Opinion
No. 16 CV 6221
03-04-2020
MICHAEL MEADOWS, et al., Plaintiffs, v. NCR CORPORATION, Defendant.
MEMORANDUM OPINION AND ORDER
Plaintiff Michael Meadows alleges that defendant NCR Corporation violated the Fair Labor standards Act by failing to pay him and his fellow customer engineers the required overtime rate. I conditionally certified a collective action. More than 1,600 people opted in to the case. Both sides conducted extensive discovery before NCR filed its motion to decertify, and that discovery reveals significant factual and legal differences among the members of the proposed collective action. In light of those differences, and for the reasons discussed further below, NCR's motion is granted. The collective action is decertified.
I. Legal Standards
Collective actions allow for the efficient resolution "of common issues of law and fact arising from the same alleged discriminatory activity." Hoffmann-La Roche Inc. v. Sperling, 493 U.S. 165, 170 (1989). District courts have wide discretion in managing collective actions, Alvarez v. City of Chicago, 605 F.3d 445, 449 (7th Cir. 2010); Weil v. Metal Techs., Inc., 925 F.3d 352, 357 (7th Cir. 2019), and courts in this district have developed a two-step conditional certification process to aid in the management of collective actions brought under the Fair Labor Standards Act. Jirak v. Abbott Labs., Inc., 566 F.Supp.2d 845, 847 (N.D. Ill. 2008) (collecting cases); 29 U.S.C. § 216(b).
During the first step, if the plaintiff makes a modest factual showing that other employees are similarly situated, the action is conditionally certified and notice is issued. DeMarco v. Nw. Mem'l Healthcare, No. 10 C 397, 2011 WL 3510905, at *1 (N.D. Ill. Aug. 10, 2011); Jirak, 566 F.Supp.2d at 847-48; Ervin v. OS Rest. Servs, Inc., 632 F.3d 971, 974 (7th Cir. 2011). The first step "requires nothing more than substantial allegations that the putative class members were together the victims of a single decision, policy, or plan." Jones v. Furniture Bargains, LLC, No. 09 C 1070, 2009 WL 3260004, at *2 (N.D. Ill. Oct. 9, 2009) (quoting Thiessen v. Gen. Elec. Capital Corp., 267 F.3d 1095, 1102 (10th Cir. 2001)). Meadows made it past this first step. [192].
The second step requires more. With the benefit of a "much thicker record than it had at the notice stage," Morgan v. Family Dollar Stores, Inc., 551 F.3d 1233, 1261 (11th Cir. 2008), courts in this district conduct a stringent, fact-specific review. Hudgins v. Total Quality Logistics, LLC, No. 16 C 7331, 2019 WL 354958, at *3 (N.D. Ill. Jan. 29, 2019). Plaintiffs bear the burden of producing sufficient evidence of "an identifiable factual nexus that binds the plaintiffs together as victims of a particular violation of the overtime laws." See id.; Russell v. Ill. Bell Tel. Co., Inc., 721 F.Supp.2d 804, 812 (N.D. Ill. 2010); Strait v. Belcan Eng'g Grp., Inc., 911 F.Supp.2d 709, 718 (N.D. Ill. 2012) ("[p]laintiffs bear the burden of demonstrating that they are 'similarly situated'"). During this step, courts ask "(1) whether the plaintiffs share similar or disparate factual and employment settings; (2) whether the various affirmative defenses available to the defendant would have to be individually applied to each plaintiff; and (3) fairness and procedural concerns." Camilotes v. Resurrection Health Care Corp., 286 F.R.D. 339, 345 (N.D. Ill. 2012). "A unified policy, plan, or scheme ... is not necessarily required ... especially if a collective action would promote judicial economy because there is otherwise an identifiable factual or legal nexus." Molina v. First Line Sols. LLC, 566 F.Supp.2d 770, 787 (N.D. Ill. 2007). See also Campbell v. City of Los Angeles, 903 F.3d 1090 (9th Cir. 2018) (at the decertification stage, plaintiffs must provide "substantial evidence that their claims arise out of a single policy, custom or practice that leads to FLSA violations").
The FLSA is interpreted "broad[ly] and comprehensive[ly] in order to accomplish the remedial purposes of the Act." Solis v. Intern. Detective & Protective Service, Ltd. 819 F.Supp.2d 740, 747 (N.D. Ill. 2011); Tony & Susan Alamo Found. v. Sec'y of Labor, 471 U.S. 290, 296 (1985). It must not be applied in a "narrow, grudging manner." Walling v. National Ice & Fuel Corp., 158 F.2d 28, 29 (7th Cir. 1946). The standards for class certification under Rule 23 have largely merged with those for certifying a collective action, Espenscheid v. DirectSat USA, LLC, 705 F.3d 770, 772 (7th Cir. 2013), and under Rule 23, the commonality inquiry requires that the class have a "common contention" that is "capable of classwide resolution—which means that determination of its truth or falsity will resolve an issue that is central to the validity of each one of the claims in one stroke." Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 350 (2011). What matters "is not the raising of common 'questions'—even in droves—but rather, the capacity of a class-wide proceeding to generate common answers apt to drive the resolution of the litigation." Id. "[W]hen multiple managers exercise independent discretion, conditions at different stores (or sites) do not present a common question." Bolden v. Walsh Const. Co., 688 F.3d 893, 896 (7th Cir. 2012).
