15 F.4th 1033

10th Cir.

2021

JEFF PETERSON, personal representative of the Estate of Andrew Peterson, on behalf of himself and all similarly situated persons, Plaintiffs - Appellants, v. NELNET DIVERSIFIED SOLUTIONS, LLC, Defendant - Appellee.

Nos. 19-1348 & 20-1217

United States Court of Appeals for the Tenth Circuit

October 8, 2021

Before MATHESON, EBEL, and MORITZ, Circuit Judges.

PUBLISH. Appeals from the United States District Court for the District of Colorado (D.C. No. 1:17-CV-01064-NYW). Christopher M. Wolpert, Clerk of Court.

Adam W. Hansen of Apollo Law LLC, Minneapolis, Minnesota (Eleanor E. Frisch of Apollo Law, LLC, Minneapolis, Minnesota; Gregg I. Shavitz of Shavitz Law Group, P.A., Boca Raton, Florida; Michael Palitz of Shavitz Law Group, P.A., New York City, New York; and Brian Gonzales of Brian D. Gonzales Law Offices, Fort Collins, Colorado, with him on the briefs), for Plaintiffs – Appellants.

Julian R. Ellis and Matthew C. Arentsen (Richard B. Benenson, Martine T. Wells, and Sean S. Cuff with them on the briefs) of Brownstein Hyatt Farber Schreck, LLP, Denver, Colorado, for Defendant – Appellee.

MORITZ, Circuit Judge.

Over 300 call-center representatives (CCRs) who work or worked at call centers operated by Nelnet Diversified Solutions, LLC (Nelnet) allege that Nelnet fails to pay them for time devoted to booting up their work computers and launching certain software before they clock in. The district court concluded that these activities are integral and indispensable to the CCRs’ principal activities of servicing student loans by communicating and interacting with borrowers over the phone and by email and therefore constitute compensable work under the Fair Labor Standards Act (FLSA) of 1938, 29 U.S.C. §§ 201–19. But it nevertheless denied the CCRs’ claim, finding that the de minimis doctrine—which applies a multi-factor balancing test to determine whether an employer must compensate employees for small or insignificant periods of time outside scheduled working hours—applied to excuse Nelnet‘s obligation to pay the CCRs for this work. After granting summary judgment to Nelnet, the district court awarded costs to Nelnet as the prevailing party.

The CCRs appeal the district court‘s de minimis ruling (Appeal No. 19-1348), and separately appeal the district court‘s order awarding prevаiling-party costs to Nelnet (Appeal No. 20-1217). We agree with the district court that the CCRs’ preshift activities are compensable work under the FLSA. But our application of the three-factor de minimis doctrine leads us to a different result. We conclude that although the CCRs’ individual and total aggregate claims are relatively small, Nelnet fails to establish the practical administrative difficulty of estimating the time at issue, which occurs with exceeding regularity. Therefore, in Appeal No. 19-1348, we reverse the district court‘s order awarding summary judgment to Nelnet. And because we reverse on the merits, Nelnet is no longer the prevailing party. Accordingly, in Appeal No. 20-1217, we reverse the district court‘s order awarding costs to Nelnet and dismiss the CCRs’ costs appeal as moot.

Background

Nelnet is a student-loan company that operates several call centers—located in Lincoln, Nebraska; Omaha, Nebraska; and Aurora, Colorado—where its employеes “service student loans and interact with debtors over the phone and through email.”¹ App. vol. 2, 494.² This case involves Nelnet employees with the job titles of Flex Advisor, Collector, or Advisor I, all of whom are essentially CCRs. Nelnet pays the CCRs “once they clock into the timekeeping system at their individual workstations.” Id.

But the CCRs must perform several preshift tasks before they can clock in. Specifically, a CCR must first wake up his or her work computer. The CCR next inserts his or her security badge into the computer and enters his or her credentials. The computer then automatically launches the specialized software program Citrix, which in turn “loads the CCR‘s personal desktop[] and Nelnet‘s Intranet[,] which contains a link to the timekeeping system.” Id. at 495. “Once the Intranet has loaded, an employee has access to the timekeeping system and may, and nearly always does, clock into the system and begin receiving payment.” Id. The district court found that

the median amount of time that the CCRs devote tо these preshift activities, which varies between the three call-center locations, is approximately two minutes per shift.

Andrew Peterson³ filed this collective action—to which over 350 individuals opted in—seeking payment from Nelnet for the time he and other CCRs devoted to booting up their computers and launching software, arguing that such activities were compensable work for which Nelnet failed to pay them, in violation of the FLSA. The parties filed cross-motions for summary judgment on whether the preshift activities were compensable work and, if so, whether the time the CCRs devoted to these activities was de minimis such that Nelnet need not compensate them for it.

The district court (a magistrate judge presiding by consent of the parties) resolved the first issue in favor of the CCRs, finding that the preshift activities were compensable. But it concluded that the time at issue was de minimis and therefore granted summary judgment to Nelnet.⁴ The CCRs appealed, arguing in Appeal Nо. 19-1348 that the time at issue is both compensable and not de minimis.

Nelnet then sought to recover certain of its costs as the prevailing party under Federal Rule of Civil Procedure 54(d). The district court awarded Nelnet approximately $58,000, including over $33,000 in e-discovery costs related to exemplification and making copies. It entered the award jointly and severally against

Peterson and the opt-in plaintiffs and denied their subsequent motion for reconsideration. The CCRs appealed, arguing in Appeal No. 20-1217 that—assuming Nelnet was entitled to summary judgment and therefore prevailed on the merits—the district court erred in awarding the e-discovery costs and entering the award jointly and severally.⁵

Analysis

The CCRs argue that the preshift activities are compensable and not de minimis, and Nelnet argues to the contrary. In considering these issues, “[w]e review the district court‘s summary[-]judgment decision de novo, applying the same standards as the district court.” Punt v. Kelly Servs., 862 F.3d 1040, 1046 (10th Cir. 2017). Under such standards, summary judgment is appropriate “if the movant shows that there is no genuine dispute as to any material fаct and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a).

