49 F.4th 1331
10th Cir.
2022

JEANNIE PARKER v. UNITED AIRLINES, INC.

No. 21-4093

United States Court of Appeals, Tenth Circuit

September 26, 2022

PUBLISH

Peter M. Katsaros, Hahn Loeser & Parks LLP (Eugene E. Endress and Rashmi D. Shivnani, Hahn Loeser & Parks LLP with him on the briefs), Chicago, Illinois, for Plaintiff-Appellant.

Jessica E. Whelan, Holland & Hart LLP, Las Vegas, Nevada (Bryan K. Benard, Holland & Hart LLP, Salt Lake City, Utah, with her on the briefs), for Defendant-Appellee.

Before HOLMES, BACHARACH, and PHILLIPS, Circuit Judges.

BACHARACH, Circuit Judge.

This case involves provisions of the Family and Medical Leave Act (FMLA), 29 U.S.C. §§ 2601–54. This statute prohibits employers from retaliating against employees for taking FMLA leave. 29 U.S.C. § 2615(a)(2). We may assume for the sake of argument that the prohibition would ordinarily apply when an employer adopts an immediate supervisor‘s recommendation to fire an employee for taking FMLA leave. With that assumption, we must decide whether the prohibition would apply when the employee obtains consideration by independent decisionmakers.

We answer no. Retaliation entails a causal link between an employee‘s use of FMLA leave and the firing. That causal link is broken when an independent decisionmaker conducts her own investigation and decides to fire the employee.

I. Ms. Parker‘s supervisor recommends the firing of Ms. Parker, and two independent decisionmakers consider the recommendation.

Ms. Parker fielded calls for United, booking flight reservations. Ms. Parker took FMLA leave because she had a vision disorder and her father had cancer. About five months after approving the leave, Ms. Parker‘s supervisor suspected that Ms. Parker was avoiding new calls by telling customers that she would get additional information, putting the customers on hold, and chatting with coworkers about personal matters while the customers waited. The supervisor characterized Ms. Parker‘s conduct as “call avoidance.”

This suspicion led to a meeting between the supervisor, Ms. Parker, and a union representative. At the meeting, the supervisоr played recordings of three calls between Ms. Parker and customers:

  1. On the first call, Ms. Parker had talked to a customer for about 4 minutes. But she stayed on the line for another 54 minutes. Ms. Parker admitted that she had “failed to disconnect the call when saying goodbye” or “watch the time,” which “resulted in a hung call for a huge amount of time.” Appellant‘s App‘x vol. 2, at 383, 390.
  2. On the second call, Ms. Parker had put a customer on hold for 15 minutes. Right after putting the customer on hold, Ms. Parker asked another supervisor for help. After getting the help, Ms. Parker and the other supervisor chatted about personal matters for over 18 minutes while the customer stayed on hold. According to another supervisor, Ms. Parker hung up on the customer. Ms. Parker denied hanging up on the customer. But she acknowledged and “regrett[ed] leaving the customer on hold for a LENGHLY [sic] amount of time and the call dropped/disconnected while [she] vented [her] home and work frustrations.” Id.
  3. On the third call, Ms. Parker had put a customer on hold for over 20 minutes and then hung up. Id. Ms. Parker explained that (1) her computer had locked up and (2) she had “spent a long time away from the monitor trying to regroup [her] emotions.” Id. at 390. When she returned to the call, she mistakenly hung up on the customer. Id.

Following this meeting, United suspended Ms. Parker while investigating her performance. During this investigation, the supervisor reviewed more of Ms. Parker‘s phone calls with customers and recommended that United fire Ms. Parker.

United‘s policies prohibited the supervisor from firing Ms. Parker. Under these policies, United had to select a manager to conduct a meeting and to allow participation by Ms. Parker, her supervisor, and a union representative. All of them could present arguments and evidence, and the manager would decide whether to fire Ms. Parker.

United applied this policy, selecting a manager to conduct the meeting. In attendance with her were Ms. Parker, the supervisor, and a union representative. The supervisor played recordings of the three calls and presented written summaries of other calls. The supervisor argued that the other calls had violated United‘s policies by unnecessarily putting customers on lengthy holds while chatting with other employees about personal matters.

Ms. Parker‘s union representative challenged the supervisor‘s account about two of Ms. Parker‘s calls. The union representative contended that

  • the customer had ended one of the calls and
  • Ms. Parker had to end the other call because of computer problems.

But the union representative acknowledged a decline in Ms. Parker‘s work performance based on her circumstances: She suffered from a vision disorder and had been taking care of her terminally ill father. Given the circumstances and Ms. Parker‘s long work history, the union representative asked United to apply its progressive discipline policy rather than to fire her. The manager sided with the supervisor, agreeing with her recommendation to fire Ms. Parker for serious policy violations.

United‘s policy allowed Ms. Parker to appeal the firing by submitting a grievance. If she were to submit a grievance, another manager would conduct the appeal through a conference call. In the conference call, the fired employee and a union representative could participate and present further arguments and evidence.

Ms. Parker invoked this procedure by submitting a grievance. She declined to participate, relying on her union representative. The union representative admitted in the conference call that Ms. Parker had “no excuse for the demonstrated behavior of call avoidance except for being under extreme mental duress.” Appellant‘s App‘x vol. 2, at 411. With this admission, the union representative asked United to give Ms. Parker another chance. The senior manager declined and concluded that United hadn‘t acted improperly in firing Ms. Parker.

