Dissenting:
As they always do, my conscientious colleagues display a thorough and scholarly grasp of the issues that arise in the settlement of class lawsuits. With all respect, however, I see this settlement and the district court’s approval of it in a different light. Thus, I respectfully dissent.
Three main worries, each of which in my view is just an illusion, appear to be driving the majority’s decision to reverse the district court’s approval of this consent decree: (1) the “possibility” that class counsel could have betrayed their clients in favor of their own fees; (2) that the “large differential” in the distribution of monetary awards between class representatives and certain identified class members, on one hand, and unnamed class members on the other, indicates something rotten in *488Denmark; and (3) that the district court inflated the $3.65 million value attached to the “largely precatory” injunctive relief. As I read this record, and as I shall attempt to demonstrate, these concerns have no foundation in fact or law.
Collusion
The Objectors correctly direct our attention to the rule that we have an obligation to police the settlement of class actions for evidence of collusion. See Fed.R.Civ.P. 23(e) (“A class action shall not be dismissed or compromised without the approval of the court....”) “The purpose of this salutary requirement is to protect the nonparty members of the class from unjust or unfair settlements affecting their rights” as well as to minimize conflicts that “may arise between the attorney and the class, between the named plaintiffs and the absentees, and between various subclasses.” Piambino v. Bailey, 610 F.2d 1306, 1327-28 (5th Cir.1980). We are to be cognizant of the “ ‘danger ... that the lawyers might urge a class settlement at a low figure or on a less-than-optional basis in exchange for red-carpet treatment for fees.’ ” In re General Motors Corp. Pick-Up Truck Fuel Products Liab. Litig., 55 F.3d 768, 819 (3d Cir.1995) (quoting Weinberger v. Great Northern Nekoosa Corp., 925 F.2d 518, 524 (1st Cir.1991)). In fact, this is the precise rationale fueling the Objectors’ challenge to the attorneys’ fees agreement: it suggests (1) that the attorneys “exploited the class action device to obtain merits large fees at the expense of the class,” and (2) that the representation was deficient.
Given the purpose of this rule, it occurs to me that if a class action merits settlement is not the product of collusion, the class et cetera has not been sold down the river in exchange for fees, and if the representation of the plaintiffs has been adequate, then the reasonableness of the fees presents itself in a different light, especially if the award is separate from the merits settlement of the case, as it is here. Thus, the first question for us is whether a showing has been made that the merits settlement suffers from some infirmity, suggesting in turn that the fees demonstrate a quid pro quo red carpet treatment in return for a betrayal of the client. I answer this question “no.”
The “possibility” of collusion in this case turns out under scrutiny to be a classic red herring. The district court carefully looked into this “possibility” and found nothing of the sort. The court allowed the Objectors to depose five individuals the Objectors claimed were participants in an alleged secret deal. These depositions uncovered no evidence of collusion, and they supported the proponent’s contention that there was no such evidence. At the end of the day on this important issue, the district court said:
Settlement agreements that are presented prior to class certification are to be considered with particular care because they are more likely to be a product of collusion and because they have not been negotiated by court-approved representatives. The Court has exercised this scrutiny and finds no evidence of collusion. In addition, the Court notes that unlike many other class actions, the plaintiffs here did not file a complaint having already agreed to a settlement. On the contrary, this case was contentiously litigated for a substantial period prior to the settlement negotiations. And the plaintiffs’ counsel spent a great amount of time preparing the case before it was filed. The nature and quality of this work gives every indication that plaintiffs’ counsel intended to pursue these claims vigorously until they reached a satisfactory result; and is wholly inconsistent with the suggestion *489that the attorneys planned to sacrifice the plaintiffs’ claims in exchange for a large fee award. It also indicates that the plaintiffs’ counsel had obtained sufficient evidence to understand the strength of the claims despite the lack of formal discovery.
The majority is certainly correct when they conclude that “[t]he district court neither abused its discretion in finding that counsel’s representation was appropriately vigorous for purposes of class certification nor clearly erred in finding that there was no overt collusion”; but then my colleagues erroneously decide this case on mere fears of the possibility of collusion— for which the district court specifically looked and found to be non-existent. What does collusion or the possibility of collusion have to do with this case and this settlement? Nothing. But, the mere specter of collusion in cases like this ends up almost as a neurosis undoing this settlement — even though villainy is manifestly non-existent.
