ORDER ON MOTION FOR FINAL APPROVAL OF CLASS SETTLEMENT
This is a class action lawsuit over arrestee strip searches at the York County Jail. On August 1, 2005, I held a final fairness hearing on a proposed settlement of the lawsuit. I now approve the notice sent to the class. I also approve the fairness, reasonableness and adequacy of the settlement in all respects but one: I cannot endorse a local government paying twice as much to women as to men, because the difference in treatment does not pass constitutional muster. Although I find the settlement is in most respects a good one, the parties have not established that a gender-based difference in treatment is substantially related to an important governmental objective , the constitutional standard that governs. To that extent only, I disapprove the settlement.
I. PROCEDURAL BACKGROUND
On October 15, 2002, individuals who had been arrested at the York County Jail ("Jail") filed this lawsuit under 42 U.S.C. § 1983 and asked to represent a class of plaintiffs. They challenged the Jail's practice of requiring that every arrestee to be held in a housing area disrobe completely and submit to a corrections officer's visual inspection while changing over into a jail uniform. The plaintiffs alleged that this practice violated the Fourth Amendment because it was a policy or custom of "strip searching" arrestees without individualized suspicion.
York County responded, maintaining (as it does to this day) that its policy was constitutional because the officers were looking for contraband in clothing and were not intentionally viewing arrestees' naked bodies.
Both parties conducted extensive discovery. York County responded to three sets of document requests and interrogatories and one request for admissions, and deposed seventeen class members. The parties deposed a total of thirty-three individuals. The plaintiffs responded to a set of interrogatories and document requests for each of the three class representatives, and responded to two additional sets of interrogatories and one additional set of document requests. Both parties retained jail experts and exchanged their reports. Pls.' Mot. for Final Approval of Class Settlement ("Mot. for Final Approval") at 8-9 (Docket Item 133); id. Ex. N, Aff. of Howard Friedman, ¶ 22.
Discovery was disputed from the beginning of the case, when the plaintiffs had difficulty obtaining electronic booking data from the Jail. The Jail initially stated that it did not have electronic booking data. The plaintiffs later discovered that the Jail did have electronic data, but that it was not in a format that the plaintiffs could access. As directed by Magistrate Judge Cohen, Report of Hearing and Order Re: Discovery Dispute at 2 (Docket Item 40), the plaintiffs' expert worked with the developer of the Jail's software and representatives from the Jail and was able to extract the data and convert it into a database. The plaintiffs used this database to find and contact potential class members. See Mot. for Final Approval at 6-7 (describing this discovery dispute).
On December 18, 2003, I certified a class under Federal Rule of Civil Procedure 23(b)(3). The class includes:
All people strip-searched at the York County Jail after October 14, 1996, under a policy or custom of conducting strip-searches without evaluating individualized reasonable suspicion:
(1) while waiting for bail to be set or for a first court appearance after being arrested on charges that did not involve a weapon or drugs or a violent felony; or
(2) while waiting for a first court appearance after being arrested on a default or other warrant that did not involve a weapon or drugs or a violent felony.
Nilsen v. York County, 219 F.R.D. 19, 20 (D. Me. 2003). York County appealed and the First Circuit affirmed the decision certifying the class. See Tardiff v. Knox County, 365 F.3d 1 (1st Cir. 2004).
In January of 2004, the plaintiffs sent York County a settlement proposal. York County did not want to discuss mediation until the First Circuit ruled on the class certification. After the First Circuit's decision affirming class certification, Judge Cohen issued a scheduling order that required the parties to conduct mediation by the end of September 2004. Report of Scheduling Conference and Revised Scheduling Order at 2 (Docket Item 79). The parties began settlement discussions with a mediator in late September, 2004, and the plaintiffs filed a notice of settlement on November 29, 2004.
On January 24, 2005, I held a preliminary hearing on the proposed settlement. Following that h earing, I approved class-wide notice of the proposed settlement after requiring the parties to provide class members with a second opportunity to opt out of the class under Federal Rule of Civil Procedure 23(e)(3). I held a final fairness hearing on August 1, 2005.
II. SUFFICIENCY OF NOTICE
Federal Rule of Civil Procedure 23(e)(1)(B) requires that the court "direct notice in a reasonable manner to all class members who would be bound by a proposed settlement." Notice meets Rule 23(e) and due process requirements if it is reasonably calculated to reach the class members and inform them of the existence of and the opportunity to object to the settlement. See 4 Alba Conte & Herbert B. Newberg, Newberg on Class Actions § 11:53, at 164-65 (4th ed. 2002); see also Wal-Mart Stores, Inc., v. Visa U.S.A., Inc., 396 F.3d 96, 114 (2d Cir. 2005).