II. Ms. Parker bore the burden to show pretext.

For a prima facie case, Ms. Parker needed to show that (1) she had taken leave authorized by the FMLA, (2) United had taken a materially adverse action, and (3) a causal connection had existed between Ms. Parker‘s FMLA leave and United‘s decision to fire her. See Metzler v. Fed. Home Loan Bank of Topeka, 464 F.3d 1164, 1170 (10th Cir. 2006).

United doesn‘t question the existence of a рrima facie case. So United needed to present “a legitimate non-discriminatory reason for the adverse employment action.” Litzsinger v. Adams Cnty. Coroner‘s Off., 225 F.4th 1280, 1287 (10th Cir. 2022) (quoting Smothers v. Solvay Chems., Inc., 740 F.3d 530, 538 (10th Cir. 2014)). United presented a legitimate reason: call avoidance. So Ms. Parker needed to show that this reason had been pretextual. Id. She could show pretext through evidence “that a discriminatory reason [had] more likely motivated [United] or that [its] proffered explanation [had been] unworthy of credence.” Zamora v. Elite Logistics, Inc., 478 F.3d 1160, 1166 (10th Cir. 2007) (quoting Stinnett v. Safeway, Inc., 337 F.3d 1213, 1218 (10th Cir. 2003)).

The district court granted summary judgment to United based on Ms. Parker‘s failure to show pretext. The court reasoned that United had fired Ms. Parker after hearing her side.

III. We conduct de novo review based on the summary-judgment standard.

We engage in de novo review of the district court‘s summary-judgment ruling, applying the same standard that applied in district court. SEC v. GenAudio Inc., 32 F.4th 902, 920 (10th Cir. 2022). Under this standard, the district court must view the evidence and draw all justifiable inferences favorably to Ms. Parker. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 255 (1986). Viewing the evidence and drawing reasonable inferences favorably to Ms. Parker, the district court could grant summary judgment to United only without a “genuine dispute as to any material fact” and United‘s showing of an entitlement “to judgment as a matter of law.” Fed. R. Civ. P. 56(a).

IV. Ms. Parker argues that the district court should have imputed the supervisor‘s retaliatory motive to United.

Ms. Parker argues that the district court erred in

  • relying on the manager‘s independence and
  • disregarding the supervisor‘s retaliatory motive.

For those arguments, Ms. Parker relies on the cat‘s paw theory. That theory imputes a supervisor‘s motive to an employer if the motive influenced the employer‘s decision. See Singh v. Cordle, 936 F.3d 1022, 1038-39 (10th Cir. 2019) (discussing the cat‘s paw theory).

The district court rejected Ms. Parker‘s reliance on the cat‘s paw theory, relying on English v. Colorado Department of Corrections, 248 F.3d 1002 (10th Cir. 2001). In English, the employer allowed the employee to contest findings by an investigator who was allegedly biased. Despite this opportunity, the employee declinеd to respond. We observed that “a plaintiff [could not] claim that a firing authority [had] relied uncritically upon a subordinate‘s prejudiced recommendation where the plaintiff had an opportunity to respond to and rebut the evidence.” Id. at 1011.

Given this observation in English, the district court reasoned that Ms. Parker could have presented evidence and arguments to rebut the recommendation of an allegedly biased supervisor. So in the court‘s view, the alleged bias could not be imputed to the employer.

Ms. Parker argues that the district court erred by skipping over a later Supreme Court opinion: Staub v. Proctor Hospital, 562 U.S. 411 (2011). Staub involved a claim of employment discrimination under another federal statute ‍​​‌​​​​‌​‌​‌‌‌​​​‌​​​​​‌​​​‌‌​‌​‌​‌​​​‌‌‌‌‌‌‌‌‌‌‍(the Uniformed Services Employment and Reemployment Rights Act, 38 U.S.C. § 4311(a)). There the Court analyzed the claim based on proximate cause: “[I]f a supervisor performs an act motivated by . . . [illegal] animus that is intended by the supervisor to cause an adverse employment action, and if that act is a proximate cause of the ultimate employment action, then the employer is liable . . . .” Staub, 562 U.S. at 422 (emphasis in original). We may assume for the sake of argument that this standard applies to FMLA claims. See Marshall v. Rawlings Co., 854 F.3d 368, 378 (6th Cir. 2017) (“The rationale for the cat‘s paw theory applies equally to FMLA retaliation claims . . . .“); Marez v. St.-Gobain Containers Inc., 688 F.3d 958 (8th Cir. 2012) (applying the cat‘s paw theory to an FMLA claim).

In Staub, the Supreme Court concluded that if an employer had conducted an independent investigation and rejected an employee‘s allegations of illegal animus by a supervisor, the employer could still incur liability under a cat‘s paw theory. 562 U.S. at 421. Despite the employer‘s investigation, “the supervisor‘s biased report may remain a causal factor if the independent investigation takes it into account without determining thаt the adverse action was, apart from the supervisor‘s recommendation, entirely justified.” Id.

