Horvath v. United States

In the United States Court of Federal Claims
No. 16-688C
Filed: August 13, 2020
FOR PUBLICATION

MICHAEL HORVATH, individually, and
on behalf of the classes of federal Secret
Service agents and federal Diplomatic Keywords: RCFC 23;
Security Service agents similarly situated Class Certification;
to him, Commonality; Typicality;
Adequacy; Civilian Pay;
Plaintiff, Overtime; LEAP;
5 U.S.C. § 5542(e); Secret
v.
Service; Diplomatic Security
UNITED STATES, Service

Defendant.

Nicholas Michael Wieczorek, Clark Hill, PLC, Las Vegas, Nevada, for the plaintiff.

Galina I. Fomenkova, National Courts Branch, Civil Division, U.S. Department of Justice,
Washington, D.C., with whom were Jane M. Brittan, Office of the Chief Counsel, U.S. Secret
Service, and Rebecca C. Wulffen, Office of Employment Law, Office of the Legal Adviser, U.S.
Department of State, of counsel, for the defendant.

MEMORANDUM OPINION

HERTLING, Judge

Secret Service Special Agent Michael Horvath, on behalf of himself and similarly
situated U.S. Secret Service and Diplomatic Security Service special agents, seeks backpay he
alleges was illegally denied by a regulation that was inconsistent with the governing statute. The
Federal Circuit agreed with the plaintiff and held the regulation unlawfully required agents’
unscheduled hours of protective duty to be “consecutive” before triggering the two additional
hours of overtime pay guaranteed by 5 U.S.C. § 5542(e). See Horvath v. United States, 896 F.3d
1317 (Fed. Cir. 2018). The Federal Circuit has remanded the case for this Court to consider
“whether class certification is appropriate.” Id. at 1322.

The Secret Service has demonstrated, based on an analysis of its time and payroll records,
that Mr. Horvath was not denied overtime pay due to the application of the now-invalidated rule.
Mr. Horvath, however, seeks more. Based on discovery from the Secret Service, he argues that
he may have been denied overtime pay on some basis other than the invalidated regulation. He
argues that the Court should certify a broader class of all Secret Service and Diplomatic Security
Service special agents who performed protective duties and may have been denied § 5542(e) pay
on account of any basis at all and not just due to the invalidated regulation.

Neither the plaintiff’s complaint nor the record in the case supports Mr. Horvath’s effort
to certify this broader class of special agents. Despite providing examples of other special agents
who, like himself, were denied pay for consecutive unscheduled hours, Mr. Horvath has failed to
establish that this allegedly broader denial of § 5542(e) pay was attributable to a policy, practice,
or unexplained systematic and widespread error that meets the requirements of Rule 23 of the
Rules of the Court of Federal Claims (“RCFC”) for the United States to have acted on grounds
generally applicable to such a class.

Mr. Horvath’s ultimate lack of damages on the narrower claim that was expected to yield
variable damages for each class member is not necessarily fatal to class certification. In this
case, however, Mr. Horvath’s interests no longer sufficiently align with those of the potential
class members to make him an adequate representative of the putative class. Mr. Horvath’s
response to the government’s analysis questions the accuracy of the time and payroll records that
putative class members would rely on to prove their claims, potentially jeopardizing the
commonality of the class’s claims. Additionally, when crafting, negotiating, and litigating
discovery requests, Mr. Horvath’s interests in proving his individual claim based on consecutive
unscheduled hours may overshadow the class interest in proving claims for nonconsecutive
unscheduled hours.

Accordingly, the plaintiff’s motion to certify a class is denied.1

I. LEGAL BACKGROUND

Special agents of the Secret Service and the Diplomatic Security Service are regularly
scheduled to work 12-hour shifts protecting the President, Vice President, Secretary of State,
other dignitaries, candidates, or families of these individuals. The Court describes the three
statutory provisions that compensate special agents for their especially long workdays and then
describes the Secret Service time-reporting practices at issue in this case used to implement these
provisions.

A. Pay Statutes and Regulations

Special agents are exempt from the overtime-pay provisions of the Fair Labor Standards
Act (“FLSA”), but they are compensated for hours worked in excess of a standard eight-hour

Mr. Horvath’s failure to meet the requirements needed to certify a class does not limit his
1

residual, individual claim for all the overtime pay that he is due. See Horvath v. United States,
No. 16-688C, 2020 WL 1487642, at *4 (Fed. Cl. Mar. 25, 2020) (denying the defendant’s
motion for summary judgment because Mr. Horvath’s amended complaint identifies a money-
mandating provision of law and his response showed a genuine issue of material fact as to
whether he was due more pay under that provision than he actually received). The quantum of
the backpay owed to Mr. Horvath remains to be resolved.

2
workday by provisions of the Federal Employees Pay Act, 5 U.S.C. §§ 5541-5550b (“FEPA”).
FEPA’s Law Enforcement Availability Pay (“LEAP”) provision, § 5545a, augments some
special agents’ base salary by 25-percent. FEPA’s overtime provision, § 5542, pays time-and-a-
half for overtime hours.

1. § 5542(d)

For special agents who receive the LEAP premium, § 5542(d) increases the overtime
threshold from eight hours to 10 hours:

(d) In applying subsection (a) of this section with respect to any
criminal investigator who is paid availability pay under section
5545a—
(1) such investigator shall be compensated under such
subsection (a), at the rates there provided, for overtime work
which is scheduled in advance of the administrative
workweek—
(A) in excess of 10 hours on a day during such
investigator's basic 40 hour workweek; or
(B) on a day outside such investigator's basic 40 hour
workweek; and
(2) such investigator shall be compensated for all other
overtime work under section 5545a.

5 U.S.C. § 5542(d). As a result, special agents who receive LEAP pay are compensated for
hours 8 and 9 of their 12-hour shifts only with the fixed-amount LEAP premium and receive
time-and-a-half overtime pay only for the last two hours of a shift.