The notice program here consisted of mailed notice, print and electronic publication notice and notice posted at the Jail. There were both a long-form and a short-form notice. There were two versions of the short-form notice. One was published in two Maine newspapers, the York County Journal Tribune and the Portland Press Herald, and the other was posted at the Jail's booking area and in each of the Jail's housing units. These short forms summarized the litigation and the terms of the settlement and described the process for obtaining a claim form.
The Jail also kept copies of the long-form notice and claim form for any prisoner who requested them. See Mot. for Final Approval, Ex. I, Aff. of Myong J. Joun ("Joun. Aff."), ¶¶ 19-24 (describing the notice program). The long form described the nature of the litigation, the class definition, the general terms of the settlement, the class members' legal rights and the process for filing a claim, and included a toll free number and a dedicated website for class members who wanted additional information. The plaintiffs' lawyers mailed this long-form notice and the claim form to class members and to anyone who requested the claim form through the toll-free number or the claims administrator. The forms were also available on the website.
The plaintiffs' lawyers took several steps to assure that the direct mail notice reached the class members. They initially obtained the class members' addresses from the Jail's electronic booking data. Mot. for Final Approval at 7. Before mailing the notice of class certification, the claims administrator updated these addresses with the change of address database maintained by the United States Postal Service. Pls.' Supplemental Consent Mem. in Support of Consent Mot. for Preliminary Approval of Class Settlement ("Supplemental Consent Mot.") at 13 (Docket Item 116); id. Ex. J, Aff. of Richard W. Simmons, ¶ 3. The post office returned as undeliverable forty-two percent of the envelopes containing notice of class certification. The claims administrator then ran those addresses through an online LEXIS database designed to provide updated addresses. Only thirty percent of the direct mail notices of settlement ultimately were returned as undeliverable. Mot. for Final Approval, Ex. D, Aff. of Estelle McLean, ¶¶ 3-7.
This settlement also received substantial media coverage in Maine and some additional coverage in New Hampshire and Massachusetts. The plaintiffs' lawyers issued a number of press releases about the settlement to various print and broadcast outlets and held a live press conference on the settlement. Ten days before the deadline for submitting claim forms, the plaintiffs' lawyers issued another press release about the settlement and the upcoming deadline. See Supplemental Consent Mem., Ex. I, Decl. of David G. Webbert, Esq., ¶¶ 11-12; Pls.' Second Supplemental Consent Mem. in Support of Mot. for Final Approval at 2 (Docket Item 141).
I find that the content of both the long-form and the short-form notices sufficiently informed the class members of the proposed settlement and their options of filing a claim, seeking exclusion or objecting to the settlement. I find that all forms of notice could be understood by the average class member. See 4 Conte & Newberg, supra, § 11:53, at 167. The distribution was excellent, satisfying both Rule 23(e) and due process. Although mailed notice alone might have been sufficient, see, e.g., Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 175 (1974); 7B Charles Alan Wright, et. al., Federal Practice & Procedure § 1797.6, at 200, the plaintiffs' lawyers made a considerable and highly successful effort beyond the mailings so as to reach as many class members as possible. They carefully chose the newspapers for publication notice based on the high concentration of class members in York County (78.5% of the class) and Cumberland County (14.3% of the class). Supplemental Consent Mem., Ex. H, Aff. of Myong J. Joun, ¶ 9. Notice provided at the Jail likely reached some class members who did not receive the mailed notice because they no longer lived at the address known to the plaintiffs' lawyers. Other class members who did not receive the mailed notice nevertheless may have learned of the settlement from the media and received more information from the toll-free number and the website. After the settlement was publicized, the claims administrator received over 316 telephone calls and more than 1,936 visits (with 685 unique network addresses) to the dedicated website. Mot. for Final Approval at 26; id. Ex. D, Aff. of Estelle McLean ¶¶ 8-12. Moreover, the efforts by the plaintiffs' lawyers to publicize the claim deadline appear to have reached a significant number of class members. By June 10, 2005, the claims administrator estimated that 12-13% of the class would participate in the settlement, based on the number of claim forms that had been filed and were expected to be approved as well as prior experience. Seven weeks later, after the additional press release about the claim deadline, a large number of additional class members filed claims and the claims administrator estimated a 17.78% participation rate. Decl. of Myong J. Joun dated July 27, 2005, ¶ 4 (Docket Item 143); Joun Aff., ¶ 16. I conclude that the content of the notice and the extensive and well-planned means of disseminating the notice satisfied Rule 23(e) and due process requirements.