Furthermore, this case, like all such cases, is unique in its facts and circumstances, and the district court’s clear understanding of all of its aspects, as expressed in the court’s Order of approval, dated September 30, 1999, explain away the “troublesome” dimensions of the settlement over which my colleagues — and the Objectors — fret. Three of these circumstances are covered by the first three Hanlon factors the district court must— and did — independently verify to determine that the consent decree is fair, adequate, and reasonable: (1) the strength of the plaintiffs’ case; (2) the risk, expense, complexity, and likely duration of further litigation; and (3) the risk of maintaining class action status throughout the trial. Hanlon v. Chrysler Corp., 150 F.3d 1011, 1026 (9th Cir.1998). The district court made the following findings and conclusions about these factors:
The Court reviews the first three factors together. The strengths of the plaintiffs’ claims and the risks of future litigation are clearly related questions. Boeing has faced a series of individual race discrimination law suits in recent years and has won every one. Although the number of named plaintiffs in this case and the descriptions of their experiences lend credence to their allegations, discrimination cases are notoriously difficult to prove, particularly when they are based on failures to promote or other discretionary decisions. There is also some risk that the class certified by the Court would not survive a challenge by Boeing, which might, for example, be able to demonstrate after more extensive discovery that the class claims are too individual to meet the commonality and typicality requirements. Finally, there is no doubt that continued litigation would be enormously burdensome and expensive for the plaintiffs as well as for Boeing. The price of discovery alone in a nationwide class action such as this one is almost always measured in millions of dollars. The objectors raise no serious challenges to these points, and the Court finds that these factors all strongly favor adopting the consent decree.
The record provides ample support for the district court’s understanding of these aspects of this complex litigation.
Accordingly, I conclude that there was no chicanery in this case, and the aspects of the consent decree to which the Objectors point as telltales of trouble are clearly dealt with and adequately explained by counsel and the district court. Judge Ber-zon’s scholarly opinion is an excellent discussion of that for which we must always be vigilantly on the lookout in these cases, but in the end, none of it applies to this *490settlement. My conclusion in this regard drives my analysis of the reasonableness of the attorneys’ fees, as I later explain.
The Differences in Damage Awards
Next, there is my colleagues’ problem with the distribution of monetary awards. This, too, was addressed, explained, and adequately handled by the district court:
More problematic is the disparity between the awards to the named plaintiffs and individually identified recipients and the awards to absent class members. The Court has considered this disparity carefully because excessive payments to named class members can be an indication that the agreement was reached through fraud or collusion. After reviewing the record, however, the Court concludes that there is nothing improper about the damages awards under the decree in this case. There are 264 named members of the class and individually identified recipients who will receive the larger awards. This is a very large group to have been brought in-line behind an allegedly collusive agreement. More importantly, this is the group of class members identified by plaintiffs’ counsel as having the strongest claims and thus as providing the strongest foundation for the lawsuit. Because they have the strongest claims it is fair that they would receive the largest awards. Also awards to named class members and others assisting in the prosecution of a class action are generally significantly larger than awards to unnamed class members to compensate the first group for the time and risks taken by involving themselves in the litigation. For these reasons, the Court concludes that the awards to the named parties are not excessive.
As a final factor supporting the damages provision of the consent decree, the Court notes that the optout provision allowed any member of the class to preserve an individual claim. Notice was provided to all class members by mail and by publication of the effects of the suit, and information about the claims process was readily available from the Clerk of the Court and from other sources. Any individual who believed that the damages he or she is likely to receive under the decree are inadequate could optout and file an individual suit. The decree provides that the statute of limitations period for such claims would be tolled during the pendency of the class action suit. This protection, as well as the other factors described above, leads the Court to conclude that the provisions for monetary relief in the decree are adequate, fair, and reasonable.