II. Facts
Meadows and his fellow customer engineers are paid by the hour to repair ATM machines and point-of-sale systems. [288-1] at 91:24-92:5; [288-8] at 16:17-17:2. (Point-of-sale systems include "registers, ... pin pads, ... printers ... [b]asically anything a retail store would use." [288-6] at 40:9-11.) Customer engineers are "home dispatched," meaning that every morning they leave their homes and go directly to the location of the first ATM or point-of-sale system they have been assigned to repair. See [288-1] at 24:3-6; [288-4] at 28:2-4. Before they can leave, they have to know where to go, and in order to learn where to go, they have to check in (on their phones or laptops) with NCR's central dispatchers. See, e.g., [281-13] at 30:12-17, 31:12-22; [281-11] at 10:18-11:15.
Bracketed numbers refer to entries on the district court docket.
This procedure is reflected in NCR's written policies. The first thirty minutes of a customer engineer's commute (in both directions) are unpaid, [288-10] at 9; [288-9] at 184:23-185:10; [281-52] at 42:2-15, so NCR's customer engineer handbook advises that, "[a]s a general rule," at least thirty minutes before the start of their shift, customer engineers should check in for their call assignments and "activate an incident" or update their "whereabouts." [288-10] at 8. See also id. at 4; [288-1] at 92:6-8 (customer engineers are required to comply with the handbook). According to the handbook, this should take "no more than one or two minutes." [288-10] at 8.
The handbook also says that "[a]ll work time is compensable time," and that customer engineers are "required to document and report all time worked." [288-10] at 7. Examples of prohibited off-the-clock work include answering work-related phone calls, reading or responding to NCR email, and processing or ordering parts. [288-11] at 15. With regard to pre-shift work, the handbook instructs that, "[b]efore the start of [their] shift, [customer engineers] are prohibited from performing any work, other than briefly checking [their] handheld mobile device 30 minutes before the start of [their] shift (this is so that you can determine the location of your first assignment)." Id. at 15. Supervisors are prohibited from "requiring, encouraging, or even suggesting" that customer engineers work off-the-clock. Id.
Meadows says these rules are often broken. According to Meadows, customer engineers often perform (and do not record time spent on) pre-shift tasks that go beyond checking the location of their first job site. For instance, many customer engineers spend time before their shift checking the details of their entire schedule for the day, reviewing other emails, and fielding work-related calls, among other things. See e.g., [281-8] at 35:23-36:25 (one customer engineer, Roy Bratt, checks his email every morning); [281-10] at 31:16-34:3 (another customer engineer checks "to see what kind of work I'm going to have on my plate for the day and see if I need to reach out to my service coordinator"); [281-11] at 10:18-11:15 (checks email, learns where to go); [281-13] at 30:13-17, 31:12-22 (learns where to go); [281-16] at 24:20-25:14 (checks schedule, email, texts); [281-25] at 86:15-87:17 (checks schedule, route map); [288-15] at 28:8-29:5, 31:3-21 (checks email, calls service planner). Some report taking longer than two minutes to complete these pre-shift routines. See, e.g., [288-15] at 29:15-20 ("5 to 15 minutes"); [281-8] at 36:4-10 ("it averages between five minutes and probably twenty to twenty-five minutes"); [281-28] at 70:21-71:7 (often only one to two minutes but sometimes five to fifteen minutes); [281-51] at 35:22-36:10 (between one to five minutes to check emails both before and after shifts). And in spite of the handbook's instructions, many do not record their pre-shift work at all because, for instance, they say recording time is a hassle, or that their managers do not like it when they record that time. See, e.g., [288-15] at 48:2-12. Other customer engineers say that NCR's time recording system is only capable of recording time in increments of seven minutes or more and that, as a result, it would be useless or impossible to record their shorter bursts of pre-shift work. See [281-20] at 44:4-10, 54:23-55:7; [288-15] at 18:6-16.
The list of pre-and-post-shift work activities includes processing parts, communicating with colleagues, printing documents, "loading keys," conversations between day-shift workers and night-shift workers, closing out tickets, key audits, and more. See [281-7] at 69:8-25 (checking for first location); [281-44] at 114:9-115:5 (phone calls, emails, looking at work orders); [281-45] at 38:5-38:18 (setting up route), 43:20-44:19 (coordinating with colleagues), 44:20-46:17 (responding to emails, coordinating with the control tower), 48:14-50:6 (ordering parts); [281-18] at 102:25-103:19 (answering phone calls and emails, ordering parts); [281-50] at 14:3-9 (phone calls); [281-38] at 10:12-22 (answer emails, phone calls); [281-8] at 60:9-18 (checking email, loading and checking keys); [281-43] at 11:24-12:24 (closing calls, processing parts); [281-46] at 11:10-12:14 (communications with colleagues); [281-12] at 88:13-89:9 (phone calls, emails, training course); [281-16] at 35:13-36:15 (parts management); [281-46] at 61:13-67:9 (conversations between night shift and day shift workers); [281-51] at 34:6-14 (emails); [281-35] at 41:18-45:21 (printing information, reading manuals); [281-20] at 56:14-24 (closing out tickets), 61:11-62:10 (key audits); [281-52] at 44:18-45:8 (paperwork), 96:21-97:5 (training, cleaning van).
