U.S. Equal Employment Opportunity Commission v. Maryland Insurance Administration

PUBLISHED

UNITED STATES COURT OF APPEALS
FOR THE FOURTH CIRCUIT

No. 16-2408

U.S. EQUAL EMPLOYMENT OPPORTUNITY COMMISSION,

Plaintiff - Appellant,

v.

MARYLAND INSURANCE ADMINISTRATION,

Defendant - Appellee.

Appeal from the United States District Court for the District of Maryland, at Baltimore.
J. Frederick Motz, Senior District Judge. (1:15-cv-01091-JFM)

Argued: October 25, 2017 Decided: January 5, 2018

Before WILKINSON, KEENAN, and FLOYD, Circuit Judges.

Vacated and remanded by published opinion. Judge Keenan wrote the majority opinion,
in which Judge Floyd joined. Judge Wilkinson wrote a dissenting opinion.

ARGUED: Philip Matthew Kovnat, U.S. EQUAL EMPLOYMENT OPPORTUNITY
COMMISSION, Washington, D.C., for Appellant. John Van Lear Dorsey, OFFICE OF
THE ATTORNEY GENERAL OF MARYLAND, Baltimore, Maryland, for Appellee.
ON BRIEF: James L. Lee, Deputy General Counsel, Jennifer S. Goldstein, Associate
General Counsel, Lorraine C. Davis, Assistant General Counsel, Office of General
Counsel, U.S. EQUAL EMPLOYMENT OPPORTUNITY COMMISSION, Washington,
D.C., for Appellant. Brian E. Frosh, Attorney General, Lisa Boylan Hall, Assistant
Attorney General, OFFICE OF THE ATTORNEY GENERAL OF MARYLAND,
Baltimore, Maryland, for Appellee.

2
BARBARA MILANO KEENAN, Circuit Judge:

The Equal Employment Opportunity Commission (the EEOC) brought this action

on behalf of three female employees against their employer alleging salary discrimination

under the Equal Pay Act (the EPA), 29 U.S.C. § 206(d). The district court granted

summary judgment in favor of the employer, the Maryland Insurance Administration

(MIA), and the EEOC appealed. Upon our review, we conclude that the EEOC

established a prima facie violation of the EPA, and that genuine issues of material fact

exist regarding whether the pay disparity was due to factors other than gender. We

therefore vacate the district court’s grant of summary judgment in favor of MIA and

remand for further proceedings consistent with this opinion.

I.

A.

MIA is an independent state agency that performs various functions related to the

regulation of Maryland’s insurance industry and the enforcement of Maryland’s

insurance laws. Md. Code, Ins. §§ 2-101, 2-102. MIA is subject to the State Personnel

Management System, a merit-based system, which establishes job categories based on the

general nature of required duties and sets corresponding levels of compensation. Md.

Code, State Pers. & Pens. §§ 6-102(1)(i), (2), (3). Although MIA, as an independent state

agency, is given discretion to set its employees’ salaries, MIA follows the hiring and

salary practices of Maryland’s Department of Budget and Management, which

3
promulgates a Standard Pay Plan Salary Schedule (the Standard Salary Schedule). Md.

Code, Ins. § 2-105.

In accordance with the Standard Salary Schedule, when a new employee is hired,

MIA assigns that employee to a grade level matching the position being filled. Each

grade level carries an assigned base salary and a specific salary range consisting of 20

separate steps. After designating a new employee’s particular grade level, MIA assigns

the new hire to an initial step placement based on prior work experience, relevant

professional designations, and licenses or certifications. In selecting a particular step

level, MIA also considers the difficulty of recruiting for the position, and, under

Maryland law, also awards a new employee credit for any prior years of service in state

employment for the purposes of determining that employee’s step in the applicable pay

grade. Md. Code, State Pers. & Pens. § 2-601. Additionally, a Maryland government

employee who transfers to a “lateral” position takes her assigned grade and step with her

to the new position.

MIA employees work within one of six units, each comprising a different area of

insurance regulation. Md. Code, Ins. § 2-102. At issue in this case is the Fraud

Investigation Division. Employees working in the Fraud Investigation Division, or Fraud

Investigators, are charged with investigating allegations of criminal insurance fraud

perpetrated by individuals. Until July 2013, the Fraud Investigator position was

classified at a grade 15 on the Standard Salary Schedule. At that time, based on an

internal job study conducted by MIA, the position was reclassified at grade 16. Under

the Standard Salary Plan, individuals hired as Fraud Investigators now are assigned to a

4
step within the grade 16 classification according to their qualifications and work

experience.

MIA advertises minimum and preferred qualifications for the position of Fraud

Investigator. To be minimally qualified for hire as a Fraud Investigator, an applicant

must have five years of fraud investigatory experience in such areas as white collar

crime, financial fraud, insurance fraud, and investigations conducted under the

supervision of prosecutors or other attorneys. Preferred or desired qualifications for the

position include designation as a Certified Fraud Examiner, as well as experience

working with attorneys and participation in court or administrative hearings.

B.

The EEOC brought the present action on behalf of three former MIA employees:

Alexandra Cordaro, Marlene Green, and Mary Jo Rogers (collectively, the claimants).

MIA hired Cordaro as a Fraud Investigator in December 2009. Cordaro had worked as a

fraud investigator for a federal credit union for over two years, and as a criminal

investigation and litigation paralegal for 12 years in the Baltimore County State’s

Attorney’s Office. MIA assigned Cordaro to grade level 15, step four, with a starting

salary of $43,495. By the time she resigned from her position at MIA about five years

later, Cordaro was earning $49,916.

Marlene Green was hired as a Fraud Investigator in November 2010. Green held a

bachelor’s degree from Johns Hopkins University, and had more than 20 years of

experience working for the Baltimore City Police Department. During the course of that

employment, Green worked for approximately 13 years in an investigative capacity. In

5
the year immediately prior to joining MIA, Green worked as an investigator for the

United States Office of Personnel Management and the Office of the State’s Attorney for

Baltimore County. MIA assigned Green to grade level 15, step four, with a starting

salary of $43,759. By February 2013, when Green resigned from MIA, her salary was

$45,503.