III. FAIRNESS, REASONABLENESS AND ADEQUACY OF THE SETTLEMENT
Under Rule 23(e), I can approve the proposed settlement only if it is fair, reasonable and adequate. Fed. R. Civ. P. 23(e)(1)(C); City P'ship Co. v. Atlantic Acquisition Ltd. P'ship, 100 F.3d 1041, 1043 (1st Cir. 1996); see also Wal-Mart Stores, 396 F.3d at 116. In evaluating the settlement, I consider the following factors: (1) a comparison of the proposed settlement with the likely outcome of litigation; (2) the stage at which the matter settled and the amount of discovery completed; (3) the reaction of the class to the settlement, including the number of objectors and the nature of the objections; (4) the quality of the plaintiffs' lawyers; (5) the conduct of the negotiations; and (6) the prospects of the case, including risk, complexity, expense, and duration. See In re Compact Disc Minimum Advertised Price Antitrust Litig., 216 F.R.D. 197, 206 (D. Me. 2003) (noting that the First Circuit has not adopted a specific test to evaluate class action settlements but that other circuits consider these factors); see also Wal-Mart Stores, 396 F.3d at 117 (considering similar factors); 4 Conte & Newberg, supra, § 11:43, at 120-21. I also evaluate the fairness of the proposed distribution of the settlement funds among the class members. There is a presumption in favor of the settlement if the parties negotiated the settlement at arms-length after conducting meaningful discovery. City P'ship Co., 100 F.3d at 1043; 4 Conte & Newberg, supra, § 11:41, at 90; see also Wal-Mart Stores, 396 F.3d at 116.
(A) Value of the Settlement
I first compare the value of the settlement to what the plaintiffs could recover after a successful trial and appeal, discounting that potential recovery by the risk, delay, and expense of continued litigation. The value of the settlement is a monetary award of $3.3 million (accruing interest), to be distributed to the class members after the payment of attorney fees, costs, claims administration expenses, and incentive payments to deponents and class representatives; and a requirement that York County maintain a written policy prohibiting the challenged strip searches.
The monetary award compares favorably to what the plaintiffs could recover at trial. York County has vigorously contested liability factually from the outset, maintaining that Jail officers avoided looking at arrestees during the changeover process. If the plaintiffs nevertheless succeed on liability at a trial, it will still be difficult for them to prove and recover damages. See, e.g., Mot. for Final Approval, Ex. B, Decl. of Charles Harvey, Esq. ("Harvey Decl."), ¶ 10 (noting his belief that the class members may receive nothing or only very small or nominal damage awards from a jury). Because of the individualized nature of the damages, it is unclear how damages would even be awarded following a liability determination. In affirming class certification, the First Circuit noted that the court could use a formula to distribute damages, the parties could agree to modest uniform damages with a special master evaluating any more serious claims, class members could pursue damage claims in a separate lawsuit or the parties could settle after a jury determined liability. Tardiff, 365 F.3d at 6-7. Any individualized determination could be problematic because plaintiffs may be hesitant to testify fully and openly about any mental, emotional or other harm that they experienced as a result of the search, even if they were willing to testify at trial that the strip search occurred. Collecting any judgment larger than the settlement would also be difficult. This settlement is funded largely by insurance, and the amount available declines as York County spends more on defense costs. See Harvey Decl. ¶¶ 7-8, 15 (indicating that because of insurance limitations the proposed $3.3 million award is the maximum amount that the plaintiffs could have obtained in a settlement). It is challenging to recover from a governmental entity amounts that are not covered by insurance.
There is also value in the settlement's requirement that York County maintain a written policy prohibiting the challenged searches. York County claims that the written policy does not reflect a change in the Jail's practice but instead codifies the existing practice. It says that the litigation was only one factor in York County's decision to change its written policy because it was already in the process of reviewing and revising many of its policies and procedures. See Def.'s Response to Pls.' Mot. for Final Approval of Class Settlement ("Def.'s Response") at 4 (Docket Item 136). But regardless of whether York County changed its practice and the extent to which the litigation caused York County to create the written policy, there is still value in requiring York County to maintain the written policy so that the allegedly unconstitutional searches do not occur in the future. This requirement is the equivalent of injunctive relief that the plaintiffs could recover following a successful liability determination at trial. See Second Am. Compl. at 8 (Docket Item 24) (requesting injunctive relief).1 Viewed as a whole, the value of the settlement compares very favorably to the plaintiffs' likely recovery at trial.
(B) Class Reaction
The reaction to the settlement by the approximately 7,500 class members has been favorable. Only one class member filed written objections, another objected orally to the settlement at the final fairness hearing, and only five class members opted out of the settlement. Because of the nature of this litigation, however, I do not consider the relatively low number of objections and opt-outs to be significant. Class members may have been embarrassed to come forward and admit that they were arrested and strip searched. It is also unlikely that class members who felt uncomfortable identifying themselves as arrestees would opt out in order to litigate their claims individually and publicly. The generally favorable response therefore is not a significant factor in settlement approval.
