Tracy Sempowich v. Tactile Systems Technology

                                        PUBLISHED

                           UNITED STATES COURT OF APPEALS
                               FOR THE FOURTH CIRCUIT


                                        No. 20-2245


TRACY SEMPOWICH,

               Plaintiff – Appellant,

        v.

TACTILE SYSTEMS TECHNOLOGY, INC., d/b/a Tactile Medical,

               Defendant – Appellee,

------------------------------

EQUAL EMPLOYMENT OPPORTUNITY COMMISSION,

               Amicus Supporting Appellant.



Appeal from the United States District Court for the Eastern District of North Carolina, at
Raleigh. James C. Dever III, District Judge. (5:18-cv-00488-D)


Argued: October 27, 2021                                     Decided: December 3, 2021


Before WILKINSON, NIEMEYER, and MOTZ, Circuit Judges.


Vacated and remanded by published opinion. Judge Motz wrote the opinion, in which
Judge Wilkinson and Judge Niemeyer joined.


ARGUED: Kathryn F. Abernethy, NOBLE LAW FIRM, PLLC, Chapel Hill, North
Carolina, for Appellant. Julie Loraine Gantz, EQUAL EMPLOYMENT OPPORTUNITY
COMMISSION, Washington, D.C., for Amicus Curiae. Kristin Berger Parker, STINSON
LLP, Minneapolis, Minnesota, for Appellee. ON BRIEF: Carroll Theresa Wright,
STINSON LLP, Minneapolis, Minnesota; Theresa Sprain, Jonathon D. Townsend,
WOMBLE BOND DICKINSON (US) LLP, Raleigh, North Carolina, for Appellee.
Sharon Fast Gustafson, General Counsel, Jennifer S. Goldstein, Associate General
Counsel, Elizabeth E. Theran, Assistant General Counsel, Office of General Counsel,
EQUAL EMPLOYMENT OPPORTUNITY COMMISSION, Washington, D.C., for
Amicus Curiae.




                                        2
DIANA GRIBBON MOTZ, Circuit Judge:

       This appeal arises from the district court’s grant of summary judgment to Tactile

Systems Technology, Inc. on former employee Tracy Sempowich’s discrimination,

retaliation, and Equal Pay Act claims. Because the court applied an incorrect legal standard

to the Equal Pay Act claim and erred in holding that there are no genuine issues of material

fact precluding summary judgment on the other claims, we must vacate its judgment and

remand for further proceedings consistent with this opinion. 1

                                          I.

                                          A.

       Tactile, a medical device company, sells compression devices to treat chronic

swelling and wounds. In 2007, Tactile hired Tracy Sempowich — a woman — as a field

sales employee, a position known at the company as a “product specialist.” Sempowich

briefly left full-time employment in 2009 but continued to work with Tactile as an

independent contractor during that time. In 2010, Tactile rehired her as a full-time product

specialist and subsequently promoted her to a senior product specialist.

       Four years later, Tactile again promoted Sempowich — then forty-nine years old —

to be the regional sales manager for the Mid-Atlantic region. In this role, Sempowich


       1
         In granting Tactile’s motion for summary judgment, the district court also granted
Tactile’s motion to strike Sempowich’s proffered expert testimony and dismissed as moot
Sempowich’s own motion for partial summary judgment and motion to strike Tactile’s
responsive statement of material facts. As discussed below, we vacate the grant of Tactile’s
motion to strike Sempowich’s proffered expert testimony and remand for further
proceedings consistent with this opinion. On remand, the district court should also
reconsider and decide Sempowich’s motions that it dismissed as moot and any motions
that may otherwise be revived.
                                               3
supervised a sales team of up to fifteen people for a region then consisting of Maryland,

North Carolina, part of South Carolina, and Virginia. Later that year, Tactile hired Greg

Seeling — a forty-six-year-old man — to be the regional sales manager for the Southern

region, consisting of Alabama, Arkansas, Florida, Georgia, Louisiana, Mississippi, part of

South Carolina, Tennessee, and West Virginia.

                                          B.