2. § 5542(e)

Section 5542(e)—the focus of this case—is an exception to § 5542(d). Section 5542(e)
reverts the overtime threshold for LEAP recipients performing protective duties from 10 hours
back to eight hours on days when an agent performs at least two hours of unscheduled overtime
work:2

2
The exception distinguishes overtime for “protective duties” from time special agents spend
investigating crimes like counterfeiting. For Secret Service special agents, “protective duties”
include protecting the President, Vice President, their immediate families; former Presidents and
Vice Presidents and their spouses and young children; visiting heads of foreign states or foreign
governments; “[o]ther distinguished foreign visitors to the United States”; “official
representatives of the United States performing special missions abroad” when the President

3
 (e) Notwithstanding subsection (d)(1) of this section, all hours of
overtime work scheduled in advance of the administrative
workweek shall be compensated under subsection (a) if that work
involves duties as authorized by section 3056(a) of title 18 or section
37(a)(3) of the State Department Basic Authorities Act of 1956, and
if the investigator performs, on that same day, at least 2 hours of
overtime work not scheduled in advance of the administrative
workweek.

5 U.S.C. § 5542(e). For a special agent performing protective duties on a shift of 12 scheduled
hours who must arrive early or stay late by at least two unscheduled hours, subsection (e) turns
hours 8 and 9, which otherwise would have only been compensated by LEAP, into overtime-pay
hours.

In summary, broken down by hour, a special agent who works a scheduled 12-hour shift
on a protective detail without additional unscheduled hours is paid:

• hours 1 through 8 at the base rate (x);

• hours 9 and 10 with the fixed-amount LEAP premium; and

• hours 11 and 12 at the overtime rate (1.5x).

directs such protection; and “Major Presidential and Vice Presidential candidates” and their
spouses. 18 U.S.C. § 3056(a).

“Protective duties” for Diplomatic Security Service special agents include “maintaining the
security and safety of . . . heads of a foreign state, official representatives of a foreign
government, and other distinguished visitors to the United States, while in the United States; the
Secretary of State, Deputy Secretary of State, and official representatives of the United States
Government, in the United States or abroad; [their family members]; foreign missions . . . and
international organizations . . . , within the United States; a departing Secretary of State . . . ; [a
presidential designee] to serve as Secretary of State, prior to that individual's appointment.” 22
U.S.C. § 2709(a)(3).

4
A special agent who works a scheduled 12-hour shift plus two additional hours on a protective
detail is paid:

• hours 1 through 8 at the base rate (x);

• hours 9 and 10 at the overtime rate (1.5x);

• hours 11 and 12 at the overtime rate (1.5x); and

• hours 13 and 14 (which were unscheduled) with the fixed-amount LEAP premium.

3. Invalidated Consecutive-Hours Rule

When Mr. Horvath filed this case in 2016, an Office of Personnel Management regulation
purported to implement § 5542(e) by requiring that a special agent’s two unscheduled hours on a
protective detail be “consecutive” in order to trigger the two additional hours of overtime pay.
See 5 C.F.R. § 550.182(b)(2) (the “consecutive-hours rule”).

B. Secret Service Pay Policies

Instead of calculating pay from records of special agents’ start and end times, the Secret
Service pays special agents based on their self-reported number of base, overtime, and LEAP
hours. These self-reported time entries reside in a database called WebTA. Separate “MARS”
reports track the start and end times of agents’ LEAP hours, but their contents do not affect a
special agent’s pay.

When completed, a special agent’s time entry for a 12-hour workday reads “8-2-2”—
eight base, two overtime, and two LEAP hours.

An agent would report a 12-hour scheduled shift lengthened by two unscheduled hours
(making it a 14-hour shift) with an 8-4-2 time-entry. An 8-4-2 time-entry nets an agent three
additional hours of base pay (two overtime hours) than he would otherwise receive from an 8-2-2
entry, subject to annual pay caps.

Under the consecutive-hours rule, a special agent who needed to start a scheduled 12-
hour shift an unscheduled hour early and stay an unscheduled hour late, despite having worked a
daily total of two unscheduled hours, would have to settle for submitting an 8-2-2 time-entry
rather than an 8-4-2 time-entry.3 The effect would be to cost the agent two hours of overtime
pay (equivalent to three hours of base pay).

3
Although 8-2-4 would technically be more accurate, it is unclear from the record whether
unscheduled LEAP hours were consistently included in special agents’ timecards except when
recording the first two unscheduled hours to indicate the reason for recording more than two

5
II. PROCEDURAL BACKGROUND

In 2016, Mr. Horvath sued on behalf of himself and similarly situated special agents of
the Secret Service and the Diplomatic Security Service seeking backpay for non-consecutive
overtime hours accrued when working on protective details. He claimed that the consecutive-
hours rule was an invalid interpretation of § 5542(e). The rule, the plaintiff alleged, deprived
him and similarly situated special agents of two hours of overtime pay each time a special agent
was required to submit an 8-2-2 time-entry despite having worked the necessary number of
unscheduled, although nonconsecutive, hours on a protective detail to allow the special agent to
submit an 8-4-2 time-entry.

The defendant moved to dismiss the complaint, and a judge of this court granted the
motion. On the plaintiff’s appeal, however, the Federal Circuit held that the consecutive-hours
rule was invalid and partially reversed the dismissal of this case, remanding for consideration of
whether to certify a class. See 896 F.3d at 1322. The case was then transferred to this judge.

Mr. Horvath amended his complaint to remove other theories of recovery for which the
Federal Circuit had affirmed the dismissal. (ECF 21.) Following discovery limited to the issue
of class certification, Mr. Horvath moved, pursuant to RCFC 23, to certify a class of as many as
3,000 Secret Service and 2,000 Diplomatic Security Service special agents who met § 5542(e)’s
requirements, including working at least two unscheduled hours in addition to their scheduled
shifts on protective details but who were not paid for “all of their [scheduled overtime].” (ECF
42 at 5.)

In addition to opposing the plaintiff’s class-certification motion, the defendant moved for
summary judgment against Mr. Horvath, arguing that Mr. Horvath had already received all the
pay to which he was entitled. Horvath, 2020 WL 1487642, at *2. The Court denied summary
judgment for the defendant, but adjudication of that motion raised certain issues relevant to
deciding the motion to certify a class.