(C) Stage of the Litigation
Two years of litigation and substantial discovery provided the parties information about the strengths and weaknesses of their positions on liability and damages before the settlement negotiations began. There were several rounds of interrogatories and document requests and both parties retained experts. The plaintiffs' lawyers communicated with over two hundred potential class members. Depositions of class members and correctional officers highlighted factual disputes over whether the Jail officers afforded the arrestees privacy while they changed over into the jail uniform, and whether the officers searched only the arrestees' clothes or whether they deliberately viewed the arrestees' naked bodies during this changeover. Mot. for Final Approval at 8-9 (detailing the discovery conducted by both parties). In response to a court order, York County provided the plaintiffs with information about its relevant insurance agreements during the class period. See Consent Confidentiality Order (Docket Item 96). The parties also had the benefit of the First Circuit's decision in Tardiff affirming class certification. Thus, this is a settlement based upon comprehensive information.
(D) Quality of Counsel and Representation
The plaintiffs' lawyers were highly qualified and experienced. Both of the lead counsel have litigated a number of civil rights claims under 42 U.S.C. § 1983, and one has litigated cases specifically alleging unconstitutional strip searches (including some class actions). See Mot. for Final Approval, Ex. A, Decl. of Howard Friedman ("Friedman Decl."), ¶¶ 6-7; id. Ex. P, Decl. of David G. Webbert ¶ 4. They diligently pursued discovery, as demonstrated by their efforts to obtain electronic booking data from the Jail to help identify and contact potential class members and the significant amount of discovery conducted prior to settlement. The lawyers were effective and thorough in their written and oral arguments and other communications with the Court. They advocated zealously and capably for the plaintiffs.
(E) Conduct of Negotiations
The settlement negotiations were extensive and were conducted at arm's length, with no sign of collusion. York County was initially reluctant to engage in settlement discussions. Both parties remained far apart after two lengthy mediation sessions before a distinguished local mediator. They had to sort through complex issues regarding York County's insurance coverage before they could reach an agreement. See Harvey Decl. ¶¶ 3-4, 7-9, 11-12. Both the mediator and York County's "regular outside counsel" report that the monetary award represents the maximum amount that the plaintiffs could have obtained in a settlement, and the outside counsel states that York County would have abandoned settlement efforts if the plaintiffs had rejected the $3.3 million offer. See Harvey Decl. ¶ 15; Mot. for Final Approval, Ex. C, Decl. of Gene R. Libby, Esq., ¶¶ 10-11.
I also note that the parties informed me at the final fairness hearing that there were no side agreements made in connection with the proposed settlement. See Fed. R. Civ. P. 23(e)(2) (requiring the parties to identify any such agreements).
(F) The Risk, Complexity, Expense and Duration of Continued Litigation
To succeed at trial, the plaintiffs would have to overcome the significant factual dispute as to whether the York County corrections officers actually viewed the arrestees' naked bodies while they changed into the jail uniform. York County would likely call a number of correctional officers to testify that they did not behave in that way, and several higher-ranking officers to testify about the Jail's training and formal policies. The plaintiffs would have called at least twenty-five class members to testify that the officers did view their naked bodies during the changeover. The plaintiffs estimate that a total of fifty to seventy witnesses would testify at trial. A trial would be lengthy and expensive, and ultimately unsuccessful for the plaintiffs if the jury credited the testimony of corrections officers over that of arrestees, always a serious risk in a trial of this sort.
Following a liability determination, the plaintiffs would still have to recover damages. At that point the defendant could move to decertify the class on the issue of damages. See Tardiff, 365 F.3d at 7 (noting this possibility). Many plaintiffs likely would not pursue individual damage claims given the cost, the effort, and the need to testify publicly about the fact that they were arrested and the harm that they experienced from the search. Moreover, eve n if the parties settled after a determination of liability, the costs of litigating the trial would have reduced the available funds (because of insurance limits) for any post-trial settlement of the damage claims. Harvey Decl. ¶ 7.
Finally, if the plaintiffs did succeed at trial, York County probably would appeal, adding more cost and delay. See Mot. for Final Approval at 39-40; Def.'s Response at 5.
(G) Allocation of the Settlement
(1) Incentive Payments to Named Class Representatives and Deponents
The plaintiffs' lawyers propose incentive payments of $6,500 to Michele Nilsen, the first class representative; $5,500 to class representative Charles Neville; $5,000 to class representative Michael Goodrich; and $500 to each class member who was deposed.2 These proposed payments are fair, reasonable and adequate to compensate the class representatives and deponents for their valuable efforts on behalf of the class.