       The above facts are undisputed, but the record is rife with other facts that are in

serious dispute. The disputed facts relate primarily to Sempowich’s performance as a

regional sales manager, about which Tactile and Sempowich have offered starkly different

testimony and documentary evidence.

       On one hand, Tactile maintains that Sempowich failed to meet the company’s

performance goals. Tactile’s Senior Vice President of Sales, Bryan Rishe, testified that

Sempowich oversaw lagging year-over-year growth, high employee turnover, and slow

hiring in her region, and that there was a “lack of professional development of [her

region’s] personnel.” He further testified that these issues “had challenged the region since

2015” and that he had discussed them with Sempowich “on a number of occasions” and

“tried to assist her with recruitment,” to no avail. Tactile also points out that Sempowich

stated on a call with Rishe that she “couldn’t grow the way” that he was “measuring [her]

on from last year with the fact that there was a lot of things that were out of [her] control

from a business perspective.” And in a business action plan, Sempowich acknowledged

that her region’s “biggest hur[d]le has been headcount and the ability for expansion,”

noting that hers was “the only tenured region that has not maximized expansion

                                               4
opportunities or had increase of territory.” Rishe testified that he and the company’s CEO

“concluded [that] a change of management was needed to turn around performance of the

Mid-Atlantic Region.”

       On the other hand, Sempowich testified and offered documentary evidence showing

that Tactile consistently viewed her as a top performer. She testified that Vice President

Rishe never “explicitly” told her she had “performance deficiencies” that she needed to

work on to keep her position. In fact, she offered documentary evidence that in two of

Tactile’s recent annual evaluations of her performance, Rishe listed her as having

“[e]xceptional [s]trengths” in people development, team building, leadership, and

planning, organization, and execution skills.       Sempowich also testified and offered

documentary evidence that Tactile repeatedly gave her awards, including a Regional

Manager Sales Leadership Award three years in a row for exceeding the revenue plan in

her region and, at the national sales meeting on January 21–24, 2018, an award for

Sustained Excellence (an honor that, according to Sempowich, no other current regional

sales manager received at that time). In addition, she testified that in January 2018, Tactile

informed her that it would give her a discretionary equity grant of $40,000 and a $10,000

salary raise effective February 1, 2018.

       Moreover, Sempowich offered evidence that Tactile viewed her not only as a top

performer but also as a better performer than Seeling. In their 2015 evaluations, Tactile

rated Sempowich as a Key Contributor (the third-highest possible rating) and Seeling only

as a Contributor (the fourth-highest); and in their 2016 evaluations, Tactile rated



                                              5
Sempowich as a Major Contributor (the second-highest) and Seeling only as a Key

Contributor (the third-highest).

       Nevertheless, on February 12, 2018, Rishe informed Sempowich that she would no

longer be a regional sales manager, that Tactile would reassign her region to Seeling, and

that Seeling would be promoted to area director (a step above regional sales manager).

Tactile offered Sempowich a newly created position as a market development manager for

its “head and neck” business, in which she would retain the same base salary. But

Sempowich viewed this offer as a demotion — she saw the new position “more like a sales

job” with only a nominal title as manager, especially because she would no longer have

any employees reporting directly to her. And the formal offer letter that Tactile later

provided did not mention a plan that would allow her to earn incentive compensation after

the expiration of a six-month guarantee.

       Ten days later — on February 22, 2018 — Sempowich submitted a complaint to

Tactile’s Human Resources department, alleging that Tactile had discriminated against her

on the basis of sex and age. On March 23, Tactile’s counsel informed Sempowich’s

counsel that if she failed to accept the offer to become a market development manager, her

employment with Tactile would cease effective March 30. Sempowich did not accept, and

on March 30 her employment ended.

       Sempowich then sued Tactile in state court, alleging: (1) Title VII disparate

treatment on the basis of sex and sex-plus-age; (2) wrongful termination under North

Carolina state law; (3) Title VII retaliation; and (4) a violation of the Equal Pay Act. Tactile

removed the case to federal court. Once in federal court, Tactile moved for summary

                                               6
judgment on each of Sempowich’s claims.             The district court granted the motion.

Sempowich then noted a timely appeal to this court.