I. Compensability

Nelnet urges us to affirm the judgment in its favor on the alternative basis that—contrary to the district court‘s conclusion—the time the CCRs devote to booting up their computers and launching software is not compensable under the

FLSA.⁶ “The FLSA requires an employer to pay employees for their work, but it does not define what kinds of activities qualify as compensable work.” Aguilar v. Mgmt. & Training Corp., 948 F.3d 1270, 1276 (10th Cir. 2020). Faced with that absence, the Supreme Court initially defined the term “work” broadly, “as ‘physical or mental exertion (whether burdensome or not) controlled or required by the employer and pursued necessarily and primarily for the benefit of the employer and his business.‘” Integrity Staffing Sols., Inc. v. Busk, 574 U.S. 27, 31 (2014) (Busk) (quoting Tenn. Coal, Iron & R. Co. v. Muscoda Loc. No. 123, 321 U.S. 590, 598 (1944)). Similarly, the Court “defined ‘the statutory workweek’ to ‘includ[e] all time during which an employee is necessarily required to be on the employer‘s premises, on duty or at a prescribed workplace.‘” Id. (alteration in original) (quoting Anderson v. Mt. Clemens Pottery Co., 328 U.S. 680, 690–91 (1946)).

But Congress narrowed these broad definitions in the Portal-to-Portal Act of 1947, 29 U.S.C. §§ 251–62. Aguilar, 948 F.3d at 1276. Thе Portal-to-Portal Act excludes from the definition of compensable work any time devoted to “walking, riding, or traveling to and from the actual place of performance of the principal activity or activities which such employee is employed to perform“—in other words, commute time. 29 U.S.C. § 254(a)(1). And it further excludes preliminary and

postliminary activities from the definition of compensable work, providing “that the time devoted to ‘activities which are preliminary to or postliminary to [the employee‘s] principal activity or activities’ is not compensable.” Aguilar, 948 F.3d at 1276 (alteration in original) (quoting § 254(a)(2)).

These exclusions require identifying an employee‘s principal activities. See id. And an employee‘s principal activities include both the principal activities themselves and “all activities which are an ‘integral and indispensable part of the principal activities.‘” Busk, 574 U.S. at 33 (quoting IBP, Inc. v. Alvarez, 546 U.S. 21, 29–30 (2005) (Alvarez)). “[A]n activity is ‘integral and indispensable . . . if it is an intrinsic element of those [principal] activities and one with which the employee cannot dispense if [the employee] is to perform his [or her] principal activities.‘” Aguilar, 948 F.3d at 1276 (alterations and omission in original) (quoting Busk, 574 U.S. at 33). Unlike the earlier judicial definitions of the term “work,” “the integral-and-indispensable inquiry does not turn on whether the employer requires the activity or whether the activity benefits the employer.” Id. at 1276; cf. Tenn. Coal, 321 U.S. at 598 (defining “work or employment” as “physical or mental exertion (whether burdensome or not) controlled or required by the employer and pursued necessarily and primarily for the benefit of the employer“). “Instead, the question is ‘tied to the productive work that the employee is employed to perform.‘” Aguilar, 948 F.3d at 1277 (emphasis omitted) (quoting Busk, 574 U.S. at 36).

Here, the parties agree that the CCRs are employed to perform the principal activity of servicing student loans, which they do by “interact[ing] with debtors over the phone and through email.” App. vol. 2, 505. Thus, as should be obvious, the preshift activities of booting up a computer and launching software are not the work that Nelnet employs the CCRs to perform. See Busk, 574 U.S. at 35 (noting that employer “did not employ its workers to undergo security screenings, but to retrieve products from warehouse shelves and package those products for shipment“). At issue, though, is whether the CCRs’ preshift activities are integral and indispensable to their principal duties such that the preshift activities are compensable under the FLSA and not excluded under the preliminary-activities exception in the Portal-to-Portal Act.

Yet Nelnet insists that we need not enter the gray space of the integral-and-indispensable test because the preshift activities here are not compensable for a more fundamental reason: They “require little to no concentration and are not properly considered work.” Aplee. Br. 25 (emphasis omitted). In support, Nelnet relies on Reich v. IBP, Inc., 38 F.3d 1123 (10th Cir. 1994) (IBP). There, we considered whether employees “should be compensated for putting on a hard hat, safety glasses, earplugs, and safety shоes.” IBP, 38 F.3d at 1125. We noted in passing, without significant analysis, “that such equipment” is both “integral” and “indispensable to these particular workers.” Id. But we nevertheless found the time devoted to donning this equipment was not compensable because it did not satisfy the Tennessee Coal definition of work that predated the Portal-to-Portal Act. Id. at 1125–26 (citing Tenn. Coal, 321 U.S. at 598); see also Alvarez, 546 U.S. at 28 (explaining that although Portal-to-Portal Act rendered ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍commute time and preliminary and postliminary activities not compensable, it “d[id] not purport to change th[e] Court‘s earlier descriptions of the term[] ‘work‘“). Specifically, we concluded that donning standard protective gear was not work at all because it “takes all of a few seconds and requires little or no concentration.” IBP, 38 F.3d at 1126.

In line with IBP, Nelnet argues that the preshift activities here are “the precise type of mechanical, pre[]shift steps this [c]ourt has deemed not work under the FLSA.” Aplee. Br. 27. We find this argument unpersuasive. As an initial matter, whether an activity is work depends on the circumstances of each case, so we are not bound by the result in IBP. See Smith v. Aztec Well Servicing Co., 462 F.3d 1274, 1285 (10th Cir. 2006) (“[T]he courts are . . . left to determine on a case-by-case basis whether an employee‘s activities are compensable under the FLSA.“).