The Value of the Injunction
With all respect, my colleagues in appropriately brush off a highly significant aspect of the record on which the district court explicitly relied in evaluating the value of this settlement, starting with the informed Declaration of Jesse Jackson who vouched in strong terms for its benefit to the affected employees. Here verbatim is what Reverend Jackson offered the district court in this regard:
In June 1998, I was asked if I would review the racial issues pertaining to The Boeing Company (“Boeing”) and, if I felt there was a problem, if I could offer some assistance. Before agreeing to do so, I reviewed a substantial number of documents regarding promotional opportunities for African-Americans at Boeing. I met many of the named plaintiffs in the class action to hear, first *491hand, what they perceive to be the roadblocks to equal opportunity at Boeing. I was impressed by the named plaintiffs, their candor, and their ability to articulate what they perceive to be the hurdles to equal opportunity. In conversations with them it was clear they were pursuing this litigation not for monetary gain for themselves or to receive a substantial monetary award for the class or particular individuals with grievous discriminatory claims but, instead, they were committed to trying to find a mechanism which would over time permit qualified minorities to have the jobs they deserved but only after full and fair competition with non-minorities. They were not looking for freebies but simply a fair opportunity to be considered and to be promoted on their own merits. After listening to these named plaintiffs, and after discussing the matter with McKAY HUFFINGTON HARRELL & DESPER (“Plaintiffs’ Class Counsel”), I agreed to participate in the process. I contacted senior management at Boeing and scheduled several meetings with them. I personally met with Phil Condit and other senior management officials. We openly and candidly discussed the racial issues at Boeing and they provided materials that I requested so that I could review the significance and magnitude of the issues at Boeing. After these meetings, I met with and worked with Plaintiffs’ Class Counsel and the named plaintiffs in defining the goals of this litigation, what could be achieved and how to achieve it. I have not been paid any compensation [for] my services and advice, nor have the organizations with which I am affiliated been paid anything for my services, and there have been no promises of. nor is there an expectation of payment in the future either for past or future services regarding this litigation or Boeing.
The proposed settlement in the Boeing litigation is, in my view, one that should make Boeing and its African-American workers proud. It not only provides significant monetary relief, relative to the potential recovery for the class in this case, but advances the equal opportunity cause to a new and somewhat novel level. One might say that the proposed consent decree is even bold. To my knowledge no other employer has agreed to provide to its African-American employees, free of charge for three years, an outside, unaffiliated law firm with African-American partners to review, assist, and advocate employment opportunities for African-Americans. I know from my personal experience that there is no language in a policy, no committee, or any other of the traditional kinds of equitable relief in employment consent decrees that can even remotely come close to the efficacy of direct involvement by an independent advocate on behalf of a disenfranchised worker. Having a lawyer available to assist an African-American worker to understand the employment process, the complaint process, and to challenge a failed promotional bid will, in my opinion, significantly increase the chances that that African-American worker, and others like him or her, will be fairly promoted in the future. This consent decree may well become the model or template for future consent decrees. Providing a truly independent advocate to champion the cause for African-American minority workers — as this consent decree does — may become the benchmark by which future employment class settlements are measured. I applaud Boeing for having the courage to provide such an advocate for its workers and Plaintiffs’ Class Counsel for taking *492on the obligation and the significant financial risks that it entails.
As I said before, I've been involved in several race-related lawsuits including, but not limited to Texaco. In those, I worked with the African-American workers, management teams, and the lawyers to resolve equal opportunity issues. There, as in this case, I facilitated the negotiation process and the settlements. From that experience, and my general experience in working with businesses on racial issues, I can say without qualification that the proposed settlement for the Boeing litigation is good for minority workers and I urge the Court to approve it.
(Emphasis added).
The Reverend Jackson’s opinion is just the beginning. An impressive array of experts also submitted declarations in favor of this settlement, experts simply unnamed and ignored by the majority.
Dr. Larry Davis, Ph.D., a chaired professor of racial and ethnic diversity at Washington University in St. Louis, Missouri, said about the proposed consent decree, and its forward-looking aspects:
This is truly a bold approach and, in my experience, would do more in the long run to advance the career opportunities of individual African-Americans at Boeing than any added verbiage to existing programs, than simply increasing their existing wages or salaries, or even giving them an automatic one-time promotion now. It provides a powerful enforcement device and information resource that is critically absent in most corporate environments. I would not be surprised if this aspect of the Boeing Consent Decree — that provides for a disinterested outside law firm to assist the employer’s workers at no charge— becomes the role model or template for further employment discrimination cases. I encourage the Court to give favorable consideration to this component of the Proposed Consent Decree.