The named class representatives sacrificed their privacy to vindicate the privacy rights of the class members. Other class members refused to act as class representatives because they wanted to protect their privacy and did not want to reveal to the public that they had been arrested and strip searched. Friedman Decl. ¶ 15. This lawsuit would not have been filed but for class representative Michele Nilsen, who contacted one of the plaintiffs' lawyers when she read a newspaper article about a case involving similar strip searches that occurred in Massachusetts. Nilsen estimated that she spent over sixty hours on this case. She answered interrogatories and responded to document requests. York County also deposed her, a process that took three hours and ten minutes (the longest of the class representatives). Her former boyfriend was also deposed. Nilsen stated that she felt that her "personal life was being examined under a microscope." Nilsen traveled from Massachusetts to attend the first mediation session. As the first named plaintiff, Nilsen was frequently mentioned in the press, although she chose not to speak with the media. Nilsen provided the Court with a copy of a newspaper article she found online (conducting a Google search for her name) that contained inaccurate and embarrassing information about her. In the article, a York County official allegedly stated that he thought that Nilsen was arrested for driving while intoxicated. Nilsen actually was arrested for driving with a suspended license. See Mot. for Final Approval at 47-48; id. Ex. E, Aff. of Michele Nilsen.
Class representative Charles Neville expended a significant amount of time and effort. He estimated that he spent thirty hours on this case. He responded to a set of interrogatories and a request for production of documents. York County also deposed him, an event that took one hour and forty-five minutes. Neville read and commented on a draft of the settlement documents. Given the significant amount of publicity the case received in Maine, it is likely that people in Neville's community (Neville lives in York County) would recognize him as a class representative and therefore realize that he had been arrested and searched. He stated that friends, acquaintances, and business associates asked him about his participation in the case. Neville is self-employed and fears that the publicity will hurt his business. See id. at 48-49; id. Ex. F, Aff. of Charles Neville.
Like Neville, class representative Michael Goodrich lives in York County and is likely to be recognized by people in his community as a class representative. He answered a set of interrogatories and responded to a request for a production of documents. York County deposed him for one hour and fifteen minutes. He also believed that he was retaliated against for his participation in this lawsuit. Specifically, he was incarcerated at York County Jail after he agreed to act as a class representative. When he needed to be strip searched at the Jail during that time, he believed that the correctional officers conducted a particularly thorough and embarrassing strip search due to his participation in this case. See id. at 49- 50; id. Ex. G, Aff. of Michael Goodrich.
The seventeen class members who were deposed deserve the $500 incentive payment to recognize the time and effort they expended in preparing for and attending the deposition. Their participation was valuable in advancing discovery in this case and helping both sides evaluate the merits of their respective positions. See id. at 50.
(2) Lack of Individualized Recovery for Class Members
I conclude that it is reasonable for the settlement to provide each class member with a modest uniform payment rather than individualized recovery based on the specific harm to each class member. Individualized recovery would require a much more complicated claims process for the class members, involving a more detailed claim form and the possible submission of medical records, a hearing on damages and a psychiatric evaluation. This more involved claims process might discourage class members from filing claims, particularly those class members who wish to maintain their privacy. See Friedman Decl. ¶ 15 (noting the importance of anonymity to class members in strip search cases and the value of allowing them to recover for a violation of their privacy without giving up their privacy in the process). Evaluating the individual harm to each class member would also incur more administrative costs, potentially reducing the amount of money available for the class members, see Harvey Decl. ¶ 16, and could significantly delay payment.
(3) Lack of Additional Payment for Multiple Searches
The proposed settlement provides only one payment to each class member regardless of the number of searches the class member experienced. One class member objected orally at the preliminary hearing and submitted a written objection to this lack of additional payment for multiple searches. He argued that it is unfair not to compensate class members for multiple searches because successive searches may be as, or more, humiliating to the arrestee. Objection to Mot. for Final Approval at 1 (Docket Item 135).
It is true that the lack of compensation for multiple searches does fail to recognize that there may be at least some additional harm from multiple searches. I also find this method of distribution troubling to the extent that it may reflect an unfair bias against repeat arrestees.3 But at a trial jurors might legitimately fail to compensate significantly for additional searches on the premise that most of the harm resulted from the shock of the first strip search, and that class members were harmed less by successive searches because they were mentally prepared for the later searches. See Bochnak Aff. ¶ 8 (noting that mock jurors believed that a person who had been previously arrested and strip searched would know what to expect if arrested and searched again and would not be as bothered by the search). Charles Neville, a class representative searched twice during the class period, confirmed this belief. He stated that although he was very upset by and remembered the details of the first search, he felt more prepared for and could not remember the details of the second search. Mot. for Final Approval, Ex. F, Aff. of Charles Neville, ¶ 6.
