d Cir.1987) (district court is entitled to some deference when it concludes whether a person has been seized within the meaning of the Fourth Amendment). These eases illustrate that while appellate courts extend due regard to those who are especially familiar with a matter, they do not forfeit the authority to independently decide issues within their province. A special master who facilitates settlement and reviews fee applications is likely to have “direct and extensive” knowledge about the particular circumstances of a case and which method will produce a reasonable fee. Gottlieb, 43 F.3d at 487 n. 10. Just as an appellate court displays some deference to those with special knowledge, a district court must extend some deference to the master’s recommendation regarding the lodestar and percentage-of-fund methods. But the judgment of the special master does not usurp that of the district judge, who retains the discretion to choose the method she finds most appropriate in light of the particular circumstances of the case.
Next we address the standard under which we review the district court’s decision to reject the master’s recommendation of methods. Perhaps we risk redundancy, because we have already stated that the district court has discretion to choose between meth[1011]*1011ods. But it is worth emphasizing that we will review the district court’s decision for an abuse of discretion. Typically when the district court disagrees with the special master, we bypass the district court’s decision altogether and review only the report of the special master. See McDonald v. United Air Lines, Inc., 745 F.2d 1081, 1089 (7th Cir. 1984); Gottlieb, 43 F.3d at 486 (“Where the district court rejects a factual finding by the master, we ... directly review the findings of the special master, thereby effectively ignoring the district court’s review of the master’s findings.”). Or, in other words, we review the district court’s decision de novo to determine whether the master has erred. In this circumstance, however, it would be foolish to circumnavigate the opinion of the district judge. No matter how thoroughly an appellate court combs the record, it has no more than a sterile and second-hand account of what occurred in the court below. Of course, the same can be said of a district court when it reviews a master’s factual findings — it too is one step removed. But an award of attorney’s fees reflects the services counsel performs from the inception of a case through its ultimate resolution. A district judge has witnessed the progress a case has made while on the court’s docket and is in the best position to evaluate the litigation in light of its entire history. Here, for instance, almost five years elapsed between the filing of Cook’s suit and final approval of the settlement agreement. During that time, Judge Manning certified the class, consolidated the case with a similar action, appointed a special master and placed her stamp of approval on the settlement. In light of involvement such as this, even when a district judge is assisted by a special master, the district judge is “far better suited than [an] appellate court[] to assess a reasonable fee.” Harman, 945 F.2d at 973. Therefore we will review the district court’s selection of the lodestar or percentage-of-fund method for abuse of discretion.
Of course, “whether this discretion has been abused depends ... on the bounds of that discretion and the principles that guide its exercise.” United States v. Taylor, 487 U.S. 326, 336, 108 S.Ct. 2413, 2419-20, 101 L.Ed.2d 297 (1988). We do not believe that Judge Manning violated any of the precepts that govern the choice between the lodestar and percentage-of-fund methods, although in some respects her opinion renders resolution of this issue more difficult than it ordinarily would be. She neither expressly rejected nor accepted the findings of the special master that related to attorney’s fees. Yet her adoption of the master’s findings is implicit in her opinion, which notes the riskiness of the litigation, the extent of the Fund’s recovery and the substantial part that Cook’s attorneys had in obtaining the result — all questions of fact that had been determined by the special master. Thus it is apparent that Judge Manning accepted Special Master MeGarr’s factual findings, as she was required to do unless the findings were clearly erroneous. See Fed.R.Civ.P. 53(e)(2).
It also is clear that Judge Manning was concerned with the fact that Cook’s litigation resulted in the creation of a common fund, an actuality not emphasized by the special master. In common fund cases, after attorneys obtain a settlement for the class, they petition the court for compensation from the fund that has been established for the benefit of the plaintiffs. Because the payment of attorney’s fees ultimately reduces the amount of the common fund, after attorneys secure a settlement, “their role changes from one of a fiduciary for the clients to that of a claimant against the fund created for the clients’ benefit.” Skelton v. General Motors Corp., 860 F.2d 250, 253 (7th Cir.1988) (internal quotation marks omitted). As the district judge (and no doubt the special master) understood, this metamorphosis on the part of counsel means that “[t]he court becomes the fiduciary for the fund’s beneficiaries and must carefully monitor disbursement to the attorneys by scrutinizing the fee applications.” Id. When, as here, the plaintiff class has independent legal representation with respect to fees, the need for vigilance by the district court inevitably diminishes. But the district judge nonetheless “must balance the competing goals of fairly compensating attorneys for their services rendered on behalf of the class and of protecting the interests of the class members in the fund.” Id. at 258. This is not to suggest that the district court should compel attorneys to sacrifice a portion [1012]*1012of their reasonable fee for the good of the class. But the district court must ensure that the plaintiffs pay no more than what is reasonable.