At least some of the managers that were deposed say that (in accordance with the handbook) they regularly tell their customer engineers to record all of their time worked. [281-27] at 77:11-23; [281-21] at 33:17-34:6; [281-15] at 67:15-68:6. At least some of the customer engineers say that they are aware that they are supposed to accurately report all of the time they spend working. [281-19] at 32:25-33:23; [281-13] at 49:7-15. Others confirmed that they are paid for the time they record, even if it amounts to overtime (and even if they bill the time recorded to an "administrative" code) that was not preapproved by a manager. [281-7] at 42:11-23; [281-43] at 84:9-85:4; [281-49] at 95:2-9; [281-16] at 28:3-28:8.
At the same time, at least one customer engineer says that his current manager lets him record no more than ten or fifteen minutes of work before leaving his house and that his previous manager did not allow him to record any time at all for pre-shift activities. [281-40] at 89:19-90:25. See also [281-12] at 10:10-11:23 (another manager says that, although he tells his current customer engineers not to answer emails or calls outside of their shift, those same policies were not enforced by his previous manager.) In addition, at least a few customer engineers reported working during their lunch break, [281-11] at 13:21-14:18 (one took "critical calls" during lunch, but acknowledged that was her choice); [288-14] at 73:22-74:9 (another customer engineer sometimes looked at upcoming job tickets during lunch), and others said they performed (and did not record) work on their days off. [281-46] at 62:8-63:10; [281-51] at 65:4-23; [281-40] at 59:10-24; [281-38] at 10:8-22; [281-8] at 54:13-55:12.
In January of 2018—a month after notice of this lawsuit issued to customer engineers around the country—NCR started measuring productivity using something called the "Five Star" program. [281-34] at 102:21-24. The Five Star program assesses customer engineers' performance using metrics like utilization, productivity, "revisits," and "re-trips," among other things, and compares their performance against that of other customer engineers around the world. [288-3] at 35:3-37:8. At least one customer engineer was under the impression that he would get a raise if he had a high Five Star rating. [288-16] at 67:17-19.
Some customer engineers reported believing that the program discourages engineers from recording time that is not related to a specific work order (e.g., reviewing general emails, fielding calls, etc.) because such time counts against some of the metrics that are used to calculate one's overall Five Star score. [281-12] at 67:23-69:11; [281-16] at 28:20-29:8, 53:5-54:11. One manager reported being evaluated based on the Five-Star ratings of the customer engineers that he oversees. [281-13] at 113:5-20; [281-12] 67:12-16. NCR's representatives, however, say that not recording time does not improve one's overall Five Star score. See [288-13] at 56:21-59:15; [281-34] at 96:23-97:21, 101:25-102:20. Instead, they say that the Five Star program accounts for, and seeks to minimize, the incentive to work off-the-clock. See [288-8] at 78:2-79:13; [288-13] at 56:3-57:23. Some managers allowed customer engineers to bill for the complained-of pre-and-post-shift work so long as they billed it as "administrative time." [281-23] at 64:3-17 (one manager said he scheduled administrative time "all the time" so that his customer engineers could do things like clean their vehicles or complete training); [281-49] at 95:2-96:13 (noting that while some managers approved administrative time, those same managers occasionally denied requests to bill administrative time); [281-11] at 52:8-25 (one customer engineer remembered being allowed to bill as administrative time hours he spent installing shelving in his van); [281-16] at 28:3-8 (one customer engineer confirmed he was paid if he recorded his time using the administrative code). At least one manager confirmed that time spent on things like training did not affect the Five Star rating at all. [288-13] at 26:1-28:6.
None of the witnesses provided a complete explanation of how the Five Star program works. For instance, Zeid Isho, a field operations senior manager for NCR, says that the calculations behind the Five Star program were "basically system generated" and that there were "a lot of algorithms and mathematical equations that go on in the background of the 5 star program." [288-8] at 77:6-21.
The customer engineers gave different reasons for not recording their time. Some said they did so as a "matter of survival," to "be a good employee," [281-19] at 43:19-25, or to "help out" their coworkers. [281-40] at 60:10-15; [281-38] at 15:22-16:11; [281-10] at 54:8-54:24; [281-41] at 75:16-76:2; [281-12] at 22:9-24; [281-37] at 57:23-58:22; [281-46] at 57:13-58:8. Others said it was in part because their managers did not allow them to use an administrative code to bill the time they spent on pre- and post-shift work. [281-37] at 68:22-72:8; [281-16] at 44:12-22. Some cited pressure from their managers to deliver results, [281-49] at 16:19-17:13, and others mentioned the Five Star program specifically. [281-41] at 64:12-65:12, 67:2-14; [281- 12] at 66:11-68:5; [281-18] at 73:18-74:5; [281-36] at 22:5-23:18; [281-16] at 28:20-29:13, 53:2-55:10. When one customer engineer was asked why he might not record time spent helping out other customer engineers on the weekend, he said, "[t]hat's a good question," and that he did "not have a good answer." [281-22] at 39:11-40:5. These differences are an illustrative (but not comprehensive) representation of those identified during discovery.