Next, in IBP, we applied the Tennessee Coal definition of “work” rather than the integral-and-indispensable test. See 38 F.3d at 1125–26. But intervening Supreme Court cases—namely, Busk and Alvarez—treat the integral-and-indispensable inquiry as the sole test governing compensability. See Busk, 574 U.S. at 31–37 (tracing history of broad judicial definition of “work” being narrowed by Portal-to-Portal Act to exclude travel time and preliminary and postliminary activities and treating integral-and-indispensable test as the only relevant inquiry for determining compensability of postliminary activity); Alvarez, 546 U.S. at 25–30, 33, 40, 42 (same). Nelnet points to no contemporary caselaw—and we have found none—purporting to apply the Tennessee Coal definition either in addition to or instead of the integral-and-indispensable inquiry. See Garcia v. Tyson Foods, Inc., 474 F. Supp. 2d 1240, 1246 (D. Kan. 2007) (speculating that if Tenth Circuit were to “revisit” IBP, it “would focus not on whether the donning and doffing constituted ‘work’ within the meaning of Tennessee Coal[] but on whether standard protective clothing and gear are ‘integral and indispensable’ to the work рerformed by production employees“). In light of intervening Supreme Court case law decided since IBP, we decline to apply the Tennessee Coal definition of “work” here.

Moreover, Nelnet‘s overall position on this point is that booting up computers and launching software is not compensable solely because it requires “little to no effort and concentration.” Aplee. Br. 27. But this argument ignores that “‘exertion’ [i]s not in fact necessary for an activity to constitute ‘work’ under the FLSA.” Alvarez, 546 U.S. at 25 (citing Armour & Co. v. Wantock, 323 U.S. 126 (1944)); see also Armour, 323 U.S. at 133 (pointing out that “an employer, if he [or she] chooses, may hire a[n] [individual] to do nothing, or to do nothing but wait for something to happen“). Thus, Nelnet‘s argument necessarily fails. See Aguilar, 948 F.3d at 1282–83 (rejecting argument that time correctional officers devoted to checking in and out keys and equipment was not compensable because such items were easy to carry).

Nelnet next argues that booting up a computer and launching software is not integral or indispensable to the CCRs’ principal activities of servicing student loans and cоmmunicating with borrowers. The district court rejected this argument, finding “that setting up the computer and loading the relevant programs to become call-ready is ‘an integral and indispensable part of the principal activities for which [the CCRs] are employed.‘” App. vol. 2, 507 (quoting Steiner v. Mitchell, 350 U.S. 247, 256 (1956)). In support, the district court noted that “pre[]shift preparation of tools or equipment is considered integral and indispensable to the principal activities when the use of such tools in a readied or activated state is an integral part of the performance of the employee‘s principal activities.” Id. at 506; see also, e.g., Steiner, 350 U.S. at 248–51, 256 (holding that preshift time changing into work clothing and postshift showering and changing out of work clothing was integral and indispensable to working in battery plant where workers were exposed to, among other things, lead dust); Mitchell v. King Packing Co., 350 U.S. 260, 261–63 (1956) (holding that sharpening knives was integral and indispensable to working in meatpacking plant because employees regularly use knives to рerform duties and dull knives “would slow down production . . . , affect the appearance of the meat as well as the quality of the hides, cause waste[,] and make for accidents“); Aguilar, 948 F.3d at 1281, 1283 (holding that checking in and out keys and equipment was integral and indispensable to prison-guard work because of “close connection” between keys and equipment and work of guarding prisoners); Kosakow v. New Rochelle Radiology Assocs., P.C., 274 F.3d 706, 715, 717–18 (2d Cir. 2001) (holding that reasonable factfinder could conclude setting up and testing X-ray processing machine was integral and indispensable because machine must be in its ready-to-use state for patients arriving at start of day).⁷

Accordingly, the district court explained that the CCRs cannot “dispense” with booting up their computers and launching software “if [they are] to perform [their] principal activities” of servicing student loans and communicating with borrowers. App. vol. 2, 506 (quoting Busk, 574 U.S. at 33). “Indeed, the very data that allows the CCRs to service student loans, [such as] borrower information and payment history, appears to reside within the computer system.” Id. at 507. And the CCRs “necessarily use computers to access electronically stored information” and to communicate with borrowers via email. Id. (quoting App. vol. 1, 207). Thus, because use of the computers was integral and indispensable to the work the CCRs were employed to perform, the district court concluded that the time devoted to preparing the computers for performance was likewise integral and indispensable. See Busk, 574 U.S. at 33 (explaining that “integral and indispensable” means activity is “an intrinsic element” of principal activities “and one with which the employee cannot dispense if [the employee] is to perform his [or her] principal activities“); id. at 36 (“The integral-and-indispensable test is tied to the productive work that the employee is employed to perform.“).

On appeal, Nelnet revives the arguments it made below, contending that these preshift activitiеs “are the modern equivalent of the historically noncompensable activities of ingress to the workstation and waiting in line to punch a time clock.” Aplee. Br. 29. In support, Nelnet relies on § 254(a)(1), which excludes commute time from compensable work. See Alvarez, 546 U.S. at 40 (holding that time waiting to don work gear “comfortably qualif[ies] as a ‘preliminary’ activity“). Nelnet also raises general policy concerns about the application of the FLSA in our modern and digital age, including during the COVID-19 pandemic, when telework is increasingly common.