Judge Manning plainly was concerned that the percentage-of-fund method resulted in excessive compensation for Cook’s attorneys. In recommending that Cook’s counsel receive 30 percent of the common fund, Special Master McGarr did not consider the hourly rates that would result from such an award: roughly $750 for each attorney and paralegal on Cook’s team, more than twice the usual hourly rate of the most expensive attorney involved in the litigation. We suspect that this figure weighed heavily on the mind of Judge Manning when she decided that the percentage approach was not necessarily this case’s Rosetta Stone. Indeed, she expressly noted that the lodestar method “provides for greater accountability and a more careful cheek on excessive fees.” Mem. Op. at 39 (citing Harman, 945 F.2d at 974).
This is not a case like Gottlieb, where the district court abused its discretion by rejecting the master’s recommendation of'the percentage-of-fund method “for reasons largely unrelated to the particular circumstances of [the] case.” 43 F.3d at 487. Here Judge Manning afforded some deference to the master, as she considered the riskiness of the litigation and the other factors that influenced his recommendations. But she also contemplated her responsibility to the plaintiff class and recognized that the lodestar method encourages accountability and guards against excessiveness. Under these circumstances, we cannot say that Judge Manning abused her discretion when she rejected the special master’s recommendation.
III. The District Court’s Choice of the Lodestar and the Resulting Award
Our analysis cannot end here, however. Cook’s attorneys argue that even if Judge Manning had the discretion to reject Special Master McGarr’s recommendation, her choice of the lodestar was nonetheless an abuse of discretion. First they contend that Judge Manning did not consider the merits of the percentage-of-fund approach, because she erroneously believed that the lodestar method applied to common fund cases as a matter of law. And it is true that the district judge’s opinion includes an unfortunate sentence: “[A]s a matter of law, we find the percentage method to be inapplicable to this case as it defeats the whole purpose of attempting to make the plaintiffs whole.” Mem. Op. at 39. But the bulk of the opinion manifestly demonstrates that the judge understood the relevant law. She explained at the outset of her fee analysis that “both the lodestar approach and the percentage approach may be appropriate in the determination of attorney’s fee awards” and that “[t]he decision of which method to employ remains within the discretion of the district court.” Id. at 38 (citing Florin v. Nationsbank of Ga., 34 F.3d 560, 565 (7th Cir.1994) (a common fund case)). In announcing her decision to utilize the lodestar method, she stated that it was “the better determinant of the attorneys’ fees for the plaintiffs counsel in this case.” Id. at 39 (emphasis added). She also noted that there are advantages to the percentage method, such as the simplicity of its administration. See id. at 38. • In the face of all this, we have no doubt that the district judge was cognizant that she had a choice of methods. We think the statement about the percentage approach’s not applying as “a matter of law” was a careless articulation of the judge’s conviction that the lodestar method would yield the most reasonable fee in this case. We decline to remand on account of what amounts to no more than a slip of the pen.
Cook’s attorneys also object to Judge Manning’s use of the lodestar on the ground that, in this particular case, it failed to generate a reasonable fee. In In re Continental, we explained that “the object in awarding a reasonable attorney’s fee ... is to give the lawyer what he would have gotten ... in an arm’s length transaction, had one been feasible.” 962 F.2d at 572. Special Master McGarr reported that in the Chicago area, contingent fee contracts typically provide that an attorney will receive between 25 and 33 percent of the total recovery. As we construe the argument of Cook’s attorneys, this factual finding is significant for potentially two reasons. First, [1013]*1013because an arm’s length transaction probably would have resulted in the attorneys receiving a percentage of the cash recovery, a fee that is determined by the lodestar method is not, a fortiori, a reasonable fee. Or, in the alternative, even if the fee can be determined pursuant to the lodestar approach, the fee is not reasonable unless it is similar to that reached under the percentage method. We consider each of these arguments in turn.