Applying Staub, Ms. Parker maintains that the district court erroneously relied on our earlier analysis in English. She points to our statement in English that “[a] plaintiff cannot claim that a firing authority relied uncritically upon a subordinate‘s prejudiced recommendation where the plaintiff had an opportunity to respond and rebut the evidence supporting the recommendation.” Appellant‘s Opening Br. 38-39 (quoting English v. Colo. Dep‘t of Corr., 248 F.3d 1002, 1011 (10th Cir. 2001)). After Staub, she argues, the opportunity to rebut a biased supervisor‘s recommendation does not foreclose a cat‘s paw theory. Staub, 562 U.S. at 421. And here, she asserts, the supervisor‘s bias led to the firing.

V. Ms. Parker did not invite the alleged error.

United argues that Ms. Parker invited any possible error by arguing in district court that English applied. We disagree.

The “invited error doctrine” prevents a party from arguing that the district court erred “in adopting a proposition that the party had urged the distriсt court to adopt.” United States v. DeBerry, 430 F.3d 1294, 1302 (10th Cir. 2005). The doctrine is based on reliance interests. Id. “Having induced the court to rely on a particular erroneous proposition of law or fact, a party may not at a later stage use the error to set aside the immediate consequences of the error.” United States v. Morrison, 771 F.3d 687, 694 (10th Cir. 2014) (quoting United States v. DeBerry, 430 F.3d 1294, 1302 (10th Cir. 2005)).

We have applied the invited error doctrine when a party

In arguing that Ms. Parker invited any error, United points to Ms. Parker‘s reliance on English in her district court briefing. There she had cited English for the proposition that she needed “to show that the decision maker [had] followed the biased recommendation of a subordinate without independently investigating the complaint.” Appellant‘s App‘x vol. 2, at 463. The district court cited English for the same point. Appellant‘s App‘x vol. 7, at 1398.1 Though Ms. Parker argues that the district court shouldn‘t have rеlied on English, she didn‘t invite the district court to reject the argument that she‘s now making.

In district court, Ms. Parker argued that the manager‘s investigation hadn‘t been independent: “[Ms. Parker‘s immediate supervisor] [had] relentlessly recommended and pursued termination and then [the manager] concurred in that decision without conducting her own separate investigation.” Appellant‘s App‘x vol. 6, at 1049–50. On appeal, Ms. Parker argues again that the manager improperly relied on the supervisor despite her bias. See Appellant‘s Opening Br. at 39-42 (arguing that “[the manager] was influenced by [the supervisor‘s] biased recommendation to terminate [Ms. Parker]“).

The district court reasoned that the manager was independent because she had allowed Ms. Parker to present evidence and arguments. Ms. Parker did not invite this reasoning even though she had cited English. So Ms. Parker did not invite the error that she now alleges.

VI. The evidence does not support Ms. Parker‘s cat‘s paw theory.

Ms. Parker argues that her use of FMLA leave had sparked retaliation from her supervisor. For the sake of argument, we can assume that Ms. Parker is right. With that assumption, we‘d need to decide whether United‘s procedures had broken the causal chain between the supervisor‘s retaliatory motive and the firing. In our view, United broke the causal chain by directing other managers to independently investigate and decide whether to adopt the supervisor‘s recommendation. See Singh v. Cordle, 936 F.3d 1022, 1038 (10th Cir. 2019) (“One way an employer can ‘break the causal chain’ between the subordinate‘s biased behavior and the adverse employment action is for another person . . . higher up in the decision-making process to independently investigate the grounds for dismissal.“).

A. Ms. Parker‘s opportunity to respond to the supervisor‘s evidence does not defeat her cat‘s paw theory.

United argues that it broke the causal chain by relying on independent decisionmakers

  • to investigate and decide whether to adopt the supervisor‘s recommendation and
  • to give fresh consideration and decide whether to reverse the decision to fire Ms. Parker.

To challenge these arguments, Ms. Parker relies on a cat‘s paw theory. But this theory doesn‘t apply when independent decisionmakers “conduct their own investigations without relying on biased subordinates.” Ward v. Jewell, 772 F.3d 1199, 1205 (10th Cir. 2014).

Ms. Parker contends that the district court erred in declining to rely on the cat‘s paw theory. For this contention, she points to the court‘s observations that (1) she “had the opportunity to present any information she chose” and (2) “the manager did in fact hear presentations from both parties.” Appellant‘s App‘x vol. 7, at 1398. Ms. Parker responds that her opportunity to present arguments would not alone prevent liability.

We agree with Ms. Parker. The inquiry involves the independence of the employer‘s investigation, not the employee‘s opportunity to respond. Staub v. Proctor Hosp., 562 U.S. 411, 421 (2011); see Lobato v. N.M. Env‘t Dep‘t, 733 F.3d 1283, 1296 (10th Cir. 2013) (rejecting an employee‘s cat‘s paw theory because “there is no indication that [the decisionmakers] ultimately relied on [the supervisor‘s] version of the facts“).

But we need not remand for the district court to apply the right test. Because our review is de novo, we can apply the right test to the undisputed evidence. See Knitter v. Corvias Mil. Living, LLC, 758 F.3d 1214, 1227 n.9 (10th Cir. 2014) (“[B]ecause our standard of review is de novo, we are free to apply the proper test here, and we may affirm on any ground supported by the record.“).

