Hartman v. Albright

UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA

_________________________________________
)
CAROLEE BRADY HARTMAN, et al., )
)
Plaintiffs, )
)
v. ) Case No. 77-cv-2019 (APM)
)
MICHAEL R. POMPEO, et al., 1 )
)
Defendants. )
_________________________________________ )
MEMORANDUM OPINION AND ORDER

I. INTRODUCTION

This case is in all respects extraordinary. Originating over forty years ago, it represents

the largest Title VII sex discrimination class action settlement in United States history. Its over

1,000 class members each received an average of $460,000—the largest per-capita recovery in a

case of its kind. Class members are women who sought employment or promotions with the

United States Information Agency, a former agency of the United States government, the relevant

components of which were incorporated into the State Department. Remarkably, the lead counsel

for the class, Bruce Fredrickson, took on the case as a 26-year-old just one year out of law school

and, now well into his sixties, has stayed on for its duration. Over the last four decades,

Mr. Fredrickson has led a team of over 120 individuals across seven law firms. In 2018, the last

of the $508 million settlement fund was distributed to class members, leaving resolution of

attorneys’ fees as the sole remaining issue.

1
Michael R. Pompeo, in his official capacity as Secretary of the United States Department of State, is substituted as
Defendant Madeleine K. Albright under Federal Rule of Civil Procedure 25(d).
 Since 1995, there have been 28 interim payments to class counsel for fees, expenses, and

interest accrued during the pendency of the case, totaling $26,570,701.19. Plaintiffs now seek an

additional $34,114,143.52, for a final total fee recovery of $75,000,000. 2 To justify this demand,

Plaintiffs primarily argue that they are entitled to a percentage of the total settlement under a

“constructive common fund” theory. Alternatively, Plaintiffs argue that an enhancement to the

lodestar is proper because the lodestar calculated for the interim fee petitions does not reflect class

counsel’s true market value and it does not adequately compensate them for delay in receiving

payment.

For the reasons that follow, the court denies Plaintiffs’ motion without prejudice. This is

a fee-shifting case—not a common-fund case—and the parties agreed to use the lodestar method—

not the percentage-of-the-fund method—to calculate the final fee award. Although the court

agrees with Plaintiffs that the interim lodestar is likely not an adequate measure of class counsel’s

true market value, the court is not in a position to award an enhancement because the lodestar, as

calculated, is itself inexact. The court is hopeful that this decision will provide a path forward for

the parties to reach an agreement on what the proper lodestar should be, as well as any

compensation for delay.

II. BACKGROUND

A. Factual Background

The underlying facts of this case have been summarized at length in numerous prior

opinions by other judges in this District and the D.C. Circuit. 3 See De Medina v. Reinhardt,

2
This amount updates all prior payments to their 2019 dollar value.
3
Multiple judges have presided over the case during its 43-year lifespan. “United States District Judge Charles R.
Richey presided over the case until his death in 1997, after which the case was assigned to United States District Judge
James Robertson, who presided over the case until his retirement from the bench in 2010. There was no permanently-
assigned presiding judge from 2010 until October 16, 2019, when [this court] was assigned [ ] the case shortly after
Plaintiffs filed their pending motion for an additional fee award.” Defs.’ Corrected Opp’n to Pls.’ Mot. for Final
Determination of Attys.’ Fees, ECF No. 1085-1, at 2.

2
686 F.2d 997, 1000–01 (D.C. Cir. 1982) (detailing the background of this protracted litigation and

reversing the District Court’s adverse decision on liability); Hartman v. Wick, 600 F. Supp. 361,

375 (D.D.C. 1984) (finding the United States Information Agency had “discriminat[ed] against

women as a class with regard to hiring”); Hartman v. Wick, 678 F. Supp. 312, 341 (D.D.C. 1988)

(as amended) (setting “forth a concrete plan for remedying victims” of the agency’s

discrimination); Hartman v. Duffey, 19 F.3d 1459, 1474 (D.C. Cir. 1994) (remanding the class

certification for additional findings); Hartman v. Duffey, 88 F.3d 1232 (D.C. Cir. 1996) (affirming

class certification and liability determinations); Hartman v. Duffey, 973 F. Supp. 199 (D.D.C.

1997) (resolving disputes over first and second interim fee petitions, and awarding enhancement

of fees for lead attorneys Bruce Fredrickson and Susan Brackshaw). Accordingly, the court will

focus its discussion on the facts most relevant to Plaintiffs’ motion for a final fee determination.

1. Pre–Consent Decree (1977–2000)

Exhibiting extraordinary dedication to their clients’ cause, Plaintiffs’ counsel worked

without any fees for the first eighteen years of this litigation. It wasn’t until July 30, 1993, that

Plaintiffs submitted their first fee petition, “seeking fees in the amount of $2,989,150.28 and

expenses of $194,610.93 incurred from the beginning of the litigation in 1977 through August 31,

1992.” Pls.’ Mem. in Supp. of Pls.’ Mot. for a Final Determination of Attys.’ Fees, ECF No. 1081

[hereinafter Pls.’ Mot.], Decl. of Lindsey B. Lang, ECF No. 1081-6 [hereinafter Lang Decl.], ¶ 15;

see also Notice of Filing of Defs.’ Corrected Opp’n to Pls.’ Mot. for Final Determination of Attys.’

Fees, ECF No. 1085, Defs.’ Corrected Opp’n to Pls.’ Mot. for Final Determination of Attys.’ Fees,

ECF No. 1085-1 [hereinafter Defs.’ Opp’n], at 10. The court at the time admonished Plaintiffs

“for having waited all th[o]se years to file an interim application for attorneys’ fees and costs after

being encouraged to do so in open Court on several occasions.” Order, Jan. 24, 1995, ECF No.

3
179, at 2 (on file at Defs.’ Exhibit 10-6, ECF No. 1083-16 [hereinafter DEX 10-6]). 4 Plaintiffs

now explain why they did not seek an interim fee award sooner. According to Plaintiffs’ lead

counsel and declarant Bruce A. Fredrickson, the delay was due to his “reluctan[ce] to seek fees

before [Plaintiffs’] position as a prevailing party was firmly established.” See Pls.’ Reply Mem.

in Supp. of Pls.’ Mot. for a Final Determination of Attorneys’ Fees, ECF No. 1087 [hereinafter

Pls.’ Reply], Suppl. Decl. of Bruce A. Fredrickson, ECF No. 1087-3 [hereinafter Suppl.

Fredrickson Decl.], ¶ 5. If “defendant[s] [had] succeeded in [their] appeal [of the liability decision]

to the D.C. Circuit or the Supreme Court,” Fredrickson elaborates, “[his] firm might [have been]

forced to repay the government hundreds of thousands of dollars in interim attorneys’ fees upon

conclusion of [the] case.” Suppl. Frederickson Decl. ¶ 8. Although Defendants were initially

found liable in 1984, see Hartman, 600 F. Supp. at 361, it was not until 1992 that Defendants first

had an opportunity to appeal that decision, see Defs.’ Opp’n at 4. On initial appeal of the liability

determination, “[t]he Court of Appeals held that it could not determine whether the class

certification decision was correct, and remanded the case for the District Court to make the

necessary findings on, and to consider possible revisions to, class certification.” Id. at 5 (citing

Hartman, 19 F.3d at 1459). Then, on remand, “[o]n November 23, 1994, the District Court

confirmed the Court’s previous class certification decision and class-wide liability findings.” Id.

(citing Hartman v. Duffey, 158 F.R.D. 525 (D.D.C. 1994)). “The Court of Appeals subsequently

affirmed those rulings in all significant respects, and the Supreme Court denied further review.”

Id. (citing Hartman, 88 F.3d at 1232, cert. denied, 520 U.S. 1240 (1997)). Once these decisions

became final, Plaintiffs’ counsel sought their first interim fee award.

4
Because many of the filings in this case pre-date the electronic docket, Defendants have obtained the paper copies
and included them as exhibits to their opposition to Plaintiffs’ motion. Labeled hereinafter as “DEX,” these exhibits
may be found at ECF No. 1083.

4
 Plaintiffs’ early fee petitions were the subject of litigation before a Special Master and the

District Court. Plaintiffs’ declarant Lindsey B. Lang explains that “procedure dictated that the

[fee] petition be filed with the Special Master, followed by the government’s objections and

plaintiffs’ reply.” Lang Decl. ¶ 13; Order of Reference to Special Master, Oct. 1, 1991 (on file at

DEX 10-1) (ordering counsel for plaintiffs to file “with the Special Master a petition for attorneys’

fees and costs every six” months). Included with each petition were affidavits attesting that the

hourly rates Plaintiffs requested corresponded to “non-contingent rates that were actually charged

for the work of each timekeeper during the year the work was performed.” Defs.’ Opp’n at 24–

25. After reviewing Plaintiffs’ submission, “[t]he Special Master [would] then issue[] a

preliminary report that either party could object to, and once the final report was issued, either

party could appeal to the district court.” Lang Decl. ¶ 13. Ms. Lang explains, “[f]ollowing this

procedure meant that the petition took a long time to reach a conclusion and often several petitions

would be pending at the same time at different stages of the process.” Id. This often resulted in

the district court deciding more than one petition at a time, and some were “decided piecemeal

with several court orders required to fully decide the full fee award.” Id.

For each of the first through fifth fee petitions, the Special Master issued reports

recommending an award of at least 90 percent of the fees sought by Plaintiffs’ counsel, and in each

instance, the court affirmed virtually all of the Special Master’s findings. See, e.g., Lang Decl.

¶¶ 15, 17, 26; see also Defs.’ Opp’n at 10–12. With the first fee petition, for example, the Special

Master recommended an award of approximately 90 percent of the fees sought and approved

Plaintiffs’ request for a quality enhancement. See Lang Decl. ¶ 15. The “[c]ourt [subsequently]

approved the Special Master’s recommendations,” and awarded a “25 [percent] quality

enhancement, or $188,968.70, for the hours devoted by Bruce Fredrickson and Susan Brackshaw

5
for the years 1980–1990.” Id. ¶ 19. But this process took time. Although the court had ordered

an interim payment of $500,000 in January 1995, see Order, Jan. 24, 1995, ECF No. 179 (on file

at DEX 10-6), as of December 1996, final resolution of Plaintiffs’ first and second fee petitions

was still pending. It wasn’t until August 4, 1997, that the court issued a final opinion on the

balance of the fees sought in those petitions. See Lang Decl. ¶ 19 (citing Hartman, 973 F. Supp.

