Kerner v. City and County of Denver

Filing 266

SUPPLEMENTAL OPINION AND ORDER AWARDING ATTORNEY FEES AND COSTS by Judge Marcia S. Krieger on 5/24/19. The Plaintiffs' motion for reconsideration 249 is denied. (dkals, )

IN THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLORADO
Honorable Marcia S. Krieger
Civil Action No. 11-cv-00256-MSK-KMT
MARIAN G. KERNER; and
ROMONA J. LOPEZ, on behalf of themselves and all others similarly situated,
Plaintiffs,
v.
CITY AND COUNTY OF DENVER,
Defendant.
______________________________________________________________________________
SUPPLEMENTAL OPINION AND ORDER AWARDING ATTORNEY FEES AND
COSTS
______________________________________________________________________________
THIS MATTER comes before the Court pursuant to the Plaintiffs’ motion (# 262)
seeking reconsideration of the Court’s October 22, 2018 Supplemental Opinion and Order (#261)
awarding attorney fees, Denver’s response (# 264), and the Plaintiffs’ reply (# 265).
The Court assumes the reader’s familiarity with the proceedings to date. The Court has
considered the question of attorney fees on two prior occasions in this case. On May 26, 2017,
the Court awarded (# 249) the Plaintiffs attorney fees in the amount of $894,044. On appeal, the
10th Circuit vacated that award, finding that this Court had not adequately considered the
reasonable number of hours expended by the Plaintiffs’ counsel, and remanded the matter for
further findings. On October 22, 2018, the Court found that a 40% reduction in all hours claimed
by all billing entities was appropriate, yielding a fee award of $928,825.
In the instant motion, the Plaintiffs take issue with three components of the Court’s
October 2018 Order: (i) the Plaintiffs contend that no reduction should have been made to the

1

number of paralegal hours claimed because Denver had not previously objected to that number
of hours; (ii) the Plaintiffs contend that, at most, a 30% reduction in the hours claimed by
counsel would be appropriate because Denver previously agreed to such a reduction; and (iii) the
Court made a mathematical error in its final calculations.
The Court rejects the first two contentions. In assessing the reasonableness of a claim for
fees, the number of hours conceded by the opposing party is a relevant consideration but not a
dispositive one. In other words, “the court’s discretion is not absolutely constrained by the
amount of a fee request put in controversy by the parties” and the court “may well decide to go
below” even the amount conceded by the non-movant. Robinson v. City of Edmond, 160 F.3d
1275, 1285-86 (10th Cir. 1998). Indeed, the 10th Circuit cited Robinson for this very proposition
in its prior order in this case. Kerner v. City and County of Denver, 733 Fed.Appx. 934, 937
(10th Cir. 2018). And though it expresses particular concern where a court applies a “blanket
reduction ratio” when dropping below a figure conceded by the non-movant – as this Court did
in the October 2018 Order – Robinson does not indicate that such a conclusion is per se error;
rather, it is “a factor to consider in deciding whether the court abused its discretion in the
magnitude of the cuts ordered.”
Nevertheless, the Court deems it appropriate here to amplify the reasoning underlying
this Court’s fee award(s). The Court begins by repeating its finding that the number of hours
claimed by the Plaintiffs in this case has always been excessive. As discussed in some detail in
the Court’s 2017 fee order, the primary dispute in this case was not liability, but damages. One
of the primary aspects of the damage controversy was a difference in methodology used by
experts retained by the parties. Despite years of dispute as to calculation of damages, the
Plaintiffs’ damage expert, Dr. Bardwell, abandoned his own damage model during the Plaintiffs’