B. The first manager relied on her own investigation in deciding to fire Ms. Parker.

Ms. Parker contends that she presented evidеnce of the manager‘s reliance on the supervisor. In support, Ms. Parker states that the manager

  • did not review most of the calls that the supervisor had criticized,
  • credited the supervisor‘s statements about some issues, and
  • deferred to the supervisor in refusing to apply United‘s progressive discipline policy.

We reject these arguments.

1. Review of the Other Calls

The manager‘s alleged failure to review most of the calls does not show a retaliatory motive. She limited her inquiry based on what Ms. Parker‘s union representative had said.

In the meeting, Ms. Parker‘s supervisor had complained about the handling of multiple calls from customers, characterizing Ms. Parker‘s conduct as a pattern of call avoidance. Ms. Parker‘s union representative responded that she had listened to

  • the 3 calls that the supervisor had relied on and
  • other calls from the same 5-week period.

The union representative added that

  • there had been “enough to see the pattern” and
  • the manager didn‘t need to listen to the other ‍​​‌​​​​‌​‌​‌‌‌​​​‌​​​​​‌​​​‌‌​‌​‌​‌​​​‌‌‌‌‌‌‌‌‌‌‍calls that the union representative had heard.

Appellant‘s App‘x vol. 2, at 396.

The other calls showed that Ms. Parker had

  • put a customer on hold for 8 minutes and 40 seconds with no activity,
  • put a customer on hold for 21 minutes and 50 seconds,
  • put a customer on hold for 6 minutes and 10 seconds with no activity,
  • taken 25 minutes to complete a customer call while spending only 2-3 minutes working,
  • read trade emails for 2-3 minutes after a call,
  • violated company policy in assigning a seat to a caller‘s wife,
  • put a customer on hold for 17 minutes and 30 seconds,
  • violated company policy by overbooking and putting a customer on hold for 10 minutes and 30 seconds,
  • kept a customer waiting for 21 minutes and 50 seconds to get a seat assignment,
  • kept a customer on hold for 12 minutes and 10 seconds, and
  • kept a customer on hold for 38 minutes and 10 seconds.

Id. at 380-83. At the meeting, Ms. Parker‘s union representative did not dispute the supervisor‘s characterization of these calls. Id. at 396.2

Given the union representative‘s admission of a pattern from the calls that she‘d heard, the manager‘s decision to play only 3 of the calls didn‘t suggest pretext.

2. Crediting the Supervisor‘s Statements

Ms. Parker also maintains that the manager credited 3 of the supervisor‘s statements even when they were obviously incorrect:

  1. The manager said that Ms. Parker had hung up on a customer during a call when the customer had been disconnected for another reason.
  2. The manager didn‘t acknowledge that another call had ended prematurely because of computer problems rather than Ms. Parker‘s neglect.
  3. The manager erroneously found that the supervisor hadn‘t known of Ms. Parker‘s FMLA leave.

The manager‘s assessment of these statements does not show improper reliance on the supervisor.

For the first call, Ms. Parker acknowledged that she had left the customer on hold for a long time “while [she] vented [her] work and home frustrations” with a coworker. Id. at 390. The manager did discount Ms. Parker‘s denial that she‘d hung up on the caller. But Ms. Parker did not deny that she had been at fault in leaving the customer on hold while chatting with a coworker about personal matters.

For the other call, Ms. Parker didn‘t show influence from the supervisor‘s retaliatory motive. The manager knew what had happened because she sampled key points throughout the call. And there‘s no evidence that Ms. Parker or her union representative had said anything in the meeting about a computer problem on this call. See id. at 396.

As to Ms. Parker‘s FMLA leave, the manager said only that the supervisor had denied awareness of Ms. Parker‘s medical condition. Id. at 403. The summary of the meeting supports the manager‘s account. See id. at 397 (reporting the supervisor‘s statement that “[she] was not aware of [Ms. Parker] using much FMLA [leave] or the condition she had“). And the manager‘s statement does not suggest a failure to investigate Ms. Parker‘s work performance. The manager made her findings based on undisputed evidence of deficiencies in Ms. Parker‘s work.

3. Policy of Progressive Discipline

The manager‘s refusal to impose progressive discipline also does not show influence from the supervisor. The manager considered the union representative‘s request, but relied on United‘s authority to forgo progressive discipline for an egregious offense. Id. at 409.

Ms. Parker argues that her offenses weren‘t egregious, pointing to a United employee‘s testimony identifying theft or violence as egregious offenses. But the employee‘s testimony did not suggest that these were the only offenses that United considered egregious. We thus reject Ms. Parker‘s reliance on the policy of progressive discipline. See Lobato v. N.M. Env‘t Dep‘t, 733 F.3d 1283, 1291 (10th Cir. 2013) (concluding that when ““progressive discipline [is] entirely discretionary,‘. . . the failure to implement progressive discipline is not evidence of pretext” (quoting Timmerman v. U.S. Bank, N.A., 483 F.3d 1106, 1120 (10th Cir. 2007))).