199).

In an attempt to streamline the interim-fee-petition process, in December 1996, the parties

“met with the Special Master and were encouraged to attempt to resolve some of the recurring

issues that were litigated in each fee petition.” Lang Decl. ¶ 21. Mr. Fredrickson explains that

because there was significant litigation over the appropriate hourly rate up until this point, and

because his firm “d[id] not have regular billing rates,” the parties agreed that Plaintiffs’ counsel’s

fees should “be computed using a market rate for similar services in the District of Columbia.”

Pls.’ Mot., Ex. 2, Decl. of Bruce A. Fredrickson, ECF No. 1081-2 [hereinafter Fredrickson Decl.],

¶ 63. “At that time, the rates in the [U.S. Attorneys’ Office’s (“USAO”)] Laffey matrix were

roughly comparable to the prevailing market rates in the District of Columbia.” Id. Accordingly,

“[o]n December 20, 1996[,] the parties agreed that, for the purposes of [the third, and all

subsequent] interim petitions, the parties would use actual rates for counsel with established billing

rates to the extent those rates did not exceed the applicable rate for the timekeeper under the Laffey

matrix, and that Laffey rates would be used for counsel without established billing rates.” Lang

Decl. ¶ 21; Fredrickson Decl. ¶ 63. 5 Mr. Fredrickson explains that although he later became aware

5
The Laffey rates are discussed in greater detail infra, Section IV.B.1.b. “The initial Laffey rates were representative
of the rates for complex federal litigation in the District of Columbia [“D.C.”] in 1982.” Lang Decl. ¶ 62. The D.C.
USAO “undertook an update of those rates for subsequent years using the Consumer Price Index for All Urban
Consumers (CPI-U) for the Washington-Baltimore, DC-MD-VA-WVA [area], as announced by the Bureau of Labor
Statistics for May of each year.” Id. ¶ 63. The Laffey matrices are available at https://www.justice.gov/usao-dc/civil-
division (last accessed Nov. 2, 2020).

6
that the “Laffey rates . . . were lagging behind prevailing market rates,” because “each petition

reserved the right to seek an enhancement to the lodestar awarded, [he] was satisfied to honor the

agreement [to use the Laffey rates] rather than introduce more litigation over rates into the interim

petitions.” Fredrickson Decl. ¶ 63.

The Laffey rates, however, remained generally consistent with market rates through the

sixth and seventh fee petitions, submitted on September 30, 1998, and May 25, 1999, respectively.

As with the previous fee petitions, Plaintiffs provided affidavits in support of the petitions,

including that of Crowell & Moring attorney Laurel Pyke Malson. Crowell & Morning had joined

as class counsel in 1992. Id. ¶ 38. Ms. Malson attested to the “non-contingent rates that were

actually charged during th[e] [relevant] period for each of the timekeepers whose time [was] sought

in Plaintiffs’ seventh fee petition,” and explained that “th[o]se rates [we]re somewhat different

from [the Laffey rates] used to calculate [the] Crowell & Moring lodestar.” Decl. of Laurel Pyke

Malson, May 20, 1999 (on file at DEX 7 [hereinafter Malson Decl.]), ¶ 42 (emphasis added). And,

more to the point, in their memorandum in support of the seventh fee petition, Plaintiffs argued

that the hourly rates in the USAO’s Laffey Matrix for the time period covered by the seventh fee

petition (January 1 – June 30, 1998) were “market rates . . . supported by the rates awarded in

recent cases in this Circuit as well as rate information provided in exhibits to previous Hartman

fee petitions.” Mem. in Supp. of Pls.’ Seventh Mot. for Interim Attys.’ Fees & Expenses (on file

at DEX 10-15), at 10–11. 6 For both petitions, the Special Master recommended Plaintiffs receive

virtually all of the $2,106,252.65 in fees sought. See Lang Decl. ¶ 28; Defs.’ Opp’n at 13.

6
Defendants note that “[t]he Seventh Fee Petition was the last formal interim fee petition submitted by Plaintiffs. All
subsequent fee requests were resolved informally. Thus, the record of this case does not include contemporaneous
data about historic market rates for periods after June 30, 1998.” Def’s Opp’n at 30.

7
 2. Post–Consent Decree (2000–2018)

With Defendants’ objections to Plaintiffs’ sixth and seventh fee petitions still pending, the

parties entered settlement negotiations on March 22, 2000, and reached agreement two days later.

See Lang Decl. ¶ 29. On March 24, 2000, the parties entered a Consent Decree that provided for

a staggering $508,000,000 settlement fund for the class. Id. It also provided that “Plaintiffs shall

be entitled to reasonable attorneys’ fees, expenses, and costs from the initiation of this case through

the final distribution of amounts in the settlement fund or the final resolution of any issues that

may arise under this Decree, whichever is later.” Consent Decree, Mar. 24, 2000, ECF No. 866

(on file at DEX 10-13 [hereinafter DEX 10-13]), at ¶ 8.

In his declaration, Mr. Fredrickson explains his motivation for deferring resolution of

attorneys’ fees until the end of litigation, and not seeking a percentage of class recovery for

attorneys’ fees during settlement negotiations, as he knew had been done in other cases at the time.

Seeking a percentage of the class recovery, which would have come out of the settlement fund, he

believed, “would have violated [counsel’s] prior commitments to the class members.” Suppl.

Fredrickson Decl. ¶ 9. “[T]o avoid any suggestion of a conflict of interest,” Fredrickson Decl.

¶ 65, Mr. Fredrickson says, he decided to “negotiate[] a consent decree [that] contemplated [an

award of] reasonable attorneys’ [fees] upon the conclusion of the case,” Suppl. Fredrickson Decl.

¶ 9. Putting “the question of attorneys’ fees to the side” until final resolution of the case, he

explains, was critical to reaching agreement. Id.

Following approval of the Consent Decree, the court “ordered the withdrawal of all

pending motions, including motions for review of the Special Master’s findings regarding the Sixth

and Seventh Petitions.” Lang Decl. ¶ 29. The parties subsequently settled on the fees owed for

those petitions, totaling $2,099,131.78. Id. ¶ 30; see also Defs.’ Opp’n at 13.

8
 Thereafter, the parties “settled twenty-one interim attorney fee requests that collectively

covered a period of approximately 20 years . . . without litigation or intervention by the Court,

aside from routine approvals of the stipulated payments.” Defs.’ Opp’n, Decl. of Daniel F. Van

Horn, ECF No. 1083-8, ¶ 15. By the terms of each petition, payment of interim fees was

without prejudice to either party’s position with respect to issues
which have been reserved pending entry of final judgment and
exhaustion of any appeals in this action and the final resolution of
the various interim fee petitions, including but not limited to issues
regarding reimbursable expenses, the appropriate interest rate, and
enhancement of the lodestar.

Pls.’ Mot. at 4–5 (quoting Joint Stipulation & Order Regarding Pls.’ Ninth Request for Interim

Attys.’ Fees & Expenses, Apr. 5, 2001, ECF No. 984 [hereinafter Ninth Fee Petition Stip.], at 1–

2); see also Suppl. Fredrickson Decl. ¶ 9. According to Ms. Lang, “[t]he pace of work was such

that for several petitions covering the late 1990’s[,] fees were sought for a six-month period at a

time.” Lang Decl. ¶ 31. And, “[b]y the Eleventh Petition (2001), each year a petition was filed

for fees incurred in the preceding calendar year.” Id.

All told, class counsel received $7,788,724.19 in fees under the eighth through twenty-

eighth fee petitions, see Lang Decl. ¶¶ 32–52, for a sum of $20,987,515.51 in fees since the first

award in 1995, see Lang Decl., Ex. B., at PDF p. 28.

3. Interest and Compensation for Delay

Defendants have paid interest on interim fees earned since 1991. Initially, Plaintiffs

“sought current market rates to compensate for delay.” Lang Decl. ¶ 14. But the court at first

rejected any payment for delay, ruling “that Congress had not waived sovereign immunity to

permit prejudgment interest on fees and expenses in this case.” Defs.’ Opp’n at 12 (citing

Hartman, 973 F. Supp. at 201). Plaintiffs sought reconsideration of that decision on the basis that

interest was authorized and should be applied retroactively, under the 1991 Civil Rights Act. See

9
Lang Decl. ¶ 14. The court ultimately agreed, “and, on February 5, 1998, ruled that prejudgment

interest would be permitted, and would begin to accrue on November 21, 1991, the effective date

of [the Act].” Defs.’ Opp’n at 12 (citing Mem. Order re Prejudgment Interest, Feb. 5, 1998, ECF

No. 506 (on file at DEX 10-9 [hereinafter DEX 10-9]); see also Lang Decl. ¶ 25. The court also

decided that “interest in this case would be calculated using the 1-year Treasury bill rate rather

than the prime rate Plaintiffs requested.” DEX 10-9, at 4–5; see also Hartman v. Duffey, 8 F. Supp.

2d 1 (D.D.C. 1998). In total, class counsel have received $1,891,040.16 in interest on the interim

fee payments. See Lang Decl., Ex. B., at PDF p. 28.

B. Plaintiffs’ Motion for a Final Fee Determination

The last payment from the $508 million settlement fund issued on June 22, 2018. See Pls.’

Mot. at 3. Then, on October 19, 2019, Plaintiffs filed their Motion for a Final Determination of

Attorneys’ Fees, as permitted under the Consent Decree. See generally id. Plaintiffs seek an

additional fee award of $34,114,143.52, for a total award of $75 million. Id. at 1. Plaintiffs justify

that sum based on a percentage of the total class recovery, see id. at 26, and assert that the roughly

eight percent fee they seek 7 is “consistent with, if not lower than, fees awarded in similar cases,”

id. at 28. Alternatively, Plaintiffs argue, if the court uses a lodestar analysis, then an enhancement

is warranted. They offer an enhancement calculation that would result in an adjusted lodestar

roughly equivalent to the $75 million sought. Id. at 43.

Defendants oppose the award of additional fees beyond what has already been paid to date.

See Defs.’ Opp’n at 1. In short, Defendants contend that the lodestar method is the proper way of

calculating fees in this case and, under that approach, Plaintiffs have received a reasonable fee and

7
Plaintiffs calculate the percentage by updating the total class member recovery (lump sum settlement + costs and
fees) to its 2019 value, and then dividing the $75 million fee sought by that amount ($880,136,968.23). See Lang
Decl., Ex.A, at PDF p. 18.