But we need not speculate about the FLSA‘s application to teleworkers or the pandemic‘s broad implications for our digital age. We need only decide the case before us, which doesn‘t concern teleworking. Indeed, as the CCRs point out, Nelnet‘s comparison of the boot-up process here to employees walking to a workstation or to factory workers standing in line at a time clock is overly categorical and would exempt a large swаth of activities as noncompensable. Yet the integral-and-indispensable inquiry applies on a case-by-case basis, and the result of that inquiry depends entirely on the work an individual is employed to perform. See Busk, 574 U.S. at 36. Moreover, the CCRs do not argue that booting up a computer and launching software is always compensable—they argue only that it is compensable in this case because it is integrally and indispensably connected to the CCRs’ principal activities of servicing student loans by interacting with customers over the phone and by email.

In any event, applying its categorical argument to the facts of this case, Nelnet analogizes “booting up and logging in to a computer to access Nelnet‘s electronic timekeeping software” to “digital steps to reach [the CCRs‘] digital workstation.” Aplee. Br. 33 (emphasis omitted). It similarly argues that the boot-up process is “the modern equivalent of the factory worker standing in line аt the physical time clock.” Id. at 35. But these analogies go too far. As the CCRs point out, they still travel to and from work in the traditional way; such travel time, of course, is not compensable. And that commute time does not somehow morph into the digital realm and continue until the moment that a CCR successfully clocks in at his or her computer. Likewise, turning on a computer, entering passwords, and launching software is not analogous to waiting in line to punch a clock, particularly when—very much unlike a time clock—the computer itself is an integral tool for the work the individual is employed to perform.

Critically, nothing in Nelnet‘s analogies and arguments adequately refutes the obvious connection between the computers and software programs and the work the CCRs are employed to perform. As the district court put it, “the CCR[s] make[] regular use of the prepared electronic tools in performing their substantive tasks. Therefore, the necessary preliminary work is intertwined with the substantive performance of the principal tasks which renders such preliminary work integral and indispensable.” App. vol. 2, 508. Indeed, Nelnet effectively concedes indispensability, noting “the necessity of booting up and logging in to access job-related programs.” Aplee. Br. 36 (emphasis added). As for the integral prong, there is a clear connection between the computers and software programs and the work the CCRs are employed to perform—the CCRs make consistent use of the computer and its programs to perform their work. So preparing those tools is integral to the CCRs’ work. See Aguilar, 948 F.3d at 1281 (“[T]he close connection between (1) the keys and equipment and (2) the nature of the officers’ work convinces us that checking out and returning these items is ‘an intrinsic element’ of providing security in the prison.” (quoting Busk, 574 U.S. at 37)); cf. Rutti v. Lojack Corp., 596 F.3d 1046, 1049, 1057 (9th Cir. 2010) (noting that time devoted to receiving assignments on handheld computer device was not integral to work of employee who installed and repaired vеhicle recovery systems).

In sum, we reach the same conclusion that the district court did: The preshift activities of booting up a computer and launching software are integral and indispensable to the CCRs’ principal duties of servicing student loans by communicating with borrowers over the phone and by email. Busk, 574 U.S. at 36; see also Jackson v. ThinkDirect Mktg. Grp., Inc., No. 16-cv-03449, 2019 WL 8277236, at *4 (N.D. Ga. Dec. 9, 2019) (explaining that “requirement of logging in and out of electronics systems needed to process calls is at least integral to the work of answering phone calls” (quoting Gaffers v. Kelly Servs., Inc., No. 16-cv-10128 (E.D. Mich. June 2, 2016)). Booting up a computer and launching software is “an intrinsic element of” servicing student loans and communicating with borrowers because the data and tools necessary to those principal duties exist on the computer. Busk, 574 U.S. at 35. Likewise, Nelnet could not have eliminated these activities “without impairing the employees’ ability to complete their work.” Id. Such integral and indispensable activities ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍are compensable under the FLSA.⁸

II. De Minimis Doctrine

The CCRs argue that the district court erred in concluding Nelnet need not compensate them for their preshift activities because the time at issue is de minimis.⁹ The de minimis doctrine provides that “insubstantial or insignificant periods of time beyond the scheduled working hours, which cannot as a practical administrative matter be precisely recorded for payroll purposes, may be disregarded.” § 785.47. At the same time, however, “[a]n employer may not arbitrarily fail to count as hours worked any part, however small, of the employee‘s fixed or regular working time or

practically ascertainable period of time [the employee] is regularly required to spend on duties assigned to [the employee].” Id.

Courts balance three factors when “determin[ing] whether work time is de minimis and therefore not compensable: ‘(1) the practical administrative difficulty of recording the additional time; (2) the size of the claim in the aggregate; and (3) whether the [employee] performed the work on a regular basis.‘” Aguilar, 948 F.3d at 1284 (quoting Castaneda v. JBS USA, LLC, 819 F.3d 1237, 1243 (10th Cir. 2016)). Before undertaking this analysis, courts “must first estimate the amount of time at issue.” Id. “There is no precise amount of time that may be denied compensation as de minimis.” Id. (quoting Reich v. Monfort, Inc., 144 F.3d 1329, 1333 (10th Cir. 1998) (Monfort)). It is the employer‘s burden to show that the de minimis doctrine applies. Kellar v. Summit Seating Inc., 664 F.3d 169, 176 (7th Cir. 2011).

A. Amount of Time

The district court accepted the estimates provided by Nelnet‘s expert, who calculated that the median time from the badge swipe to the initiation of Citrix was “0.5 minutes in Omaha, 0.9 minutes in Lincoln, and 1.02 minutes in Aurora.” Id. at 495. The expert further calculated that “[t]he median [ten]th percentile” time from the initiation of Citrix to clock-in was “1.1 minutes at Omaha, 1.3 minutes in Lincoln, and 1.25 minutes in Aurora.” Id. Adding these times together results in 1.6 minutes in Omaha, 2.2 minutes in Lincoln, and 2.27 minutes in Aurora.