This proposed uniform distribution is significantly easier and less costly to administer than a system that would require determining how many searches within the class definition each class member experienced. Thirty percent of the class members were booked multiple times within the class period. It is difficult to ascertain how many of these bookings involved searches within the class definition because the Jail's records indicate only indirectly whether a booking involved the arrestee changing over into a jail uniform-just one criterion for qualifying. Supplemental Consent Mem., Ex. H, Aff. of Myong J. Joun, ¶¶ 3-4 & Chart 1. The plaintiffs' lawyers spent three hours carefully reviewing the Jail records for the objector to determine how many of his bookings fit the class definition. This review revealed that only one of the objector's five bookings during the class period involved a search that fit the class definition. See Decl. of Howard Friedman, Esq. (Docket Item 139). An appeal of the initial determination that a booking did not fit the class definition could also require much effort and expense. If the settlement provided additional compensation for multiple searches, the plaintiffs' lawyers would need to have someone spend a substantial amount of time and effort evaluating which bookings met the class definition. The plaintiffs' lawyers are currently agreeing to cap their administrative expenses, but these expenses could increase significantly if they had to undertake this review for all of the class members who were booked more than once during the class period. Pls.' Supplemental Mem. in Support of Mot. for Att'y's Fees and Litig. Expenses at 16, 17 n.17 (Docket Item 134). Any increase in administrative expenses would reduce the recovery to the class members because the $3.3 million paid by York County covers all costs associated with this class action, including the payment to class members, attorney fees, and administrative expenses.
So despite some concerns, I conclude that the failure to compensate for additional searches is justifiable.4 My role is not to dictate the terms of settlement, but only to reject the settlement if it is not fair, reasonable and adequate. Evans v. Jeff D., 475 U.S. 717, 726-27 (1986) ("Rule 23(e) wisely requires court approval of the terms of any settlement of a class action, but the power to approve or reject a settlement negotiated by the parties before trial does not authorize the court to require the parties to accept a settlement to which they have not agreed"); In re Warner Commc'n Sec. Litig., 798 F.2d 35, 37 (2d Cir.1986) ("[I]t is not a district judge's job to dictate the terms of a class settlement; he should approve or disapprove a proposed agreement as it is placed before him and should not take it upon himself to modify its terms"). I am ultimately persuaded by the considerable administrative costs that compensation for multiple searches would require.5 I do not base this conclusion on endorsing any unfair bias that jurors might harbor toward the rights of repeat arrestees. But I do consider that jurors legitimately may award less money for additional searches based on the belief that later searches are less damaging. Uniform distribution does fail to recognize that class members who were subjected to multiple searches may have experienced some additional harm from the later searches. But given the substantial administrative costs of compensating for that additional harm, I find that this distribution is reasonable.
(4) Gender-Based Distribution Formula
Under the proposed settlement, each woman would receive twice as much as each man. One objector challenged this double payment to women, claiming that it is unfair and unconstitutional. The plaintiffs' lawyers have argued that the double payment to women is justified because women have a legally-recognized privacy interest in their upper bodies. According to the plaintiffs' lawyers, this anatomical and legal difference between men and women means that each woman experienced "two legally recognized strip searches, one of [her] breasts and one of [her] lower body." See Mot. for Final Approval at 37-38. In their brief, the plaintiffs' lawyers contended that the fact that women experienced "two . . . searches" also prolonged the experience, causing women more humiliation and emotional distress. Id. at 38. But at the final fairness hearing, the plaintiffs' lawyer relied only on the legal privacy distinction (two legal injuries) to support the payment differential. He argued that if the search policy had required the arrestees to remove only their shirts, then women would be able to recover but men would not, and therefore that all women experienced two actionable searches regardless of the individual harm to each woman. As further support for the double payment, the plaintiffs' lawyers have said that juries in strip search cases typically award more to females. See id. at 38-39 (citing Harvey Decl. ¶ 18).