10
are not entitled to an enhancement on any ground. See generally id. Briefing on Plaintiffs’ motion

concluded on February 28, 2020. See Pls.’ Reply. On September 23, 2020, the court heard oral

argument on Plaintiffs’ motion.

III. LEGAL STANDARD

“In a certified class action, the court may award reasonable attorney’s fees and nontaxable

costs that are authorized by law or by the parties’ agreement.” Fed. R. Civ. P. 23(h). One such

law that authorizes the award of attorneys’ fees is Title VII. It provides that the court “in its

discretion, may allow the prevailing party . . . a reasonable attorney’s fee (including expert fees)

as part of the costs.” 42 U.S.C. § 2000e-5(k). “In general, a trial court enjoys substantial discretion

in making reasonable fee determinations.” Swedish Hosp. Corp. v. Shalala, 1 F.3d 1261, 1271

(D.C. Cir. 1993). “[A] reasonable fee is one that is ‘adequate to attract competent counsel, but that

does not produce windfalls to attorneys.’” West v. Potter, 717 F.3d 1030, 1033–34 (D.C. Cir.

2013) (quoting Blum v. Stenson, 465 U.S. 886, 897 (1984)). The Supreme Court has cautioned

that “[a] request for attorney’s fees should not result in a second major litigation.” Hensley v.

Eckerhart, 461 U.S. 424, 437 (1983). Instead, “[p]arties . . . should make a conscientious effort,

where a fee award is to be made, to resolve any differences.” Blum, 465 U.S. at 902 n.19.

IV. DISCUSSION

The court begins its analysis by considering the proper method for calculating fees in this

case. Because the court concludes that this is a fee-shifting case, it then turns to analyze the

lodestar and Plaintiffs’ argument that an enhancement to the lodestar is appropriate. The court

ends with a discussion of Plaintiffs’ proposed calculation of an enhanced lodestar.

11
 A. Lodestar Versus Percentage of the Fund

Any inquiry into the reasonableness of attorneys’ fees begins with determining the proper

method for calculating fees. Courts generally calculate attorneys’ fees in one of two ways: the

lodestar method or the percentage-of-the-fund approach, depending on whether the case involves

a fee-shifting provision or a fee-spreading arrangement.

“Fee shifting” is an exception to the general rule in the American legal system that “[e]ach

litigant pays his own attorney’s fees, win or lose, unless a statute or contract provides otherwise.”

Hardt v. Reliance Standard Life Ins., 560 U.S. 242, 253 (2010). Fee shifting may arise “(1) when

a statute grants courts the authority to direct the losing party to pay attorney’s fees; (2) when the

parties agree in a contract that one party will pay attorney’s fees; [or] (3) when a court orders one

party to pay attorney’s fees for acting in bad faith.” In re Home Depot Inc., 931 F.3d 1065, 1078

(11th Cir. 2019); Alyeska Pipeline Serv. Co. v. Wilderness Soc’y, 421 U.S. 240, 275 (1975)

(Marshall, J., dissenting). Generally speaking, fee-shifting cases use the lodestar method for

determining a reasonable fee. See Perdue v. Kenny A., 559 U.S. 542, 551 (2010). The lodestar is

equal to the “hours reasonably expended on the litigation multiplied by a reasonable hourly rate,”

Hensley, 461 U.S. at 433. The Supreme Court has described the “lodestar” method as the “guiding

light of our fee-shifting jurisprudence.” Perdue, 559 U.S. at 551.

The “percentage-of-the-fund method,” on the other hand, “is the appropriate mechanism

for determining the attorney fees award in common fund,” or what some courts refer to as, “fee-

spreading” cases. Swedish Hosp. Corp., 1 F.3d at 1271; Home Depot, 931 F.3d at 1079 n. 12

(explaining that “[i]t would arguably be more helpful to describe the difference as between fee-

shifting (where the fee shifts to the other party) and fee-spreading (where the fee is spread among

the benefited party)” than as between fee-shifting and “common fund”). As its name suggests,

12
under the percentage-of-the-fund method, the attorneys whose efforts benefitted the fund are

entitled to a reasonable fee to be paid from the fund as a whole. Boeing Co. v. Van Gemert, 444

U.S. 472, 478 (1980). In this Circuit, “common fund class action awards fall between twenty and

thirty percent.” Swedish Hosp., 1 F.3d at 1272. The rationale for this doctrine rests on the

equitable notion that the payment of “fair and just allowances for expenses and counsel fees”

should be spread among all the beneficiaries of the fund lest they be unjustly enriched by the

attorneys’ efforts. Trustees of the Internal Improvement Fund v. Greenough, 105 U.S. 527, 536

(1881); see also Cent. R.R. & Banking Co. of Ga. v. Pettus, 113 U.S. 116, 126–27 (1885);

Boeing, 444 U.S. at 478. Because the attorneys’ fees come from the fund, which would otherwise

be payable to the attorneys’ clients, it is the clients, not the losing party, that pays the attorneys.

In this sense, a common fund fee award is generally not considered to be an exception to the

American Rule. See generally William B. Rubenstein, 5 Newberg on Class Actions § 15:25, at 60

n.3 (5th ed.) (“Courts occasionally state that fees taken from a common fund represent another

exception to the American Rule, . . . [b]ut that is not quite right: when fees are extracted from a

common fund to pay class counsel, the class members’ recoveries are reduced accordingly and

hence those class members themselves are paying their own fees.” (citations omitted)).

The parties vigorously disagree over into which category this case falls: fee-shifting or

fee-spreading. Plaintiffs argue that the court should treat the settlement as a “constructive common

fund,” warranting a percentage-of-the-fund fee award. See Pls.’ Mot. at 34–35; see also Pls.’

Reply at 5. Defendants, on the other hand, contend that this is a fee-shifting case, see Defs.’ Opp’n

at 18, and therefore the lodestar method is the correct one for calculating that fee award, see id. at

16. As explained below, the court agrees with Defendants.

13
 1. The Parties’ Intent

To start, the parties agree that proper computation of the final fee award in this case is a

question of contract interpretation. See Oral Argument Tr., ECF No. 1097 [hereinafter Hr’g Tr.],

at 9:10-14 (Plaintiffs’ counsel agreeing that “this case, in the first instance, [is] a question of

contract interpretation”), 37:16-19 (government counsel agreeing that “this is a matter of contract

interpretation”). The relevant contract is the Consent Decree entered into in 2000. Thus, the task

before the court is to discern the parties’ intent, in 2000, with respect to calculating a final fee

award.

Settlement agreements are contracts, and as with interpreting any contract, the court begins

with its terms. See Keepseagle v. Perdue, 856 F.3d 1039, 1047 (D.C. Cir. 2017); cf. U.S. v. ITT

Cont’l Banking Co., 420 U.S. 223, 238 (1975) (“Since a consent decree . . . is to be construed for

enforcement purposes basically as a contract, reliance upon certain aids to construction is proper,

as with any other contract.”). The court must first “determine whether the disputed language is

unambiguous.” Armenian Assembly of Am., Inc. v. Cafesjian, 758 F.3d 265, 278 (D.C. Cir. 2014).

If the relevant clause is subject to more than one reasonable interpretation, the court must consider

“what a reasonable person in the position of the parties would have thought the disputed language

meant.” Id. (quoting Tillery v. D.C. Contract Appeals Bd., 912 A.2d 1169, 1176 (D.C. 2006)).

The text of the Consent Decree provides little insight into the parties’ intent concerning the

final fee award. Paragraph 8 of the 2000 Consent Decree provides:

Plaintiffs shall be entitled to reasonable attorneys’ fees, expenses,
and costs from the initiation of this case through the final
distribution of amounts in the settlement fund or the final resolution
of any issues that may arise under this Decree, whichever is later.
Reasonable attorneys’ fees, expenses, and costs shall include
reasonable fees and expenses associated with administering the
settlement, including but not limited to notifying the Class Members
of this Decree, establishing and maintaining a qualified settlement

14
 fund, distributing the fund to the Class Members, complying with
legal obligations governing such matters, and retaining appropriate
third parties to perform such tasks. The parties shall endeavor to
reach agreement on all of Plaintiffs’ claims for attorneys’ fees,
expenses, and costs, but each party reserves the right to litigate any
claims that cannot be resolved informally.

DEX 10-13 at 6–7. On its face, the Consent Decree says nothing about the method that should be

used to calculate attorneys’ fees. It says only that “Plaintiffs shall be entitled to reasonable

attorneys’ fees”—the term “reasonable” is nowhere defined. Id.

When a settlement agreement’s plain text is ambiguous, the court can turn to its structure

for clues. See Ohio Power Co. v. F.E.R.C., 744 F.2d 162, 168 (D.C. Cir. 1984). On that score,

the Consent Decree is illuminating. The Consent Decree, by its very terms, is not a common fund

settlement. The parties did not contemplate that the final award of attorneys’ fees would be

deducted from the overall class recovery, such that the fees award would be “spread[ ] . . .

proportionately among those benefited by the suit.” Boeing Co., 444 U.S. at 478. “Fee spreading”

is an essential feature of a common fund settlement. See supra, at 12–13. The parties instead

agreed that the final fee award would be paid by Defendants over and above the fees paid from the

settlement fund. See, e.g., Joint Pre-Hearing Br. in Supp. of Consent Decree, ECF No. 904 (on

file at DEX 10-14), at 9–10 (“Plaintiffs’ reasonable attorneys’ fees, expenses and costs from the

initiation of this case through the final resolution of the case will be paid by the Defendants over

and above the $508 million.”). That the parties agreed Defendants would be the source of the final

fees payment is consistent with principles of fee shifting, not the common fund doctrine. See

Home Depot, 931 F.3d at 1079 (observing that “the key distinction between common-fund and

fee-shifting cases is whether the attorney’s fees are paid by the client (as in common-fund cases)

or by the other party (as in fee-shifting cases)”). Thus, the settlement structure is strong evidence

15
that the parties did not contemplate that Plaintiffs would be paid a final fees award as a percentage

of the benefit received by the class, as Plaintiffs now demand.

Other evidence confirms this understanding of the Consent Decree. See Ohio Power Co.,

744 F.2d at 168 (“[W]hen the meaning of [a] contract cannot be determined from its text and

structure or from the application of canons of contract interpretation,” the court may consider

extrinsic evidence “to discern the meaning that the parties intended to attribute to the ambiguous

formulation.”). Take the terms of the parties’ twenty-plus fee stipulations following the Consent

Decree. Each expressly contemplates that the final fee award would be calculated using the

lodestar method. For example, the “Joint Stipulation and Order Regarding Plaintiff’s Ninth

Request for Attorneys’ Fees and Expenses”—the first such stipulation following the parties’ entry

in the Consent Decree—states that the

stipulation is without prejudice to either party’s position with
respect to issues which have been reserved pending entry of final
judgment and exhaustion of any appeal in this action and the final
resolution of the various interim fee petitions, including but not
limited to issues regarding reimbursable expenses, the appropriate
interest rate, and enhancement of the lodestar.