NCR operates in all fifty states, and splits the states into more than 70 geographic territories. [172-2] ¶ 4; [172-4] ¶ 4. Each customer engineer operates within one territory, and each territory has a designated territory manager that oversees as many as fifty customer engineers. Id. Territory managers control the work schedules of their customer engineers. [281-24] at 15:20-16:8, 38:13-39:8, 39:12-23; 49:22-51:1; [281-23] at 31:22-32:8, 74:3-14; [288-10] at 7 ("[m]anagers have discretion to set schedules and control the hours of work" and take into consideration "[v]arious factors, such as workloads, operational efficiency, and staffing needs"). In California and Washington, customer engineers' entire commute is compensated, [281-23] at 61:6-15, and some customer engineers in New York are paid according to collective bargaining agreements negotiated by their union representatives. [281-3] at 22-23. Other customer engineers are on special assignments that require them to be on call 24 hours a day (and are paid differently as a result). [281-10] at 47:15-48:6; [281-23] at 45:2-24.
III. Analysis
At this stage of the conditional certification inquiry, Meadows is required to identify a factual or legal nexus that binds him and the other plaintiffs together as "victims of a particular violation of the overtime laws." Hudgins, 2019 WL 354958 at *3; Camilotes, 286 F.R.D. at 345; Russell, 721 F.Supp.2d at 812. At the last stage (conditional certification), I found that NCR had not yet "eliminate[d] the possibility that [it] ... had an unwritten policy or an unofficial practice of encouraging off-the-clock work." [192] at 5. Relying on a thin record and resolving reasonable inferences in Meadows's favor, it remained possible that NCR was "requiring the performance of significant tasks before and after shifts while requiring CEs to clock in and clock out based on their arrival and departure times." [192] at 4.
Based on the additional evidence both sides dug up during discovery, I find that Meadows has failed to produce sufficient evidence that this case should continue as a collective action. See Hudgins, 2019 WL 354958 at *3; Camilotes, 286 F.R.D. at 345. The two most compelling nationwide factual or legal nexuses Meadows has identified are the handbook and the Five Star program. Page four of the handbook makes clear that its rules apply to a uniform, identifiable subset of employees, [288-1] at 4, but one would have to stop reading there to conclude the policy bound those employees "as victims" of the violations alleged. Hudgins, 2019 WL 354958 at *3; Camilotes, 286 F.R.D. at 345; Russell, 721 F.Supp.2d at 812. Page fifteen, titled "'Off the Clock' Work is Prohibited," describes most of the practices complained about, makes clear that each is strictly prohibited, and includes a general prohibition against any and all off-the-clock work by instructing customer engineers to record any and all time worked. [288-1] at 15.
Even if the handbook underestimates the amount of time it takes customer engineers to check the location of their first job site each morning, it also unequivocally states that all time spent working must be recorded. See [288-1] at 15 (customer engineers are "prohibited from performing any work, other than briefly checking [their] handheld mobile device" each morning, and must record all time worked). Nothing in it explicitly prohibits or excepts the few minutes it takes to determine the location of the customer engineer's job; that time, if worked, should be recorded. See id.
Meadows identified one customer engineer that believed that NCR would not compensate any time spent on a task that amounted to less than three minutes. [288-1] at 56:12-57:1. When asked whether he believed that was an NCR policy, he said he did not know. Id. Another testified that a former manager had told him that any time spent on tasks that amounted to less than either seven or five minutes (the customer engineer could not remember which) was not compensable. [281-20] at 44:3-25. See also [288-15] at 17:6-16 (a third customer engineer testified that NCR's time recording software did not register entries of five minutes or less); [288-1] at 62:3-63:11 (one of NCR's corporate representatives testified that any work performed at home is compensable and should be recorded so long as it took more than two minutes to complete). This testimony is insufficient at this stage to identify a factual or legal nexus that binds the members of the collective together as victims of any policy or practice, written or unwritten. The first two customer engineers' experiences were dependent on their interactions with their individual managers, which suggests that collective treatment of these claims would not be both efficient and fair. With regard to the third customer engineer's testimony (and that of NCR's corporate representative), Meadows's legal support amounts to an argument that decertification should not occur just because individualized damages calculations will be necessary. See [288] at 18-19. That may be true, but during this second stage, Meadows had the burden of identifying a sufficient factual nexus that tied the members of the collective together as victims. He has not described the policies or mechanisms that allegedly prohibited recording small amounts of time in sufficient detail to survive a motion to decertify. Having had the benefit of discovery, he was required to produce more than the disjointed testimony of a few witnesses in order to justify the continuance of a nation-wide class. See also Sandifer v. U.S. Steel Corp., 678 F.3d 590, 593 (7th Cir. 2012), aff'd, 134 S. Ct 870 (2014) ("[i]t is only when an employee is required to give up a substantial measure of his time and effort that compensable working time is involved"); Comcast Corp. v. Behrend, 569 U.S. 27, 28 (2013) (class certification analysis, like collective certification analysis, "will frequently overlap with the merits of the plaintiff's underlying claim") (cleaned up).