(Emphasis added).
Dr. A1 Black, Ph.D., a senior lecturer of sociology at the University of Washington agreed:
This seems to be a novel approach to a historical problem — affirmative plans designed to defeat discrimination without an active enforcement division or monitoring group giving it real teeth. If such “monitor” provides free legal service to its constituents, reviews their employment history and applications for job opportunities, assists them with job applications, and investigates problems should opportunities become unlawfully denied, then tremendous results can occur. This can be a powerful enforcement vehicle and a tremendous opportunity for African-American employees at Boeing. Such approach can have the effect of establishing real meaning to the affirmative action rhetoric commonly regurgitated by large corporations and not monitored or enforced.
Traditional affirmative action programs have established the right framework with which to begin building processes that equalize employment conditions. The success of these programs is largely dependent upon how they are implemented and monitored. I believe that a dedicated outside monitoring source would further tremendously the success of these types of programs.
Gary Smith, a senior partner with the Ivy Planning Group, a management consulting firm serving such clients as IBM, Morgan Stanley, Xerox, the U.S. Postal Service, and Chase Manhattan Bank saw this settlement as did his professional counterparts:
*493Based On My Professional Judgement, There Is Incredible Value In The Creation Of A System Whereby Employees Have Access To Independent Counsel, Willing To Devote Their Time And Resources To Measuring And Monitoring The Success Of The Employers’ Commitment To Diversity Issues, And Who Are Available To Gather Career Related Information For The Worker, To Assist In The Maze Of Varying Application Or Promotion Criteria, And To Confront The Employer If The Worker’s Job/Promotion Application Is Wrongly Rejected.
The success of any diversity initiative is dependent on how much the employer polices itself. In today’s corporate environment, effective diversity efforts are being moved beyond H.R. organizations and made the responsibility of line managers. Human Resources and EEO resources are never sufficient. The management team will always choose to devote its limited resources and money to activities that more directly generate income. Putting these two phenomenon together means that even the best affirmative action plan cannot achieve optimal results. However, if an organization is willing to allow an outside firm to assist in these efforts, incredible value is bestowed upon the employee, with bottom line improvements being passed on to the organization.
Kerby Collins, the Internal Civil Rights Manager for the Washington State Department of Transportation was of a similar mind:
When and where internal complaint procedures and adequate monitoring responsibilities fail, as claimed by the African-American employees at The Boeing Company, I believe the most productive and powerful tool available to complainants, would be access to independent legal counsel charged with the responsibility of monitoring the organization’s compliance with the law, or in this case, a Consent Decree. While many organizations may not welcome this type of “policing” function, if organizations are willing to demonstrate this commitment, in the long run it will result in improved organizational behavior and in turn, profitability.
How my colleagues can choose to dismiss the evidence regarding the importance and real value of the injunctive relief against future job discrimination is beyond me. The majority seems simply — and inappropriately — to disbelieve class counsel when they say that “Boeing will also pay $750,000 to retain class counsel for three years after the settlement becomes final to provide free legal assistance to African-American employees seeking career advancement at Boeing.” There is absolutely nothing in this record, I repeat, nothing to suggest that we should not take class counsel at their word about their promises of free legal assistance to these employees, assistance paid for by Boeing.
My colleagues express similar concern about attaching value to what they dismiss as “changes already ... implemented,” as though matters in the works somehow do not count, or have no value. Yet, they overlook why the changes which they discount came to pass. To quote Boeing in its brief to this court:
Objectors also attack certain elements of the equitable relief as “illusory” because they allegedly were not brought about through the class action litigation. This statement, however, is factually unfounded. The equitable relief provided by the Consent Decree was carefully tailored to address the concerns raised by plaintiffs in the two Seattle lawsuits, which also were raised by many African-American employees during the pe*494riod of threatened litigation that led up to those lawsuits. Thus, while some of the equitable relief may have been “in the works” prior to the effective date of the Consent Decree, that relief was clearly the result of the pressure brought to bear on Boeing by the lawsuits. Some background of the events leading up to the litigations helps illustrate this point.