Mary Jo Rogers transferred to the Fraud Investigation Division from another

position within MIA in July 2011. Rogers earlier had worked for eight years as a police

officer and a detective with the Baltimore County Police Department. Immediately

before being hired at MIA, Rogers worked for an insurance company as a special

investigator and an adjuster. MIA assigned Rogers to grade level 15, step five, with a

starting salary of $46,268. By November 2013, Rogers’ salary was $50,300.

During their tenure at MIA, Cordaro, Rogers, and Green learned that their salaries

were lower than the salaries of certain male Fraud Investigators. In early 2014, after

unsuccessfully seeking to correct these disparities, Cordaro and Rogers filed complaints

against MIA with the EEOC. It is unclear from the present record what transpired during

the EEOC process.

The EEOC later filed the present lawsuit against MIA on behalf of the three

claimants, alleging gender-based salary discrimination in violation of the EPA. During

6
the proceedings in the district court, the EEOC identified as comparators four male Fraud

Investigators: Bruno Conticello, James Hurley, Donald Jacobs, and Homer Pennington. 1

MIA hired Conticello as a Fraud Investigator in November 2010. Conticello held

both a bachelor’s and a master’s degree in criminal justice, and had nearly 20 years of

investigative experience working for various insurance companies and Maryland’s Office

of the Inspector General. He also had obtained a Certified Fraud Examiner designation. 2

MIA assigned Conticello to grade level 15, step 10, with a starting salary of $49,842. By

late 2012, his salary was $51,561.

James Hurley was hired as a Fraud Investigator in November 2006. Hurley most

recently had worked as an investigator with an underwriting insurance organization. He

also had worked previously with MIA for a total of three years as a Fraud Investigator

and as an investigator in MIA’s property and casualty department. Altogether, Hurley

had about 10 years of insurance-related investigative experience when he was re-hired at

MIA in 2006. Hurley also had worked previously as a claim adjuster for several

1
The EEOC also identified nine other male Fraud Investigators who earned higher
starting wages than Cordaro, Green, and Rogers, but the EEOC did not designate these
other employees as comparators. We focus our analysis only on the four Fraud
Investigators specifically identified as comparators.
In addition, the EEOC identified as comparators two other males who work in a
different unit within MIA, the Compliance and Enforcement Section, but who have
similar job responsibilities as Fraud Investigators: Jeffrey Gross and Maurice Xenos.
Because our disposition of this appeal does not rely on a determination whether these two
individuals are valid comparators, we do not address their qualifications and starting
salaries.
2
The parties do not describe the particulars of the Certified Fraud Examiner
designation, and the record does not provide any further information.

7
insurance companies, and had earned the designation of Certified Fraud Examiner. At

the time of his most recent hiring at MIA, Hurley was assigned to grade 15, step six, with

a starting salary of $45,298. By the time he left MIA in October 2012, Hurley’s salary

was $49,678.

MIA hired Donald Jacobs as a Fraud Investigator in May 2007. Jacobs had 11

years of experience as a Natural Resources Officer with the state of Maryland, primarily

engaged in conducting marine patrols, and had worked for three years as an investigator

in the Office of the Public Defender in Baltimore. This investigatory experience did not

relate to fraud or white collar crime. Jacobs’ starting salary at MIA was $45,298, based

on his assignment to grade level 15, step six. When Jacobs left MIA in June 2010, his

salary was $47,705.

Homer Pennington was hired as a Fraud Investigator in August 2007. Pennington

had worked in the criminal investigation unit of the Baltimore Police Department for

approximately 22 years before joining MIA. He earned the designation of Certified

Arson Investigator, but neither his resume nor the record specify the requirements for

acquiring this certification. Pennington was hired at grade level 15, step five, with a

starting salary of $45,360. By May 2013, when Pennington left MIA, his salary was

$47,194.

The parties filed cross-motions for summary judgment in the district court. The

court denied the EEOC’s motion and granted MIA’s motion. In dismissing the EEOC’s

claim under the EPA, the district court effectively held that the claimants had failed to

meet their prima facie evidentiary burden. The court concluded that the male Fraud

8
Investigators identified by the EEOC were not valid comparators because they were hired

at higher steps than were the claimants. Alternatively, the court held that MIA had

shown that the disparity in pay between the claimants and the male comparators was

attributable to their relative experience and qualifications. The EEOC appeals the district

court’s grant of summary judgment to MIA. 3

II.

On appeal, the EEOC asserts that the district court erred in awarding summary

judgment in favor of MIA. The EEOC argues that it made a prima facie showing of an

EPA violation by identifying four male Fraud Investigators who earned higher starting

salaries than the three claimants, who were assigned lower salaries than the male

comparators despite performing identical work. In addition, the EEOC contends that

MIA did not establish as a matter of law that the disparity in pay between the claimants

and the male comparators was due to the comparators’ credentials and prior work

experience.

In response, MIA argues that the EEOC failed to identify valid comparators and,

thus, did not make a prima facie showing of an EPA violation. 4 According to MIA, the

3
The EEOC does not appeal the district court’s denial of its motion for summary
judgment. Rather, the EEOC contends only that the district court erred in granting
summary judgment to MIA.
4
We observe that MIA does not directly address the EEOC’s argument that the
four male Fraud Investigator comparators, who earned higher salaries than the claimants
and performed identical work, comprise prima facie evidence of an EPA violation. As
(Continued)

9
identified males were not valid comparators, because MIA hired them at higher steps on

its pay scale than the steps to which the claimants were assigned. Alternatively, MIA

argues that even if the EEOC made a prima facie showing of an EPA violation, MIA

established as a matter of law that any pay disparity between the claimants and the

comparators was based on gender-neutral reasons involving the comparators’ prior

experience and credentials that each claimant lacked. We disagree with MIA’s

arguments.

A.

We review the district court’s award of summary judgment de novo. Matsushita

Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587–88 (1986); Carnell Constr.

Corp. v. Danville Redevelopment & Hous. Auth., 745 F.3d 703, 716 (4th Cir. 2014). We

consider the evidence and all inferences fairly drawn from the evidence in the light most

favorable to the EEOC. Carnell, 745 F.3d at 716.

The EPA prohibits gender-based discrimination by employers resulting in unequal

pay for equal work. 5 29 U.S.C. § 206(d)(1); Corning Glass Works v. Brennan, 417 U.S.