Ninth Fee Petition Stip. at 1–2 (emphasis added). Subsequent fee stipulations, up until the last,

contain an identically-worded reservation of rights. See Defs.’ Surreply in Opp’n to Pl’s Mot. for

Final Determination of Att’ys’ Fees, ECF No. 1088-1, at 1–2 & n.1; Joint. Stip. & Order Regarding

Pls.’ Twenty-Eighth Request for Att’ys’ Fees & Expenses, ECF No. 1079. Plaintiffs cannot now

genuinely assert that they intended to seek a final award on a percentage-of-the-fund basis when

for nearly two decades they have reserved the right to ask for an “enhancement of the lodestar.”

Finally, and perhaps most telling, is Mr. Fredrickson’s personal account of the negotiations

over the Consent Decree. He makes clear that he deliberately chose not to negotiate a percentage-

of-the-fund award for attorneys’ fees at the time of the settlement and that he always envisioned a

16
final award of fees based upon the lodestar method. In his initial declaration, Mr. Frederickson

states that, in 1999 when negotiating the settlement,

I knew that class counsel in some other cases had pursued
simultaneous negotiations concerning both the merits and their own
fees, and had secured a percentage of the recovery as a fee. I also
knew that courts had approved the resulting fees. I believed,
however, that class counsel in Hartman should avoid any suggestion
of a conflict of interest and should negotiate only about the merits,
reserving the amount of the fees for later discussion or judicial
decision.

Frederickson Decl. ¶ 65. As for the final fee award, Mr. Frederickson attests that he intended to

pursue such an amount as an “enhancement to the lodestar at the conclusion of the litigation.” Id.

Mr. Frederickson amplifies this history in his supplemental declaration. He explains that, “from

the start of the representation of the class, I informed the class members that we would work on a

contingency fee basis, accept no portion of any recovery of any awards to class members, and rely

upon our opportunity under Title VII to obtain our reasonable attorneys’ fees directly from the

defendant.” Suppl. Fredrickson Decl. ¶ 9. By referencing a fee award under Title VII,

Mr. Frederickson plainly is referring to Title VII’s fee-shifting provision, which uses the lodestar

methodology to calculate fees.

Mr. Frederickson’s candor is commendable. He makes plain that, from the outset, he

understood the Consent Decree reserved for Plaintiffs the right to seek a final fee award based on

an enhancement of the lodestar, not through a percentage-of-the-fund approach. That is the

methodology the court therefore must apply to carry out the parties’ intent. 8

8
It bears noting that simply because parties agree to a contractual fee-shifting arrangement does not mean that the
law concerning statutory fee shifting automatically applies. The Eleventh Circuit emphasized this point in Home
Depot, observing that “[t]his case . . . is a contractual fee-shifting case, and the appropriate method for such a case is
not clearly governed by any binding precedent.” 931 F.3d at 1081–82. The court in Home Depot nevertheless upheld
the trial court’s application of the lodestar method because, although class counsel believed it to be a common-fund
case, several of the claims were brought under fee-shifting statutes, and the parties agreed that the district court had
discretion to choose either the percentage or the lodestar method. See id. at 1082. Unlike in Home Depot, this court

17
 2. Constructive Common-Fund Cases

Plaintiffs attempt to avoid this result by characterizing the Consent Decree as having

established a “constructive common fund,” under which Plaintiffs reserved the right to seek a

percentage recovery. Pls.’ Mot. at 26. But this argument does not match the evidence and falls

under its own weight.

The constructive common-fund doctrine is based on the equitable notion that, even in cases

where attorneys’ fees and class settlements are paid from separate funds, so long as the two

amounts are negotiated as a “package deal” such that “[t]he defendant is concerned, first and

foremost, with its total liability,” Home Depot, 931 F.3d at 1080, common fund principles should

apply, see generally In re Heartland Payment Sys., Inc., 851 F. Supp. 2d 1040, 1072 (S.D. Tex.

2012) (providing an overview of the constructive common-fund doctrine). This is because “as a

practical matter, defendants undoubtedly take into account the amount of attorney’s fees when

they agree on an amount to pay the class.” Home Depot, 931 F.3d at 1080; see also Brytus v.

Spang & Co., 203 F.3d 238, 246 (3d Cir. 2000) (“[C]onsideration of the attorney’s fees was likely

factored into the amount of settlement.”).

The cases Plaintiffs cite illustrate this feature of a constructive common fund. For example,

in Vista Healthplan, Inc. v. Warner Holdings Co. III, the court approved a settlement agreement

that included separate funds for attorneys’ fees and class recovery, and found that the agreement

qualified as a constructive common fund where both amounts were determined at the time of

settlement. 246 F.R.D. 349, 363–64 (D.D.C. 2007). In Ingram v. Coca Cola Co., the court

reviewed and approved a settlement agreement in which Coca Cola agreed to a cash payment

totaling $103.5 million, comprising several funds—a “Compensatory Damages Fund

has no choice because the parties here made their intentions clear: Plaintiffs’ final fee award would be calculated
using the lodestar method.

18
(approximately $58.7 million); [a] Make–Whole Relief Back Pay Fund (approximately $24.1

million); and [a separate payment of] specified attorneys’ fees (approximately $20.7 million).”

200 F.R.D. 685, 694 & n.16 (N.D. Ga. 2001). There, the court explained that “[w]hile attorneys’

fee constitute[d] 20% of th[o]se current cash payments, the fee percentage [would] become[] much

smaller as additional, future relief to the class [was] factored in.” Id. And in In re Chrysler-

Dodge-Jeep Ecodiesel Marketing Sales Practices and Products Liability Litigation, the court

approved a settlement agreement wherein the parties agreed that class counsel would receive

“$59 million in attorneys’ fees . . . to be paid by the Defendants in addition to the compensation

available to the Class,” after it found that amount to be “reasonable, whether a percentage method

or lodestar method [was] used.” No. 17-md-02777-EMC, 2019 WL 2554232, at *2 (N.D. Cal.

May 3, 2019).

The “package deal” feature of these cases is absent here. Plaintiffs’ final fee award was

not coupled with the settlement of class-wide liability, such that Defendants would have

understood their “total liability” at the time of resolution. Home Depot, 931 F.3d at 1080. To the

contrary, the parties expressly declined to fix a final fee award under the Consent Decree. What’s

more, Mr. Frederickson’s declaration makes clear that there was no “package deal” with respect

to class recovery and attorneys’ fees. He says, “[i]nstead, [they] negotiated a consent decree which

contemplated reasonable attorneys’ [fees] upon the conclusion of the case.” Suppl. Fredrickson

Decl. ¶ 9. “[H]ad [we] tried to negotiate fees at the same time as we negotiated the class recovery,”

Mr. Frederickson explains, “[Plaintiffs] would have demanded a fund much greater than $508

million” and “negotiations would have faltered.” Id. ¶ 11. Nothing about Mr. Frederickson’s

understanding of the settlement resembles a constructive common fund.

19
 Plaintiffs also assert that they negotiated a constructive common fund “[b]ecause the fees

are to be borne by the defendant rather than the class, [as] a valuable addition to the settlement,”

and “[w]hen the defendant’s payments are added together, the full value of the settlement is

known.” Pls.’ Mot. at 26. But under that logic, every fee-shifting case would be transformed into

a common fund case merely by establishing a settlement fund. “It would be virtually impossible

to contract for fee-shifting, . . . absent perhaps some magic-word requirement.” Home Depot, 931

F.3d at 1081.

Finally, Plaintiffs try a different tack, appealing to the court’s natural instinct to want to

execute an administratively efficient calculation of a final fee award. They point to the

D.C. Circuit’s ruling in Swedish Hospital and its observations that the lodestar approach

“encourages significant elements of inefficiency” and that “a percentage-of-the-fund approach is

less demanding of scarce judicial resources than the lodestar method.” 1 F.3d at 1268, 1269; Pls.’

Reply at 10. Swedish Hospital is, of course, a different case, as Plaintiffs recognize, because it

involved a common fund. And, although the Circuit’s observations concerning the efficiency of

the percentage-of-the-fund method are surely true, that is not what the parties here negotiated. The

court’s role is to effectuate the parties’ intent, and they agreed to a fee-shifting arrangement that

would use the lodestar method to calculate the final award. The court turns now to making that

determination.

B. Whether to Enhance the Lodestar

Calculating the lodestar is, in theory, a straightforward exercise. See Perdue, 559 U.S. at

551 (lauding the “lodestar method as readily administrable”). As noted, the lodestar is achieved

by multiplying the “the hours reasonably expended on the litigation multiplied by a reasonable

hourly rate.” Hensley, 461 U.S. at 433. Using that simple mathematical equation here produces

20
an interim lodestar of $22,878,155.67. See Lang Decl., Ex. A, at PDF p. 19. That amount

represents the sum of all 28 interim attorneys’ fee awards. Id. But Plaintiffs assert that “this is the

rare and exceptional case warranting enhancement of the lodestar.” Pls.’ Mot. at 36. For their

part, Defendants assert that Plaintiffs are entitled to no more than what they have been paid to date.

“[T]he burden of proving that an enhancement [to the lodestar] is necessary must be borne

by the fee applicant.” Perdue, 559 U.S. at 553. There is a “strong presumption” that the “lodestar

method yields a . . . sufficient [fee],” id. at 546, 552, because “the lodestar figure includes most, if

not all, of the relevant factors constituting a ‘reasonable’ attorney’s fee,” Pennsylvania v. Del.

Valley Citizens’ Council for Clear Air, 478 U.S. 546, 566 (1986) (“Delaware Valley I”). To

overcome the presumption of reasonableness, “a fee applicant seeking an enhancement must

produce ‘specific evidence’ that supports the award,” Perdue, 559 U.S. at 553 (quoting Blum, 465

U.S. at 899, 901), and “an enhancement may not be awarded based on a factor that is subsumed in

the lodestar calculation,” id. For example, “the novelty and complexity of a case generally may

not be used as a ground for an enhancement because these factors presumably are fully reflected

in the number of billable hours recorded by counsel.” Id. (cleaned up). Although the Supreme

Court has “never sustained an enhancement of a lodestar amount for [superior attorney]

performance,” id. (emphasis added) (citing Delaware Valley I, 478 U.S. at 565; Blum, 465 U.S. at

897; and Hensley, 461 U.S. at 435), the Court has held open the possibility that an enhancement

may be appropriate in “rare” and “exceptional” circumstances, so long as the fee applicant provides

“specific evidence that the lodestar fee would not have been adequate to attract competent

counsel,” id. at 554.