We have never held that a district judge must choose the method that is most prevalent in the marketplace. Indeed, such a holding would run counter to our statement that the district judge has discretion to choose between the lodestar and percentage-of-fund approaches. That we leave the matter to the district court’s discretion implies that the judge may wish to consider several factors; evidence about the market’s preference is only one. In Harman, we considered whether district courts should be required to utilize the percentage method in common fund eases. See 945 F.2d at 974-75. In concluding that it was premature to banish the lodestar approach, we noted that it helps prevent overeompensation and encourages accountability. See id. at 974. The inference, of course, is that, depending on the case, the percentage approach may result in excessiveness, and that district courts may wish to consider this as they choose between methods. Judge Manning had information about which approach the market favors, and no doubt these data figured into her fee analysis. But she was not obligated to regard the market data as the sole determinant in her choice between methods.
Moreover, if substantially similar results must be reached under the lodestar and percentages approaches, then Harman might as well have eliminated the lodestar altogether. Judge Manning’s selection of a 1.5 multiplier is at the root of Cook’s attorneys’ objection to the fee award; presumably they believe that a multiplier of about 3 would have been more appropriate. The choice of a multiplier, however, is left to the district court’s discretion, see id. at 976, and whether Judge Manning has abused this discretion depends upon “the principles that guide its exercise.” Taylor, 487 U.S. at 336, 108 S.Ct. at 2419. In this regard, we have explained that, in order to determine a multiplier, the district court must attempt “a retroactive calculation of the probability of success as measured at the beginning of litigation.” Skelton, 860 F.2d at 258. Our cases hold that the district court must award a multiplier when attorney’s fees are contingent upon the outcome of the case (i.e., there is the possibility that the attorney will not receive any fee). See In re Continental, 962 F.2d at 569; Harman, 945 F.2d at 976. We also have speculated that a multiplier of 2 may be a sensible ceiling. See Skelton, 860 F.2d at 258. Finally, we have suggested that “[a]fter establishing the lodestar and multiplier, the judge may find it useful to compare the percentage of the fund to contingent arrangements negotiated in other eases of the same type.” Harman, 945 F.2d at 974. But we have never ordered the district judge to ensure that the lodestar result mimics that of the percentage approach. See id. (noting that the comparison “is a reasonable, though not essential ‘check’ for the judge.”); see also In re Continental, 962 F.2d at 572. Perhaps if we were engaging in de novo review, we would choose a different multiplier in light of the “check” against the market information provided by Special Master McGarr. But this does not mean that we can find that Judge Manning violated the principles governing her discretion to choose a multiplier.
Cook’s counsel rely heavily on In re Continental, 962 F.2d 566, in which we remanded for a recalculation of the amount due to the class counsel. There, like here, the district judge awarded the attorneys roughly one-half of what they had requested. See id. at 568. But apart from this superficial similarity, In re Continental is not analogous to the case at bar. The district judge in In re Continental refused to award any multiplier, even though the litigation was contingent. See id. at 569. He capped every attorney’s billing rate at $175 per hour, even though some attorneys had regular billing rates of almost twice that, see id. at 568, and refused to compensate paralegal services at their market rate. See id. at 569. In addition, he [1014]*1014reduced the time logged for legal research by 40 percent. See id. at 570. In contrast, Judge Manning accepted the hours reported as “representative” and calculated the lodestar according to the attorneys’ regular billing rates. She also awarded a multiplier to reflect the riskiness of the litigation, albeit a multiplier that was lower than the one Cook’s counsel would have preferred. But as we have already explained, Judge Manning’s choice of a multiplier was not an abuse of discretion. And we remind Cook’s counsel that Judge Manning’s fee award effectively compensated each attorney and paralegal at the rate of $360 per hour — $35 more than the normal billing rate of the most expensive attorney involved in the litigation.
For its part, the Fund challenges the district court’s authority to employ any sort of multiplier, even one of 1.5. But as the Fund recognizes, in Florin, 34 F.3d at 564, we held that risk multipliers are appropriate in cases that are initiated under ERISA and settled with the creation of a common fund. We do not believe that Florin conflicts with Fleischmann Distilling Corp. v. Maier Brewing Co., 386 U.S. 714, 87 S.Ct. 1404, 18 L.Ed.2d 475 (1967), as the Fund suggests. In Fleischmann, the Supreme Court addressed whether attorney’s fees could be awarded in cases brought pursuant to the Lanham Act.2 In concluding that fee awards were impermissible, the Supreme Court noted that Congress had “meticulously detailed” the remedies available to a plaintiff who proves a violation of the Lanham Act. Id. at 719, 87 S.Ct. at 1407-08. Because attorney’s fees were not included in the list of remedies and congressional attempts to amend the Act had been unsuccessful, the Court found that Congress intended to “mark the boundaries” of the power to award monetary relief in cases arising under the Lanham Act. Id. at 721, 87 S.Ct. at 1408-09. Therefore an award of attorney’s fees was inappropriate. See id.