For purposes of a motion to decertify, I do not need to decide the merits of Meadows's claims, but cannot ignore them completely. Otherwise, Meadows and his fellow customer engineers could claim to be bound by a factual nexus that bears no relationship whatsoever to the legal violations alleged in his complaint. See Comcast, 569 U.S. at 28 (class certification analysis, like collective certification analysis, "will frequently overlap with the merits of the plaintiff's underlying claim because a class determination generally involves considerations that are enmeshed in the factual and legal issues comprising the plaintiff's cause of action") (cleaned up). Meadows cannot survive a motion to decertify by pointing to a patently lawful written policy that applies uniformly to the entire collective. The handbook is not a sufficient nexus because it does not tie Meadows and the other customer engineers together as victims of any violations.
The only other written nationwide policy alleged is the Five Star program. That program did not come into existence until two months after this collective action was conditionally certified. See [281-34] at 102:21-24. There are instances when prejudice would result from declining to decertify a collective based on a policy that did not exist when the collective was conditionally certified, see Blakes v. Illinois Bell Tel. Co., No. 11 CV 336, 2013 WL 6662831, at *9 (N.D. Ill. Dec. 17, 2013) (denying modification because modifying the certified collective could have "ensnare[d] opt-in plaintiffs in a theory of liability the implicates them in a practice of hiding time from their employer" that was not present when the class was certified), but those circumstances are not present here. The claims that were conditionally certified were based in part on the idea that some customer engineers were not reporting their time. No unfair surprise would be visited upon the members of the collective if Meadows is allowed to advance those same theories based on both the Five Star program and the handbook (as opposed to simply the handbook or some other unwritten policy or practice, as was proposed during the collective certification stage). Neither side has asked to expand the collective to include persons not contemplated at the conditional certification stage, either. NCR has not demonstrated that it would suffer any prejudice, in part because it designed the Five Star program and has access to the information it needs to defend itself against the claim.
In any event, Meadows's new claims fail because the Five Star program does not bind the customer engineers as victims, either. Despite having access to comprehensive discovery, Meadows failed to produce evidence that explains how, exactly, the Five Star program works. That in turn makes it difficult to determine whether the program encourages violations of the FLSA in the ways Meadows alleges, and whether NCR will have individualized defenses. A few of NCR's customer engineers seem to think they are able to improve their productivity metrics by not recording work that they perform outside of their shifts, see, e.g., [281-12] at 67:2-22; [281-16] at 28:20-29:18, 53:5-54:11, but NCR's managers say the program eliminates any such incentive by balancing a "utilization" metric against a "productivity" metric. See [288-8] at 76:11-79:13; [288-13] at 56:21-58:16 (under-reporting time might benefit a customer engineer with regard to one of the metrics but would hurt them under others); [281-34] at 96:24-97:21, 101:18-102:20. If Meadows wished to rely on the Five Star program as a factual nexus to survive this motion to decertify, the burden was on him to produce evidence sufficient to show it tied him and his fellow customer engineers together as victims of violations of the overtime laws. Hudgins, 2019 WL 354958 at *3; Camilotes, 286 F.R.D. at 345. Instead, what he has produced amounts to the type of "substantial allegation" that sometimes survives the first stage of the certification process, see Jones, 2009 WL 3260004 at *2, but that falls short of the identification of a legal or factual nexus that is necessary to survive a motion to decertify. See, e.g., Camilotes, 286 F.R.D. at 345; Russell, 721 F.Supp.2d at 812; Campbell v. City of Los Angeles, 903 F.3d 1090, 1118 (9th Cir. 2018) (at the decertification stage, plaintiffs must provide substantial evidence that their claims arise out of a single policy, custom or practice that leads to FLSA violations).
Even if there were sufficient evidence that the Five Star program works the way that Meadows says it does, I find persuasive the decisions of other courts that have declined to find that programs like the Five Star program are a sufficient legal or factual nexus simply because of the risk that they might incentivize employees to work off-the-clock in an effort to maximize their score. See Brickey v. Dolgencorp., Inc., 272 F.R.D. 344, 347-48 (W.D.N.Y. 2011) ("[t]he Court declines to hold that facially-lawful policies, which encourage store management to make productive use of employees' time ..., can form the equivalent of a common policy or plan that violates the law merely because they indirectly might encourage the minimization of overtime"); Eng-Hatcher v. Sprint Nextel Corp., No. 07 CIV 7350, 2009 WL 7311383, at *4 (S.D.N.Y. Nov. 13, 2009). Encouraging efficiency during working hours is not the same thing as encouraging unpaid overtime work, and a policy that does the former does not bind together as victims the people subject to that policy.
