Full Opinion

FIRST DISTRICT COURT OF APPEAL STATE OF FLORIDA _____________________________ No. 1D19-2008 _____________________________ MAC PAPERS, INC., Appellant, v. GARY E. BOYD, Appellee. _____________________________ On appeal from the Florida Commission on Human Relations. October 16, 2020 MAKAR, J. In this age discrimination case, the administrative law judge made two inter-related legal errors that justify vacating the judgment in favor of Gary Boyd, a then 67-year-old salesman who agreed to be suspended by his employer, Mac Papers, Inc., after his arrest for domestic battery, but who thereafter refused to enter a proposed return-to-work agreement and was thereafter terminated on June 7, 2017, the date the parties stipulated was the triggering event leading to this litigation. Boyd, a thirty-year employee working on commissions in the Orlando, Florida, area for the office products company, was arrested, jailed (missing two days of work), and suspended pending resolution of his criminal case in Seminole County, Florida. His suspension was pursuant to mutually agreed-upon terms. The initial criminal charges against Boyd were for kidnapping/false imprisonment and domestic battery of his wife on April 9, 2017. The terms of suspension required that Boyd not service or visit customers while wearing the electronic ankle monitoring device he was required to wear; he could again service customers if the monitor was no longer mandated or the criminal charges dropped by June 1, 2017; and he would potentially be “re-hirable” if the charges were “dismissed,” but he would be terminated if they were not by that date. 1 During the suspension period, a number of his accounts were assigned to other employees, some who were younger than Boyd. About six weeks after his arrest, Boyd agreed to the entry of a permanent injunction against him in the civil domestic battery case, resulting in the ankle monitor’s removal on May 25, 2017. As to the criminal matter, he was referred to a pre-trial diversion program, a condition of which was the completion of a nine-month Batterer’s Intervention Program that required him to perform community service, not use or possess drugs or alcohol, and be subject to random testing for alcohol use. The ankle monitor issue having been resolved, Boyd’s supervisors met with him on June 2, 2017, to discuss the conditions under which he would be permitted to work, which were summarized in a return-to-work agreement that, when signed by Boyd, would allow him to begin working the following day; many of the conditions mirrored those required to resolve his civil and criminal actions. One of the conditions of returning to work was that Boyd was not to contact reassigned accounts, a condition that he admitted he violated almost immediately. Boyd never returned a signed return-to-work agreement and, because of his breach of the no-contact requirement, was terminated on June 7, 2017, the date on which both Boyd and Mac Papers jointly stipulated in writing that he was fired. As to his termination date, the “Parties’ Stipulated Facts” stated that “[o]n Wednesday, June 7, 2017, Boyd’s employment was terminated and he received a letter explaining the reason why; specifically, he contacted a reassigned account.” 1 The charges weren’t formally dropped until July 25, 2017, two months after the state attorney had required Boyd to enter the Batterer’s Intervention Program. 2 Boyd sought a right to sue letter from the Florida Commission on Human Rights based on his claim of age discrimination, but the agency reviewed the matter and issued a “no reasonable cause determination.” Boyd, thereafter, filed a petition for relief, resulting in discovery and a trial before an administrative law judge. In support of his petition, Boyd had no direct evidence of age discrimination and he did not claim disparate impact based on age. Instead, he relied on the indirect method of proof attributable to McDonnell Douglas Corp. v. Green, 411 U.S. 792, 793 (1973), which uses a burden-shifting framework. Under this method, Boyd’s sole claim of age discrimination was that he was treated differently compared to a younger Mac Papers employee whom he asserted was situated similarly to him. Boyd selected Don Swift, who worked out of the company’s Atlanta, Georgia office, as the lone “similarly situated comparator” from Mac Papers’ workforce, which had approximately 150 employees in sales manager positions spread out over nine Southeastern states. Boyd and Swift were alike in that both were salespersons who had committed domestic battery and that their return to the workplace was conditioned on random testing, mental health counseling, and the like. But that ends their similarity. Swift never had to wear an ankle monitor; Boyd did, resulting in his agreed-to suspension until the monitor was no longer required. Swift did not engage in additional misconduct requiring further disciplinary action; Boyd did, admitting that he violated the back-to-work agreement by telling customers about his restrictions, which created a problem for Mac Papers. Both had different supervisors (Boyd worked in Central Florida while Swift worked in Atlanta, Georgia). Picking a single comparator with inadequate, irrelevant, or superficial similarities falls short of what the law requires. Courts require that comparators be meaningful, which explains why the Eleventh Circuit—which reviewed the oftentimes discordant caselaw on the topic—recently decided en banc that comparators must be “similarly situated in all material respects.” Lewis v. City of Union City, Georgia, 918 F.3d 1213, 1218 (11th Cir. 2019) 3 (rejecting “nearly-identical standard” as too rigid and rejecting “not useless” as too lax). Florida state courts have not yet affirmatively addressed whether and how Lewis applies, but they have held that federal cases generally guide the analysis. Fla. State Univ. v. Sondel, 685 So. 2d 923, 926 (Fla. 1st DCA 1996). With Lewis and its progeny as our guideposts, Swift fails as a valid comparator. Consistent with Lewis, a “comparator’s misconduct must be similar in all material respects.” McPhie v. Yeager, No. 19-14914, 2020 WL 3468286, at *2 (11th Cir. June 25, 2020) (applying Lewis). But Swift was not similarly-situated in all material aspects such as: their supervisors (they had different ones), the conditions arising from their domestic battery claims (Boyd’s ankle monitor), and their ongoing conduct (Boyd engaged in admittedly improper conduct a second time). See id. (noting that a valid comparator “will share the plaintiff’s employment or disciplinary history. In other words, the plaintiff and his comparators