In support of its motion for summary judgment, the defendant collected and reviewed
office schedules and queried the MARS database of separate LEAP-hour reports and found that
Mr. Horvath had no 8-2-2 entries that could have been entered as 8-4-2 entries but for the
provisions of the consecutive-hour rule. See id. The consecutive-hours rule, the defendant
reasoned, only denied pay to special agents whose unscheduled hours were worked in a 1:1 split,
i.e., those who worked one unscheduled hour at the beginning of their shift and a second

overtime hours. (See ECF 68 at 7 (“WebTA only determines what the special agent will actually
be paid, and there is no question that the LEAP enhancement is based on an aggregate average
that is triggered by an annual certification; it is not paid on an hourly basis. Whether and when
the special agent actually worked two hours of unscheduled overtime such that § 5542(e) might
apply is only reflected in the combination of the agent’s MARS report and schedule.”).)

6
unscheduled hour at the end of their shift. The defendant determined that Mr. Horvath had been
fully paid for any 1:1 split workday that he had recorded. (See ECF 44 at 6-7.)

In response, Mr. Horvath used the schedules and separate LEAP-hour reports produced
by the Secret Service to identify multiple days for which he had submitted 8-2-2 entries despite
separately reporting at least two LEAP hours outside of his scheduled shift. Horvath, 2020 WL
1487642, at *3. (See ECF 45 at 10.) He argued that the defendant’s records showed that he
could have submitted 8-4-2 entries on those days. The timecard entry for one of these days
reflects that the entry was submitted as 8-4-2, and a payroll official asked Mr. Horvath to
“conform” the entry to 8-2-2. (ECF 45 at 9-10 (citing Exhibit 3.)) None of the examples
identified by Mr. Horvath, however, showed unscheduled hours worked in a 1:1 split that had
been reported as 8-2-2.

Mr. Horvath does not dispute that the Secret Service’s records show no underpaid 1:1-
split workdays for him. Instead, he argues that special agents were instructed not to record hours
if they would not be paid. (ECF 45 at 8.) He identifies deposition testimony from other Secret
Service special agents suggesting that unscheduled hours had to be justified according to criteria
other than the invalidated consecutive-hours rule before they could be counted in an agent’s
timecard totals and thereby affect the number of overtime hours paid. (ECF 45 at 9.)

The Court did find that Mr. Horvath had demonstrated a genuine issue of material fact as
to whether he had been denied § 5542(e) pay on days when he worked consecutive unscheduled
hours, and, accordingly, denied summary judgment to the defendant. Horvath, 2020 WL
1487642, at *4. In deferring consideration of whether to certify a class, the Court specifically
noted that the survival of Mr. Horvath’s individual claim was not dispositive of whether a class
should be certified. Id. at *4 n.4.

Mr. Horvath’s identification of denied § 5542(e) pay for consecutive hours in response to
the defendant’s arguments regarding class certification raised a question as to the scope of the
class claims, that special agents were denied pay for their two § 5542(e) overtime hours on
grounds other than the consecutive-hours rule. Although some special agents were denied §
5542(e) pay because their hours were not consecutive (i.e., they had worked 1:1 splits), other
special agents, like Mr. Horvath, might have been told that their consecutive unscheduled hours
did not trigger § 5542(e) for some other reason. The plaintiff’s amended complaint, however,
includes no claims related to such factors and makes no allegations regarding any laws,
regulations, policies, or practices beyond the consecutive-hours rule.

Mr. Horvath’s description of the class claims in his amended complaint alludes to the
possibility that special agents have been underpaid in ways not yet known without additional
discovery. As just noted, however, the dismissed theories of the original complaint (ECF 1 ¶¶
14-18, 23-27) and the remaining theory of the amended complaint (ECF 21 ¶¶ 19-25) fail to
allege any specifics of an overtime-denying policy or practice broader than the invalidated
consecutive-hours rule such that further consideration of such a claim or discovery to support it
would be proper.

7
 Following the denial of the defendant’s motion for summary judgment and the deferral of
the plaintiff’s motion for class certification, the Court directed the parties to submit supplemental
briefs on the plaintiff’s motion to certify a class and to address specific questions identified by
the Court (ECF 62). The parties filed their supplemental briefs (ECF 65, 66) and reply briefs
(ECF 67, 68), and the Court heard oral argument on July 29, 2020.

III. DISCUSSION

Certification of an opt-in class pursuant to RCFC 23 requires several elements, often
summarized by this Court as the following: (1) numerosity, (2) commonality, (3) typicality, (4)
adequacy, and (5) superiority. See, e.g., Barnes v. United States, 68 Fed. Cl. 492 (2005)
(considering the “predominance” of common questions as part of “commonality”).

Some early interpretations of Rule 23 of the Federal Rules of Civil Procedure (“FRCP”),
the analogous, but quite different,4 rule on class actions in federal district courts, prohibited a
court from evaluating the merits of a named plaintiff’s claim and limited themselves to the
pleadings. Other courts effectively required probable success on the merits before certifying a
class. The Supreme Court has since rejected both approaches, limiting the factors courts
consider to those listed in FRCP 23 but requiring a “rigorous” inquiry that will “[f]requently . . .
entail some overlap with the merits of the plaintiff’s underlying claim.” Wal-Mart Stores, Inc. v.
Dukes, 564 U.S. 338, 351 (2011) (citing Gen. Tel. Co. of the Sw. v. Falcon, 457 U.S. 147, 161
(1982)); see Gross v. United States, 106 Fed. Cl. 369, 373 (2012) (requiring Falcon’s “rigorous
analysis” under RCFC 23).