C. United‘s appellate procedure would have broken the causal chain even if the manager‘s earlier decision hadn‘t.

Even if the manager‘s decision had been tainted, Ms. Parker did not stop there. She appealed by filing a grievance, triggering a new opportunity to contest the firing before another manager. Ms. Parker declined to participate, relying instead on her union representative. That representative didn‘t question the earlier

  • “call avoidance and a lapse in good judgment,”
  • finding of “egregious behavior” resulting “in significant customer disservice,” or
  • assessments of particular calls.

Appellant‘s App‘x vol. 2, at 409, 411-12.

United sought summary judgment based in part on the senior manager‘s decision. In moving for summary judgment, United insisted that Ms. Parker had lacked any evidence of the senior manager‘s bias.3 See Appellant‘s App‘x vol. 1, at 47. Ms. Parker responded that “she [had] identif[ied] ample evidence of pretext.” See id. vol. 3, at 443. But she cited no evidence of the senior manager‘s bias. See id. at 444–59 (discussing evidence questioning the supervisor‘s motives but not the mоtives of the senior manager).

In her reply brief, Ms. Parker argues that

  • the senior manager decided the appeal more than two months after the firing and
  • the senior manager relied on the first manager‘s tainted findings.

We reject these arguments. Though Ms. Parker had already been fired, she admitted that the grievance could have resulted in reinstatement. Oral Arg. at 5:26–5:56. And we‘ve held that the causal chain is broken when an independent decisionmaker reviews the firing after it‘d taken place. Singh v. Cordle, 936 F.3d 1022, 1039 (10th Cir. 2019) (“The causal chain can even be broken by an independent review that takes place after the adverse action.“); see Thomas v. Berry Plastics Corp., 803 F.3d 510, 517 (10th Cir. 2015) (holding that the claimant‘s “virtually immediate post-termination review process—which was designed to identify and unwind termination decisions that violated company practices and policies—sufficiently constrained any retaliatory animus that [the immediate supervisor] may have possessed“).

Ms. Parker points out that the senior manager‘s decision came 85 days after she‘d been fired. We‘ve not yet addressed the significance of a delay between the firing and an appellate decision upholding the firing. When we held that a post-termination review process had broken the chain of causation, the review process had taken place only 2 days after the firing. Thomas, 803 F.3d at 517. But we didn‘t say anything to suggest that a greater delay would have changed the result.

We can assume for the sake of argument that if the delay had prejudiced Ms. Parker, the appeal to United might not have broken the causal chain. Even with this assumption, however, Ms. Parker couldn‘t prevail because she hasn‘t alleged prejudice from the 85-day delay. With no alleged prejudice from the delay, United‘s appellate procedure would have broken the causal chain even if the first manager‘s decision hadn‘t.4

VII. We direct Ms. Parker to file some documents under seal.

Ms. Parker moves to file certain documents under seal. She‘d attached these documents when responding to United‘s motion for summary judgment. The district court allowed Ms. Parker to file the documents under seal. Ms. Parker makes the motion based on United‘s preference, not

her own. Responding to the motion, United urges the continued sealing ‍​​‌​​​​‌​‌​‌‌‌​​​‌​​​​​‌​​​‌‌​‌​‌​‌​​​‌‌‌‌‌‌‌‌‌‌‍of Ms. Parker‘s Exhibits I, J, Z, AA, BB, CC, EE, and FF.5

The public enjoys a common law right of access to judicial records. JetAway Aviation, LLC v. Bd. of Cnty. Comm‘rs, 754 F.3d 824, 826 (10th Cir. 2014). But the right is not absolute. Id. The Court may order the sealing of documents if competing interests outweigh the public‘s interest. Id. For example, we have allowed sealing of documents reflecting a party‘s finances and business practices. See Sorenson Commc‘ns, Inc. v. FCC, 659 F.3d 1035, 1041 n.4 (10th Cir. 2011).

United contends that eight exhibits contain proprietary information, and Ms. Parker has not rebutted this contention. We thus conclude that United‘s interests support the sealing of these exhibits (L, J, Z, AA, BB, CC, EE, and FF). So we direct Ms. Parker to file these documents under seal.

* * *

We affirm the grant of summary judgment to United, and we grant in part and deny in part Ms. Parker‘s motion for leave to file documents under seal.

Parker v. United Airlines, Inc., No. 21-4093

HOLMES, J., Concurring.

With the exception of Part VII of the principal opinion—which I join in full—I respectfully concur in the judgment. Like the principal opinion, I conclude that Ms. Parker‘s appellate challenge—brought under the Family Medical Leave Act (FMLA), 29 U.S.C. §§ 2601–2654—fails on the merits because her showing of retaliation under a cat‘s paw theory is inadequate. Yet I reach that outcome by a path that I respectfully assert is more judicially modest and thus more appropriate—viz. by a path that entails less parsing of the record in order to opinе on matters that ultimately are not dispositive. For purposes of considering Ms. Parker‘s cat‘s paw theory, I make two analytical assumptions. First, unlike Part V of the principal opinion, I only assume arguendo that Ms. Parker did not invite the district court to err through her invocation of our decision in English v. Colorado Department of Corrections, 248 F.3d 1002, 1011 (10th Cir. 2001). Second, on the merits, I assume without deciding that Ms. Parker‘s supervisor acted with prohibited retaliatory animus.