On appeal, the CCRs argue that the amount of time from badge swipe to clock-in is actually four to six minutes. In so doing, they challenge the methodology and statistical choices of Nelnet‘s expert. But we agree with Nelnet that the CCRs waived any challеnge to Nelnet‘s expert‘s estimates or his underlying methodology and statistical choices. Nelnet detailed its expert‘s estimates in its summary-judgment motion. In response, the CCRs neither obtained their own expert nor disputed any of Nelnet‘s expert‘s estimates or methodologies. Instead, the CCRs also relied on these estimates from Nelnet‘s expert. The CCRs insist that they only “accept[ed] the findings of the expert with respect to what the records show.” Supp. App. 37 (emphasis added). But they did more than that: They expressly stated in their summary-judgment motion that “[f]or purposes of this [m]otion, [they] accept[] the calculation” of the amount of time at issue. App. vol. 1, 112. Accordingly, in ruling on summary judgment, the district court expressly identified the estimates from Nelnet‘s expert as undisputed facts. Likewise, the final pretrial order included the expert‘s estimates as facts to which “[t]he parties stipulate[d] and agree[d].” Supp. App. 186. Thus, at no point prior to this appeal did the CCRs raise any objection to the estimаtes provided by Nelnet‘s expert.

Because the proper place for the CCRs’ challenge to the expert‘s estimates and methodology was at the district court, we decline to consider their challenge to Nelnet‘s expert, which they raise for the first time on appeal and without any plain-error argument.¹⁰ See Richison, 634 F.3d at 1130–31. Accordingly, for the purpose of estimating the amount of time at issue before conducting the de minimis analysis, we accept the undisputed overall estimate from Nelnet‘s expert, which Nelnet characterizes as “2.27 minutes (or less).” Aplee. Br. 42; see also Aguilar, 948 F.3d at 1284.

B. Practical Administrative Difficulty

The first factor in the de minimis analysis is “the practical administrative difficulty of recording the additional time.” Aguilar, 948 F.3d at 1284–85 (quoting Monfort, 144 F.3d at 1333). Stated differently, this factor considers whether the time “cannot as a practical administrative matter be precisely recorded for payroll purposes.” § 785.47. The district court found that this factor “weigh[ed] heavily” in Nelnet‘s favor. App. vol. 2, 515. In so doing, it accepted Nelnet‘s position—supported by testimony from two Nelnet employees—that linking the badge swipe to the timekeeping system would require “either procuring a custom-ordered software to link the two or undergoing the laborious cross-checking.” Id. at 514. The district court further found that the CCRs “fail[ed] to set forth admissible evidence that”

alternatives, such as using a punch clock, conducting cross-checking, creating custom software, or estimating the time at issue, “would not be burdensome.” Id. at 515.

On appeal, Nelnet primarily argues that this factor weighs in its favor because “it would be administratively difficult to translate the time stamps from scanning the . . . badge or initiating Citrix to accurate, usable time records.” Aplee. Br. 44. But even if we assume that to be true, it does not adequately address why Nelnet could not simply estimate the amount of time at issue. When the amount of time at issue can be reasonably estimated, the practical administrative burden tends to be low and weighs against a de minimis finding. Fоr instance, in Aguilar, we weighed this factor in favor of the employees in part because “it [wa]s possible to estimate the average time” that the employees devoted to passing through the security screening, which they did every day. 948 F.3d at 1285; see also Kellar, 664 F.3d at 176–77 (noting that because plaintiff “testified that she typically performed the same kinds of activities every day, it would have been possible to compute how much time [she] spent on compensable activities“).

Nelnet only weakly suggests that estimating the amount of time at issue would be administratively burdensome. It primarily points to and defends the district court‘s conclusion that “to get the undisputed times at issue in this case, Nelnet‘s expert had to do precisely the same laborious cross-checking task the Ninth Circuit rejected in Corbin [v. Time Warner Entertainment-Advance/Newhouse Partnership, 821 F.3d 1069 (9th Cir. 2016)].” App. vol. 2, 514. Yet Corbin is strikingly dissimilar, and Nelnet‘s reliance on it is both misplaced and deeply flawed. At issue there was one

instance of one minute of uncompensated work; in that context, the Ninth Circuit found a high administrative burden of recording the time based on the difficulty of cross-checking various timestamps to both discover and add up the rare occasions on which an employee engaged in a single minute of uncompensated work. Corbin, 821 F.3d at 1081–82. We decline to import Corbin‘s brief analysis of this factor wholesale into this case, which—entirely unlike the time at issue in Corbin—involves the same set of tasks performed by every CCR at the start of every shift.

Instead, these facts are far more like those presented to the Fourth Circuit in Perez, 650 F.3d 350. There, the court concluded that “the time expended in [preshift and postshift donning and doffing] is not so miniscule that it would be difficult to measure,” as demonstrated by the fact that “both experts . . . were able to measure the [mean] amount of time required by employees to don and doff protective gear before and after their work shifts.” Id. at 372, 374. Here, similarly, the amount of time devoted to booting up and logging on “is not so miniscule that it would be difficult to measure,” as demonstrated by the fact that Nelnet‘s expert in this case was able to do so. Id. at 374.