So long as this rigorous analysis is conducted, the Court has wide discretion in deciding
whether the requirements of RCFC 23 have been satisfied. See Shipes v. Trinity Indus., 987 F.2d
311, 316 (5th Cir. 1993) (“district court has wide discretion in deciding whether to certify a
proposed class” under FRCP 23); De La Fuente v. Stokely-Van Camp, Inc., 713 F.2d 225, 232
(7th Cir. 1983); Milonas v. Williams, 691 F.2d 931, 938 (10th Cir. 1982) (“trial court's
conclusions as to whether the class representative has demonstrated that the numerosity,
commonality, typicality, and adequacy of representation requirements have been met will not be
disturbed absent a showing of abuse of that discretion.”) (internal quotation omitted); Mazus v.
Dep't of Transp., Com. of Pa., 629 F.2d 870, 876 (3d Cir. 1980) (“district court has a good deal
of discretion in determining whether or not to certify a class” under FRCP 23); Montgomery v.

RCFC 23 “is modeled largely on the comparable [Federal Rule of Civil Procedure
4

(“FRCP”) 23], [although] there are significant differences between the two rules.” Rules
Committee Notes to RCFC 23, 2002 Rev. Although RCFC 23’s effect is limited to the creation
of an opt-in rather than an opt-out class, RCFC 23 shares most of FRCP 23’s language and many
of its core concepts. The Court considers prior decisions of judges of this court interpreting
RCFC 23 as persuasive authority. The Court will also consider other federal courts’ application
of FRCP 23 when these other courts’ reasoning is unaffected by the opt-in/opt-out difference
between the two rules, as well as this court’s more limited authority to grant declaratory and
injunctive relief.

8
Rumsfeld, 572 F.2d 250, 255 (9th Cir. 1978) (“refusal of the district judge to certify a class . . . is
a matter within the discretion of the trial court”). Although RCFC 23 differs from FRCP 23 in
important respects, both use the same permissive (“may”), not mandatory (“shall”), language in
allowing a court to certify a class. Accordingly, the Court finds these precedents regarding a trial
court’s discretion under FRCP 23 applicable to RCFC 23.

Mr. Horvath argues that the Federal Circuit has determined the defendant’s liability and
all that remains is to analyze the time and pay records of the Secret Service and the Diplomatic
Security Service to calculate the individual damages awards due to each special agent.5 The
defendant, on the other hand, argues that there is no common question remaining, and if there are
common questions, the individual “liability” determination for each special agent predominates
over them. Further, the defendant argues that Mr. Horvath is not an adequate class representative
because he was not denied pay by the invalidated consecutive-hours regulation.

Mr. Horvath’s responses to the defendant’s attack on the viability and typicality of his
claim and his adequacy as a class representative presented three counter-arguments by which the
plaintiff proposes to expand the class claims beyond the consecutive-hours rule; broaden the
effect of the consecutive-hours rule beyond 1:1 splits; and diversify the relevant evidence beyond
the Secret Service’s time and payroll records.

The Court finds that RCFC 23’s commonality, typicality, and adequacy requirements
require the rejection of all three of the plaintiff’s counter-arguments and the denial of the motion
to certify a class, even on the narrower claims originally implicated in Mr. Horvath’s amended
complaint and motion for class certification.

As a preliminary, but necessary, matter, the Court addresses whether Mr. Horvath’s
apparent lack of damages from the consecutive-hours rule affects his standing.

A. Standing

To satisfy the Constitution’s Article III case-or-controversy requirement, a plaintiff must
maintain standing to sue throughout the course of litigation. Campbell-Ewald Co. v. Gomez, 577
U.S. ___, 136 S. Ct. 663, 669 (2016), as revised (Feb. 9, 2016) (quoting Arizonans for Official
English v. Arizona, 520 U.S. 43, 67 (1997) (“We have interpreted this requirement to demand
that ‘an actual controversy . . . be extant at all stages of review, not merely at the time the

5
As a class definition, Mr. Horvath proposes all individuals who were (1) employed as
Secret Service or Diplomatic Security Service special agents during at least one pay period
between June 10, 2010, and September 12, 2018; (2) performed “protective duties” and received
LEAP; (3) performed regularly scheduled overtime on the same day they performed at least two
hours of unscheduled overtime; and (4) were not paid the overtime rate for all regularly
scheduled overtime performed on those days. (See ECF 42 at 5.) “Unscheduled” overtime is
defined as overtime work that was “not scheduled in advance of the administrative work week.”
5 U.S.C. § 5542(d)(1).

9
complaint is filed.’”). Although the defendant has not argued that Mr. Horvath’s apparent lack
of 1:1-split hours moots his claim, the Court considers this question sua sponte to ensure its
subject-matter jurisdiction. See RCFC 12(h)(3).

In the context of RCFC 23, a representative plaintiff’s standing to bring his individual
claim satisfies the test to maintain standing on behalf of the class, so long as the representative’s
individual claim and the putative class’s claims do not “‘implicate a significantly different set of
concerns.’” Ramona Two Shields v. United States, 820 F.3d 1324, 1331 (Fed. Cir. 2016)
(quoting Gratz v. Bollinger, 539 U.S. 244, 265 (2003)).

Standing requires (1) an “injury in fact” (2) “fairly traceable to the challenged action of
the defendant” (3) “likely” to be “redressed by a favorable decision.” Friends of the Earth, Inc.
v. Laidlaw Envtl. Servs. (TOC), Inc., 528 U.S. 167, 180–81 (2000). Mr. Horvath alleges that he
received less § 5542(e) pay than he was entitled to receive and seeks a judgment in the amount of
the pay he was denied. Mr. Horvath has standing to sue for his individual claim.