       We review a district court’s grant of summary judgment de novo, “applying the

same legal standards as the district court and viewing all facts and reasonable inferences in

the light most favorable to the nonmoving party.” Ballengee v. CBS Broad., Inc., 

 (4th Cir. 2020). Of course, summary judgment is appropriate only if “the movant

shows that there is no genuine dispute as to any material fact and the movant is entitled to

judgment as a matter of law.” Fed. R. Civ. P. 56(a).

                                          II.

       To survive a motion for summary judgment on a Title VII disparate treatment claim,

a plaintiff must either proceed under the mixed-motive framework or the McDonnell

Douglas burden-shifting framework. See Perkins v. Int’l Paper Co., 

, 206 n.4

(4th Cir. 2019); see also McDonnell Douglas Corp. v. Green, 

 (1973). Here,

the district court held that Sempowich failed to present the direct evidence required to

proceed under the mixed-motive framework, a contention that Sempowich has not

challenged on appeal. Accordingly, the court analyzed her claims under the McDonnell

Douglas burden-shifting framework.

       Under the burden-shifting framework, a plaintiff must first offer a prima facie case.

Lettieri v. Equant, 

 (4th Cir. 2007). To do so, a plaintiff must show that

(1) she is a member of a protected class; (2) her employer took an adverse action against

her; (3) she had been fulfilling her employer’s legitimate expectations at the time of the

adverse action; and (4) the adverse action occurred under circumstances that raise a

                                                7
reasonable inference of unlawful discrimination, including because the employer left open

the position or replaced the plaintiff with someone outside the protected class. Id.; Bing v.

Brivo Sys., LLC, 

, 616 n.8 (4th Cir. 2020). Once a plaintiff makes out a prima

facie case, the burden shifts to the employer to put forth a nondiscriminatory explanation

for its actions. Lettieri, 

. If the employer does so, the burden then shifts

back to the plaintiff to show that the employer’s explanation was “actually a pretext for

discrimination.” 

 (quoting Hill v. Lockheed Martin Logistics Mgmt., Inc., 

,

285 (4th Cir. 2004) (en banc)).

       The district court held that Sempowich failed to make out a prima facie case because

she was not fulfilling Tactile’s legitimate expectations at the time the company reassigned

her. It then held that, even if she had made out a prima facie case, Tactile had put forth a

nondiscriminatory explanation for reassigning her, and no rational jury could find that

Tactile’s explanation was a pretext for discrimination.

                                            A.

       On Sempowich’s prima facie case, the only factor at issue before this court is

whether Sempowich was fulfilling Tactile’s legitimate expectations at the time it took an

adverse action against her. To satisfy this factor, a plaintiff need not “show that [s]he was

a perfect or model employee. Rather, a plaintiff must show only that [s]he was qualified

for the job and that [s]he was meeting [her] employer’s legitimate expectations.” Haynes

v. Waste Connections, Inc., 

 (4th Cir. 2019).

       Tactile asserts that we may solely consider the “perception of the [employer]” on

this factor, “not the self-assessment of the plaintiff,” and that “an employer is free to set its

                                                 8
own performance standards.” Br. of Appellee at 21–22 (first quoting Hawkins v. PepsiCo,

Inc., 

 (4th Cir. 2000); then quoting Beall v. Abbott Lab’ys, 

 (4th Cir. 1997)). It is not clear that Tactile is correct — although we have held that

we must focus on the employer’s perception in the context of the pretext stage, we have

not so held with respect to a plaintiff’s prima facie case. See, e.g., Hawkins, 203 F.3d at

279–80; Beall, 130 F.3d at 619–20; DeJarnette v. Corning Inc., 

 (4th

Cir. 1998). But even assuming that we must focus on just the employer’s perception at the

prima facie stage, a plaintiff may still introduce “evidence that demonstrates (or at least

creates a question of fact) that the proffered ‘expectation’ is not, in fact, legitimate at all.”

Warch v. Ohio Cas. Ins. Co., 

 (4th Cir. 2006).

       Viewing the evidence in the light most favorable to Sempowich, there is an issue of

material fact as to whether Tactile’s asserted expectations were legitimate or genuine.