Perhaps recognizing that Perez‘s holding supports the CCRs’ argument here, Nelnet ignores the persuasive language cited above. Instead, Nelnet points to Perez‘s concurring opinion to support its generalized argument that estimating the amount of time at issue here imposes a practical administrative burden. Yet that concurring opinion agreed with the majority‘s conclusion that the de minimis doctrine did not apply to the preshift and postshift donning and doffing. See id. at 378. At the same time, the concurring opinion would have applied the de minimis doctrine to an additional set of tasks—lunchtime donning and doffing—that the majority had concluded were not compensable under circuit precedent. See id. at 370; id. at 379 (Wilkinson, J., concurring). In the process of reaching that result, the concurring opinion noted the parties’ “dueling experts” and their “staggeringly different conclusions” based on “computations out to no fewer than three decimal places.” Id. at 378–79. Nelnet highlights this discussion, implying that even estimating the amount of time at issue would impose a practical administrative burden.¹¹

But aside from conclusory statements about the difficulty of its expert‘s work in this case and its general citation to the Perez concurrence, Nelnet has not established any particular practical administrative barrier to its ability to estimate the amount of time at issue in this case. Indeed, Nelnet‘s conclusory allegations stand in stark contrast to facts that support finding a significant practical administrative burden in other cases. For instance, in Singh v. City of New York, employee fire-alarm inspectors brought an FLSA claim seeking compensation for the additional commute time engendered by the requirement that they carry inspection documents, which sometimes caused them to miss a train or bus. 524 F.3d 361, 365–66 (2d Cir. 2008). The Second Circuit applied the de minimis doctrine to this time, finding in part that “as a practical administrative matter, ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍it would be difficult for the [c]ity to record and monitor the additional commuting time” and that “[t]he task of creating a reliable system to distinguish between ordinary and additional commuting time for each individual inspector on a daily basis would be challenging, if not impossible.” Id. at 371. Here, by contrast, Nelnet already has available data from which the time at issue can be estimated. And the entire amount of time is compensable, so Nelnet need not draw any distinctions to develop an estimate from its existing data.

Similar to Singh, the Ninth Circuit in Lindow v. United States applied the de minimis doctrine to employees’ time spent reading a log book and exchanging information in part because the employer “would have had difficulty monitoring th[e] pre[]shift activity.” 738 F.2d 1057, 1064 (9th Cir. 1984). The court there emphasized the “wide variance” in the amount of preshift time employees spent on compensable activities as opposed to social activities and further noted that not all employees even performed these duties before their shifts. Id. at 1063. Likewise, in Monfort, we noted that employees devoted varying amounts of time to taking on and off various types of safety gear and concluded that “it would be administratively difficult to record the actual time each worker engaged in these activities.” 144 F.3d at 1334.

In contrast, Nelnet already monitors the actual time that the CCRs devote to booting up their computers and launching software. And importantly, employees have no ability to impact the amount of time it takes to boot up their computers and software—that time is generally similar for all employees at a particular location, who are all performing the exact same set of tasks. Thus, unlike the cases discussed above, the preshift time does not vary greatly from employee to employee.

As such, Nelnet‘s conclusory allegations about the burden of the expert‘s work for this particular case do not significantly undercut the reality that it is entirely possible—аnd not significantly burdensome—for Nelnet to estimate the amount of time that the CCRs devote to these compensable preshift activities based on precise data that it already maintains and controls. And critically, Nelnet provides no other evidence to support its assertion that estimating the amount of time at issue would be administratively burdensome. See Kellar, 664 F.3d at 176 (noting that employer “bears the burden to show that the de minimis doctrine applies” (italics omitted)).

Thus, because Nelnet failed to carry its burden of showing why it would be administratively burdensome to estimate the amount of time at issue, this factor weighs in the CCRs’ favor and against a finding that the time at issue is de minimis. See Aguilar, 948 F.3d at 1285 (weighing this factor in employees’ favor because employer “already records the majority of the time at issue and could reasonably estimate the time that it does not record“); Perez, 650 F.3d at 374 (weighing this factor in employees’ favor because experts were able to estimate time, so “thе time expended in these activities is not so miniscule that it would be difficult to measure as a practical matter“).

C. Aggregation

The second factor in the de minimis analysis is “the aggregate amount of compensable time.” Aguilar, 948 F.3d at 1285 (quoting Monfort, 144 F.3d at 1334). “[B]oth the aggregate claim for each individual [employee] as well as the aggregate claim for all the [employees] combined” are relevant to the inquiry. Id.; see also Monfort, 144 F.3d at 1334 (looking at both “the aggregate for an individual” and “an aggregate based on the total number of workers“); Lindow, 738 F.2d at 1063 (noting that “[c]ourts have granted relief for claims that might have been minimal on a daily basis but, when aggregated, amounted to a substantial claim” and that “courts in other contexts have applied the de minimis rule in relation to the total sum or claim involved in the litigation” (italics omitted)). The district court weighed this factor “strongly” in Nelnet‘s favor. App. vol. 2, 517.

On appeal, Nelnet unhelpfully conflates this factor with the threshold consideration of the amount of time at issue and doеs little more than assert that the CCRs’ “aggregate claims[] easily fall[] within the permissible range of de minimis time.” Aplee. Br. 48. It cites several cases in support, but none offer any guidance on how to weigh the aggregate claims at issue here. See Lindow, 738 F.2d at 1064 (noting that “aggregate claim may be substantial” but offering no further explanation); Singh, 524 F.3d at 371 (noting that “the aggregate claims are quite small” because activities occurred only “on occasional days” but offering no monetary estimate); Lyons v. Conagra Foods Packaged Foods LLC, 899 F.3d 567, 584 (8th Cir. 2018) (approving de minimis finding without discussing size of aggregate claims); Chambers v. Sears Roebuck & Co., 428 F. App‘x 400, 418 (5th Cir. 2011) (unpublished) (noting in dicta that one “minute or so” would be de minimis without discussing size of aggregate claims).