must be sufficiently similar that they cannot be reasonably distinguished.”) (citation omitted). That Mac Papers had one corporate-level director of human resources who oversaw all regional managers matters little because she wasn’t either Boyd’s or Swift’s direct supervisor. Id. at *3 (noting that although office manager “had the ultimate disciplinary authority over them all, he was not their direct supervisor and he delegated that supervision because of the large size of the Sheriff’s Office”). Based on the lack of similarities, and the material differences between the two employees, Swift was a legally inadequate comparator to Boyd, thereby negating Boyd’s effort to establish a prima facie case under the McDonnell Douglas framework. Lewis, 918 F.3d at 1231 (finding that plaintiff “must demonstrate—as part of her prima facie case—that she was treated differently from other individuals with whom she was similarly situated in all material respects”). Though fact-finders are typically given much deference in determining whether a comparator is similarly-situated, the absence of evidence showing a similar disciplinary record can render a discrimination claim nonactionable as a matter of law. Reed v. Freedom Mortg. Corp., 869 F.3d 543, 549–50 (7th Cir. 2017) (finding that plaintiff’s discrimination “claim fails as a matter of law” due to lack of “evidence that any of these employees had a similar history of violations of the Attendance Policy or a similar disciplinary record”). 4 The degree to which the comparator analysis was flawed is attributable in large measure to the failure to enforce the parties’ stipulation. The parties stipulated that the adverse employment action, Boyd’s termination, occurred on June 7, 2017, when Boyd was fired due to his admitted violation of the company’s directive to not contact reassigned accounts. A meaningful analysis for purposes of age discrimination required that Boyd’s termination on the stipulated date be compared to a sufficiently younger employee who was “similarly situated in all material respects” but not terminated. Swift did not meet this standard because there is no evidence that he engaged in post-criminal-incident misconduct, which Boyd admitted he did; Swift did not contact clients or engage in similar misconduct post-suspension, making him an inapt comparator. That Swift had one significant act of misconduct, and Boyd had two, renders them dissimilar as to their employment histories. See Hester v. Univ. of Alabama Birmingham Hosp., 798 Fed. Appx. 453, 457 (11th Cir. 2020) (noting that plaintiff and comparator “did not share a similar employment history. [Plaintiff’s] personnel counseling record reflected that his false statement was one of the two reasons for his firing. [The comparator] never gave any statement, false or not, about his alleged misconduct. And that highlights a key difference in their employment histories.”). Both Swift and Boyd were subject to termination following their arrests, if they acted improperly; Boyd engaged in further misconduct, Swift didn’t, making them dissimilar in a highly material respect, one that the parties stipulated was the grounds for Boyd’s termination on June 7, 2017. Had Boyd’s theory been that his suspension by itself was based on age discrimination (putting aside that Boyd agreed to the terms of his suspension), his subsequent post-suspension misconduct would not be relevant in a comparator analysis, which focuses on the specific adverse employment action (i.e., suspension versus termination). But that is not his claim. 2 2 Even a suspension-centric theory of age discrimination would lack vitality for the obvious reason that Boyd had to wear an ankle monitor as a condition of the charges against him (thus precipitating his suspension) and Swift did not have such a condition. Simply stated, Mac Papers can’t be said to have treated Swift more favorably by allowing him, for example, to work with 5 The comparator analysis was doomed once the trial court allowed the litigation to wander from the stipulated termination date, morphing into an entirely different theory, i.e., that Boyd was “constructively terminated” on June 1, 2017, because the terms of the return-to-work letter were “intolerable” and not to Boyd’s liking. Once the stipulation was breached, and an earlier termination date was judicially fashioned, the comparator analysis became meaningless unless the comparator had—like Boyd had done—engaged in a post-suspension act of misconduct. This Court recently “wholeheartedly” endorsed that pretrial stipulations “are binding upon the parties and the court and should be strictly enforced.” Delgado v. Agency for Health Care Admin., 237 So. 3d 432, 437 (Fla. 1st DCA 2018) (citation omitted); see generally Esch v. Forster, 168 So. 229, 231 (Fla. 1936) (discussing importance of judicial enforcement of parties’ stipulations). What happened in this case demonstrates the importance of strictly enforcing stipulations: moving goalposts creates unfairness to litigants and spawns unintended consequences (such as the ill-fated comparator analysis here). Finally, our panel queried whether a remand was appropriate so that the administrative law judge could reevaluate the comparator analysis in light of Lewis, which is an option the Eleventh Circuit has employed. See Johnson v. Miami-Dade Cty., 948 F.3d 1318, 1322–23 (11th Cir. 2020) (remanding to reevaluate “comparators evidence under the new standard that we announced in Lewis”). The parties did not support doing so, perhaps because Lewis was filed as supplemental authority in the administrative proceeding, leaving our panel to either affirm or reverse. We therefore conclude that the combination of a breached stipulation and an inadequate comparator analysis requires reversal, particularly so when the evidence shows that material differences existed that support different treatment. See Lewis, 918 F.3d at an ankle monitor because Swift didn’t have one. See Herrera v. United Airlines, Inc., 754 Fed. App’x 684, 694–95 (10th Cir. 2018) (no evidence to conclude that comparators who briefly wore ankle monitors at work were similarly situated where employer didn’t know they were doing so). 6 1222–23 (“Treating different cases differently is not discriminatory, let alone intentionally so.”). REVERSED. LEWIS and ROBERTS, JJ., concur. _____________________________ Not final until disposition of any timely and authorized motion under Fla. R. App. P. 9.330 or 9.331. _____________________________ Leonard V. Feigel, Kevin E. Hyde, and Dabney D. Ware of Foley & Lardner LLP, Jacksonville, for Appellant. Richard L. Wilson, Orlando, for Appellee. 7