Furthermore, though I see no need to opine on whether the Supreme Court‘s cat‘s paw holding in Staub v. Proctor Hospital, 562 U.S. 411 (2011)—which involved a different employment statute than the one at issue here—applies in all material respects to the resolution of Ms. Parker‘s cat‘s paw arguments, like our prior cases, I have no difficulty in determining that, as refined in Staub, the “underlying principles of agency upon which subordinate bias theories are based” apply with full force here. Simmons v. Sykes Enters., Inc., 647 F.3d 943, 949 (10th Cir. 2011) (conсluding that “the underlying principles of agency” discussed in Staub “apply equally” to claims under the Age Discrimination in Employment Act of 1967 (ADEA), 29 U.S.C. § 621 et seq., notwithstanding important differences between the ADEA and the statute Staub addressed); see Singh v. Cordle, 936 F.3d 1022, 1038 (10th Cir. 2019) (interpreting the import of Staub and applying that case in the Title VII context); Lobato v. N.M. Env‘t Dep‘t, 733 F.3d 1283, 1294–95 (10th Cir. 2013) (same). Then, applying those agency principles, I would hold that Ms. Parker cannot prevail under a cat‘s paw theory because she failed to establish that the final appellate reviewer of her termination relied uncritically on the assumedly biased supervisor‘s recommendation. Accordingly, with the one exception previously noted, I respectfully concur in the judgment of the principal opinion.

I. Background Legal Principles

I briefly elaborate on the principles developed in our precedents following Staub that are relevant to the resolution of Ms. Parker‘s challenge based on a cat‘s paw theory.

We have interpreted Staub‘s import in the context of statutes other than the one at issue in Staub. In Lobato, for example, we addressed claims alleging, inter alia, employment discrimination based on race and subsequent retaliation in violation of Title VII of the Civil Rights Act, 42 U.S.C. §§ 2000e-2(a)(1), 2000e-3(a). See 733 F.3d at 1294–95. There, a human resources employee at the plaintiff‘s employer had investigated both the plaintiff‘s claims that his supervisor discriminated against him, as well as the supervisor‘s accusations that the plaintiff had lied on his résumé and on a reimbursement request. Id. at 1286–87, 1291–92, 1296. The employee sent a report to management summarizing her investigation. Id. at 1287. Shortly thereafter, the plaintiff was fired. Id. at 1287–88. The decisionmaker‘s stated reasons for termination were, inter alia, that the plaintiff had lied on his employment application and on a reimbursement request. Id. at 1288. In his Title VII claim, the plaintiff invoked the cat‘s paw theory, claiming that his supervisor impermissibly influenced the decisionmaker‘s termination decision. Id. at 1294.

We held that the plaintiff failed to establish liability under a cat‘s paw theory. Id. at 1296. Under Staub, we explained that “a ‘necessary’ element to a subordinate bias claim is the deсisionmaker‘s uncritical ‘reli[ance]’ on facts provided by a biased supervisor.” Id. at 1294 (alteration in original) (quoting Staub, 562 U.S. at 421). “If there is no such reliance—that is to say, if the employer independently verifies the facts and does not rely on the biased source—then there is no subordinate bias liability.” Id. We concluded that the decisionmaker in Lobato did not rely uncritically on the supervisor‘s biased recommendation. See id. at 1294, 1296. Rather, the decisionmaker conducted its own investigation into the plaintiff‘s conduct and determined independently that the plaintiff had falsified his résumé and a reimbursement request, warranting the adverse action—thereby breaking the causal chain and negating the plaintiff‘s cat‘s paw theory. See id. at 1296.

We also addressed cat‘s paw liability under Title VII in Singh v. Cordle, 936 F.3d 1022 (10th Cir. 2019). In Singh, a university department dean recommended against renewing a non-tenured professor‘s contract due to the professor‘s allegedly deficient performance and lack of collegiality. Id. at 1031. The university provost reviewed the recommendation, and, although he disagreed regarding the professor‘s performance, he determined that nonrenewal was warranted based on the professor‘s lack of collegiality alone. Id. Following notice from the provost that his term would end at the close of the following academic year, the professor filed a petition before a grievance committee. Id. at 1031-32. He claimed that the dean recommended nonrenewal based on discriminatory animus and that the provost unjustifiably followed the dean‘s recommendation. Id. at 1032. After hearing the professor‘s case, the committee found that the nonrenewal decision was not based on animus and recommended nonrenewal to the university president, who affirmed the decision. Id. In his Title VII claim, the professor invoked the cat‘s paw thеory, alleging that the dean‘s animus proximately caused the nonrenewal determination. Id. at 1038.

We again held that the plaintiff failed to establish liability under the cat‘s paw theory. See id. As we explained, “[o]ne way an employer can ‘break the causal chain’ between the subordinate‘s biased behavior and the adverse employment action is for another person or committee higher up in the decision-making process to independently investigate the grounds for dismissal.” Id. (quoting Thomas v. Berry Plastics Corp., 803 F.3d 510, 516 (10th Cir. 2015)). “[I]f the employer‘s investigation results in an adverse action for reasons unrelated to the [biased] supervisor‘s original biased action, . . . then the employer will not be liable.” Id. at 1038–39 (alteration in original) (quoting Staub, 562 U.S. at 421).