In Perdue v. Kenny A., the Supreme Court identified three possible circumstances in which

lodestar enhancement might be warranted. First, “an enhancement may be appropriate where the

21
method used in determining the hourly rate employed in the lodestar calculation does not

adequately measure the attorney’s true market value, as demonstrated in part during the litigation.”

Id. at 554–55. “This may occur,” the Court explained, “if the hourly rate is determined by a

formula that takes into account only a single factor (such as years since admission to the bar).” Id.

at 555. In such a case, “to provide a calculation that is objective and reviewable, the trial judge

should adjust the attorney’s hourly rate in accordance with specific proof linking the attorney’s

ability to a prevailing market rate.” Id. Second, “an enhancement may be appropriate if the

attorney’s performance includes an extraordinary outlay of expenses and the litigation is

exceptionally protracted.” Id. In that case, “the enhancement must be calculated using a method

that is reasonable, objective, and capable of being reviewed on appeal, such as by applying a

standard rate of interest to the qualifying outlays of expenses.” Id. Third, “there may be

extraordinary circumstances in which an attorney’s performance involves exceptional delay in the

payment of fees.” Id. at 556. Compensation for that delay should generally be “made either by

basing the award on current rates or by adjusting the fee based on historical rates to reflect its

present value.” Id. (cleaned up).

Plaintiffs argue that this case involves the first and third of these scenarios. That is, that

the interim lodestar does not reflect “the true market value of the services performed by counsel,”

Pls.’ Mot. at 40, and “[t]he use of current rates to compute the lodestar” is appropriate to

compensate Plaintiffs for delay, id. at 42. The court addresses these arguments in turn, and agrees

in part.

22
 1. True Market Value

Plaintiffs advance two arguments for why the rates used to calculate the interim fee awards

do not reflect the true market rate.

a. Junior lawyers paid at lower rates

First, they maintain that, “for much of this case, attorneys at a very junior level undertook

work normally done by more senior attorneys,” and therefore “[t]he rates awarded for their services

were [ ] much lower than if a more senior attorney had undertaken the tasks.” Pls.’ Mot. at 37.

Plaintiffs liken this case to McKenzie v. Kennickell, 875 F.2d 330 (D.C. Cir. 1989), which they say

is “the only case where the D.C. Circuit has upheld a multiplier for quality of representation,” Pls.’

Mot. at 37. 9 Similar to this case, the “young attorneys [in McKenzie] stayed active on the case for

its fifteen-year duration, which the District Court [in McKenzie] found to be an ‘extremely rare’

occurrence.” Id. (quoting McKenzie v. Kennickell, 684 F. Supp. 1097, 1107 (D.D.C. 1988)). The

district court found that “[s]uch continuity promote[d] tremendous efficiency and necessarily

reduce[d] the ultimate expenditure of hours,” McKenzie, 684 F. Supp. at 1107, and so “the lodestar

fee could not possibly reflect the benefits derived from [counsel’s] extensive experience and

intimate knowledge with this litigation,” id. For this reason, the district court granted a 25-percent

enhancement for quality of representation, id., which the D.C. Circuit affirmed, 875 F.2d at 339.

Plaintiffs ask the court to apply similar logic here.

9
Plaintiffs have already received an enhancement to the lodestar based on McKenzie for a portion of the case. In the
Special Master’s report on Plaintiffs’ First Petition, he recommended that Mr. Frederickson and Ms. Brackshaw
“receive a 25% enhancement on the lodestar rates applicable to them for the years 1980–1990,” Special Master’s Rep.
on Mot. by Pls. for Interim Attys.’ Fees, Feb. 18, 1994, ECF No. 126 (on file at DEX 10-2), at 73, because, “in th[o]se
years, . . . [they] carried the laboring oar in this complex class action and achieved success that is remarkable for two
lawyers who had begun their work on the case early in their careers,” id. at 75. Relying on McKenzie, the court
approved the Special Master’s recommendation in its opinion resolving Plaintiffs’ motion for final resolution of the
First and Second Petitions. See Hartman, 973 F. Supp. at 202. As a result, Plaintiffs received an enhancement totaling
$188,968.70. See Defs.’ Opp’n at 12; Lang Decl., Ex. A, at PDF p. 18.

23
 Analogizing this case to McKenzie is problematic though, because the D.C. Circuit

“overrule[d]” McKenzie only two years later in King v. Palmer, 950 F.2d 771, 773 (D.C. Cir. 1991)

(en banc), and it is unclear how much of McKenzie, if any, survived. The panel in McKenzie made

two holdings. It first held that a court could enhance the lodestar based on objective proof of the

difficulty of ascertaining counsel in similar cases, as opposed to the actual difficulty faced by the

plaintiff in the case at hand. See McKenzie, 875 F.2d at 338–39. Additionally, as discussed, the

panel affirmed a 25-percent enhancement of the lodestar based on the quality of work performed

by junior lawyers who had stayed on for the duration of the case. See id. at 338–39. In overruling

McKenzie, the en banc court in King rejected the panel’s enhancement of the lodestar based on

objective evidence of the difficulty of ascertaining counsel, and instead required a prevailing

plaintiff to produce, at least, evidence of actual difficulty. See King, 950 F.2d at 773, 778.

Although the en banc court did not address the panel’s affirmance of the 25-percent enhancement,

King at least casts doubt on the continued viability of McKenzie’s enhancement for quality of

performance. Indeed, no circuit or district court decision since King has cited McKenzie to support

the enhancement of the lodestar. If anything, the trend in this Circuit has been away from

enhancements. As the Circuit observed in Swedish Hospital Corp., since King, “we have generally

disavowed the use of enhancement, in recognizing that enhancing factors are reflected in the

original lodestar.” 1 F.3d at 1267 n.3. McKenzie therefore does not help Plaintiffs’ cause.

b. USAO Laffey Matrix rates are below true market rates

Plaintiffs’ second reason for seeking an enhancement is more convincing. They contend

that “the rates used to determine the lodestar for the interim petitions were not market rates that

reflected the true value of the attorneys’ services,” because “they were based on the rates contained

in the Laffey matrix.” Pls.’ Mot. at 38. Recall that, starting in 1996, “the parties agreed that, for

24
the purposes of [the third, and all subsequent] interim petitions, the parties would use actual rates

for counsel with established billing rates to the extent those rates did not exceed the applicable rate

for the timekeeper under the Laffey matrix, and that Laffey rates would be used for counsel without

established billing rates.” Lang Decl. ¶ 21; Fredrickson Decl. ¶ 63. “The practical effect of this

agreement,” Plaintiffs say, “was that Laffey rates were used for all subsequent petitions” and that

those rates do not represent their true market value. Lang Decl. ¶ 21.

A brief history of the Laffey rates is necessary to elucidate Plaintiffs’ argument. Laffey v.

Northwest Airlines, Inc. was a Title VII employment discrimination class action involving female

flight attendants. See 572 F. Supp. 354, 371 (D.D.C. 1983), aff’d in part, rev’d in part on other

grounds, 746 F.2d 4 (D.C. Cir. 1984), overruled in part, Save Our Cumberland Mountains, Inc. v.

Hodel, 857 F.2d 1516 (D.C. Cir. 1988). The case established the Laffey fee matrix, which

“recommends a presumptive maximum hourly rate for Washington, D.C.-area attorneys engaged

in ‘complex federal litigation.’” Makray v. Perez, 159 F. Supp. 3d 25, 31 (D.D.C. 2016) (citing

Laffey, 572 F. Supp. at 372). “The original Laffey matrix presented a grid which established hourly

rates for lawyers of differing levels of experience during the period from June 1, 1981 through

May 31, 1982.” Salazar v. District of Columbia, 123 F. Supp. 2d 8, 13 (D.D.C. 2000). The rates

were determined by “inquiring into the billing rates of firms in Washington, D.C., which were

engaged in active litigation practice in the federal courts.” DL v. District of Columbia, 924 F.3d

585, 589 (D.C. Cir. 2019). “The Court of Appeals accepted the 1981–1982 matrix in [Save Our

Cumberland Mountains], 857 F.2d at 1525, and the parties to that case updated it through May 31,

1989, as part of a settlement.” Salazar, 123 F. Supp. 2d at 13 (citing Covington v. District of

Columbia, 839 F. Supp. 894, 898 (D.D.C. 1993)). To calculate fees in subsequent years, the

USAO updated the Laffey rates for inflation by using the Consumer Price Index for All Urban

25
Consumers (“CPI-U”) 10 of the United States Bureau of Labor Statistics. See USAO Laffey Matrix–

2014–2015. 11

Over the last several decades, much ink has been spilled in this Circuit on the validity of

the Laffey rates. While many disputes have “revolved around whether a case [is] sufficiently

complex to warrant Laffey rates,” DL, 924 F.3d at 589 (citing as example Reed v. Dist. of

Columbia, 843 F.3d 517, 525–26 (D.C. Cir. 2016)), as the years have gone on, the focus has turned

to the accuracy of the underlying data and how to properly update the matrix to account for

inflation, see, e.g., id. at 590; Eley v. Dist. of Columbia, 793 F.3d 97, 101 (D.C. Cir. 2015); Salazar

v. District of Columbia, 809 F.3d 58, 64–65 (D.C. Cir. 2015). Specifically, plaintiffs’ attorneys

have increasingly argued that the inflationary index used to update the 1983 Laffey rates, which

measures inflation for commodities generally, has “failed to capture the true rate of inflationary

change” for legal services. DL, 924 F.3d at 589. “[L]ess than 0.325 percent of the data in the CPI-

U involves legal services,” Eley, 793 F.3d at 101 (cleaned up), which is why many “critics [of the

USAO Laffey Matrix] have advocated, to some degree of success, for a competing Laffey Matrix

(LSI Laffey Matrix) that uses the Legal Services Index [“LSI”] of the Bureau of Labor Statistics

to adjust for inflation,” id. (citing Salazar v. Dist. of Columbia, 123 F. Supp. 2d at 15 (finding that

the LSI Laffey Matrix “more accurately reflects the prevailing rates for legal services in the D.C.

community”)). 12 Plaintiffs here advance a similar argument. They contend that the USAO Laffey

rates used to calculate their interim fee awards failed to keep pace with “prevailing market rates

10
“The CPI-U measures inflation across 100,000 commodities including food, fuel, and housing for a given
geographic area.” Eley v. Dist. of Columbia, 793 F.3d 97, 101 (D.C. Cir. 2015) (cleaned up).
11
U.S. Dep’t of Justice, Laffey Matrix – 2014-2015, http://www.justice.gov/sites/default/files/usao-
dc/legacy/2014/07/14/Laffey%20Matrix_2014–2015.pdf (last visited Oct. 11, 2020).
12
The court finds it peculiar that neither party mentioned the “LSI Laffey matrix” in their briefs, despite the
D.C. Circuit recently finding it to be more representative of complex federal litigation rates in Washington, D.C. See
DL, 924 F.3d at 590; accord Feld v. Fireman’s Fund Ins., No. 12-1789-JDB, 2020 WL 1140673, at *6 (D.D.C. Mar.
9, 2020) (“[T]he D.C. Circuit has recently confirmed that the applicable matrix for “complex federal litigation” in
D.C. is the LSI Laffey Matrix.” (citing DL, 924 F.3d at 589)).