2

rebuttal presentation. He adopted many features of Denver’s expert’s model (reducing the
damages Plaintiffs claimed in their case in chief by more than half). This last-minute change by
Dr. Bardwell rendered a great deal of what had come before – not just during the trial, but in
pretrial proceedings as well – irrelevant. Because Dr. Bardwell’s initial damage model and
calculations were abandoned by the end of the case, it follows that many of the hours that the
Plaintiff’s counsel and staff expended in preparing and presenting that abandoned model cannot
be said to have been reasonably incurred, and should therefore be excluded from a fee award.
For the same reasons that attorneys do not generally bill for researching or developing factual or
legal theories that eventually prove fruitless and are never presented to a court, it seems
inappropriate for the Plaintiffs’ counsel to bill for time spent developing and presenting a
comprehensive damage model that they abandoned by the end of trial. This is particularly true
because the opposing expert opinion and model had been available to the Plaintiffs long before
the trial began.
These circumstances are indeed extraordinary and the Court has attempted to grapple
with them in various ways. In its 2017 Order, after initially assessing the record and concluding
that any fee award the Court would calculate was likely to be below the amount of fees that
Denver was concededly willing to pay, the Court cut the process short and simply awarded the
amount proposed by Denver. When the 10th Circuit took issue with that approach, this Court
returned to the question in its October 2018 Order. Because this Court had rejected Denver’s
proposed hourly rates, the Court elected not to consider Denver’s arguments regarding the
number of hours that were tied to those rates. As this Court has previously held on several
occasions (including in its 2017 Order), a reasonable hourly rate and the reasonable number of
hours incurred on any given task are matters that exert some influence upon each other. An

3

experienced attorney claiming a particularly high hourly rate will be expected to work with a
level of efficiency and economy that would not be expected from a less-experienced attorney
charging a lower hourly rate. Thus, once the Court rejected Denver’s proposed hourly rates,
Denver’s position on the reasonable number of hours incurred by the Plaintiffs became less
persuasive.1
Instead, the Court considered its own experience as trier of fact when evaluating the
reasonableness of the Plaintiffs’ claimed hours. Robinson, 160 F.3d at 1285-86. It has long been
clear that a line-by-line examination of the Plaintiffs’ billing records would be a futile task, both
because of the length of this litigation and the inability to surgically excise vast amounts of time
that were devoted to developing and presenting Dr. Bardwell’s since-abandoned methodology.
Lacking precise tools to determine the number of hours reasonably expended by the Plaintiffs,
the Court was left with the blunt instrument of blanket reductions. Contrary to the Plaintiffs’
argument here, such an approach has repeatedly been permitted by the 10th Circuit. See
generally Case v. Unified School Dist., 157 F.3d 1243, 1250 (10th Cir. 1998), citing Mares v.
Credit Bureau of Raton, 801 F.2d 1197, 1203 (10th Cir. 1986).
Needless to say, there can be no precise means to fix a specific percentage for a
reduction. The Court is mindful that it may not simply “eyeball the fee request and cut it down
by an arbitrary percentage.” Robinson, 160 F.3d at 1281. At the same time, the nature of an

1

Put differently, the Court ascribes some merit to Denver’s argument here that its
concession to a 30% reduction in certain claimed hours was made “in tandem with” its argument
that the remaining hours claimed by the Plaintiffs should be billed at a reduced rate. Because
rates and reasonable hours are somewhat elastically-connected, the Court’s rejection of Denver’s
argument for reduced rates necessarily suggests that the number of hours claimed by the
Plaintiffs at the higher hourly rates should be more closely scrutinized than they might have been
at the lower rates Denver proposed. The Plaintiffs having benefitted from a higher rate than
Denver urged cannot now complain that the Court reduced the number of hours more
aggressively than Denver would have.
4