Applying these principles, we concluded that subsequent levels of independent review “broke ‍​​‌​​​​‌​‌​‌‌‌​​​‌​​​​​‌​​​‌‌​‌​‌​‌​​​‌‌‌‌‌‌‌‌‌‌‍the causal chain” between the dean‘s recommendation and the nonrenewal determination. Id. at 1039–41. We concluded that the provost did not rely uncritically on the dean‘s recommendation, as he did not base his decision on alleged performance deficiencies that the dean had reported. Id. at 1039. And we concluded that “the grievance committee conclusively broke the causal chain between [the dean‘s] alleged animus and [the] [p]laintiff‘s nonrenewal” by reviewing evidence the plaintiff had submitted and independently concluding that “nonrenewal was justified.” Id. at 1039.

Guided by Staub, our decisions in Lobato and Singh therefore establish that a plaintiff challenging an adverse action under the cat‘s paw theory must demonstrate, at a minimum, that the entity conducting the final layer of review relied “‘uncritical[ly]’ . . . on facts provided by a biased subordinate.” See Singh, 936 F.3d at 1038 (quoting Lobato, 733 F.3d at 1294). Showing “uncritical reli[ance]” at the final layer of review is an essential element of cat‘s pаw liability that the plaintiff bears the burden to establish. See Lobato, 733 F.3d at 1294 (alteration in original) (quoting Staub, 562 U.S. at 421) (explaining that “the decisionmaker‘s” uncritical reliance is “a ‘necessary’ element to a subordinate bias claim” (emphasis added) (quoting Staub, 562 U.S. at 421)); see also Singh, 936 F.3d at 1039 (explaining that the “[p]laintiff had to show . . . a causal chain between [the supervisor‘s] allegedly biased input and the decision not to reappoint [the] [p]laintiff” (emphasis added)).

Notably, even “an independent review that takes place after the adverse action” can “break the causal chain” if the reviewer is authorized to reverse the decision. See Singh, 936 F.3d at 1039 (emphasis added) (citing Thomas, 803 F.3d at 517). Thus, in Singh, the grievance committee‘s independent determination “conclusively broke the causal chain” even though it occurred after the provost had formally notified the plaintiff of his upcoming nonrenewal. See id. at 1031–32, 1039 (noting that the provost notified the plaintiff in February 2014 of his decision not to renew the plaintiff‘s appointment and the final layer of review reached a determination, at the earliest, in November 2014). Similarly, in Thomas, a final “independent termination review process” that “was designed to identify and unwind termination decisions” “broke the causal chain between [a supervisor‘s] purported retaliatory animus and [the plaintiff‘s] termination,” even though the employer had “officially terminated” the plaintiff before the final review occurred. See id. at 513, 516–18 (emphasis added).

II. Analysis

Applying the foregoing principles, Ms. Parker‘s arguments regarding the final appellate reviewer, see Aplt.‘s Reply Br. at 25–26, fail to establish United‘s liability under a cat‘s paw theory.

In particular, our precedents flatly contradict Ms. Parker‘s position that the appellate reviewer‘s decision is “irrelevant” merely because it occurred after she was terminated. Id. at 25; see also Aplt.‘s App., Vol. VII, at 1296 (Summ. J. Hr‘g, dated Jan. 11, 2021) (asserting that the appellate review decision “is irrelevant“). So long as a reviewer is authorized to reverse the adverse action, even an independent review that occurs after the adverse action can break the causal chain. See Thomas, 803 F.3d at 517 (explaining that the “post-termination review process,” which “was designed to . . . unwind termination decisions that violated company practices and policies,” broke the causal chain between the biased recommendation and the termination (emphasis added)); see also Singh, 936 F.3d at 1032, 1039 (holding that an independent review, which occurred several months after the plaintiff received notification of nonrenewal, “conclusively broke the causal chain“). Here, Ms. Parker conceded at oral argument that the appellate reviewer could have reversed her termination. Oral Arg. at 5:20–5:55. Ms. Parker is therefore incorrect in asserting that the appellate reviewer‘s decision is legally irrelevant under a cat‘s paw theory.1

Moreover, Ms. Parker‘s argument that the appellate reviewer relied uncritically on her supervisor‘s biased recommendation is unpersuasive. See Aplt.‘s Reply Br. at 26 (noting that the appellate reviewer “uncritically upheld [the assumedly biased supervisor‘s] recommendation that [Ms.] Parker be terminated“). The appellate reviewer explained that she relied not only on information stemming from the initial stage of review of the supervisor‘s termination recommendation—that is, from the Investigative Review Meeting (IRM)—but also on information Ms. Parker‘s union representative presented at the appellate review meeting, including his concessions regarding Ms. Parker‘s misconduсt. See Aplt.‘s App., Vol. II, at 411–12 (Letter from Laurie Ledonne, Sr. Hum. Res. Manager, United Airlines, to Jeannie Parker, Plaintiff-Appellant, dated Feb. 27, 2019) (explaining that the reviewer reached her decision “[a]fter reviewing the facts that were presented at the [IRM] and the information presented at the [appellate review]“). As explained in her letter, the appellate reviewer focused in part on the “three calls cited for call avoidance.” Id. at 411. With respect to these calls, the union representative “stated that he could not negate the call avoidance and a lapse in good judgment on [Ms. Parker‘s] part.” Id. He further conceded that “there was no excuse for the demonstrated behavior of call avoidance except for being under extreme mental duress.” Id. These concessions featurе prominently in the appellate reviewer‘s determination, providing strong evidence that the appellate reviewer “independently verifie[d] the facts” supporting Ms. Parker‘s termination and did not simply rely on information from the supervisor or the IRM. See Lobato, 733 F.3d at 1294.