We begin our review with the district court‘s assessment of the aggregate claim across all the CCRs combined. On this point, the district court noted “the [p]arties agree that the lost wages total approximately $30,000“—or, as the expert more specifically calculated, $31,585—and then found “that in absolute terms[,] the aggregate amount of the claim strongly supports a de minimis finding.”¹² App. vol. 2, 517 (italics omitted). To reach this conclusion, the district court relied on Aguilar v. Management & Training Corp., No. CV 16-00050, 2017 WL 4804361 (D.N.M. Oct. 24, 2017), rev‘d, 948 F.3d 1270. There, the district court weighed the aggregation factor in the employer‘s favor because the aggregate claim—an indeterminate amount less than $355,478—was substantially less than the $1.6 million claim at issue in a prior de minimis case. See id. at *18 (citing Monfort, 144 F.3d at 1333–34). But we have since reversed that conclusion, specifically rejecting such cursory treatment of the aggregate factor and finding “err[or] in treating Monfort as if it set a baseline below which all claims are negligible.” Aguilar, 948 F.3d at 1285–86. We accordingly question the district court‘s conclusion here that the overall $31,585 was not significant; as we stated in Aguilar, “more moderately sized claims are not automatically negligible.” Id. At the same time, we acknowledge that an overall claim of $31,585 is less significant than the amount at issue in Aguilar, which was “at least $355,478,” and possibly higher. Id. at 1285 & n.10; see also Perez, 650 F.3d at 374 (noting that in case involving 280 employees, individual claims for $425 per year or $2,550 over six years were “significant“).

But our analysis of this factor does not turn solely on the aggregate size of the claim for all the employees who opted into this collective action. See Aguilar, 948 F.3d at 1285. We must also consider the aggregate claim for a single employee. See id. We do so both because precedent dictates as much and because not all FLSA actions are collective in nature; a single employee can bring an FLSA claim for unpaid compensation, so the size of a single claim must also have consequence, even in a collective action. See id.

To calculate the size of the claim per employee, the district court first divided the total aggregate claim by the number of opt-in plaintiffs to arrive at “$84 per plaintiff over the collective period, from July 15, 2014 to April 25, 2018.” App. vol. 2, 517–18. Yet this estimate (which, further divided by the almost-four-year collective-action period, comes to $22 per year) does not accurately capture the size of an individual employee‘s claim. As the district court itself recognized, “not every plaintiff worked full time during the entirety of the collective period.” Id. at 518. So merely dividing the total claim by the number of employees necessarily undervalues the individual claims at issue in this litigation.

The district court also appeared to discount the size of the individual claims because they amounted to “cents, rather than dollars, per day.” Id. As a factual matter, this appears to be correct: In its response brief, Nelnet relies on its expert‘s report to summarize that the amount of money at issue for an individual CCR is “$0.48 per shift.”¹³ Aplee. Br. 48 n.15. Yet the district court did not aggregate the impact of $0.48 per shift across the relevant time periods to create an estimate of the size of an individual claim for an employee who worked at Nelnet for a year or for the entire collective-action period.

And the aggregate imрact of $0.48 per shift—or, assuming five shifts per week, $2.40 per week—is not necessarily negligible. See Lindow, 738 F.2d at 1063 (“Courts have granted relief for claims that might have been minimal on a daily basis but, when aggregated, amounted to a substantial claim.“). Indeed, for a CCR who worked for Nelnet for one full year of the collective-action period, the aggregate claim would be approximately $125. And rounding the collective-action period up to four years (the actual period is about two and a half months less than that), a CCR who worked for Nelnet for the entire period would have an aggregate claim of approximately $500.

We are unprepared to dismiss individual claims of approximately $125 per year—particularly ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍for “low[-]wage workers” who earn $13.50 an hour, Aplt. Br. 54—as insignificant. See Perez, 650 F.3d at 374 (considering employee‘s hourly pay rate when concluding that aggregate claim was “significant“). That said, neither the total aggregate claim of $31,585 nor a prоperly estimated individual claim of approximately $125 per year is extremely large, particularly when compared to other de minimis cases. Cf. Aguilar, 948 F.3d at 1285–86 (weighing aggregate claim of over $355,478 in employees’ favor); Perez, 650 F.3d at 374 (finding individual claims of $425 per year to be “significant“). Accordingly, this factor does not weigh in favor of either party.

D. Regularity of Time

The district court then weighed the third factor, “the regularity with which the [employees] perform this work,” in the CCRs’ favor, finding that the preshift activities here occur regularly. Aguilar, 948 F.3d at 1286. Neither party quarrels with this conclusion. Indeed, Nelnet almost entirely ignores this factor, including separate sections in its brief for the other two factors, but not for this one. And yet this is a critical factor; as the CCRs assert, “[t]his point cannot be overstated.” Aplt. Br. 55. Courts often apply the de minimis doctrine when employees do not regularly engage in the activities at issue. See, e.g., Corbin, 821 F.3d at 1073, 1082 (applying de minimis doctrine to one instance of one minute of uncompensated work in part because “the uncompensated time at issue here is not ‘regular’ at all“); Lindow, 738 F.2d at 1063–64 (collecting cases on regularity factor and applying de minimis doctrine in part because employees “did not always perform these duties before their shifts“); Singh, 524 F.3d at 371 (applying de minimis doctrine in part because employees engaged in activities at issue “on occasional days“).

By contrast, in Aguilar, we weighed this factor against applying the de minimis doctrine when “most officers perform most of these activities during most shifts.” Aguilar, 948 F.3d at 1286; see also Perez, 650 F.3d at 374 (finding time not de minimis in part because it was “undisputed that these activities of donning and doffing at the beginning and the end of the employees’ work shifts occur regularly each workday“). Here, the compensable work at issue is even more regular than the work at issue in Aguilar: Every CCR performs the same preshift activities before every shift for approximately the same amount of time (depending upon the location), without fail. Thus, we conclude this factor weighs strongly in the CCRs’ favor and against finding the time at issue de minimis.

E. Balancing

After weighing these three factors, the district court found the time de minimis based on the small amount of time at issue and “the serious administrative burden” of recording it. App. vol. 2, 518. In so doing, the district court neither referred back to the regularity of the work nor mentioned the possibility of estimating the amount of time at issue. Similarly, Nelnet now defends that conclusion, almost entirely ignoring the strong regularity with which the CCRs perform these activities and omitting from its balancing discussion the possibility of avoiding practical administrative burden by simply estimating the amount of time at issue.