Sempowich presented substantial evidence that they were not. If an employer genuinely

believed that one of its employees was performing poorly on metrics the employer

perceives as critical (as Tactile claims here), it seems likely that it would at the very least

not rate the employee’s performance highly or give her awards, a salary raise, or an equity

grant. And yet there is evidence that Tactile (1) consistently rated Sempowich’s overall

performance highly (and notably, higher than the employee who it reassigned to her

position); (2) repeatedly gave her awards, including one for Sustained Excellence three

weeks before it told her that it would reassign her position; (3) told her that it would give

her a salary raise three weeks before it told her that it would reassign her position; and



                                               9
(4) gave her a discretionary equity grant twelve days before it told her that it would reassign

her position.

       We have previously held that similar evidence was sufficient to survive summary

judgment. In Haynes v. Waste Connections, we held that a plaintiff had “raise[d] the

reasonable inference . . . that he was performing at a satisfactory level” because there was

evidence that the employer had given the plaintiff bonuses for the period in question and

told him “mere weeks before his termination” that “everything looks good” and that he had

“nothing to worry about.” 

. As in Haynes, there is evidence that Tactile

signaled to Sempowich that it viewed her overall performance positively.

       Tactile argues that here we cannot consider the performance evaluations, awards,

salary raise, or equity grant because they relate to Sempowich’s performance prior to the

time it reassigned her. But some of Sempowich’s evidence relates to events that occurred

three weeks or even twelve days before Tactile informed her of the reassignment. And the

two annual evaluations that Sempowich points to are Tactile’s most recent annual

evaluations of her; not only is there no more-recent negative annual evaluation for us to

consider, but there is no more-recent annual evaluation at all. Somewhat inexplicably,

Vice President Rishe nevertheless testified that he believed Sempowich’s supposed

weaknesses had been apparent “since 2015.” Given the awards, salary raise, and equity

grant that Tactile has given Sempowich since 2015, a reasonable factfinder could find this

testimony not credible. And if Tactile asserts that it reassigned Sempowich’s position due

to weaknesses she supposedly demonstrated throughout most of her tenure, it cannot



                                              10
simultaneously argue that the court should ignore evidence about her performance during

that same period.

       This does not mean that Sempowich is entitled to summary judgment or even that a

factfinder will ultimately find in her favor on the disparate treatment claims. Our holding

is simple — a court cannot grant a party summary judgment when there are genuine issues

of material fact, and here the record reveals factual disputes as to one of the key elements

of Sempowich’s prima facie case.

                                            B.

       The record is similarly replete with genuine issues of material fact that go to the

heart of the pretext issue. The district court erred in analyzing pretext not only by failing

to account for those disputes but also by incorrectly applying the same-actor inference. 2

       First, the district court erred in holding that there is no genuine issue of material fact

as to pretext. “[T]o show pretext, a plaintiff may show that an employer’s proffered

nondiscriminatory reasons for the termination are inconsistent over time, false, or based on


       2
         Sempowich also argues that the district court incorrectly applied the pretext-plus
standard that the Supreme Court abrogated in Reeves v. Sanderson Plumbing Products,
Inc., 

 (2000). In Reeves, the Supreme Court held that “[i]n appropriate
circumstances, the trier of fact can reasonably infer” pretext “from the falsity of the
[employer’s] explanation,” but that there may be instances in which such evidence is
insufficient, for example, “if the record conclusively revealed some other,
nondiscriminatory reason for the employer’s decision.” 

 at 147–48. The record is
unclear as to whether the district court failed to follow Reeves. But regardless, the district
court erred in holding that Sempowich could not rely solely on the falsity of Tactile’s
articulated explanation to show pretext. This is so because this is not a case in which “the
record conclusively revealed some other, nondiscriminatory reason for the employer’s
decision.” 

. As a result, we need not address other evidence that Sempowich
proffers to attempt to show a discriminatory motive.

                                              11
mistakes of fact.” Haynes, 

. “Once the plaintiff offers such circumstantial

evidence, the case must be decided by a trier of fact and cannot be resolved on summary

judgment.” 