As the Court concludes below, Mr. Horvath’s original, narrow claim challenging the
consecutive-hours rule was typical of the class’s claim such that it did not “implicate a
significantly different set of concerns.” Although the defendant has shown that Mr. Horvath may
be unlikely to prove individual damages from the claim he is pursuing on behalf of the class, it is
not clear that this showing moots Mr. Horvath’s original, narrow challenge to the consecutive-
hours rule. In cases in which a defendant’s offer of judgment to the representative plaintiff alone
has potentially mooted the representative plaintiff’s claim while the motion for class certification
was pending, the Supreme Court has suggested in dicta that a trial court’s standing analysis could
relate back to the plaintiff’s claim as it stood when the plaintiff moved to certify the class. See
Sosna v. Iowa, 419 U.S. 393, 402 n.11 (1975). The Third, Fifth, and Tenth Circuits have applied
Sosna to relate standing back to when a motion to certify a class was filed. See, e.g., Holmes v.
Pension Plan of Bethlehem Steel Corp., 213 F.3d 124, 135 (3d Cir. 2000); Reed v. Heckler, 756
F.2d 779, 787 (10th Cir. 1985). But see William Rubenstein, Newberg on Class Actions § 2:11
(5th ed.) (noting that relating standing back to the filing of the motion rather than the ruling on
class certification is the minority position among the circuits, but that other circuits have
sometimes achieved the same result by applying or analogizing to other exceptions to the
mootness rule); Zeidman v. J. Ray McDermott & Co., Inc., 651 F.2d 1030, 1050 (5th Cir. 1981).

In the absence of controlling Federal Circuit authority to the contrary, the Court also
applies Sosna to find that standing still exists to allow it to decide Mr. Horvath’s RCFC 23
motion. As the relevant Supreme Court precedents explain, once a class has been certified, the
representative plaintiff’s standing is no longer essential for a case or controversy because FRCP
23—and presumably RCFC 23—but not the FLSA’s collective-action procedure, creates a class
with legal status independent of the class representative. See Genesis Healthcare Corp. v.
Symczyk, 569 U.S. 66, 75 (2013).

B. Expanded Class Claim for All § 5542(e) Pay

Whether RCFC 23’s requirements are satisfied depends here on the scope of the class
claims, whether they are limited to unpaid 1:1 splits or include § 5542(e) pay denied on any
10
ground. RCFC 23’s commonality requirement demands that the class claims be limited to
§ 5542(e) pay denied by the terms of the consecutive-hours rule.

RCFC 23 provides for certification of an opt-in class when, among other things, litigating
the class claims together would answer common questions of law or fact that predominate over
individual questions. When the question common to the putative class members is factual, it
must be answerable by generalized, not individualized, evidence. See Gross v. United States,
106 Fed. Cl. 369, 379 (2012) (holding that whether a question was “common” to a class of
agency employees seeking Sunday premium pay depended on whether their entitlement to the
pay could be proven by reference to their status in agency payroll records—"generalized
proof”—or would require the court to inquire into their respective region’s Sunday-scheduling
practices and examine the type of work each employee performed on Sundays).

This commonality inquiry further includes RCFC 23(b)’s requirements that (1) the
United States has acted (or refused to act) on “grounds generally applicable to the class; and” (2)
legal or factual questions “common to the class predominate over any questions affecting only
individual members.” RCFC 23(a)(2), 23(b)(2)-(3) (emphasis added). In requiring both that the
defendant acted on generally applicable grounds and that a question common to class members
predominates, RCFC 23 is—by its terms—more restrictive than its sister-rule, FRCP 23, which
provides a disjunctive test: it requires that a defendant act either on generally applicable grounds,
or that common legal or factual questions predominate for a class action to be maintained. See
FRCP 23(b)(2)-(3). This Court’s rule is conjunctive, requiring both conditions.

A class definition including all special agents who were denied § 5542(e) pay, regardless
of whether they worked any 1:1-split overtime hours, does not meet RCFC 23’s commonality
requirement. Although individual claims for backpay, like Mr. Horvath’s individual claim that
this Court upheld on summary judgment, need not identify the agency’s grounds for
underpayment, linking the claims together in a class action requires that the underpayment be
based on grounds generally applicable to all prospective class members.

These generally applicable grounds could be a policy or practice generally applicable to
the class or even a common error shown to have had a widespread effect. See Barnes, 68 Fed.
Cl. at 496-97 (certifying a class of employees when the agency had conceded in an earlier,
similar case that its denial of pay to 39 similar plaintiffs may have been a data-entry error). The
consecutive-hours rule provides a ground generally applicable to all special agents who worked
1:1-split unscheduled hours on protective details and, therefore, could support a class-wide claim
that it formed the basis for those special agents’ underpayments. Mr. Horvath, despite showing
that other class members were also underpaid on days when they did not work 1:1 splits, neither
alleges nor points to evidence of a generally applicable policy or practice that caused
underpayment when agents worked consecutive unscheduled hours.

The amended complaint alleges no policies, practices, or errors other than the
consecutive-hours rule that might have prohibited agents from reporting their unscheduled hours
and claiming their overtime pay. Mr. Horvath points to deposition testimony that unscheduled
hours were excluded from time entries based on criteria that were variable or unclear to Secret
Service special agents. The amended complaint and the deposition testimony on which the
11
plaintiff relies stop short, however, of describing any of these additional criteria. Mr. Horvath
points to another employee’s correction of Mr. Horvath’s timecard entry containing the remarks
“forced 8-2-2.” This notation might support the inference that there existed an “8-2-2 policy”
broader than the consecutive-hours rule, but the amended complaint fails to allege such a policy.
The sparse reference on which the plaintiff relies offers no direct indication such a policy existed
and, if it did, what the policy’s content was and how it served as grounds for underpayment
common to the class. The defendant represents that § 5542(d) itself is the only so-called “8-2-2
policy” that the Secret Service has. Mr. Horvath’s failure to allege a policy broader than the
consecutive-hours rule and his citation only to evidence that allows, but does not require, an
inference such other policies existed—without describing how their terms apply to the class—are
insufficient to controvert the defendant’s representation.

Mr. Horvath suggests that, through a court-imposed condition requiring him to discover
and prove the existence of such a broader policy before entering judgment on behalf of putative
class members affected by the existence of this policy, the Court may certify a class now to
enable such discovery. The Court rejects this suggestion. Although courts frequently refer to the
judge-made doctrine of “conditional” certification in the context of collective actions under the
FLSA, this case is not a FLSA action, and neither RCFC 23 nor its sister-rule, FRCP 23, allows
for conditional certification of a class. See William Rubenstein, Newberg on Class Actions
§ 7:33 (5th ed.) (noting that “[FRCP] 23 long enabled a court to certify a class ‘conditionally,’
but in 2003, Congress removed this provision from the Rule, while simultaneously expanding the
time permitted for the certification decision”); RCFC 23, Rules Committee Notes, 2004 Revision
(noting this Court’s adoption of the 2003 changes to FRCP 23 in RCFC 23).