The parties have cited only one case where a court has approved a strip search settlement in which the female class members received double the amount paid to men. See id. at 38, citing Eddleman v. Jefferson County, Kentucky, No. 3:91CV-144, at 6-7 (W.D. Ky May 21, 1999); see also Mot. for Final Approval, Ex. I, attachment 8. There, the trial judge was initially concerned by the disparity, but ultimately approved the double payment to women, accepting the magistrate judge's reasoning that the disparity reflected the fact that juries typically award women more than men in strip search cases. Eddleman, No. 3:91CV-144-J, at 6; see also Eddleman v. Jefferson County, Kentucky, No. 3:91CV-144-J, at 27-28 (January 29, 1999) (Mot. for Final Approval, Ex. I, attachment 9) (decision by the magistrate judge recommending settlement approval) (noting that female litigants in strip search cases historically recover higher verdicts and citing some examples). In one other case, a settlement apparently was approved where women receive d 1.5 times as much as men, but the court did not address the issue. See Joun Aff., ¶ 39 & attachment 15, citing Moser v. Anderson, No. C-93-634-B (D.N.H. Aug. 1999) (Joint Stipulation of Settlement.6 A larger number of courts, however, have approved settlements with no differential in the payments to men and women.7
For a judge to approve a local government paying twice as much to women as to men, constitutional standards must be addressed. Under the equal protection clause of the Fourteenth Amendment, a gender-based distinction by government can survive only if it is substantially related to the achievement of an important governmental objective. United States v. Virginia, 518 U.S. 515, 533 (1996); Laro v. New Hampshire, 259 F.3d 1, 10 (1st Cir. 2001). Merely approximating the likely outcome of gender-based differences at trial is not an important governmental objective unless the plaintiffs provide a legitimate reason why juries in strip search cases award women more than they award men. Otherwise, replicating this disparity in jury verdicts may only perpetuate gender stereotypes if juries in strip search cases award more to women based on a stereotypical view that women are more fragile and need more protection than men. See, e.g., J.E.B. v. Alabama, 511 U.S. 127, 138-41 (1994) (noting that distinctions based on gender stereotypes reinforce prejudicial views of the relative abilities of men and women and violate the Equal Protection clause).
Does recognizing that in American culture women, but not men, have a privacy interest in their upper bodies qualify as an important governmental objective? If it does, the formula here-a double payment to women-is not substantially related to achieving that objective. The plaintiffs' lawyers have not shown how the legal violation of women's privacy interests in their upper bodies justifies a double payment to women for the strip search.8 At the York County Jail, corrections officers performed only one search of the entire body, in which each arrestee, male or female, had to remove all of his or her clothes and submit to a visual inspection. There were not two separate searches of the upper and lower bodies of the arrestees. The plaintiffs' lawyers have failed to show that any increased humiliation or emotional distress for women resulting from the additional legal violation was twice as much as that experienced by men, thereby warranting double the recovery. Indeed, there is no empirical evidence at all about different emotional distress that men and women might suffer. Individuals obviously differ in their sensitivity to physical privacy intrusions, but it is an overbroad generalization to conclude that women suffer twice as much emotional or legal injury as men do in a strip search. I conclude that the proposed gender-based allocation of this settlement does not meet constitutional standards because the differential is not substantially related to the achievement of an important government interest.
Obviously, I cannot approve as reasonable a settlement that does not pass constitutional muster. I therefore disapprove the double payment to women under the proposed settlement.
I have already addressed some of the issues raised by the two objectors. I now address the remaining issues they have raised. One objector challenged what he believed was the improper negotiation of the plaintiffs' attorneys fees prior to settlement. Mem. Regarding Fairness at 4-5 (Docket Item 140). But the plaintiffs' lawyers in this case did not negotiate their fees with the defendant prior to settlement. They negotiated for a common fund of $3.3 million, which includes payments to class members, attorney fees, costs, administrative expenses, and incentive payments to class representatives. They have filed a motion for attorney fees and expenses. Because I am disapproving the settlement over the gender-based differential, I do not yet address the attorney fees request.
The sole objector at the final fairness hearing stated that she objected at first to the amount of recovery (she thought that each class member would receive only a few hundred dollars based upon her premise that every class member would participate in the distribution of funds), but seemed to withdraw this objection based upon what the lawyers described as the actual amount that each class member would receive under the proposed settlement (about $2,800 to women; about $1,400 to men). In any event, for the reasons I have described, the amount of the overall settlement is fair, reasonable and adequate. More important to this objector was her eloquently expressed concern that the settlement did not include an admission of wrongdoing by York County. I recognize the importance to the class members of an admission of wrongdoing and the symbolic importance it would have . Nevertheless, I cannot dictate the terms of the settlement, only determine whether the settlement as proposed is fair, reasonable and equitable. I find that, given the other positive attributes of the settlement, the lack of such an admission does not make the settlement unreasonable, unfair or inadequate.
I DISAPPROVE the settlement but only to the extent that it provides for a gender-based allocation of the settlement proceeds.9
1. The settlement does not provide the declaratory relief that the plaintiffs requested in the lawsuit, a declaration that the challenged practice is unconstitutional. See id. The defendant does not admit liability, and claims that it did not employ the challenged practice.
2. The written objector objected to these incentive payments, then later withdrew his objection. Under Federal Rule of Civil Procedure 23(e)(4)(B), he needed court approval to withdraw his objection. I therefore address his objection now, just as I would assess the fairness of the incentive payments in any event.