The Fund argues that Fleischmann necessitates that we overrule Florin. In Florin we held that in ERISA common fund eases, the award of attorney’s fees is made pursuant to the principles of equity, not ERISA’s fee-shifting provision. See 34 F.3d at 563. In the Fund’s view, Fleischmann dictates that because ERISA has a fee-shifting provision, see 29 U.S.C. § 1132(g), attorney’s fees may be recovered pursuant to that provision or not all. Under Fleischmann, the Fund contends, the resort to equitable principles violates boundaries demarcated by Congress. If the Fund is correct and attorney’s fees in ERISA common fund cases may be granted only pursuant to the fee-shifting provision, Judge Manning does indeed lack the authority to utilize a risk multiplier. In City of Burlington v. Dague, 505 U.S. 557, 112 S.Ct. 2638, 120 L.Ed.2d 449 (1992), the Supreme Court held that fees awarded pursuant to a fee-shifting provision cannot be enhanced with a multiplier.
But because the Fund’s interpretation of Fleischmann is untenable, the Fund’s argument cannot withstand scrutiny. A ease that finds that a court cannot award attorney’s fees unless authorized by statute is not the equivalent of a case that holds that a court cannot award attorney’s fees according to equitable principles when the relevant statute plainly contemplates that attorney’s fees will be a remedy. Moreover, even if Fleisch-mann could be stretched so far, its application is not absolute. Fleischmann only stated that “when a cause of action has been created by a statute which expressly provides the remedies for vindication of the cause, other remedies should not readily he implied.” 386 U.S. at 720, 87 S.Ct. at 1408 (emphasis added). As we explained in Florin, there is ample reason to conclude that Congress (and the Supreme Court in Dague) did not intend to foreclose resort to equitable powers in common fund cases. See Florin, 34 F.3d at 563-65 (discussing statutory language, the purpose of ERISA and the policy considerations in Dague). Finally, Fleischmann itself suggests that in common fund cases, a court may exercise its equitable powers to award attorney’s fees, regardless of whether the remedy has been prescribed by Congress. The Fleischmann plaintiff sought to rely on Sprague v. Ticonic Nat’l Bank, [1015]*1015307 U.S. 161, 59 S.Ct. 777, 83 L.Ed. 1184 (1939), which involved a variation of the common fund situation. See 386 U.S. at 719, 87 S.Ct. at 1407-08. In declining to grant attorney’s fees, the Fleischmann Court expressly noted that the considerations that had supported the award of attorney’s fees in Sprague were inapplicable. See id. at 720, 87 S.Ct. at 1408. For all these reasons, we are confident that the reasoning of Florin is sound.
Under Florin, Judge Manning’s use of a multiplier was not merely permissible, but mandated, because Cook’s counsel “had no sure source of compensation for their services.” 34 F.3d at 565 (quoting In re Continental, 962 F.2d at 569). Contrary to the Fund’s contention, the application of a multiplier did not result in Cook’s attorneys receiving double compensation for the riskiness of the litigation. As the Supreme Court has explained, the degree of risk in a particular case depends upon the merits of the claim and the difficulty of establishing those merits. See Dague, 505 U.S. at 562, 112 S.Ct. at 2641. The unenhanced lodestar usually accounts for the difficulty of prevailing on the merits, because it reflects the number of hours that are necessary to overcome the difficulty and the hourly rates of attorneys who are skilled enough to do so. See id. But the unenhanced lodestar does not reflect the factual and legal merits of the claim — the fact that at the outset of the litigation, no matter how dazzling the array of legal talent or how many hours will eventually be logged, there is nonetheless the possibility of no recovery. See Hannan, 945 F.2d at 976. The Fund emphasizes that Dague holds that the risk of no recovery should not factor into the computation of the lodestar. But as we explained in Florin, the policy considerations in Dague “have little force in common fund cases.” Florin, 34 F.3d at 564. In fee-shifting cases, the attorney’s fee is an additional burden on the defendant; the plaintiffs recovery is not reduced by the amount of fees that the defendant is ordered to pay. In common fund cases, however, attorney’s fees are assessed against the fund; the defendant’s liability cannot exceed the amount the defendant has agreed to compensate the plaintiff. See id. Therefore, in contrast to Dague and other fee-shifting cases, defendants in common fund cases cannot be characterized as subsidizing unsuccessful lawsuits against other defendants. See id. at 565. And we perceive “no injustice in requiring plaintiff class members to shoulder the burden of compensating counsel for prosecuting the class’ ease without any assurance of compensation.” Id. In sum, Judge Manning did not err when she enhanced the lodestar by a multiplier of 1.5.