188, 195 (1974). The plaintiff creates a presumption of discrimination under the EPA

when she establishes a prima facie case. Brinkley-Obu v. Hughes Training, Inc., 36 F.3d

previously noted, MIA only challenges the male Enforcement Officers the EEOC
identified as comparators, an issue that we need not reach in this appeal.
5
The EPA is written in gender-neutral terms so that it is available to remedy
discriminatory actions against both men and women. Brinkley-Obu v. Hughes Training,
Inc., 36 F.3d 336, 350 n.32 (4th Cir. 1994).

10
336, 344 (4th Cir. 1994). We review the issue whether a plaintiff has made the required

prima facie showing using a burden-shifting framework. 6 Id.; see also Corning Glass,

417 U.S. at 196.

A plaintiff establishes a prima facie case of discrimination under the EPA by

demonstrating that (1) the defendant-employer paid different wages to an employee of the

opposite sex (2) for equal work on jobs requiring equal skill, effort, and responsibility,

which jobs (3) all are performed under similar working conditions. Corning Glass, 417

U.S. at 195. An EPA plaintiff need not prove that the employer acted with discriminatory

intent to obtain a remedy under the statute. Ryduchowski v. Port Auth. of N.Y. & N.J.,

203 F.3d 135, 142 (2d Cir. 2000); Miranda v. B & B Cash Grocery Store, Inc., 975 F.2d

1518, 1533 (11th Cir. 1992) (stating that the EPA “prescribes a form of strict liability”);

Sinclair v. Auto. Club of Okla., Inc., 733 F.2d 726, 729 (10th Cir. 1984).

Once a plaintiff has made the required prima facie showing, under the EPA, the

burdens of production and persuasion shift to the defendant-employer to show that the

wage differential was justified by one of four affirmative defenses listed in the statute.

Brinkley-Obu, 36 F.3d at 344. These affirmative defenses are: (1) a seniority system; (2)

a merit system; (3) a pay system based on quantity or quality of output; or (4) a disparity

6
The EPA burden-shifting framework is distinct from the McDonnell Douglas
burden-shifting framework that we apply when reviewing claims brought under Title VII.
See McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973). The McDonnell Douglas
framework proceeds in three steps in which the burden of persuasion remains with the
plaintiff throughout the entire analysis. Tex. Dep’t of Cmty. Affairs v. Burdine, 450 U.S.
248, 253 (1981).

11
based on any factor other than gender. 29 U.S.C. § 206(d)(1); Corning Glass, 417 U.S. at

195. To avoid liability, the defendant must prove one of these four affirmative defenses.

29 U.S.C. § 206(d)(1); Brinkley-Obu, 36 F.3d at 344. Accordingly, if the employer fails

to establish as a matter of law, for purposes of summary judgment, or to persuade a fact

finder, that one or more affirmative defenses are applicable in a particular case, the

plaintiff will prevail. See Brinkley-Obu, 36 F.3d at 344.

While an employer may be entitled to summary judgment on an EPA claim if the

employer establishes an affirmative defense as a matter of law, the burden on the

employer necessarily is a heavy one. 7 See Mickelson v. N.Y. Life Ins. Co., 460 F.3d 1304,

1312 (10th Cir. 2006); Stanziale v. Jargowsky, 200 F.3d 101, 107–08 (3d Cir. 2000). The

EPA prohibits disparities in pay between men and women “except where such payment is

made pursuant to” one of the four statutory affirmative defenses. 29 U.S.C. § 206(d)(1)

(emphasis added).

We agree with the Third and Tenth Circuits’ explanation that this statutory

language requires that an employer submit evidence from which a reasonable factfinder

could conclude not simply that the employer’s proffered reasons could explain the wage

disparity, but that the proffered reasons do in fact explain the wage disparity. See

Stanziale, 200 F.3d at 107–08; Mickelson, 460 F.3d at 1312 (quotation and citation

omitted). Thus, because the employer in an EPA action bears the burden of ultimate

7
In contrast, in a Title VII case, the employer need only proffer a legitimate,
nondiscriminatory reason for the challenged action, and is not required to establish that
the cited reason in fact motivated the employer’s decision. McDonnell Douglas, 411
U.S. at 802.

12
persuasion, once the plaintiff has established a prima facie case the employer will not

prevail at the summary judgment stage unless the employer proves its affirmative defense

so convincingly that a rational jury could not have reached a contrary conclusion. See

Stanziale, 200 F.3d at 108; Tenkku v. Normandy Bank, 348 F.3d 737, 741 n.2 (8th Cir.

2003) (“At the summary judgment stage of the proceedings, the employer’s justification

for the differences is irrelevant, unless it is strong enough to establish one of the statutory

affirmative defenses as a matter of law.”).

B.

We turn now to consider the issue whether the claimants made a prima face

showing of discrimination under the EPA. To satisfy their burden, the claimants were

required to show that (1) they were paid less than one or more males, for (2) performing

work of substantially equal skill, effort, and responsibility, and that (3) such work was

performed under similar working conditions. Corning Glass, 417 U.S. at 195. We hold

that the claimants have made this required showing.

First, the record plainly establishes, and MIA does not dispute, that the claimants

were paid less than the male comparators. Cordaro’s and Green’s respective starting

salaries, of $43,495 and $43,759, were lower than the starting salaries of all four male

comparators. Rogers’ starting salary of $46,268 was lower than Conticello’s starting

salary of $49,842. 8

8
The fact that there were other male MIA employees hired at the same grade and
step as the claimants at about the same time does not change our evaluation of the male
comparators identified by the EEOC. The EPA does not dictate to a plaintiff what
(Continued)

13
 The record also shows that the claimants and the male comparators performed

substantially equal work. It is undisputed that the claimants and the four male

comparators each held the same position as Fraud Investigators. Though sharing a job

title and a job description is not dispositive of this issue, see Brobst v. Columbus Servs.

Int’l, 761 F.2d 148, 155 (3d Cir. 1985), nothing in the record suggests that the male Fraud

Investigators performed work requiring skill, effort, or responsibility different from that

required by the work performed by the female Fraud Investigators. Indeed, MIA

affirmatively admitted before the district court that the claimants performed “identical

job[s] as other Insurance Fraud Investigators.”