 Here, Sempowich has introduced a good deal of evidence suggesting that

Tactile’s explanations for its decisions were false or inconsistent over time. Much of this

evidence has already been discussed above in considering Sempowich’s prima facie case.

But she has also introduced other evidence that supports her assertion of pretext. For

example, the fact that Tactile replaced her with Seeling is certainly evidence of pretext.

This is so because a jury might well conclude it unlikely that an employer who reassigned

an employee solely because it believed that she performed poorly would replace her with

an employee whose performance it consistently rated as worse.

       Sempowich has also introduced evidence that challenges the accuracy of the

statistics Tactile used to measure its chosen performance standards. Sempowich testified

that Tactile removed Maryland and part of Virginia from her region in January 2017, even

though she had previously told Vice President Rishe that “most of the Region’s significant

sales growth over the next two years would likely come from [those] territories.” And yet

Tactile’s year-over-year growth statistics may have failed to account for this external

factor. See Miles v. Dell, Inc., 

 (4th Cir. 2005) (observing, in a different

statutory context, that a rational jury could find that by reducing an employee’s sales

territory and increasing her quotas, the employer’s actions “set her up for the failures that

he later used to make the case for her termination”). The same logic applies here — an

employer may be free to choose its own performance standards, but it cannot measure those



                                             12
standards with distorted data that fails to account for factors over which the plaintiff had

no control. 3

       Tactile argues that none of this evidence shows that its stated reasons were false or

inconsistent, but rather merely reflects that Sempowich disagrees with the usefulness of its

chosen performance standards. Tactile points out that in Hawkins v. PepsiCo, Inc., we held

that a plaintiff must offer evidence that an employer’s assessment is “dishonest or not the

real reason for her termination,” rather than merely “dispute[] the merits of [the

employer’s] evaluations.” 

. It is true that courts must defer to the

company’s business judgment with regard to legitimate criteria it chooses to measure

successful employee performance. 

 We are not free to substitute criteria of our own.

But Sempowich has done more than challenge the criteria or merits of Tactile’s

evaluations. Sempowich has done what the plaintiff in Hawkins failed to do — “supply

evidence that [her employer] actually believed her performance was good.” 

(emphasis added).     This evidence is the employer’s own words and actions — the



       3
          Sempowich also offered an expert report, and later a supplemental declaration, by
a certified public accountant who “investigate[d] what is actually being measured by”
Tactile’s year-over-year growth statistics. The district court excluded both the report and
declaration, holding that the report was not relevant and that Sempowich had not previously
disclosed the opinions laid out in the declaration. The court abused its discretion in doing
so. The report is clearly relevant because it could help a factfinder understand the accuracy
of Tactile’s statistics — in the report, the expert stated that Tactile’s statistics “did not
isolate and track any particular action over which a Regional Sales Manager could
reasonably be expected to exert control.” And Sempowich did previously disclose the
opinions in the declaration by outlining them in the earlier report, in which the expert
“opine[d] on whether Tactile’s year-over-year growth statistics are reproducible and
statistically valid” by stating that, “[d]espite working with several different types of reports
generated by Tactile, [she] was unable to reproduce the statistics shown in Exhibit 38.”
                                              13
performance ratings, awards, salary raise, and equity grant. Each of these pieces of

evidence indicates that Tactile not only thought that Sempowich was performing

satisfactorily, but that her performance was of such a high quality that it deserved repeated

praise.

          The district court also erred in applying the same-actor inference to dispose of

Sempowich’s claim. Under the same-actor inference, if the plaintiff’s “hirer and the firer

are the same individual and the termination of employment occurs within a relatively short

time span following the hiring, a strong inference exists that discrimination was not a

determining factor.” Proud v. Stone, 

 (4th Cir. 1991) (emphasis added).

But Rishe did not reassign Sempowich “within a relatively short time span” after he rehired

her — far from it. In Proud, the time span was less than six months. 

. Here,

Rishe reassigned Sempowich approximately eight years after he rehired her and four years

after he promoted her. Moreover, there is a genuine issue of material fact as to whether

Rishe rehired and promoted Sempowich under protest: although Rishe ultimately approved

rehiring Sempowich, she testified that he told her that he was “not in favor” of doing so.