Apart from the issue of the multiplier, the Fund raises an objection to Judge Manning’s computation of the lodestar. Here the Fund challenges the nuts-and-bolts of the lodestar calculation, specifically, whether Judge Manning included excessive hours when she multiplied the number of hours expended by each attorney by the attorney’s reasonable hourly rate. This is a question about the district court’s “methodology,” which is subject to de novo review. See Hannan, 945 F.2d at 973.
After review of the fee application and the district court’s opinion, we believe that a remand is unwarranted. The Fund argues that too little work was done by associates, that excessive hours were logged for conferencing and after settlement offers had been made, and that the Fund was billed for time spent preparing the fee petition. Special Master McGarr found that a “very small part of the total hours expended” may have been excessive, but declined to pursue the matter further because it was not relevant to the percentage-of-fund approach. See Report at para. 83. Judge Manning noted the objections, but concluded that the hours were “representative” of the work performed by Cook’s counsel. See Mem. Op. at 41. In light of this conclusion, we are disinclined to pursue the only issue raised by the Fund that might have necessitated a remand — the possibility that the lodestar included hours expended on the fee application. See Mills v. Eltra Carp., 663 F.2d 760, 761 (7th Cir. 1981) (attorneys cannot bill common fund for services related to the fee petition); Lindy Bros. Builders, Inc. v. American Radiator & Standard Sanitary Corp., 540 F.2d 102, 111 (3d Cir.1976) (same). Here Judge Manning [1016]*1016was concerned with the total number of hours logged and whether the resulting fee was a reasonable one given the services of Cook’s attorneys. Therefore it is unlikely that a minor change in the hourly totals would have an effect on the final amount of the fee award. Given the events in this particular case, and our review of the fee application, we are satisfied that the hours included in the lodestar are representative. As we explained in In re Continental, the judge determining the fee award has to “play surrogate client.” 962 F.2d at 572. And we suspect that Judge Manning was a particularly diligent surrogate for the plaintiffs. She was plainly determined to preserve as much of the fund as possible while still providing Cook’s counsel with a reasonable fee. Indeed, given Judge Manning’s manifest concern for the plaintiff class, the Fund lacks credibility when it suggests that the district judge did not adequately guard against ex-cessiveness.
IV. The Incentive Award
Having resolved the issues surrounding the attorney’s fees, we address the final matter raised by the Fund — the propriety of Archie Cook’s $25,000 incentive award. Because a named plaintiff is an essential ingredient of any class action, an incentive award is appropriate if it is necessary to induce an individual to participate in the suit. See In re Continental, 962 F.2d at 571. In deciding whether such an award is warranted, relevant factors include the actions the plaintiff has taken to protect the interests of the class, the degree to which the class has benefitted from those actions, and the amount of time and effort the plaintiff expended in pursuing the litigation. See Spicer v. Chicago Bd. Options Exchange, Inc., 844 F.Supp. 1226, 1267 (N.D.Ill.1993). Here these factors are readily satisfied. Cook brought a suit that resulted in structural reforms to the Health & Welfare Fund as well as a cash recovery of more than $13 million. In findings that were well-supported by the evidence, Special Master McGarr noted that Cook spent hundreds of hours with his attorneys and provided them with an “abundance of information.” Most significantly, the special master found that, in filing the suit, Cook reasonably feared workplace retaliation. In light of the benefit Cook bestowed on his class, the risks he faced in bringing the case and the time he spent pursuing it, Judge Manning did not err when she approved a $25,000 incentive award. Accordingly, the case is
Affirmed.