The fact that other male employees at MIA performed substantially identical work

but made less money than the claimants does not require us to reach a different

conclusion. An EPA plaintiff is not required to demonstrate that males, as a class, are

paid higher wages than females, as a class, but only that there is discrimination in pay

against an employee with respect to one employee of the opposite sex. See Fowler v.

Land Mgmt. Groupe, Inc., 978 F.2d 158, 161 (4th Cir. 1992) (“A female plaintiff bears

the burden of proof of establishing a prima facie case by showing [among other things]

that [ ] her employer pays her a lower wage than a male counterpart . . . .” (emphasis

added)). The undisputed facts in the present record establish that each claimant earned

comparator she must choose, but requires only that, for the purposes of establishing a
prima facie case, she choose a comparator of the opposite sex who performs substantially
equal work for greater pay. 29 U.S.C. § 206(d); see Corning Glass, 417 U.S. at 195
(identifying the elements of an EPA prima facie case).

14
less than at least one male comparator performing substantially equal work. These

undisputed facts alone satisfy the EEOC’s prima facie burden.

Our conclusion is not altered by the fact that the male comparators were hired at

higher step levels than at least one of the claimants, allegedly based on their background

experience, relevant professional designations, and licenses or certifications. Such

experience and other factors are relevant only to any affirmative defense asserted by

MIA, not to the issue whether the EEOC has satisfied its prima facie burden. Instead, as

we have explained, at this initial stage we ask only whether the claimants and the

identified comparators worked jobs requiring “equal skill, effort, and responsibility,” and

whether each claimant was paid less than one or more comparators. See 29 U.S.C. §

206(d)(1); Fowler, 978 F.2d at 161. As both these requirements are met here, the EEOC

has established a prima facie case of wage discrimination under the EPA. 9

C.

Because the claimants established a prima facie case of discrimination under the

EPA, MIA was not entitled to summary judgment unless a rational jury could not have

rejected MIA’s proffered reasons for the wage disparities. See Stanziale, 200 F.3d at 107.

Here, MIA asserts one affirmative defense, namely, that a factor other than the

9
The dissent appears to take issue with Congress’s determination of what
constitutes a prima facie case of discrimination under the EPA. We are not free to
disregard the policy choices made by Congress but, instead, must ensure that both private
and public entities comply with the law as written.

15
comparators’ gender justifies the wage disparity. 10 29 U.S.C. § 206(d)(1). In support of

this defense, MIA offers two allegedly gender-neutral reasons for the disparity: (1)

MIA’s use of the state’s Standard Salary Schedule, which classifies each position to a

grade level and assigns each new hire to a step within that grade level; and (2) the

comparators’ experience and qualifications. We conclude that MIA has not adduced

evidence sufficient to require a factfinder to conclude that either of these two reasons

actually occasioned the undisputed wage disparities.

MIA cannot shield itself from liability under the EPA solely because MIA uses the

state’s Standard Salary Schedule and awards credit for prior state employment or a lateral

transfer within the state employment system. See Brinkley-Obu, 36 F.3d at 339 n.3, 340–

41, 353 (rejecting an employer’s argument that a jury verdict in the employee-plaintiff’s

favor should be reversed because of the employer’s gender-neutral compensation

system); Aldrich v. Randoph Cent. Sch. Dist., 963 F.2d 520, 525 (2d Cir. 1992).

Although the Standard Salary Schedule is facially neutral, MIA exercises discretion each

time it assigns a new hire to a specific step and salary range based on its review of the

hire’s qualifications and experience. A fact finder faced with the present record could

10
During the district court proceedings, MIA argued that its use of the Standard
Salary Schedule constituted a “merit system” for purposes of establishing a merit-based
defense under the EPA. See 29 U.S.C. § 206(d)(1). On appeal, however, MIA’s brief
focuses solely on the “factor other than sex” defense. We therefore do not address any
merit-based defense argument previously raised because such argument has been
abandoned. See Fed. R. App. P. 28(a)(6), (8) (stating that appellant’s brief must contain
the appellant’s “contentions and the reasons for them, with citations to the authorities and
parts of the record on which the appellant relies”); Edwards v. City of Goldsboro, 178
F.3d 231, 241 n.6 (4th Cir. 1999).

16
have determined that, when exercising this discretion, MIA at least in part based its

assignment of the claimants’ step levels on their gender with a resulting diminution of

their assigned starting salary. Therefore, while MIA uses a facially gender-neutral

compensation system, MIA still must present evidence that the job-related distinctions

underlying the salary plan, including prior state employment, in fact motivated MIA to

place the claimants and the comparators on different steps of the pay scale at different

starting salaries. See Stanziale, 200 F. 3d at 107–08.

MIA additionally argues that the pay disparities are justified by the difference

between the experience and qualifications of the comparators and the claimants. As an

example, MIA notes that Hurley and Conticello both earned Certified Fraud Examiner

designations, which MIA had advertised as a preferred designation during the hiring

process. Further, MIA observes that Hurley, Conticello, and Jacobs previously had

worked for the state of Maryland, and that Pennington had over 20 years of law

enforcement experience.

We agree that qualifications, certifications, and employment history fall within the

scope of the fourth affirmative defense which, as stated above, encompasses “a

differential based on any factor other than sex.” 29 U.S.C. § 206(d)(1); see, e.g., Ritter v.

Mount St. Mary’s Coll., 814 F.2d 986, 993 (4th Cir. 1987). And, certainly, the factors

cited by MIA could explain the wage disparity between the claimants and the

comparators. Nevertheless, as we have emphasized, a viable affirmative defense under

the EPA requires more than a showing that a factor other than sex could explain or may

17
explain the salary disparity. Instead, the EPA requires that a factor other than sex in fact

explains the salary disparity.

The present record does not show, as a matter of law, that the reasons proffered by

MIA do in fact explain the salary disparities. In particular, the record does not contain

any contemporaneous evidence showing that the decisions to award Hurley, Jacobs, and

Pennington their respective starting salaries were in fact made pursuant to their

aforementioned qualifications. And, although there is some contemporaneous evidence

regarding Conticello’s hiring, this evidence is not sufficient to hold as a matter of law that

his starting salary was assigned because of a factor other than sex. Although an official

at MIA recommended that Conticello be hired at a higher starting salary than the typical

investigator due to his prior experience, no evidence shows that the decision setting

Conticello’s salary was actually made on that basis.