Meadows might still prevail if he could show an unwritten practice or policy that resulted in the Five Star program (or the handbook, or any other written policy) being applied in a way that violates the FLSA. See Russell, 721 F.Supp.2d at 815 ("lawful written (or verbal) policies will not shield the company from liability if plaintiffs can show other company-wide practices that may have been contrary to those policies and violated the FLSA"); Brand v. Comcast Corp., No. 12-1122, 2012 WL 4482124, at *5 (N.D. Ill. Sept. 26, 2012) (holding that collective actions may be certified based on allegations that a company has an unwritten policy that violates the FLSA). This was the inference that justified conditional certification. The certification could remain in place if Meadows identified circumstantial evidence that there was a "pattern to the frequency, manner or duration of ... pre-shift and post-shift activities." Elder v. Comcast Corp., No. 12 C 1157, 2015 WL 3475968, at *9 (N.D. Ill. June 1, 2015). Another way might be to produce evidence of a nationwide culture or attitude that encouraged overtime work without pay.
But the evidence he has produced shows the opposite. Each individual customer engineers' description of their reasons for working off-the-clock varied considerably. So did the work they performed. Some worked off the clock for personal reasons, some did so because they did not believe their individual managers would apply NCR's policies as written, and only some believed they might obtain a higher Five Star rating. Some checked their email, some prepared their tools, and others made calls. Meadows has produced evidence that customer engineers were performing plenty of off-the-clock work, but that evidence has not revealed any pattern that supports an inference that NCR had a common policy or practice, the legality of which would be efficient to assess via a collective action.
Another problem with Meadows's theory is that individual managers retain the discretion to approve both overtime work and other billing practices (such as the use of an "administrative" code) that significantly reduce the pressure customer engineers might feel to work off-the-clock. As a result, whether Meadows and his fellow customer engineers were encouraged to work off-the-clock depends on the individual circumstances of each manager-engineer relationship and interaction. "Alleged FLSA violations stemming from the enforcement decisions of individual supervisors, rather than a company-wide policy or plan" are not appropriate for collective treatment. Adair v. Wisconsin Bell, Inc., No. 08-C-280, 2008 WL 4224360, at *7 (E.D. Wis. Sept. 11, 2008); Salinas v. O'Reilly Auto., Inc., No. CIV.A.3:04-CV-1861-B, 2005 WL 3783598, at *1 (N.D. Tex. Nov. 17, 2005) (plaintiff had failed to make the minimal evidentiary showing required for conditional certification when the plaintiff's evidence showed only "that a handful of [defendant's] managers have committed possible FLSA violations in a variety of different ways"). The questions that need to be answered in such circumstances are not the sort of broadly applicable inquiries that help drive collective action litigation forward in a productive, efficient way. See Boelk v. AT & T Teleholdings, Inc., No. 12-CV-40-BBC, 2013 WL 261265, at *12 (W.D. Wis. Jan. 10, 2013) ("[a]lthough plaintiffs and other technicians submitted declarations stating that they often worked through meal breaks because of the meal break restrictions, the productivity rating system or a combination of both, it is clear from the deposition testimony of plaintiffs and other technicians that the reason for doing so depended on the circumstances, which varied on a day-to-day basis").
For similar reasons, I grant the motion to decertify insofar as it addresses claims of work that occurred during mealtime and on days off. Meadows has not pointed to any evidence that suggests claims pertaining to off-the-clock work during meal hours or on days off should be treated any differently than pre-shift and post-shift work. No aspect of the Five Star program or the handbook applies in any way that is unique to mealtime hours or days off. His mealtime work claims center around the allegation that customer engineers received calls, text messages, and emails during their lunch breaks, but do not include facts that suggest NCR required them or encouraged them to answer those calls, texts, or emails. See [288] at 11, 21. And again, as was the case with customer engineers that reported working pre-shift and post-shift, each individual customer engineer's explanation of their decision to work off-the-clock points to individualized rationalizations motivated by things like the demands of their schedule and the expectations of their individual managers—not the existence of a nationwide policy or practice, the legality of which could be determined all at once.
The factors sometimes used to assess motions to decertify support decertification here: "(1) whether the plaintiffs share similar or disparate factual and employment settings; (2) whether the various affirmative defenses available to the defendant would have to be individually applied to each plaintiff; and (3) fairness and procedural concerns." Camilotes, 286 F.R.D. at 345 (N.D. Ill. 2012); Mielke v. Laidlaw Transit, Inc., 313 F.Supp.2d 759, 762 (N.D. Ill. 2004).