3. Part of the offered justification is that a jury would not award additional money for multiple searches because jurors would believe that class membe rs who had been arrested more than once are less deserving because of the "pattern of lawlessness." See Mot. for Final Approval at 34-37; Supplemental Consent Mem., Ex. G, Aff. of Elizabeth Bochnak ("Bochnak Aff.") ¶ 9 (describing a study in which mock jurors stated that individuals who were arrested and strip searched more than once were "less deserving of damages than a person arrested only once"). At the final fairness hearing the plaintiffs' lawyer also stated that jurors may think that class members who experienced multiple arrests and searches were "causing the problem" themselves, presumably by the conduct that led to the arrest. Such justifications have little to do with the emotional injury actually suffered by class members.
4. I have read and considered the strip search cases cited by the plaintiffs' lawyers in which other courts have approved settlements that do not compensate for additional searches. See Joun Aff., ¶¶ 33-34, 39, 41 (citing cases) & attachments (opinions and orders from these cases); see also id. ¶ 35 (noting a case where the court preliminarily approved such a settlement). Although these cases indicate that other courts believe that the lack of compensation for additional searches is not per se unreasonable, they are of only limited use. The cited cases approved uniform distribution without discussing it, and there are a number of other cases where courts have approved greater recovery for multiple searches, see id. ¶¶ 36-38 & attachments.
5. Capping recovery at a certain numbe r of searches, perhaps three, would not save much time and effort. Most class members with repeat bookings were booked only two or three times during the class period, so it still would be necessary to review all these class members' bookings to determine how many qualified. It might also be necessary to review a large number of the bookings for class members who were booked more than three times, just to find up to three qualifying searches. The plaintiffs' lawyers, for example, reviewed all five of the objector's bookings and found only one qualifying search.
6. The plaintiffs lawyers provided a Joint Stipulation with an attached distribution formula but state that the Court endorsed this formulation in its final settlement award. See Joun Aff., ¶ 39.
7. The plaintiffs' lawyers have listed several such cases in their affidavit. Joun Aff., ¶¶ 33-35, 37-38 & attachments 5-7, 10, 14 (citing strip search cases where courts approved-without discussion of the issue-settlement awards that paid males and females equal amounts). See Maneely v. City of Newburgh, No. 01-CIV-2600 (S.D.N.Y. Mar. 29, 2005) (Order for Preliminary Approval of Settlement) (Joun Aff., attachment 7) (preliminary approval of class action settlement with no differentiation made in payments to male and female class members); Kahler v. County of Rensselaer, et. al., No. 03-cv-1324 (N.D.N.Y. Sept. 23, 2004) (Final Approval Order) (Joun Aff., attachment 5) (final approval of proposed settlement award where men and women shared equally and without regard to the number of strip searches); Doan v. Watson, No. NA 99-4-C-B/S, 2002 WL 31730917 (S.D. Ind. Dec. 4, 2002) (Order Granting Final Approval of Class Action Settlement) (Joun Aff., attachment 6) (same); Williams v. County of Los Angeles, et. al., No. CV 97-03826-CW (C.D. Cal. Dec. 6, 2002) (Order Re Settlement, Release and Dismissal of Claims in Class Action) (Joun Aff., attachment 10) (order approving settlement where men and women received the same award); Tyson v. City of New York, No. 97 Civ. 3762 (S.D.N.Y. Dec. 2000) (Stipulation of Settlement) (Joun Aff., attachment 14) (stipulation of settlement where men and women received the same award).
8. Moreover, the parties have explicitly rejected the proposition that multiple searches should result in greater recovery.
9. If the parties decide to amend the settlement to remove the cause for my disapproval, I would require a new opportunity for female class members who have filed claims to opt out. Because the class-wide settlement notice in this case already satisfied the requirements of Rule 23(e)(1)(B), as I discussed above, I would require notice of the amendment and the opt-out right only to female class members who have already filed claims, because they are the only class members who would be negatively affected by such an amendment. Cf. In re Integra Realty Res., Inc., 262 F.3d 1089, 1111 (10th Cir. 2001) (holding that the district court did not abuse its discretion by failing to notify class members of a new right to opt out because the right to opt out "merely expanded the rights of class members" and did not create "a risk that unfavorable terms would be forced upon some class members"); Manners v. Am. Gen. Life Ins. Co., No. Civ.A. 3-98-0266, 1999 WL 33581944, at *13 (M.D. Tenn. Aug. 11, 1999) (finding that notice of settlement amendments was unnecessary because the amendments enhanced the relief provided to the class members and the original notice satisfied constitutional standards); White v. Nat'l Football League, 836 F. Supp. 1458, 1468-69 (D. Minn. 1993) (directing notice of amendments to the proposed settlement only to those class members affected by the amendments).