Mr. Horvath likens this case to Barnes v. United States, in which Judge Allegra held that
the issue of whether an agency’s denial of night-and-weekend pay was systematic and
widespread was a common question that predominated over individual questions. 68 Fed. Cl. at
496-97. This comparison suggests that, if the denial of § 5542(e) pay was systematic and
widespread, it could serve here as the grounds generally applicable to the class. This case,
however, differs from Barnes in critical respects.

The Barnes plaintiffs, unlike Mr. Horvath, alleged that the denial of night-and-weekend
pay was systematic and widespread. Barnes, 68 Fed. Cl. at 496. The variation in job titles and
consistency in underpayment among the three named plaintiffs in that case—a member of
facility staff, a computer operator, and a registered nurse—supported that allegation. See id. at
493. Indeed, the plaintiffs in Barnes sought to certify a class of 5,000 potential members across
21 job-titles. See id. Moreover, as the court noted in Barnes, the Navy had conceded that “data
entry errors” might have denied this pay to 39 named plaintiffs in a case that it had settled two
years earlier. Id. at 497 (referring to Abrams v. United States, 57 Fed. Cl. 439 (2003)). Against
this background, the court found that the Barnes class included members who would have been
included in the Abrams class, had one been certified. Id. at 498.

Although Mr. Horvath points to four other Secret Service special agents whose records
suggest were denied § 5542(e) pay (ECF 65 at 12), this evidence amounts to considerably less
support than the plaintiffs supplied in Barnes. Perhaps if Mr. Horvath had alleged this broader
class definition from the start or taken discovery on this issue, he might have amended his
12
complaint and used class-certification discovery to show a systematic and widespread denial of
overtime broader than the terms of the consecutive-hours rule. At this point in the case, with the
defendant’s evidence that Mr. Horvath suffered no direct injury from the application of the
consecutive-hours rule, his argument for a broader, unspecified policy causing harm looks more
like a last-minute addition meant to ensure that the class claims are typical of Mr. Horvath’s, not,
as RCFC 23 requires, the other way around.

The plaintiff’s failure to allege and establish generally applicable grounds for the denial
of pay triggered both by 1:1 splits and by consecutive unscheduled overtime hours on protective
details leaves Mr. Horvath to argue that denying pay to multiple employees mandated by the
same provision creates generally applicable grounds and a common question. As for a common
question, such an argument was roundly rejected by the Supreme Court, which wrote,
“[c]ommonality requires the plaintiff to demonstrate that the class members have suffered the
same injury. This does not mean merely that they have all suffered a violation of the same
provision of law.” Dukes, 564 U.S. at 349-50 (citation omitted). Rather, the “claims must
depend upon a common contention . . . that it is capable of classwide resolution—which means
that determination of its truth or falsity will resolve an issue that is central to the validity of each
one of the claims in one stroke.” Id. at 350. To be sure, the employment discrimination claims
at issue in Dukes posed different questions of predominance than are at issue here and raised
them in the context of FRCP 23’s provision for an opt-out, rather than an opt-in, class. Despite
any significance these differences might have, the Supreme Court’s requirement that the
plaintiffs identify a common policy in that case is consistent with RCFC 23’s requirement that
the defendant act on grounds commonly applicable to the class.6

For the foregoing reasons, the Court finds that certifying a class for the plaintiff’s
expanded claim for all § 5542(e) pay is not appropriate.

C. Original Class Claim for Underpaid 1:1 Splits

A narrower class seeking § 5542(e) pay denied by the terms of the consecutive-hours rule
satisfies commonality and typicality, but Mr. Horvath’s conflicting interests prevent him from
being an adequate representative of the class he originally proposed.

1. Commonality

The defendant opposes the plaintiff’s motion to certify the narrower class implicated in
Mr. Horvath’s motion, arguing that determining “liability” for backpay is a predominant
individual question because it requires comparison of each special agent’s time entries,
schedules, LEAP reports, and pay caps to find which agents, if any, are entitled to more pay than
they received. These individual determinations of “liability” for each special agent, the

6
Mr. Horvath’s attempt to characterize the denial of § 5542(e) pay as caused by the
consecutive-hours rule similarly lacks support in the record. Moreover, it relies on a legal theory
distinct from the class claims, raising typicality issues.

13
defendant argues, predominate over the only common issue—the facial validity of the
consecutive-hours rule—which the Federal Circuit has already decided. Mr. Horvath responds
that “liability” has already been decided by the Federal Circuit, and the questions the defendant
characterizes as individualized determinations of “liability” are merely damages calculations.
The need for individual damages calculations, Mr. Horvath argues, are commonplace in wage-
and-hour class actions and are no reason to deny class certification.

The damages due to putative class members who worked 1:1 splits but did not receive §
5542(e) pay is a common question that can be answered on a class-wide basis with factual
analysis of the Secret Service’s time and pay records. Generalized proof requires uniformity in
the kind of evidence used to prove the class’s claims, not uniformity in the results.

“The ‘[m]ere fact that damage awards will ultimately require individualized fact
determinations is insufficient, by itself’ to defeat a class action.” Curry v. United States, 81 Fed.
Cl. 328, 334 (2008) (quoting McCarthy v. Kleindienst, 741 F.2d 1406, 1415 (D.C. Cir. 1984)).
Although some factual questions, such as the tasks each of a class of employees performed and
the conditions in which they were performed or highly subjective subsidiary conclusions such as
individuals’ diagnoses, may be too individualized for class resolution, the determination of
individual damages “ based on information found on the face of government records, and [that
can] be computed mechanically, with relatively little controversy” is not too individualized for
class-wide determination. Curry, 81 Fed. Cl. at 334 (internal quotation omitted) (distinguishing
Jaynes v. United States, 69 Fed. Cl. 450, 457–58 (2006), and Black v. United States, 24 Cl. Ct.
471, 477–78 (1991)).