Based on our above review of the de minimis factors, we reach a different conclusion. Perhaps most critically, we do not find that Nelnet has established any “serious administrative burden” in estimating the amount of time at issue based on precise data that it already maintains and controls. Id. Additionally, the steady regularity with which the CCRs perform these activities weighs heavily against applying the de minimis doctrine in this case. And the relatively low amount of the individual and total aggregate claims doesn‘t favor either party under the circumstances of this case.

Balancing these three factors, we ultimately conclude that the relatively small size of the claims is not enough to outweigh the regularity of the work and the absence of any significant practical administrative burden in estimating the amount of time involved. We therefore hold that Nelnet has not shown that the time at issue is de minimis. Accordingly, we reverse the district court‘s order granting summary judgment to Nelnet and remand for further proceedings, beginning with the entry of partial summary judgment for the CCRs on the issues of compensability and the inapplicability of the de minimis doctrine.

Conclusion

We conclude that when booting up their computers and launching software, the CCRs are preparing tools that are both necessary to and an intrinsic part of their principal activities. Therefore, these preshift tasks are compensable principal activities under the FLSA. Moreovеr, because Nelnet‘s ability to estimate the amount of time at issue and the consistent regularity with which the CCRs perform these activities weigh more heavily than the relatively small size of the claims in this case, the de minimis doctrine does not apply to excuse Nelnet‘s obligation to pay its employees for their work. As such, in Appeal No. 19-1348, we reverse the district court‘s order awarding summary judgment to Nelnet and remand for further proceedings consistent with this opinion.

And because Nelnet is no longer the prevailing party, we also reverse the district court‘s order awarding costs to Nelnet under Rule 54(d). See Ctr. for Legal Advoc. v. Earnest, 320 F.3d 1107, 1112 (10th Cir. 2003) (“Since we now reverse the grant of summary judgment, [a]ppellees are no longer considered the prevailing party . . . and that [costs] order must be reversed.“). Thus, the CCRs’ arguments in Appeal No. 20-1217—that the district court erred in awarding costs related to e-discovery and in entering the award jointly and severally against all the CCRs—are mоot, as is their requested relief (remand for entry of a correct cost judgment). We dismiss that appeal accordingly.

Notes

We take these facts—which are undisputed except where noted in our analysis below—from the district court‘s summary-judgment order.

Our citations to the parties’ briefs and appendices refer to the briefing and appendices filed in Appeal No. 19-1348.

Andrew Peterson has since died, and we allowed his estate, through personal representative Jeff Peterson, to substitute for him.

The district court further declined to exercise supplemental jurisdiction over “Peterson‘s individual state[-]law claim under the Colorado Wage Claim Act.” App. vol. 2, 519. That ruling and claim are not before us in either appeal.

The district court‘s costs order noted a potential clerical error in its underlying summary-judgment order, which referred only to Peterson when it should have also referred to all the opt-in plaintiffs. After oral argument, we directed a limitеd remand to the district court for the purpose of correcting this error. See Fed. R. Civ. P. 60(a). The district court then filed its corrected summary-judgment order with this court.

An “appellee can generally seek affirmance on any ground found in the record“; a cross-appeal is only required “if [appellee] seeks to enlarge its rights and gain ‘more than it obtained by the lower-court judgment.‘” (quoting ).

The district court also cited (unpublished), which held that checking in and out specialized tools was a compensable activity. The CCRs likewise cite this case in their opening brief. As we did in Aguilar, we decline to rely on Von Friewalde because it was decided before Busk and relied on one of the rationales that Busk expressly rejected. .

This conclusion accords with the Department of Labor‘s 2008 fact sheet that the CCRs submitted to the district court. There, the Department stated that “[a]n example of the first principal activity of the day for agents/specialists/representatives working in call centers includes starting the computer to download work instructions, computer applications, and work-related emails.” App. vol. 1, 187. The district court declined to defer to this fact sheet. To the limited extent the CCRs challenge that lack of deference, their single-sentence briefing is inadequate. See (declining to consider inadequately briefed argument). In light of this inadequate challenge, we decline to consider what kind of deference, if any, is due to this fact sheet. We simply note that it supports and is consistent with our independent analysis.

The CCRs argue in their opening brief that “[t]he de minimis doctrine contravenes the plain text of the FLSA and should be repudiated.” Aplt. Br. 45 (emphasis omitted). In response to Nelnet‘s assertion that the CCRs forfeited this argument by failing to make it below and waived it on appeal by failing to argue for plain error, the CCRs back away from the argument, insisting in their reply brief that they are not “mounting a facial attack on the de minimis doctrine.” Rep. Br. 21; see also (setting forth forfeiture and waiver rules). Given this context, we decline to consider the merits of the argument here. But we pause to notе that the de minimis doctrine doesn‘t appear to have any textual basis in the FLSA and instead originated with a passing statement in Anderson regarding “the application of a de minimis rule . . . [w]hen the matter in issue concerns only a few seconds or minutes of work beyond the scheduled working hours.” ; see also (noting that application of de minimis doctrine to FLSA claims “began with Anderson“); (Wilkinson, J., concurring) (noting “no obvious statutory derivation” for de minimis doctrine); 29 C.F.R. § 785.47 (codifying Anderson in regulation).

Indeed, perhaps recognizing that they missed their opportunity to raise this dispute, the CCRs place the majority of their challenges to the expert‘s methodology and conclusions in the fact section—rather than the argument section—of their opening brief.

Of course, one materially distinguishing feature worth noting is that unlike Perez, this case does not involve dueling experts; the only expert here is Nelnet‘s.

In so finding, the district court excluded employees who did not opt into this action. The CCRs do not challenge that decision; their briefing obliquely refers to 3,499 affected employees, but they do not argue that the district court erred in considering only the opt-in plaintiffs.

The CCRs do not meaningfully dispute ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍this per-shift estimate.