In late 1996 and early 1997, two groups of African-American employees from Boeing’s Everett, Washington and Auburn, Washington facilities came forward and expressed concerns regarding alleged race discrimination. Boeing agreed to interview each employee who raised a complaint, consider common issues raised, and take appropriate action to address the concerns. These two groups grew to approximately 165 employees, and a number of those employees are now included in the Seattle Class Action as HR’s.
The concerns — and litigation threats — expressed by the two groups centered around perceived unfairness in promotions, especially: (1) the ERT (“Employee Request for Transfer”) system used by Boeing as an element of selecting persons for promotions within the hourly ranks; (2) perceived unfairness in the selection of entry-level managers; and (3) perceived ineffectiveness of the EEO investigation and corrective action function. While Boeing did not believe that these systems were discriminatory, the company nevertheless began to develop improved processes in response to these concerns and threats of litigation. Equitable relief to address these concerns — as well as additional relief addressing other issues — is now found in the Consent Decree. When a defendant takes voluntary action to address claims raised by plaintiffs in litigation, its actions are hardly considered “illusory” — under appropriate circumstances plaintiffs in such cases may be considered “prevailing parties” entitled to recover attorneys’ fees. See, e.g., Stivers v. Pierce, 71 F.3d 732, 751 (9th Cir.1995) (citing Farrar v. Hobby, 506 U.S. 103, 113 S.Ct. 566, 121 L.Ed.2d 494 (1992)).
Attorneys’ Fees
My colleagues’ problems with the attorneys’ fees stem from getting lost in unfamiliar trees and a consequent failure to see the forest for what it is. This settlement promises clear future value to African-Americans working for one of our nation’s largest and most important employers. Measured against this value, much of which is intangible, the modest attorneys’ fees agreement appears to me on the record to be quite appropriate. As I have pointed out, it is a stand alone aspect of the settlement. In dollar terms, it does not subtract from the value of the settlement, which is only a small part of what the settlement accomplishes. The merits settlement itself was (1) free of collusion and chicanery, (2) generous in its opt-out provisions, (3) fair in its monetary provisions, (4) forthcoming in its look to the future, and (5) vigorously litigated. Nevertheless, the attorneys’ fees issue ends up as the proverbial tail wagging the dog to death, even though the dog is not a dog at all, but a viable solution to a serious problem demanding prompt resolution. If the settlement itself were truly suspicious and indicative of a betrayal, then a different approach might be in order; but if the settlement stands, in my view, so do these fees. It may not match up perfectly with other methods of measuring whether fees are appropriate, but no matter whether fees here are low or high, they do not subtract from the relief obtained by the plaintiffs. My colleagues say that the *495method used in this ease to determine attorneys’ fees “allows too much leeway for lawyers representing a class to spurn a fair, adequate and reasonable settlement for their clients in favor of inflated attorneys’ fees.” This problem is not part of this case, and I do not see how something that might have happened but did not must torpedo this hardbargained positive outcome. In the end, this case has been decided based on possibilities, not realities.
Conclusion
The standard of review we are bound to employ is highly differential, as it should be. As the majority admits, “We have repeatedly stated that the decision to approve or reject a settlement is committed to the sound discretion of the trial judge because he is exposed to the litigants, and their strategies, positions, and proof. Hanlon, 150 F.3d at 1026 (citation and internal quotation marks omitted). Accordingly, a district court’s final determination to approve the settlement should be reversed ‘only upon a strong showing that the district court’s decision was a clear abuse of discretion’ ...” Hanlon, 150 F.3d at 1026 (quoting In re Pac. Enters. Sec. Lit., 47 F.3d 373, 377 (9th Cir.1995)). Here, not only do the majority fail to adhere to this deferential standard, adopting instead a standard of “somewhat uneasy with the settlement as a whole”; but in my view, they do so in a case where the district court’s approval of the settlement and of the attorneys’ fees was clearly an appropriate exercise of discretion. The district court judge responsible for this case is highly experienced, capable, and astute, one over whose eyes no one pulls the wool. It is a rare settlement that will delight all parties, but this settlement has much to say for it. Accordingly, I dissent from a decision that will have the effect of unnecessarily delaying full implementation of this efficacious solution for four years— if not more — from the date the district court found it to be appropriate.