And when Rishe learned that Sempowich wanted to be promoted to regional sales manager,

he told her that he “didn’t think [she] wanted to get into management.” Under these

circumstances, the same-actor inference cannot support a grant of summary judgment to

the defendant.




                                             14
       Because the record is replete with genuine issues of material fact as to both the prima

facie case and pretext, we vacate the grant of summary judgment on the disparate treatment

claims and remand for further proceedings consistent with this opinion. 4

                                           III.

       Sempowich also claimed that Tactile violated Title VII by retaliating against her for

submitting her discrimination complaint. To establish a prima facie case of retaliation

under the burden-shifting framework, a plaintiff must show: “(i) ‘that [she] engaged in

protected activity,’ (ii) ‘that [her employer] took adverse action against [her],’ and

(iii) ‘that a causal relationship existed between the protected activity and the adverse

employment activity.’” Foster v. Univ. of Md.-E. Shore, 

 (4th Cir. 2015)

(alterations in original) (quoting Price v. Thompson, 

 (4th Cir. 2004)).

The burden then shifts to the employer to demonstrate that “its purportedly retaliatory

action was in fact the result of a legitimate non-retaliatory reason.” 

 If the employer

does so, the burden shifts back to the plaintiff to show that “the employer’s purported

nonretaliatory reasons ‘were not its true reasons, but were a pretext for discrimination.’”

 (quoting Hill, 354 F.3d at 285).


       4
          The district court also granted Tactile’s motion for summary judgment on
Sempowich’s state law wrongful termination claim “for the same reason” as her Title VII
disparate treatment claims. Because we vacate the grant of summary judgment on the Title
VII disparate treatment claims, we also vacate the district court’s holding as to the state
law wrongful termination claim and remand for further proceedings consistent with this
opinion. We need not reach aspects of the state law claim that the district court did not
address. See Hulsey v. Cisa, 

 (4th Cir. 2020) (“It is the general rule, of
course, that a federal appellate court does not consider an issue not passed upon below,”
especially if doing so “would require extensive analysis of issues never addressed by the
district court.” (quoting Singleton v. Wulff, 

 (1976))).
                                             15
      As to the prima facie case, the district court assumed without deciding that

Sempowich engaged in protected activity and that Tactile took an adverse action against

her. But it then held that no rational jury could find that a causal relationship existed,

reasoning that (1) temporal proximity alone cannot establish a causal relationship; and (2)

no temporal proximity existed in Sempowich’s case. The court erred on both counts.

       First, the court erred by holding that temporal proximity alone cannot establish a

causal relationship. We have made abundantly clear that temporal proximity suffices to

show a causal relationship. We explained this in Strothers v. City of Laurel, 

(4th Cir. 2018). A plaintiff may establish a causal relationship “simply by showing that

(1) the employer either understood or should have understood the employee to be engaged

in protected activity and (2) the employer took adverse action against the employee soon

after becoming aware of such activity.” 

 (emphasis added).

       Second, the district court erred by holding that there was not temporal proximity in

Sempowich’s case. The court reasoned that, because Vice President Rishe told Sempowich

on February 12 that Tactile would reassign her region, and Sempowich did not submit her

internal discrimination complaint until February 22, Tactile’s adverse actions could not

have been caused by Sempowich’s internal complaint. But there is a genuine issue of

material fact as to whether Rishe made clear on February 12 that her employment with

Tactile would end if she did not accept the reassignment. It was not until March 23 —

about a month after Sempowich submitted her internal complaint — that Tactile stated that,

if she did not accept the offer of reassignment, her “employment with Tactile [would] end

effective March 30.”

                                            16
       Tactile relies largely on Francis v. Booz, Allen & Hamilton, Inc., 

 (4th

Cir. 2006), to assert that there is no temporal proximity. In that case, we held that, “[w]here

timing is the only basis for a claim of retaliation, and gradual adverse job actions began

well before the plaintiff had ever engaged in any protected activity, an inference of

retaliation does not arise.” 

 at 309 (quoting Slattery v. Swiss Reins. Am. Corp., 

 (2d Cir. 2001)). But in Sempowich’s case, the adverse actions were far from

“gradual”; nor did they begin “well before” she engaged in protected activity. A reasonable

factfinder could find that after Sempowich submitted an internal complaint, Tactile decided

to take the more drastic approach of telling her that her employment would end if she failed

to accept the reassignment offer.