In this case, the timecard entries are already stored in a centralized database that can be
queried. The separate LEAP reports are also stored in a centralized database that can be queried.
The relevant weekly schedules are neither centralized nor in a format that can be queried, but
nothing prevents their collection and en masse transcription into a database that can be queried.
Regardless of local variations in the schedules’ formats or medium of distribution, each schedule
must at least contain the common data “fields” needed, including a unique identifier for the
agent, the date, the start- and end-times of the shift, and a description or notes that a transcriber
or reviewer could code as whether the work constituted protective duty. This case is not one in
which some class members are trying to prove their claims with stock prices, with others relying
on misrepresentations, and still others relying on an expert’s regression analysis to prove the
same elements (as would be the case for the plaintiff’s effort to certify a class for unpaid
overtime arising from any cause beyond the consecutive-hours rule).

The remaining fact-finding is more a question of damages than, as the defendant argues, a
question of legal liability. Unlike class actions that allege patterns of predicate conduct or
injuries with subjective elements, like discrimination cases, calculating the number of days on
which special agents were scheduled for certain hours of protective duty, separately reported a
1:1 split, and entered their timecards as “8-2-2” does not require the application of law to each
individual agent’s facts. It merely requires the application of certain less accessible facts
(schedules) to other more accessible facts (time entries and separate LEAP-hour reports) for each
special agent. The defendant’s emphasis on the number of special agents, the number of offices,
and the number of days involved misses the point. Unless the special agents all had a different
14
definition of what constitutes an hour, the sheer number of rows in the eventual combined dataset
does little, if anything, to complicate the final calculation. Moreover, the amount of manual
data-entry required is most directly attributable to the Secret Service’s practice of maintaining
separate schedules in each of its many offices, and may be positively related, but will not be
directly proportional, to the number of special agents opting into the class. (See ECF 66 at 4
(noting the Secret Service’s 45 “field” offices, 88 smaller “resident” offices, and 7 protective
divisions).)

The calculation of backpay for Diplomatic Security Service special agents may be
another matter, or it may not be. The plaintiff had the burden to show that the common issue of
fact extended to the claims of Diplomatic Security Service special agents but has not pointed to
evidence that generalized proof exists to show the number of days that each Diplomatic Security
Service special agent was denied the § 5542(e) pay when working a 1:1 split. Had the Court
been prepared to certify a class of Secret Service special agents, additional discovery might have
been appropriate to permit Mr. Horvath to sustain his effort to include Diplomatic Security
Service special agents within the class. Because, as discussed below, the Court declines to
certify a class of Secret Service special agents, further discovery related to the Diplomatic
Security Service is not appropriate. Accordingly, the Court finds the plaintiff has failed to carry
his burden of supporting the inclusion of Diplomatic Security Service special agents in the class.

2. Typicality

The inquiry underlying the typicality element is focused on whether the representative
plaintiff’s claim is typical of the putative class members’ claims. See Curry, 81 Fed. Cl. at 335-
36. A representative plaintiff’s claim does not need to be identical to the class members’ claims.
The claims of both merely must “aris[e] from the same governmental action and [be] premised
upon a common theory of recovery.” Geneva Rock Prod., Inc. v. United States, 100 Fed. Cl.
778, 790 (2011) (named plaintiff’s claims shared the same “essential characteristics” as class
members’ claims because both owned property affected by the government’s notice and both
alleged that the notice effected a taking of their property); see Machella v. Cardenas, 653 F.2d
923, 927 (5th Cir. 1981) (“[The named plaintiff’s] sole claim of injury under the 1969 Act is
predicated on an entirely different legal theory from that on which potentially successful
members of the class (seeking the $1,800 forgiveness, not backdating) would rely.”).

When comparing the essential characteristics of Mr. Horvath’s claim to the essential
characteristics of the class’s claims, the Court must distinguish between the predicate facts that
make up the claims and the outstanding facts whose discovery and evaluation for entitlement to
money damages—on a class basis—are the reasons for this suit.

Mr. Horvath’s claim appeared typical of the class’s claims when he filed it, suffered
dismissal, successfully appealed, amended the complaint on remand, and moved for class
certification. Until the defendant moved for summary judgment, while Mr. Horvath’s class-
certification motion was still pending, all indications were that Mr. Horvath had been subject to
the consecutive-hours rule and denied pay according to that rule’s terms. If the defendant had
not reviewed Mr. Horvath’s schedules, separate LEAP reports, and time entries, effectively
calculating the damages for Mr. Horvath’s original individual claim, Mr. Horvath would still
15
have an unresolved claim identical to the class members’ unresolved claims, many of which may
themselves result in no damages, just as Mr. Horvath’s has. Indeed, the process by which the
Secret Service evaluated Mr. Horvath’s original, narrow claim is analytically the same process
required to evaluate the claims of any class member who would choose to opt into the class.

In the narrow category of case in which the only common question is a factual one and its
answer lies with the defendant, the defendant’s choice to answer the question for the
representative plaintiff alone should not render a representative plaintiff’s otherwise typical
claim atypical of the putative class members’ claims. When the only difference between Mr.
Horvath’s original, narrow claim and the class claims is that Mr. Horvath’s claim was evaluated
first, the Court will not say that Mr. Horvath’s claims and the class claims no longer share the
same essential characteristics. Comparing Mr. Horvath’s known lack of damages to the
unknown amounts of individual damages resulting from the class’s otherwise identical claims is
not a valid comparison and is no basis for finding RCFC 23’s typicality requirement unsatisfied.