Moreover, the claimants each had extensive prior investigative or law enforcement

experience. At the very least, their prior experience is relevant to the position of Fraud

Investigator and to the decisions fixing the claimants’ starting salaries, and creates an

issue of fact for the jury to decide whether MIA in fact objectively weighed the

comparators’ qualifications as being more significant than the claimants’ qualifications.

For these reasons, we conclude that the district court erred in granting summary

judgment in favor of MIA. Viewing the evidence in the light most favorable to the

claimants, a jury would not be compelled to find that the reasons proffered by MIA were,

in fact, the reasons for the disparity in pay awarded to the claimants and the comparators.

18
MIA’s affirmative defense must be weighed by a finder of fact against the other evidence

in the case.

III.

For these reasons, we vacate the district court’s order granting summary judgment

to MIA, and remand the case for further proceedings consistent with this opinion. 11

VACATED AND REMANDEd

11
On remand, among other things, the district court should address in the first
instance whether Green’s claim is timely. See 29 U.S.C. § 255(a); Nealon v. Stone, 958
F.2d 584, 591 n.5, 593 (4th Cir. 1992).

19
WILKINSON, Circuit Judge, dissenting:

The majority refuses to so much as mention a state’s sovereign interest in its own

civil service. The place of state governments in our Republic has quite passed it by.

Respect for states qua states fails to merit even the slight courtesies of lip service. The

majority instead sounds a requiem of silence: states as constitutional entities have quietly

expired.

And yet ours is not a Constitution of federal power only. The EEOC should not be

able to bring a case as marginal as this one against a state. State workforces are highly

regulated and regimented, and state law provides remedies for gender discrimination in

all its forms. Simply put, state civil service systems are not hotbeds of gender bias, as this

feeble suit makes all too clear.

I recognize that the EEOC is not acting in an adjudicative or regulatory capacity in

this case. By bringing suit, though, the agency has consigned the state’s sovereign interest

in its own workforce wholly to the hands of federal authority. Here, a federal agency is

bringing suit, the federal courts are deciding the suit, and federal law is providing the

applicable rule of decision. In combination, this assertion of federal authority diminishes

to an unacceptable extent the proper role of states in our constitutional system. Perhaps

this federal takeover could be pardoned if this were a lawsuit of substance. But this is a

thin, slight action.

Even if a federal statute can constitutionally be held to apply to the states, states

can still utilize the Tenth Amendment to vindicate in litigation their constitutionally

protected interests. At a very minimum, a case like this should not proceed to trial absent

20
clear and convincing evidence of the state’s misconduct. The lax preponderance burden

and summary judgment standard may be properly protective of a plaintiff’s trial rights in

most instances. But here they operate only to further subordinate a state’s status.

For a federal agency to bring this tenuous case raises serious constitutional

questions, and makes one wonder if Washington’s overlords even know what dual

sovereignty is all about. There are remedies for this kind of overreach, which I shall

promptly address.

I.

Ours is a federal system of government. It was carefully designed to strike a

balance between the need for enumerated federal authority and respect for the residual

sovereignty of the states. One half of that balance the majority now leaves wholly in

shadow. In fact, the majority does not even hint that this suit raises the question of the

extent to which Congress’s power may constitutionally extend over state civil service

systems. The answer to this question is fundamental to our constitutional structure, and it

is one the Supreme Court has struggled with for decades.

A.

For present purposes, our story begins with National League of Cities v. Usery,

426 U.S. 833 (1976), a case involving the Fair Labor Standards Act (FLSA). The FLSA

initially covered only private employers and was upheld as a valid exercise of Congress’s

commerce power. See United States v. Darby, 312 U.S. 100, 115–16 (1941). Congress

later amended it to also cover state employers. An assortment of states, cities, and their

representatives challenged these amendments as unconstitutional. See National League of

21
Cities, 426 U.S. at 836–37. They argued that “our federal system of government imposes

definite limits upon the authority of Congress to regulate the activities of the States as

States by means of the commerce power.” Id. at 842.

The Supreme Court agreed. It explained that Congress’s commerce power, while

broad, was nonetheless subject to distinct limitations. Id. at 841–43. One such limitation

is the structure of our federal system as reflected in the Tenth Amendment’s reservation

of powers to the states. Id. at 842–43. The Court stressed that “the States as States stand

on a quite different footing from an individual or a corporation when challenging the

exercise of Congress’ power to regulate commerce.” Id. at 854. States are themselves

“coordinate element[s] in the system established by the Framers for governing our

Federal Union.” Id. at 849. Allowing Congress to regulate “States in their capacities as

sovereign governments” under the guise of the commerce power “would impair the

States’ ‘ability to function effectively in a federal system.’” Id. at 852 (quoting Fry v.

United States, 421 U.S. 542, 547 n.7 (1975)). For these reasons, the Court held that

Congress lacked authority under the Commerce Clause to extend the FLSA’s minimum-

wage and overtime provisions to the States “in areas of traditional governmental

functions.” Id.

In Garcia v. San Antonio Metropolitan Transit Authority, 469 U.S. 528 (1985), the

Supreme Court by a 5–4 vote abandoned the project of limiting federal interference with

traditional state functions. That case presented the question whether “municipal

ownership and operation of a mass-transit system” was a “traditional governmental

function.” Id. at 530. Rather than answer this question, the majority opted to jettison the

22
entire inquiry. It reasoned that the task of identifying traditional governmental functions

was “difficult, if not impossible” because the Court had been unable to craft a well-

defined “organizing principle” in close to a decade. Id. at 539. And despite recognizing

that Congress has only limited powers in our constitutional system, the majority

determined that the Court had little role to play in enforcing those limits. It deemed the

“basic limit on the federal commerce power” to be more properly policed through the

“political process” and “state participation in federal governmental action” than in the

courts. Id. at 556. Based on this analysis, the Court overruled National League of Cities

and concluded that the FLSA could be constitutionally applied to state transit systems. Id.

at 531, 554.