across-the-board reduction is that it is not susceptible to specific quantification. Here, as
discussed above, the Court was particularly troubled by the fact that the Plaintiffs’ damages case
was fundamentally reshaped mid-trial, and both the importance of the damages methodology and
differences in the methodology used by the expert witnesses was known to the parties long
before the trial occurred.
Issues related to damages substantially predominated over all other issues in this case.
By means of example, of the 59-page transcript of the Court’s oral ruling (#223) at the
conclusion of trial, 22 pages were devoted to analyzing the question of liability and 37 pages
were devoted to findings and conclusions relating to damages. Even then, owing to the parties’
shifting damage models, the precise quantification of damages required further briefing by both
sides and, ultimately, a supplemental 16-page written opinion (# 249) by detailing additional
findings by the Court regarding damages.2
The decision by the Plaintiffs to abandon one damage model and adopt a different one so
late in a case that reflects a troubling lack of focus in pre-trial and trial proceedings. No
explanation is offered as to why Dr. Bardwell changed his analytical approach and adopted
Denver’s at trial rather than at an earlier point in the pre-trial process. Few attorneys would bill
their clients for extensive fees where much of what counsel prepared as to the central issue in the
case was abandoned at trial and reframed on-the-fly. Certainly, the Plaintiffs are entitled to
recover some degree of fees incurred in preparing their original damage model, as it appears that

2

Another illustrative metric is that, of the roughly 1,000 entries in the Plaintiffs’ lead
counsel’s billing records, Docket # 245-2, the word “damages” appears in nearly 150 of those
entries. The Court also notes that, by the time of summary judgment motions, Denver had
essentially conceded the disparate impact claim as to 8 of the 21 job categories at issue, leaving
the issue of damages as the sole remaining issue to be resolved as to those plaintiffs. Docket #
188 at 14 n. 8.
5

Dr. Bardwell was able to salvage and port over to his revised model various data sets,
assumptions, and calculations that were prepared as part of his initial model. But because
damages were the central focus of much of this case and the Plaintiffs so dramatically changed
damage approaches mid-trial, the Court finds that a substantial reduction in claimed hours is
warranted. Based on the Court’s familiarity with the proceedings in this case, an across-theboard reduction in hours that approaches 40% suffices to reflect these circumstances.
On top of this, the Court has previously remarked upon other factors contributing to a
necessary reduction in claimed hours, including inefficiencies in the presentation of evidence and
the Plaintiffs’ limited degree of success relative to the huge damage figures they initially
proffered. Although these factors suggest that relatively modest adjustments to the number of
hours claimed (or, in the case of degree of success, subsequent adjustment to the lodestar figure)
are appropriate, those modest adjustments nevertheless contributed to nudge the Court towards
the 40% figure that it ultimately selected. All tolled, the Court is confident that the 40% across
the board reduction reflects a careful consideration of the circumstances of this case as they bear
upon the reasonable number of hours expended. For these reasons, the Court sees no basis to
modify its October 2018 Order.
The Court also rejects the Plaintiffs’ argument that the October 2018 Order contained a
mathematical error. The Court’s calculations are set forth in detail at page 2 of that order, and it
is not clear which particular term in the equation the Plaintiffs suggest is erroneous. From the
Court’s review of the Plaintiffs’ calculations in the instant motion, it appears that the alleged
error arises from the Plaintiffs using Denver’s brief as the source of the actual number of hours
the Plaintiffs claim. But the Court derived its calculations from the chart in the Plaintiffs’ own

6

initial fee motion: Docket # 245 at page 20.3 That motion indicates that the number of hours
claimed for Kenneth Padilla is 2,336 (not 2,359 as claimed in the current motion); 733 hours for
Joaquin Padilla (rather than 737 as claimed in the current motion), and 698 for Mr. Moya.
Because the Court based its calculations on the hours recited in the Plaintiffs’ motion, not
Denver’s response, the Court finds no calculation error requiring correction.
Accordingly, although the Court has reconsidered its October 2018 ruling in light of the
Plaintiffs’ arguments herein, the Court finds no reason to modify that order. The Plaintiffs’
motion for reconsideration (# 249) is DENIED.
Dated this 24th day of May, 2019.
BY THE COURT:

Marcia S. Krieger
Senior United States District Judge

3

Although the Plaintiffs contemplated submitting requests for additional fees incurred in
briefing the fee motion, they never submitted any additional billing records reflecting time
incurred after Docket #245.
7