26
for complex federal litigation in the District of Columbia.” Pls.’ Mot. at 41. As a consequence,

the interim lodestar understates the true market value of their work. This argument has merit.

It is well established that the USAO Laffey rates have failed to keep pace with the true rate

of inflation, which is why “[w]hen the two [are] pitted against each other, courts frequently [find]

the LSI Laffey matrix more persuasive.” DL, 924 F.3d at 589; see also Salazar, 809 F.3d at 65

(affirming the district court’s choice to apply the LSI Laffey matrix over the USAO’s). To

demonstrate the divergence between USAO Laffey rates and true market rates, Ms. Lang includes

as an Exhibit to her declaration the 2011 ALM Survey of Law Firm Economics (“2011 SLFE”),

which tracked general inflation and billing rates from 1985 to 2011. See Lang Decl., Ex. E, at

PDF p. 86. The survey shows that while the CPI-U (the measure of inflation used to update the

Laffey matrix) increased 110% during that time period, billing rates for fifth-year associates rose

207% over the same time, and rates for partners with 25 to 29 years of experience rose 195%. Id.

This difference is due to the fact that the CPI-U-updated USAO Laffey matrix is based on the

flawed assumption that “the rate for legal services in the Washington, D.C. area increases in

lockstep with the overall rise in the cost of all goods and services, including pizza delivery and

cleaning services for the area.” Eley v. District of Columbia, 999 F. Supp. 2d 137, 153 (D.D.C.

2015), vacated & remanded on other grounds, 793 F.3d 97. On the contrary, a comparison of the

Bureau of Labor Statistics data for the CPI-U and the LSI show that the “cost of legal services

nationally has far outstripped the increase in overall prices.” Id. 13 In fact, “[t]he nationwide cost

of legal services has jumped ninety-one percent, nearly twice as much as the general CPI, since

13
As the district court in Eley explained, this data can be verified on the Bureau of Labor Statistics website at
http://data.bls.gov/cgi-bin/srgate. For the legal services index, enter “CUUR0000SEGD01” into the text box; click
the “Next” button; under the Specify Year Range button, select “1997” from the “From:” drop down menu; click the
“Retrieve Data” button. For the general CPI, enter “CUUR0000SA0” into the text box; click the “Next” button; under
the Specify Year Range button, select “1997” from the “From:” drop down menu; click the “Retrieve Data” button.
See Eley, 999 F. Supp. 2d at 153.

27
1997.” Id. In his declaration, Crowell & Moring, LLP Partner Thomas P. Gies speaks to this

divergence from his personal experience:

When I first worked on this case in 1990, my standard billing rate
was $255 and the Laffey rate for an attorney with my level of
experience was $250, or 2% lower. The next year my rate was $275,
but the Laffey rate was $265, or 3.7% lower or $10 per hour below
market. In 1994 my rate was $290 but the Laffey rate was still $265
or 9% lower or $25 per hour below market. The gap continued to
widen with each year. My current standard billing rate is $1,020,
and this is the rate that I am regularly paid by clients that are billed
hourly for my services. The current USAO rate applicable to me is
$637, or 38% lower or $383 per hour below market.

Pl.’s Mot., Ex. 4, Decl. of Thomas P. Gies, ECF No. 1081-4, ¶ 49. The case law and evidence thus

conclusively establish that, over the life of this case, the USAO Laffey rates did not keep up with

market rates for complex litigation in Washington, D.C.

Precisely when those rates began to diverge is somewhat less certain. Plaintiffs offer some

evidence. The Lang Declaration reproduces a chart from the 2011 SLFE, which is reproduced

below.

28
Lang Decl. ¶ 64. The chart shows that rates for both partners in their twenty-fifth to twenty-ninth

years of practice and fifth-year associates began to diverge from the CPI factor used to annually

increase the USAO Laffey Matrix rates around 1998. That divergence continued to grow through

2011. The 2011 SLFE survey data thus shows that the USAO Laffey Matrix rates used to calculate

Plaintiffs’ interim fee awards have not always reflected counsel’s “true market value.” Purdue,

559 U.S. at 555.

But this rationale for an enhancement only goes so far. Plaintiffs seek a lodestar

enhancement for their work over the entire life of the case. The record, however, shows that such

a full-term enhancement, if applied, would overcompensate Plaintiffs’ counsel. For at least the

first 21 years of this case, Plaintiffs’ counsel either received the rates they asked for, or were

compensated with USAO Laffey rates that were virtually identical to then-prevailing market rates.

Through December 31, 1994, class counsel’s fee petitions, which the court largely approved, were

29
supported by affidavits stating that the requested hourly rates corresponded to actual billing rates

charged by Plaintiffs’ counsel from fee-paying clients for similar legal services. See Defs.’ Opp’n

at 24–27; see also, e.g., Mem. in Supp. of Pls.’ Second Mot. for Interim Attys.’ Fees & Expenses,

DEX 3, at 12. In the ensuing years, the parties agreed to compute the lodestar for the third and

subsequent interim petitions using either actual billing rates or rates from the Laffey matrix,

whichever was lower. The record shows that as late as June 30, 1998, those Laffey rates provided

a close approximation of market rates. See DEX 7, 1999 Malson Decl. ¶¶ 41–42 (stating that the

non-contingent rates actually charged by a Crowell & Moring associate during that time were only

“somewhat different” than the hourly Laffey rate applied to the seventh fee petition (covering 1/1

to 6/30/1998)). 14 In sum, for the services rendered from the start of the case through June 1998—

the first two decades—Plaintiffs’ counsel were paid based on their true market value.

In their reply brief, Plaintiffs seek to downplay the representations made in these early

petitions about market rates. They emphasize that those rates were for “noncontingent matters

where fees [were] paid promptly.” Pls.’ Reply at 19 n.19. But the Supreme Court in Dague

rejected such a distinction, explaining that “an enhancement for contingency would likely

duplicate in substantial part factors already subsumed in the lodestar.” 505 U.S. at 562.

Specifically, the Court explained, “[t]he risk of loss in a particular case (and, therefore, the

attorney’s contingent risk) is the product of two factors: (1) the legal and factual merits of the

claim, and (2) the difficulty of establishing those merits. The second factor, however, is ordinarily

reflected in the lodestar—either in the higher number of hours expended to overcome the difficulty,

14
The seventh fee petition was the last formal interim fee petition submitted by Plaintiffs. All subsequent fee requests
were resolved informally. Thus, the record of this case does not include contemporaneous data about historic market
rates for periods after June 30, 1998. See Defs.’ Opp’n at 30.

30
or in the higher hourly rate of the attorney skilled and experienced enough to do so.” Id. Stated

differently, the additional risk associated with a contingency fee is already baked into the lodestar.

In the end, this means that, for the period at least through the seventh fee petition, Plaintiffs

cannot reasonably contend that the lodestar “does not adequately measure the attorney’s true

market value.” Perdue, 559 U.S. at 545–55. Accordingly, no enhancement to the lodestar for that

period is warranted. As the Supreme Court said in Perdue, “[t]here [is] nothing unfair about

compensating [ ] attorneys at the very rates they requested.” Id. at 557 n.7. But for the time period

encompassing the eighth through the final interim fee petition—when the USAO Laffey rates were

below the prevailing market rates for complex civil litigation in Washington, D.C.—Plaintiffs have

the better argument that the interim lodestar does not reflect their true market value.

c. Defendants’ responses

For their part, Defendants resist applying any enhancement to the interim fees lodestar for

any period of time. They concede an eventual divergence between Laffey rates and market rates,

but insist that no enhancement is required for two reasons. First, they assert that for much of the

life of the case, the USAO Laffey rates were comparable to market rates for complex federal

litigation in Washington, D.C. Defs.’ Opp’n at 27–29. Second, they maintain that “any significant

divergence . . . did not occur until well after the litigation in Hartman was over.” Id. at 31. More

to the point, they contend that Plaintiffs’ attorneys were not engaged in complex federal litigation

after the parties entered the 2000 Consent Decree, and therefore they are not entitled to prevailing

market rates for such work. Id.

As to their first argument, Defendants cite Berke v. Bureau of Prisons, 942 F. Supp. 2d 71

(D.D.C. 2013), “as one of many cases” they say supports their position that the Laffey rates were

“relevant evidence of the ‘prevailing’ rates in the Washington, D.C. area” at least as of 2012. Id.

31
at 30. In Berke, the court found that rates charged by five attorneys with the law firm Ballard

Spahr in a civil rights case in 2012 were “very much ‘in line’ with [the Laffey rates].” Id. (quoting

Berke, 942 F. Supp. 2d at 78). Comparing Ballard Spahr to Plaintiffs’ counsel at Crowell &

Moring, Defendants argue that the findings of Berke apply with equal force to this case through

2012. Id. at 30 n.12.

The court is not convinced. As discussed, the court finds the 2011 SLFE and Bureau of

Labor Statistics CPI data persuasive. It shows that the “nationwide cost of legal services has

jumped . . . nearly twice as much as the general CPI, since 1997.” Eley, 999 F. Supp. 2d at 153.