Justices Rehnquist and O’Connor in dissent predicted that, like the reign of

National League of Cities before it, Garcia’s rule would be finite. Justice Rehnquist

predicted that the vision of federalism espoused in National League of Cities would “in

time again command the support of a majority of this Court.” Id. at 580 (Rehnquist, J.,

dissenting). Justice O’Connor agreed. She forecast that the Court would “in time again

assume its constitutional responsibility” for “oversee[ing] the Federal Government’s

compliance with its duty to respect the legitimate interests of the States.” Id. at 581, 589

(O’Connor, J., dissenting).

True to these predictions, in the years since Garcia, the Supreme Court has

proceeded to chip away at that case’s underlying rationale. The Court has rejected

discrete attempts at federal overreach and exhibited its commitment to protecting the

place of states in the constitutional scheme. See, e.g., United States v. Morrison, 529 U.S.

23
598, 617–19 (2000) (holding that the Violence Against Women Act’s civil remedy for the

victims of gender-motivated violence exceeded Congress’s commerce power because

regulation of “intrastate violence . . . has always been the province of the States”); Alden

v. Maine, 527 U.S. 706, 754 (1999) (holding that the FLSA exceeded Congress’s Article

I powers by abrogating state sovereign immunity in state courts); Printz v. United States,

521 U.S. 898, 919, 933 (1997) (holding that the Brady Handgun Violence Prevention Act

violated the Tenth Amendment by conscripting state law enforcement officers to perform

background checks on prospective handgun purchasers); Seminole Tribe of Florida v.

Florida, 517 U.S. 44, 71–73 (1996) (holding that the Indian Gaming Regulatory Act

violated the Eleventh Amendment by abrogating state sovereign immunity in federal

courts); United States v. Lopez, 514 U.S. 549, 564–68 (1995) (holding that the Gun-Free

School Zones Act’s prohibition on carrying firearms in school zones exceeded

Congress’s commerce power because it regulated a noneconomic activity in an area

“where States historically have been sovereign”); New York v. United States, 505 U.S.

144, 177 (1992) (holding that the Low-Level Radioactive Waste Policy Act exceeded

Congress’s commerce power because it compelled states to regulate the disposal of

radioactive waste according to Congress’s instruction).

Overruling Garcia is not a step any court would take lightly. Nor is it one we are

even empowered to take. See Rodriguez de Quijas v. Shearson/Am. Express, Inc., 490

U.S. 477, 484 (1989) (“[T]he Court of Appeals should follow the case which directly

controls, leaving to this Court the prerogative of overruling its own decisions.”). But it is

not improper for us to observe that in the wake of Garcia, the states’ corporeal interest in

24
our constitutional order has been left remarkably denuded. A state’s right to manage its

own workforce lies at the heart of this interest. This organic interest of the states differs

from the more peripheral interests implicated by many of the abovementioned holdings.

The organic interest of governments protects their ability to proceed as autonomous

bodies on a day-to-day basis. Or, to use an indelicate phrase, it relates to the states’

control over their own bodily functions.

To say that the sole constraints on federal oversight of state organic functions are

political is to ignore centuries of growth in the sheer size of the federal government,

centuries of expansion in the scope of federal control and areas of federal oversight, and

centuries of experience with the appetite of one part of government to enhance its own

power at another part’s expense. Much of this federal growth was not only necessary but

also salutary. Much of it helped to bind us economically as one nation and expand to all

our most basic protections of civil rights and liberties. But applauding a development

need not mean the loss of all sense of qualification and degree. The expanded federal

role, for all its value, risks leaving states abandoned by the wayside. The concentration of

federal power risks leaving liberty moribund and the vitality envisioned by the Framers

dispiritingly inert. Instead of both checks and balances, we are now left with the prospect

of irreversible centuries of neither.

B.

The states’ vulnerability to federal overreach is, as noted, alarming where their

sovereign interests are at their zenith: in the management of their own workforces. States

have passed extensive, gender-neutral regulations to govern their civil services. Such

25
regulations can and often do include state remedies to address gender discrimination in

all of its forms: the matter of pay inequity has not gone unnoticed.

Maryland is no exception in this regard. Like the federal government, Maryland

has its own Grade/Step system and its own Equal Pay for Equal Work Act, which

prohibits “paying a wage to employees of one sex or gender identity at a rate less than the

rate paid to employees of another sex or gender identity if both employees work in the

same establishment and perform work of comparable character or work on the same

operation, in the same business, or of the same type.” Md. Code, Lab. & Empl. § 3-

304(b)(1). This prohibition expressly applies to government employers, including the

state itself. Id. at § 3-301(b)(1). The Maryland legislature has also prohibited sex

discrimination of all kinds in state employment, Md. Code, State Pers. & Pens. § 2-

302(b)(1)(xi), and established an extensive Equal Employment Opportunity Program for

civil servants that in many ways mirrors the federal EEOC, id. at § 5-201 to -215. These

regulations both reflect the state’s robust sovereign interest in this area and the utilization

of that sovereign interest to achieve gender pay equity.

Maryland courts have hardly been reluctant to enforce these provisions, further

obviating the need for unwarranted federal overreach. Indeed, Maryland courts have

recognized that the EPA established only “a baseline rule” that states may and do

supplement in promoting pay equity, meriting robust enforcement of Maryland’s Equal

Pay for Equal Work Act. Gaskins v. Marshall Craft Assocs., Inc., 678 A.2d 615, 618–20

(Md. Ct. Spec. App. 1996). Other examples of Maryland courts enforcing employment

discrimination laws are simply too numerous to list, but even a brief sampling illustrates

26
the state’s willingness to take seriously these claims. See, e.g., Manikhi v. Mass Transit

Admin., 758 A.2d 95 (Md. 2000) (finding that a state government employee had a valid

claim for a hostile work environment based on sexual harassment); Molesworth v.

Brandon, 672 A.2d 608 (Md. 1996) (reinstating a jury verdict finding wrongful discharge

based on sex discrimination); Edgewood Mgmt. Corp. v. Jackson, 66 A.3d 1152 (Md. Ct.

Spec. App. 2013) (upholding jury damages award to plaintiff alleging she was terminated

on the basis of her sex). The active partnership between Maryland’s legislature and courts

reflects a healthy state system, not a hive of discrimination in need of a federal agency’s

intrusion.