In the rare situations in which the issue presented here has arisen under FRCP 23, other
courts have agreed with this conclusion. See Richards v. FleetBoston Fin. Corp., 238 F.R.D.
345, 351 (D. Conn. 2006) (refusing to consider in its FRCP 23 decision expert analysis showing
that some members of challenged benefit plan profited rather than lost money); Ventura v. New
York City Health & Hosps. Corp., 125 F.R.D. 595, 600 (S.D.N.Y. 1989) (finding typicality
although “unlike most employees, plaintiff was [drug] tested twice” and “unlike some members
of the proposed class, plaintiff's test results were negative”); Ouellette v. Int'l Paper Co., 86
F.R.D. 476, 480 (D. Vt. 1980) (“that the damages, if any, may reach a de minimis level at some
point among the class members does not make the named plaintiffs' claims atypical”); Sanders v.
Faraday Labs., Inc., 82 F.R.D. 99, 101 (E.D.N.Y. 1979) (finding typicality despite the
defendant’s argument that the representative plaintiff sold his holdings at a profit and thus would
be unable to prove damages at trial); In re Wal-Mart Stores, Inc. Wage & Hour Litig., No. C 06-
02069 SBA SMITH, 2008 WL 413749, at *11 (N.D. Cal. Feb. 13, 2008) (holding that
representatives’ potential inability to prove any damages at trial did not undermine the typicality
of their claims, although noting that as to one of the representatives there was no dispute that he
was underpaid by at least ten dollars).

This narrow holding should not be confused for a rule that pre-certification summary
judgment is always inappropriate, or that a court applying RCFC 23 must turn a blind eye to
merits-related facts that are relevant to RCFC 23’s standards.7 Rather, the Court holds that when

7
When a legal flaw affecting all the claims is apparent, even dismissal prior to class
certification may be appropriate. Similarly, other kinds of factual flaws in the class claims or
even a factual flaw in the representative plaintiff’s claim may support pre-certification summary
judgment in circumstances that do not significantly prejudice the named plaintiff or the class.
See Thompson v. Cty. of Medina, Oh., 29 F.3d 238, 241 (6th Cir. 1994) (“[G]iven that many of
the violations claimed by [the named plaintiffs] had been remedied, or never existed, and that
neither plaintiffs nor the members of the class were prejudiced by the order of the court's rulings,

16
the amount of damages is the merits issue and some class members are expected not to have any
damages, whether the named plaintiff suffered damages is not relevant to the RCFC 23-typicality
analysis, and thus, it is not the kind of merits issue that overlaps with the rigorous inquiry that
RCFC 23 requires.

This outcome is consistent with the refusal of judges of this court to base class
certification on conditions that the defendant can alter at will. In Barnes, this court rejected the
defendant’s argument that the defendant’s decision to concede liability could deprive the case of
a common issue and prevent class certification.

The Court rejects the defendant’s argument that its decision to calculate the named
plaintiff’s damages ahead of the class members’ can deprive the case of typicality.

3. Adequacy

The “adequacy” element has two parts: (1) class counsel must be qualified, experienced
and generally able to conduct the litigation; and (2) the class members must not have interests
that are antagonistic to one another. King, 84 Fed. Cl. at 127; Curry, 81 Fed. Cl. at 338; Barnes,
68 Fed. Cl. at 499. The Court finds that “class counsel ha[s] experience litigating similarly
complex class actions and will devote time and energy to the litigation—indeed, [he] ha[s]
already pursued a successful appeal of the Court's opinion granting the government's motion to
dismiss.” Mercier v. United States, 138 Fed. Cl. 265, 274 (2018). The Court has no difficulty in
finding that the plaintiff’s counsel more than satisfies the test for adequacy.

Mr. Horvath’s interests, however, are not sufficiently aligned with the interests of the
class to satisfy RCFC 23’s adequacy requirement. Although Mr. Horvath’s no-damages claim
for pay denied by the consecutive-hours rule is nominally typical of the claims of members of the
prospective class as limited by this opinion, Mr. Horvath’s real interest in pursuing his claim on
such limited terms is minimal. Mr. Horvath’s interest now lies in showing that the Secret
Service’s records did not reliably record all of his 1:1 splits.

In contrast, the putative class’s interest and claim to commonality lie in using the Secret
Service’s records as common, generalized evidence to measure the individual damages of each
class member, and not in attacking the accuracy of those records. If Mr. Horvath pursues his
interest and is allowed to use other types of evidence, such as personal time and attendance

the district court acted well within its discretion in concluding that it should decide the motion
for summary judgment first.”); Wright v. Schock, 742 F.2d 541, 543–44 (9th Cir. 1984) (“Under
the proper circumstances—where it is more practicable to do so and where the parties will not
suffer significant prejudice—the district court has discretion to rule on a motion for summary
judgment before it decides the certification issue.”).

17
records, to prove that he had additional, unrecorded 1:1 splits, he risks undermining the common
answer to a common question that supports certification of this class in the first place.

Additionally, Mr. Horvath’s individual claim for unpaid, consecutive, unscheduled hours
on protective details remains. This claim has the potential to overshadow the class claims in that
it may require broader discovery requests than what would be needed to prove the class
members’ 1:1-split claims. The Court cannot expect Mr. Horvath’s counsel, in crafting,
negotiating, and litigating such requests, to balance effectively the competing interests of Mr.
Horvath in proving his broader claim and those of the class members in pursuing the same
analysis that, in the light of the uncontroverted facts adduced by the defendant’s payroll records,
will yield no recovery for Mr. Horvath.

Because of the potential conflicts between Mr. Horvath’s interest in pursuing his
remaining personal claim and the class members’ interest in pursuing putative claims under the
invalidated consecutive-hours rule, the Court finds, in the exercise of its discretion under RCFC
23, that Mr. Horvath is unable to represent the interests of class members adequately. On this
basis, the Court denies the plaintiff’s motion to certify a class of Secret Service special agents for
underpaid 1:1 splits.

IV. CONCLUSION

The plaintiff’s motion to certify a class is denied.8 The Court will file an order
concurrent with this opinion denying the motion and setting a schedule for further proceedings in
the case.

s/ Richard A. Hertling
Richard A. Hertling
Judge

8
The plaintiff has also sought an order regarding the tolling of the statute of limitations for
prospective class members. (ECF 42 at 3-4.) As the defendant points out, the precise nature of
the plaintiff’s request is unclear. (ECF 44 at 27-28.) Given the denial of the motion to certify a
class, the Court declines to rule on this additional aspect of the plaintiff’s motion, finding it
better left to resolution in any future case brought by a putative class member.

18