The 2011 SLFE data also shows that in Washington, D.C., market rates for senior partners and

mid-career associates began to outstrip the CPI inflationary index in 1998. See supra, at 29. This

data identifies the divergence between prevailing market rates and USAO Laffey rates as occurring

more than a decade before Defendants concede a divergence occurred. Defendants’ reliance on

various district court cases from 2007 to 2011 that used the USAO Laffey Matrix to compute fees

does not compel a different result, see Defs.’ Opp’n at 30–31, as none of those cases took a hard

look at whether the USAO Matrix during that time period reflected actual market rates for complex

civil litigation in Washington, D.C., see, e.g., Heller v. District of Columbia, 832 F. Supp. 2d 32,

48 (D.D.C. 2011) (relying on the “the frequency with which the USAO Laffey Matrix rates are

applied to be strong evidence of both their prevalence and their reasonableness”).

Defendants’ second argument—that Plaintiffs’ counsel are not entitled to rates for complex

federal litigation during the post–Consent Decree phase of the case—is equally unconvincing.

Defendants contend that “[a]fter the parties signed the Consent Decree on March 21, 2000, the

litigation between Plaintiffs and Defendants ceased, and the [case] entered a distinct and lengthy

non-litigation phase.” Defs.’ Opp’n at 31. The relative simplicity of that work, Defendants

32
maintain, does not justify complex rates. See id. That assertion, however, is flawed for two

reasons.

First, it mischaracterizes the record. Following approval of the Consent Decree on July 12,

2000, see Mem. & Order, July 12, 2000, ECF No. 917, the in-court legal work did not cease. In

fact, by Defendants’ own telling,

[f]our women (the “Petitioners”) who had unsuccessfully petitioned
to be included in the class, appealed from that decision [approving
the Consent Decree], contending that the original class notice was
constitutionally defective. That contention was rejected by the
Court of Appeals, which summarily affirmed the District Court’s
approval of the Consent Decree on March 15, 2001, and denied
Petitioners’ rehearing petitions on June 14, 2001. Petitioners’ [then
filed a] petition for a writ of certiorari[, which] was denied by the
Supreme Court on January 7, 2002. See Dillon v. Powell, 2001 WL
41046 (D.C. Cir.), cert. denied, 534 U.S. 1078 (2002).

Defs.’ Opp’n at 9. Thus, for at least a year and a half following entry of the Consent Decree, class

counsel were engaged in what would be considered traditional litigation. They also were involved

in additional work to make the settlement meaningful to the class, including individual remedy

proceedings, which Plaintiffs note “were often highly complex, involving interpretation of federal

regulations and their application to individual claimants’ employment history.” Pls.’ Reply at 20.

Defendants do not dispute this characterization.

Second, Defendants’ position finds no support in the law. The only case that Defendants

cite for the proposition that “it is proper for the Court to analyze [the post–Consent Decree] phase

of the case separately to determine a reasonable attorney fee award” is In re Olson, 884 F.2d 1415

(D.C. Cir. 1989), but that case is inapposite. Defs.’ Opp’n at 33. Olson involved the Independent

Counsel Reauthorization Act, which authorized recovery of legal fees that would not have incurred

“but for” the operation of the Act. 884 F.2d at 1419. The court’s discussion of phases in that case

was limited to whether the fees sought for each phase met the “but for” standard. Id. at 1419–22.

33
No portion of Olson supports the proposition advanced here that an otherwise concededly complex

case is divisible into complex and non-complex phases for purposes of awarding fees.

Moreover, the closest Supreme Court precedent conflicts with Defendants’ position. In

Delaware Valley I, a case arising under the Clean Air Act, the losing party argued that certain

“work performed after issuance of the consent decree” was not fairly categorized as part of the

litigation and therefore was not compensable. Delaware Valley I, 478 U.S. at 553, 557–58. The

Court rejected that argument. In doing so, the Court analogized the Clean Air Act’s fee-shifting

provision with that of the Civil Rights Act of 1976, 42 U.S.C. § 1988, and agreed with those courts

that had held under § 1988 that “post-judgment monitoring of a consent decree is a compensable

activity for which counsel is entitled to a reasonable fee.” Id. at 559. The Court also found no

pre- and post-settlement distinction on the facts. Phase II and IX of the case involved, respectively,

monitoring of the consent decree and hearings before a federal agency. The Court acknowledged

that those phases “were not ‘judicial’ in the sense that they did not occur in a court room or involve

‘traditional’ legal work such as examination of witnesses or selection of jurors for a trial.” Id. at

558. Yet, the Court held that “the work done by counsel in th[o]se two phases was as necessary

to the attainment of adequate relief for their client as was all the earlier work in the courtroom

which secured [their client’s] initial success in obtaining the consent decree” and therefore was

compensable all the same. Id. So, too, here. The work done by class counsel after entry of the

Consent Decree in March 2000 was “as necessary to the attainment of adequate relief” as their

prior work. If the post-settlement phase of the case was less complicated, it surely involved fewer

hours, which would be reflected in a reduced lodestar.

In light of the foregoing, the court finds that Plaintiffs have provided “specific evidence”

that the rates used to calculate the lodestar for Plaintiffs’ eighth through twenty-eighth interim fee

34
petitions (covering years 1998–2018) “[did] not adequately measure the attorney[s’] true market

value.” Perdue, 559 U.S. at 554–55. An enhancement of the lodestar for those petitions is

therefore necessary to “approximate the fee that the prevailing attorney would have received if he

or she had been representing a paying client who was billed by the hour in a comparable case.”

Id. at 551.

2. Delay

The second ground that Plaintiffs say justifies an enhancement of the interim fees lodestar

is delay in receiving payment. In Perdue, the Court explained that “enhancements to compensate

for delay in reimbursement for expenses must be reserved for unusual cases,” because “when an

attorney agrees to represent a civil rights plaintiff who cannot afford to pay the attorney, the

attorney presumably understands that no reimbursement is likely to be received until the successful

resolution of the case.” Id. at 555. In elaborating on what might constitute an “unusual case,” the

Court stated that “there may be extraordinary circumstances in which an attorney’s performance

involves exceptional delay in the payment of fees,” id. at 556, and in those cases, “[c]ompensation

for [ ] delay is generally made either by basing the award on current rates or by adjusting the fee

based on historical rates to reflect its present value,” id. (cleaned up). It also suggested that “an

enhancement may be appropriate where an attorney assumes these costs in the face of

unanticipated delay, particularly where the delay is unjustifiability caused by the defense.” Id.

In one sense the delay in this case has been “exceptional,” but in another sense, not at all.

Plaintiffs’ counsel have waited more than 40 years to receive a final fee award. And it was not

until 18 years after the case commenced that they received their first fee payment. Those are long

periods of delay by any measure. But the significance of those delays is tempered substantially by

three factors. The first is that, starting in 1995, Plaintiffs began to seek and receive interim fee

35
payments every six to twelve months. See Lang Decl. ¶ 31. Plaintiffs have received 28 interim

payments, totaling nearly $23 million. Second, Plaintiffs were awarded prejudgment interest on

all payments after 1991, thus compensating them to some degree for delay. See DEX 10-13, at

4–5; Hartman, 8 F. Supp. 2d 1 (finding interest rate on one-year Treasury bills was appropriate).

Third, the two long periods of delay of 18 and 40 years are largely attributable to choices made by

Plaintiffs, not impropriety or foot-dragging by the defense. Plaintiffs acknowledge that they

waited until 1993 to file their first fee petition because they did not wish to be awarded fees until

the issue of liability was finally resolved by the Circuit. See Suppl. Fredrickson Decl. ¶¶ 5–8.

And, as for waiting more than 40 years for a final fee award, Plaintiffs agreed to that course as part

of the Consent Decree in 2000. Perhaps they never conceived that they would have to wait two

decades for a final resolution, but at the same time, Plaintiffs never sought to accelerate any final

fee issues. So, although Plaintiffs have experienced delay in receiving final payment, whether this

case involves the type of “extraordinary circumstances” that warrant an enhancement for

“exceptional delay” is far from certain. Perdue, 559 U.S. at 556.

None of this is to say that Plaintiffs are not entitled to some additional fees to account for

delay, for two reasons. One, as discussed, because the interim fee calculations for two decades

relied on the USAO Laffey Matrix, and those matrix rates began to diverge from prevailing market

rates in or around 1998, Plaintiffs have been receiving payments based on sub-market rates for

20-plus years. Plaintiffs are entitled to additional fees to make up for the shortfall, and any such

make-up payment should reflect its delayed receipt. Two, Plaintiffs assert that the interest on fee

payments made thus far inadequately compensates them for delay. See Pls.’ Reply at 19 n.19

(arguing “interest on a submarket lodestar does not make plaintiffs whole”). Defendants respond

that Plaintiffs have not challenged the prejudgment interest rates in their current fee motion, and

36
the prejudgment interest payments are “now final and no longer subject to appeal, reconsideration,

or review.” Defs.’ Opp’n at 41. But at oral argument, Defendants conceded that the appropriate

interest rate was an issue preserved for adjudication of the final fee award. See Hr’g Tr. at 52:14–

53:19. That concession is consistent with the terms of each interim fee petition, in which each

party expressly reserved the right to contest, among other things, “the appropriate interest rate.”

Pls.’ Mot. at 4–5 (quoting Ninth Fee Petition Stip. at 1–2). Thus, it remains a live question what

type of an enhancement is necessary to compensate Plaintiffs’ counsel for any deficiency arising

from an inadequate interest rate.

C. Calculating an Appropriate Enhancement, or the “True Lodestar”

Thus far, the court has concluded that Plaintiffs’ total interim fee award, or interim lodestar,

falls short of a “reasonable” fee two reasons, and a possible third: (1) below-market USAO Laffey

Matrix rates were used to calculate interim fee awards for over 20 years, (2) there has been a delay

in receiving whatever amount is appropriate to compensate them for the shortfall caused by use of

those reduced rates, and (3) the interest thus far received on interim fee awards arguably has been

inadequate to compensate for delay. The question then becomes how to enhance the interim

lodestar to make up for these shortfalls. Or, perhaps more accurately stated: What is the “true

lodestar” in this case? Recall that Perdue held that where the original lodestar “does not adequately

measure the attorney’s true market value, . . . the trial judge should adjust the attorney’s hourly

rate in accordance with . . . [the] prevailing market rate.” 559 U.S. at 554–55. The true lodestar

in this case thus would reflect prevailing market rates when the fees were incurred.

37
 By Plaintiffs’ calculation, the true lodestar in this case is $75,232,463.96—more than three

times the total interim fee payment. See Pls.’ Mot. at 43; Lang Decl. ¶ 75 & Ex. D. How they get

to this amount is both labor intensive and creative.