And so the question thus arises: does the disregard of the Tenth Amendment and

the disparagement of our federal system lead to a debilitating duplicativeness in law that

brings but limited public benefit? One need not—and indeed, may not—go so far as to

revive National League of Cities to correct what has become a gross constitutional

imbalance. But it does no violence to Garcia to require that when a federal agency seeks

to supervise through litigation a state’s management of its own workforce, it must make a

clear and convincing case for doing so rather than proceeding under the preponderance of

the evidence threshold. This clear and convincing standard is a workable approach that

would serve to screen out deprivations of state sovereignty for trivial purposes, while

leaving serious problems open to the corrective of federal law. Steering this middle

course would restore for state governments in this narrow and discrete area of paramount

state interest some of the constitutional protection of which they have been inexcusably

deprived.

27
 II.

A.

Violations of the EPA require that the employer have “discriminate[d] . . . between

employees on the basis of sex.” 29 U.S.C. § 206(d)(1). Therefore, there is no violation

under the EPA if the salaries paid to employees are “made pursuant to (i) a seniority

system; (ii) a merit system; (iii) a system which measures earnings by quantity or quality

of production; or (iv) a differential based on any other factor other than sex.” Id.

(emphasis added). The so-called disparities in pay identified by the EEOC here are all

easily explained by neutral factors unrelated to gender, and these explanations are not in

genuine dispute.

The majority contends that “the burden on the employer necessarily is a heavy

one,” Majority at 11, conveniently overlooking the fact that not just one, but no fewer

than four affirmative defenses were provided by Congress in the statute. They were

provided for a reason, namely to prevent courts from reshuffling gender-neutral salary

structures to suit their policy preferences. And while the burden to establish an

affirmative defense rests with the one raising it, the ultimate burden of proof almost

always rests with “the party seeking relief.” Schaffer v. Weast, 546 U.S. 49, 57–58

(2005).

The majority gets wound up in the question of a prima facie case, see Majority at

14 n.9, but the question here is whether there is an issue of triable fact regarding the

state’s alleged pay inequity. The civil rights statutes of course all have shifting proof

schemes. But the Supreme Court has cautioned that there is a danger in an appellate court

28
getting wrapped around the axle of a proof scheme, which serves basically as a vehicle

for the orderly presentation of evidence before the district courts. See U.S.P.S. Bd. of

Governors v. Aikens, 460 U.S. 711, 716 (1983). In fact, the Court has emphasized that we

should not “treat discrimination differently from other ultimate questions of fact” or

otherwise “evade[] the ultimate question of discrimination vel non.” Id. at 714–16. On

that ultimate question, plaintiffs, as noted, bear the burden. And a modest respect for state

prerogatives with regard to the management of their own workforces requires that the

quantum of evidence needed to satisfy that burden must be a clear and convincing one.

Any diluted standard brings the Tenth Amendment and residual state authority under our

Constitution up short.

In discussing the elements of a prima facie case, the majority quite overlooks the

fact that a state’s interest comes in by way of a Tenth Amendment defense, which it does

not surrender even in those instances where a statute may be constitutionally applied to it.

Regardless, this case is so weak that it would falter even on the permissive terms the

majority has set for it. As the district court recognized, summary judgment was

appropriate in these circumstances. Its decision must be affirmed.

B.

The differences in pay identified by the EEOC are all readily explainable by state

policies crediting prior state employment and by differences in the experience and

qualifications of the individuals involved. There is no dispute that each of these

explanations is a neutral factor unrelated to sex. States, which have a clear interest in

serving their citizens well, may legitimately design a workforce composed of qualified

29
individuals best able to deliver superior service. States are able to attract and retain such a

workforce by tailoring their compensation schemes to the experience and credentials of

the people they hire. The proposition is so obvious that to state it risks embarrassment.

With respect to the case at hand, it is important for an insurance agency to be

staffed by people who are specialized in the types of claims typical of that agency’s work.

Because the majority skims over the relevant differences among the employees, I

describe each of the state’s justifications for its employees’ salaries in turn. 1

The state of Maryland has an overarching policy of rewarding dedicated civil

servants by ensuring that their salaries reflect their years of state employment. First,

Maryland law governing lateral transfers authorizes officials to freely reassign state

employees to “another position of equal grade and service.” Md. Code, State Pers. &

Pens. § 7-602(a)(1). By contrast, the State Personnel and Pensions Code provides for

reassignment to a “position of a lower grade in any unit” only when an employee

“appl[ies] for a voluntary demotion.” Id. at § 7-602(c)(1). The record reflects that MIA

applied these laws in a neutral manner to employees hired as lateral transfers, assigning

them to grades and salaries equivalent to those held in their prior positions. Mary Jo

1
In EPA cases, an employee alleging wage discrimination (a “complainant”) must
show that her pay is lower than that of a fellow employee (a “comparator”) who performs
“work substantially equal in skill, effort and responsibility under similar working
conditions.” Strag v. Board of Trs., 55 F.3d 943, 948 (4th Cir. 1995). This case involves
three complainants (Mary Jo Rogers, Marlene Green, and Alexandra Cordaro) and six
comparators (Bruno Conticello, Jay Hurley, Donald Jacobs, Homer Pennington, Jeffrey
Gross, and Maurice Xenos). The parties dispute whether Gross and Xenos are valid
comparators, and neither the majority nor I decide this question. However, I address their
qualifications to demonstrate that summary judgment is appropriate regardless.

30
Rogers, one of the complainants, was informed of her new position in MIA in a letter

explaining: “Since this is a lateral transfer, your classification will remain . . . Grade 15

step 5 which equates to an annualized salary $ 47,018.” J.A. 92. Similarly, when the

Associate Commissioner of the Insurance Fraud Division, Carolyn Henneman,

recommended hiring Bruno Conticello, one of the comparators, her letter to the HR

Director noted that he would need to be “offered a somewhat higher salary than the

typical line investigator” in part because he was “already being paid significantly higher

than that line salary” at his prior state job. J.A. 328.

Maryland law also required MIA to take prior state employment into account in

setting salaries in other ways. For example, former employees who are reinstated in the

civil service are entitled to “receive credit for time employed before separation” when

they are assigned to a pay grade. Md. Code, State Pers. & Pens. § 2-601(c)(1). And

Maryland employees transferring from a contract position to a budgeted position within

the same principal unit “shall be given credit for service in the contractual position for the

purpose of establishing” their pay grade. Id. at § 13-304.