With regard to the first factor, courts typically consider the "location, job duties, supervision, and policies or practices that bind the plaintiffs' claims together," in addition to whether there was a uniform compensation plan and common supervision. Camilotes, 286 F.R.D. at 345 (N.D. Ill. 2012); Blakes, 2013 WL 6662831, at *10. NCR's customer engineers' factual and employment settings are unique. They start their day at home and travel to remote locations spread about their geographic territory. They do not congregate together in the same place in order to perform their assigned tasks. They are all doing something similar once they arrive—fixing ATM machines and point-of-sale systems—but they are performing their functions in isolated places, tethered to NCR only by their communications with their managers and central dispatchers. Their schedules are dictated by their individual managers and the unique repair needs of the customers in their geographic region. Their individual managers retain discretion to approve overtime pay and to authorize them to record time spent on work unrelated to any individual assignment. There is no common supervision (at least not of all or even most of the members of the nationwide collective action). Each of the customer engineers may have been provided with work-supplied email addresses, phones, and laptops, and each may have used those items to communicate with their managers and their central dispatchers, but the fact that all of them communicated via the same medium is of little use in determining whether the content and context of those communications bound them together as victims. The customer engineers did many different types of off-the-clock work, see [281-7] at 69:8-25 (checking for first location); [281-44] at 114:9-115:5 (phone calls, emails, looking at work orders); [281-45] at 43:20-44:19 (coordinating with colleagues); [281-8] at 60:9-18 (loading and checking keys), for many different reasons. See [281-19] at 43:19-25 (as a "matter of survival" and to "be a good employee"); [281-37] at 68:22-72:8 (NCR did not allow them to use a certain billing code); [281-49] at 16:19-17:13 (managers expected results); [281-22] at 39:11-40:5 (unsure why). Some of the collective action members were also paid differently because they were subject to collective bargaining agreements or were on special assignments to work for a single client, among other reasons. Their factual and employment settings were different. The first factor weighs in favor of decertification.
With regard to the second factor, some of the defenses NCR has put forth are unsound. Determining whether NCR violated the FLSA with respect to customer engineers in New York might depend on the terms of the collective bargaining agreements in place there, but that is not a sufficient reason to avoid a collective action here, in part because the FLSA rights Meadows seeks to assert are "independent of the collective-bargaining process." Barrentine v. Arkansas-Best Freight Sys., Inc., 450 U.S. 728, 745 (1981). See also Brooklyn Sav. Bank v. O'Neil, 324 U.S. 697, 707 (1945) (waiver is not a valid defense to an FLSA action); Caserta v. Home Lines Agency, Inc., 273 F.2d 943, 946 (2d Cir. 1959) (estoppel is not a valid defense because the obligation to comply with the FLSA "is the employer's and it is absolute").
But when decisions made by individual managers affect the pressures that employees feel to work off the clock, see Camilotes, 286 F.R.D. at 352, Kuznyetsov v. W. Penn Allegheny Health Sys., Inc., No. CIV.A. 10-948, 2011 WL 6372852, at *6 (W.D. Pa. Dec. 20, 2011) ("[t]he individual supervisors had discretion with regard to whether a meal break was missed and how a meal break deduction would be cancelled"), and because it remains to be seen whether those managers (and by extension, NCR) had actual or constructive knowledge of any work performed off-the-clock, Camesi v. Univ. of Pittsburgh Med. Ctr., No. CIV.A. 09-85J, 2011 WL 6372873, at *9 (W.D. Pa. Dec. 20, 2011), the analysis of at least some of NCR's affirmative defenses requires further individualized inquiry. Meadows says that evidence that NCR knew, or should have known, that customer engineers were performing work off-the-clock "can be found in NCR's own policies, admissions, and data." [288] at 23. But Meadows carries the burden at this stage, and he has failed to cite to the policies, admissions, or data that might show NCR had such knowledge. Vague and unsubstantiated assertions that NCR's affirmative defenses do not require individualized inquiry are insufficient. The second factor, too, weighs in favor of decertification.
With regard to fairness and procedural concerns, this factor weighs in favor of decertification because this case presents a "myriad of individualized factual issues" that will require complex proceedings and extensive judicial resources without resulting efficiencies, and because continuing this action as a collective will risk significant unfairness to NCR. See Camilotes, 286 F.R.D. at 353. Meadows has offered very little in the way of thoughtful trial planning. He has not explained how to minimize or eliminate the difficulties associated with managing a collective action of nearly 1,600 customer engineers in a case where his claims depend on using the collective testimony of those engineers to encourage the trier of fact to draw an inference about the existence of a nationwide, unwritten policy or practice. Meadows mentions things like "class-wide data" (which could help simplify damages calculations but would not help establish liability, see [288] at 15), the use of subclasses (which would also not help establish liability because the unwritten policies or practices Meadows's alleges exist would have affected all customer engineers equally), and bifurcation (which, again, would simplify the damages portion of the case but would not avoid the complexities underlying Meadows's theories of liability), but does not elaborate in any helpful way as to how those mechanisms would simplify the initial determination of liability. See [288] at 7. In circumstances like these, where Meadows's theories are largely dependent on circumstantial inferences drawn from voluminous individual testimony, "[g]enerically stating that unspecified mechanisms may exist is not sufficient to allay manageability and fairness concerns." Camilotes, 286 F.R.D. at 354.