Plaintiffs first tabulated the individual hours worked by each lawyer and paralegal/law

clerk (“paraprofessional”) who devoted time to the case from its inception. That total is an

astonishing 116,783.23 hours: over 45,000 hours for partners, over 51,000 hours for associates,

and nearly 20,000 for paraprofessionals. See Lang Decl., Ex. D., at PDF p. 66. Next, Plaintiffs

estimated an hourly rate for each of those three categories of timekeepers. That hourly rate is

drawn from the 2011 SLFE, 15 which was the basis for the USAO Matrix that replaced Laffey in

2015. Although that Matrix was deemed invalid by the D.C. Circuit in DL v. District of Columbia,

924 F.3d 585, Plaintiffs rely on its underlying data. As Plaintiffs explain:

In 2017, [during the DL litigation] the United States endorsed the
2011 SLFE as containing “one of the most complete, accurate, and
up-to-date sets [o]f economic statistics and financial data available
about law firms.” Statement of Interest of the United States, DL v.
District of Columbia, Case No. 1:-05-cv-01437, ECF No. 564, at 3
n.2 (Apr. 27, 2017). Though the new [USAO] matrix itself has been
invalidated by the D.C. Circuit, DL v. District of Columbia, 924 F.3d
585 (D.C. Cir. 2019), the underlying data provides relevant rates for
complex federal litigation in the District of Columbia as recently as
2010. The SLFE gathered data on rates for mid-level associates and
partners and gave rate values for the lower quartile, median, upper
quartile, and ninth decile. Given the complexity of this case and the
superior skill manifested by class counsel, as well as the original
Laffey characterization of the proper market rate as that for the most
experienced, highly respected, and capable attorneys, the ninth
decile rate for the District of Columbia is the most appropriate for
the timekeepers in this case.

Pls.’ Mot. at 43.

15
Discussed supra, at 27.

38
 To unpack that description a bit, to identify the true market rate for their work, Plaintiffs

rely on the Washington, D.C.-specific data gathered by the 2011 SLFE. The 2011 SLFE identified

timekeeper rates in 2010 based on various years of experience (the survey identified nine

experience bands) across different types of law practices (not just complex civil litigators), and it

isolated four different data points of rates: lower quartile, median, upper quartile, and ninth decile.

See Lang Decl., Ex. E, at PDF pp. 78–80, 104. The survey did not have data for every years-of-

experience grouping for Washington, D.C. firms, but it did have robust data for associates with

four or five years of experience and for partners with 16 or more years of experience, which were

broken down into four timekeeper categories (one for associates and three for partners). See id. at

PDF p. 104. The survey also provided composite rates for just two categories of Washington, D.C.

attorneys: “Associate/Staff Lawyer” and “Partner/Shareholder-Equity/Non-Equity.” See id. at

PDF p. 100. Plaintiffs used those composite rates, calculated at the ninth decile, on the assumption

that complex civil litigators would command the highest rates in Washington, D.C. Id. ¶ 73. As

for paraprofessionals, because the 2011 SLFE did not capture their rates, Plaintiffs used the

average rate for those timekeepers charged in 2010 by two firms that worked on the matter, Steptoe

& Johnson and Crowell & Morning. Id. ¶ 74. The resulting 2010 hourly rates were $204.50 for

paraprofessionals, $450 for associates, and $675 for partners. See id., Ex. D. at PDF p. 66.

Plaintiffs then multiplied those rates by the total hours worked for each timekeeper category to

come up with a “9th Decile 2010 Lodestar” of $57,871,126.13. Id. Finally, Plaintiffs updated the

2010 lodestar to 2019 dollars by applying an inflation factor of 1.3. Id. ¶ 75. 16 The resulting true

16
Plaintiffs use the Producer Price Index – Office of Lawyers (“PPI-OL”), also published by the Bureau of Labor
Statistics, to adjust for inflation. See Lang Decl. ¶ 10.

39
lodestar in 2019 dollars, according to Plaintiffs, is $75,232,463.96. Id. A chart summarizing these

calculations is reproduced below.

Position Hours 2010 9th Decile SLFE 9th Decile 2010
DC Rate Lodestar
PARAPROFESSIONAL* 19,833.75 $ 204.50 $ 4,056,001.88
ASSOCIATE 51,670.11 $ 450.00 $ 23,251,549.50
PARTNER 45,279.37 $ 675.00 $ 30,563,574.75
TOTAL 116,783.23 $ 57,871,126.13
adjusted to 2019 Rates using
PPI/OL, dividing August 1.300 $ 75,232,463.96
2019 number by Dec. 2010
number

Id., Ex. D., at PDF p. 66.

Plaintiffs’ true lodestar calculation suffers from at least three problems. First, and most

glaringly, nowhere do Plaintiffs justify the logic of applying updated 2010 rates to all hours worked

on the case. That approach might be justifiable if Plaintiffs had not received a penny in interim

fees and were seeking to fully compensate themselves for delay by basing the final award on

current rates. But, of course, Plaintiffs have received multiple interim fee awards along the away,

and they cannot enlarge the final award by applying current rates or an inflationary adjustment to

hours worked for which compensation has not been delayed. Second, Plaintiffs’ approach fails to

account for the fact that, at least through 1998, they received compensation at or near market rates

for the hours they committed to the case. They were not undercompensated due to a below-market

rate for those years; they received exactly what they sought. Third, as for the years in which the

USAO Laffey Matrix Rate diverged from prevailing market rates, arbitrarily using 2010 rates and

applying an inflationary multiplier to the hours worked in those years would result in

overcompensation. That is particularly true for the earlier years of that divergence, when the delta

between Laffey rates and the prevailing market rate was less pronounced.

40
 Without applying the updated 2010 rates to all hours worked in the case—an approach the

court has now rejected—Plaintiffs’ justification for using three composite rate categories falls

apart. See Lang Decl. ¶ 72 (explaining that “use of these general rate categories provides a good

approximation of what the lodestar would be” because “although the . . . rate will be high for

associates in their earliest involvement,” since many of them “stayed with the case for many

years,” the use of the composite rates “for all years gives a reasonable approximation of the

lodestar” (emphasis added)). Plaintiffs offer the expert opinion of Professor William Rubenstein,

the author of the leading treatise on class actions, Newberg on Class Actions, to support their

methodology. See Pls.’ Mot., Expert Report of Prof. William B. Rubenstein, ECF No. 1081-1

[hereinafter Rubenstein Report]. In his expert report, Professor Rubenstein says that “given the

40-year arc of this case, re-calculating a new lodestar hour by hour, time-keeper by time-keeper,

would be unduly burdensome, if not outright impossible, and the use of the three-rate approach is

a sensible and efficient approach.” Rubenstein Report ¶ 38(a). That is no doubt true to some

extent. But convenience cannot justify overcompensating class counsel to the tune of millions of

dollars. Additionally, having reviewed cases in this Circuit where the court adopted the LSI-

adjusted matrix, for example, it does not seem like it would be unduly burdensome to apply that

matrix, or something similar, which contains rates at a greater level of specificity than the 2011

SLFE, and which has already been found reasonable for complex federal litigation. See DL, 924

F.3d at 589. To the extent the LSI-adjusted matrix requires information pertaining to years of

experience to discern the hourly rate, Exhibit C of Ms. Lang’s declaration appears to provide that

information. It details, by fee petition, the hours billed by each attorney, including their years

since graduation from law school. See Lang Decl., Ex. C, at PDF pp. 36–51. Thus, the necessary

data for an LSI-adjusted matrix calculation, though perhaps cumbersome, would appear to exist.

41
 Professor Rubenstein also points to the NFL litigation in which he served as an expert to

support the reasonableness of Plaintiffs’ composite rates by showing that the average hourly rate

under Plaintiffs’ calculation ($574.17) is well below the “blended rate” the court used in the NFL

litigation ($623.05). See Rubenstein Report ¶ 39. But unlike this case, the NFL litigation involved

work done by more than twenty law firms, and the hourly billing rates by partners across the firms

diverged by $850. In re Nat’l Football League Players’ Concussion Injury Litig., No. 12-MD-

02323-AB, 2018 WL 1635648, at *8–9 & n.8 (E.D. Pa. Apr. 5, 2018). Relying on an approach

endorsed by the Third Circuit, the NFL court chose to average the rates of all partners, associates,

and paralegals to arrive at a “blended rate” for the purposes of the lodestar cross-check. Id. at *9.

But in this case, the court is not calculating an updated lodestar for a cross-check, so this

comparison is not useful. As the NFL court explained, “[s]ince the lodestar cross-check is ‘not a

full-blown lodestar inquiry,’ the evaluation can be based on summaries and less precise

formulations.” Id. at *8 (quoting In re Rite Aid Corp. Sec. Litig., 396 F.3d 294, 307 n.16 (3d Cir.

2005)). Indeed, that is in part the problem with Plaintiffs’ suggested calculation overall—it is

provided primarily as a cross-check on the percentage-of-the-fund award they seek and is not

precise enough.

* * *

In light of the foregoing, Plaintiffs need to go back to the drawing board. They bear the

burden of “identifying a factor that the lodestar does not adequately take into account and proving

with specificity that an enhanced fee is justified.” Purdue, 559 U.S. at 546. Although it is apparent

that an adjustment to the lodestar for the eighth through twenty-eighth fee petitions (covering years

1998–2018) is necessary to “approximate[ ] the fee that the prevailing attorney would have

received if he or she had been representing a paying client who was billed by the hour in a

42
comparable case,” the court lacks the information necessary to “adjust the attorney’s hourly rate

in accordance with specific proof linking the attorney’s ability to a prevailing market rate.” Id. at

551. Furthermore, although some additional compensation is appropriate to account for delay of

amounts unpaid, Plaintiffs have not proposed “a method that is reasonable, objective, and capable

of being reviewed on appeal” to calculate such amount. Id.

Although the court denies Plaintiffs’ request for a final attorneys’ fee award at this juncture,

the court hopes that its rulings will assist the parties in reaching a resolution.

V. CONCLUSION AND ORDER

For the reasons set forth above, the court denies without prejudice Plaintiffs’ Motion for a

Final Determination of Attorneys’ Fees, ECF No. 1080. The parties shall appear for a status

hearing on November 13, 2020, at 10:30 a.m., to discuss further proceedings in this matter.

Dated: November 3, 2020 Amit P. Mehta
United States District Court Judge

43