                                           IV.

       Unlike her Title VII claims, Sempowich’s final claim — an asserted violation of the

Equal Pay Act — does not turn on whether there was a genuine issue of material fact. On

the Equal Pay Act claim, there are few, if any, material facts in dispute. Rather, the claim

turns on the appropriate legal standard for determining whether an employer has violated

the Act.

       To establish a prima facie case of an Equal Pay Act violation, a plaintiff must

demonstrate 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.” EEOC v. Md. Ins.

Admin., 

 (4th Cir. 2018).



                                              17
         In Sempowich’s case, the district court assumed without deciding that Sempowich

had demonstrated the second and third factors of her prima facie case. But it then held that

Sempowich had failed to demonstrate the first factor, even though neither party disputes

that, in 2015, 2016, and 2017, Tactile paid Seeling a higher annual base salary than

Sempowich. The court reasoned that, in 2016 and 2017, Sempowich earned more in sales

commissions than Seeling, such that when combining their salaries and commissions,

Sempowich earned more in total wages than Seeling in each of those two years.

         This dispute thus centers on the proper metric for determining wage discrimination

under the Equal Pay Act. Sempowich argues that the proper metric is the rate at which an

employer pays the plaintiff. Amicus the Equal Employment Opportunity Commission

(EEOC) agrees. In contrast, Tactile argues that the proper metric is the employee’s total

wages.

         The text of the Equal Pay Act unambiguously states that an employer may not

“discriminate . . . between employees on the basis of sex by paying wages to employees . . .

at a rate less than the rate at which he pays wages to employees of the opposite sex.”

(d)(1) (emphasis added). This critical portion of the statute says nothing

about total wages; it places all the emphasis on wage rates. As a result, we need not even




                                             18
decide whether we should defer to the EEOC’s interpretation of the statute — the statute

itself makes clear that wage rate is the proper metric. 5

       The district court incorrectly stated that total wages is the proper metric under the

regulations. This error apparently arose from a misreading of the EEOC’s definition of the

term “wages.” The regulations define “wages” as including “all forms of compensation

. . . whether called wages, salary, profit sharing, expense account, monthly minimum,

bonus . . . or some other name.” 

. As a result, the district court

reasoned that “wages” must include commissions, and thus that the proper metric is to

compare total wages. But this definition is beside the point. The term “wages” includes

commissions because, just as with salary, an employer could not pay commissions to a

female employee at a lower rate than a similarly situated male employee. This does not

mean that all types of remuneration should be combined into one lump sum when

comparing the earnings of a male and female employee.

       Rather, the statute and the EEOC’s regulations make clear that an employer violates

the Equal Pay Act if it pays female employees at a rate less than that of similarly situated

male employees. A hypothetical illustrates the point: “As a matter of common sense, total

remuneration cannot be the proper point of comparison. If it were, an employer who pays


       5
         We note, however, that the EEOC’s regulations reach the same conclusion. Under
those regulations, “an employer would be prohibited from paying higher hourly rates to all
employees of one sex and then attempting to equalize the differential by periodically
paying employees of the opposite sex a bonus.” 

. Although this
regulation refers to bonuses rather than commissions, the logic is the same: an employer
may not pay a female employee a lower salary than a similarly situated male employee and
then hope to avoid liability if the female employee works hard enough to earn extra money
through commissions or bonuses.
                                              19
a woman $10 per hour and a man $20 per hour would not violate the [Equal Pay Act] . . .

as long as the woman negated the obvious disparity by working twice as many hours.”

Ebbert v. Nassau County, No. 5 Civ. 5445, 

, at *3 (E.D.N.Y. Mar. 31,

2009).

                                         V.

         For the foregoing reasons, we vacate the judgment of the district court and remand

for further proceedings consistent with this opinion.

                                                            VACATED AND REMANDED




                                              20