Meadows has presented no plans for managing the large number of witnesses that would need to testify or the potentially voluminous documents that would likely be necessary to establish liability on a class-wide basis for the theories he relies on. If the case were as simple as reading a handbook or the terms of the Five Star program—or if the case depended only on the testimony of a few executives that were allegedly responsible for the unlawful unwritten practices or policies—this type of sweeping assurance might be enough. But what Meadows proposes is relying on the testimony of many different customer engineers to establish a pattern of experience so uniform as to create an inference that such an unlawful, unwritten policy or practice exists. That is a valid theory as a matter of law, but as a matter of trial procedure and judicial efficiency, it presents significant difficulties. Meadows's briefing does not suggest there are ways to eliminate those difficulties.
At most, Meadows provides unpersuasive assurances that he and his fellow customer engineers would limit their presentation to the trial testimony of a representative subset of the collective. His briefing says only that, at some point in the future, "comparisons between a representative sample of Plaintiffs' ES Mobility and Workday data will reveal an average amount of time Plaintiffs performed off-the-clock work." [288] at 25. Meadows gives no reason why that sampling could not have been explained in greater detail as part of his response, but one reason may be that such sampling will not prove as easy as Meadows says it will. Sample evidence will support group adjudication only if "each class member could have relied on that sample to establish liability if he or she had brought an individual action." Tyson Foods, Inc. v. Bouaphakeo, 136 S. Ct. 1036, 1046 (2016). See also Solsol v. Scrub, Inc., No. 13 CV 7652, 2017 WL 2285822, at *8 (N.D. Ill. May 23, 2017). The problem for Meadows is that, to the degree the deposition testimony hints at the existence of an unwritten policy or procedure, that unwritten policy or procedure played out only in the context of individual manager-engineer relationships around the country. Sampling would be of little help because the other members of the collective could not rely on representative testimony to establish that the same violation occurred in their case; whether it did would depend on whether their managers approved overtime or took other actions to alleviate any pressures they might have felt to work off-the-clock.
There are some instances when sample evidence can be a trustworthy and efficient proxy for a larger body of evidence but, at the least, such sampling usually requires the use of statistical methods to create a random assortment of collective members as witnesses. See Espenscheid, 705 F.3d at 774. Meadows has not provided any insight into what type of sampling methods he might use or how he might go about selecting a truly random, representative sample of customer engineers to testify. His ability to do that has been at least partially undermined by the fact that only thirty-six of the eighty-one plaintiffs that NCR subpoenaed actually appeared for their depositions. See [277] at 4. That low turnout casts doubt on Meadows's ability to obtain testimony from anyone other than the representatives most interested in winning the case—a subset that is not representative of the collective as a whole. Nor has Meadows explained how the selection of representative candidates would assuage concerns that other customer engineers were underreporting their time for any of the many benign reasons that were suggested during their depositions. See Espenscheid, 705 F.3d at 774 ("plaintiffs have not explained how their representative proof would distinguish such benign underreporting from unlawful conduct by DirectSat").
That each customer engineer might require an individualized damages hearings is not a sufficient reason in and of itself to decertify. Rule 23 permits class certification even if separate hearings will be necessary to establish damages, so long as liability can be established for the class as a whole. See Butler v. Sears, Roebuck & Co., 727 F.3d 796, 800 (7th Cir. 2013); Arreola v. Godinez, 546 F.3d 788, 801 (7th Cir. 2008) ("the need for individual damages determinations does not, in and of itself, require denial of his motion for certification"). But special masters are only helpful "once liability is established," see Espenscheid, 705 F.3d at 775, and here, the factual complexities extend not only to the issue of damages but to the threshold issue of liability as well. Again, having failed to identify a nationwide written policy or practice that binds the customer engineers together as victims, Meadows theory depends on producing enough customer engineers to testify that a trier of fact will be able to infer the existence of a nationwide unwritten policy or practice that violates the FLSA. The use of circumstantial evidence to encourage this type of inference is permissible, but Meadows has not marshaled enough evidence to suggest that it is possible here.
Given the factual complexities the evidence has revealed, the importance of the relationship between each customer engineer and his or her manager, the discretion each manager retained to award overtime work and discourage off-the-clock work, the lack of sufficient evidence of any nationwide unwritten policy or practice that might explain the numerous instances of off-the-clock work that Meadows has identified, and the absence of any meaningful explanation for how Meadows might overcome these factual issues via efficient trial management or thoughtful sampling, the three factors together favor decertification. See Camilotes, 286 F.R.D. at 353. Meadows has failed to identify a factual or legal nexus that binds him and the other plaintiffs together as "victims of a particular violation of the overtime laws." Hudgins, 2019 WL 354958 at *3; Camilotes, 286 F.R.D. at 345; Russell, 721 F.Supp.2d at 812. He may continue to pursue his claim individually, but the FLSA collective action must be decertified.
IV. Conclusion
NCR's motion to decertify, [280], is granted. A status hearing is set for March 18, 2020, at 9:30 a.m. ENTER:
/s/_________
Manish S. Shah
United States District Judge Date: March 4, 2020