The complainants and the comparators were not similarly situated with respect to

prior state employment. Rogers was the only complainant who was already employed by

the state when she was hired as a Fraud Investigator. By contrast, among the

comparators, Bruno Conticello, Maurice Xenos, and Jeffrey Gross were all employed by

the state at the time they were hired as budgeted Fraud Investigators or Enforcement

Officers. As noted above, Conticello’s salary reflected his lateral transfer. Xenos and

Gross were both hired after acting as contract employees, and they were therefore

31
statutorily entitled to credit for that time under § 13-304. Comparators Hurley and Jacobs

were not employed by the state at the time they were hired, but the EEOC acknowledges

that they nonetheless had creditable prior state work experience, affecting their starting

salaries. EEOC Reply Br. 25.

Differences in the extent of creditable prior state employment thus explain much

of the variation in salaries among both the male and female employees: The EEOC does

not dispute that, among the complainants, only Rogers had prior state work experience at

least equivalent to that of Conticello, Hurley, or Jacobs. And, predictably enough, Rogers

was more highly compensated than the other two complainants. Not only that, she was

given a higher starting salary than either Hurley or Jacobs.

Of course, prior state employment was not the only factor involved in setting

starting salaries at MIA. Beyond prior state employment, there were also important

differences in the qualifications and credentials of the complainants and comparators.

Comparators Conticello and Hurley were both Certified Fraud Examiners, and Gross held

an insurance producer (agent) license, credentials none of the complainants had achieved.

Comparator Pennington had training from the National Fire Academy on arson

investigations, highly relevant training in the insurance field that none of the

complainants had. Xenos had a special expertise in title insurance, allowing him to serve

a role at MIA that none of the complainants could have served. All of these credentials,

held by the comparators and not by any of the complainants, are indisputably relevant to

the work of MIA. And while all of the employees had at least some relevant experience,

not all of the employees had equal prior experience. For example, comparator Pennington

32
had thirty years of experience in law enforcement, more than any of the complainants.

These are important components of MIA’s mission, and expertise in these matters on the

part of MIA employees is preferred.

For each comparator, then, the salary assigned by MIA is explained by prior state

experience, specialized credentials, other exceptional work experience, or a combination

of these factors. Conticello’s salary was based on matching his prior salary after he

laterally transferred into MIA, and is additionally justified by his Certified Fraud

Examiner qualification. Xenos’s and Gross’s salaries were enhanced by their creditable

prior state employment and their abovementioned specialized expertise. Comparators

Hurley, Jacobs, and Pennington all had lower starting salaries than complainant Rogers,

and they made more than complainants Green and Cordaro based on the creditable prior

state employment of Hurley and Jacobs, Hurley’s Certified Fraud Examiner qualification,

and Pennington’s arson training and lengthy law enforcement experience. Gender is not

required to understand any of these decisions.

These conclusions are further bolstered by looking to the salaries of other MIA

employees. While the EEOC identifies a number of men who were paid more than

Cordaro, the lowest-paid complainant, it ignores the numerous examples of men who

were paid the same amount or less than she was paid. Michael Stefanowitz and Thomas

Bradford were each hired at lower starting salaries than Cordaro’s. Todd Young, a Fraud

Investigator at MIA, started prior to Cordaro, and with a higher starting salary. But his

salary was later reduced by the same temporary salary reduction that affected Cordaro’s

starting salary, so at the time she was hired, they had identical salaries. Following a

33
reclassification of all Fraud Investigators to a higher compensation grade in 2013,

Cordaro’s compensation was identical to the male Fraud Investigators Williams Johns,

Thomas Zanfardino, and Edward Spragg. And notably, the record indicates that the

highest starting salary offered to any new Fraud Investigator during the relevant period

was $51,809, an amount offered to both Mark Bilger, a man, and Novia Maduro, a

woman. The overall salary history of the department, then, strengthens the impression

given by examining only the complainants and comparators: MIA assigned salaries based

on an individualized assessment of each new employee, not based on gender.

Given the large number of individuals involved here and the variation in each

individual’s background, it is easy to become focused on minutiae in this case. When

considered together, though, the brushstrokes of each salary decision paint a clear picture

of a state agency taking valid factors unrelated to gender into account when setting

salaries. MIA necessarily took prior state employment into account: the comparators had

more creditable state employment history than the complainants. MIA took expertise and

relevant experience into account: the comparators had more credentials and relevant

experience than the complainants. These straightforward explanations are more than

sufficient to foreclose an EPA claim against the agency. The majority says the question is

whether these explanations “in fact” explain the differentials as opposed to whether they

“could.” Majority at 11. And the majority is certainly correct that discriminatory intent is

not required in an EPA action. Id. at 10. But the explanations provided by the MIA do in

fact explain the differentials in this case. Therefore, summary judgment in favor of MIA

was appropriate in this case, even on the majority’s own terms.

34
 III.

I return briefly to where I began. The majority treats the state of Maryland as just

any ole defendant, and that is not right. States are not mere abstract entities reserved for

civics lectures. They discharge important tasks of governance and they house human

heartbeats. The Framers made them a significant part of our constitutional architecture,

which the majority has no right to dismantle.

The majority has now allowed a federal agency possessed of no more than a

transparently thin and insubstantial case to abrogate a state’s interest in managing its own

workforce. It is foundational to both our civil and our criminal justice systems that the

ultimate burden resides with the party bringing suit. How much more must this be the

case when the most basic ingredients of state autonomy are implicated. The structure of

our Constitution—adumbrated specifically in the Tenth Amendment but infused

throughout that document—would seemingly require that pretrial dismissals be more,

rather than less, available to state governments attempting to vindicate the place that the

Constitution has assigned to them. The intrusion here is not justified by the evidence,

under the law, or by the state’s weighty interest in managing its own employees.

Requiring a federal agency to meet its ultimate burden here by clear and convincing

evidence, not a mere preponderance, would seem one modest step in restoring the dignity

of our fifty states in the constitutional design. The clear and convincing standard allows

the federal government to proceed in every case of real wrongdoing while deterring the

assertion of federal authority in less-than-marginal instances such as this. Because the

35
majority decision diminishes the rightful place of our states in every way conceivable or

possible, I respectfully dissent.

36