In attempting to establish uncritical reliance on the supervisor‘s allegedly biased recommendation, Ms. Parker relies on speculation and conjecture and presents only conclusory arguments. That is not enough to carry her burden to establish United‘s liability under a cat‘s paw theory. As with the manager who presided over the IRM, there is no indication that Ms. Parker deposed the United manager who served in the role of appellate reviewer to determine her rationale or the specific materials she considered or other specifics of her investigatiоn. Thus, Ms. Parker is left in the problematic position of resorting to speculation and conjecture, not hard evidence, in attacking the substance and underlying methodology of the appellate reviewer‘s decision.2

More specifically, Ms. Parker offers no evidence to refute the appellate reviewer‘s description of her methodology, which evinces independent review. See Aplt.‘s App., Vol. II, at 411–12 (discussing consideration of information presented at the IRM and during the appellate review, including the union representative‘s presentation and concessions regarding Ms. Parker‘s call avoidance). Without hard evidence to back it up, Ms. Parker‘s argument that the appellate reviewer‘s decisionmaking process in refusing to unwind her termination was infected by impermissible subordinate bias is thus speculаtive and conjectural—and, consequently, unpersuasive. See Aplt.‘s Reply Br. at 25-26 (asserting that the appellate reviewer only listened to three of the calls and questioning factors referenced in the appellate reviewer‘s letter to support her decision).

In sum, Ms. Parker failed to establish that “the supervisor‘s biased report . . . remain[ed] a causal factor” in the appellate reviewer‘s decision to uphold her termination. Staub, 562 U.S. at 421. Therefore, her FMLA retaliation claim—predicated on a cat‘s paw theory—cannot prevail.

Based on the foregoing, with the one exception previously noted, I respectfully concur in the judgment.

Notes

1
United also points to the district court‘s statement that “[a] plaintiff cannot claim that a firing authority relied uncritically upon a subordinate‘s prejudiced recommеndation where the plaintiff had an opportunity to respond to and rebut the evidence supporting the recommendation.” Appellant‘s App‘x vol. 7, at 1398 (quoting English v. Colo. Dep‘t of Corrs., 248 F.3d 1002, 1011 (10th Cir. 2001)). But Ms. Parker didn‘t make this statement in district court. Citing Thomas‘s conclusion that a “virtually immediate post-termination review process” broke the causal chain, 803 F.3d at 517, Ms. Parker further claims that the appellate review is irrelevant due to the amount of time that elapsed following the adverse action. See Aplt.‘s Reply Br. at 25 (explaining that the appellate decision did not occur until 85 days after her termination). However, I agree with the principal opinion‘s assertion that nothing in Thomas “suggest[s] that a greater delay would havе changed the result.” Principal Op. at 19. Our decisions do not address whether the timing of the decision of the final allegedly independent reviewer is a relevant factor in determining whether the decision breaks the causal chain (that is, cannot be said to uncritically rely on the biased subordinate‘s recommendation), and I have no reason to believe that it is. Indeed, in Singh, the final layer of review, which “conclusively broke the causal chain,” occurred months after the provost notified the plaintiff of the nonrenewal determination. See 936 F.3d at 1031-32, 1039. Nevertheless, concerning a related matter, I decline to join any suggestion in the principal opinion that insofar as any delay in the appellate reviewer‘s decision prejudiced Ms. Parker, “the appeal to United might not have broken the causal chain.” Principal Op. аt 19. The principal opinion appropriately does not decide whether prejudice is a relevant factor; instead, it merely assumes that even if prejudice were relevant, Ms. Parker alleges no such prejudice here. See id. at 19–20. But the principal opinion provides no legal foundation for the suggestion that prejudice may be relevant, and I am not aware of any. If the final layer of appellate review is authorized to reverse the completed adverse action suffered by the plaintiff, I do not understand why alleged prejudice to the plaintiff stemming from delay in the appellate reviewer‘s decision would have any relevance to the resolution of the dispositive question of whether the appellate reviewer uncritically relied on the biased subordinate‘s adverse-actiоn recommendation.
2
In her written statement, the union representative acknowledges that she was “not stating that United [was] at fault for [Ms. Parker‘s] actions [or] denying what happened on the calls.” Appellant‘s App‘x vol. 4, at 753. At the summary-judgment hearing before the district court, Ms. Parker engaged in similar speculation and conjecture regarding whether the decision of the manager that presided over the IRM was impermissibly infected with the assumedly prohibited bias of her supervisor. See Aplt‘s. App., Vol. VII, at 1312–15 (Summ. J. Hr‘g, dated Jan. 11, 2021). She implied that because the record was silent regarding the particular documents that the manager relied on both before and after the IRM in making her decision, as well as about other details of her decisionmaking process, there at least was a genuine dispute of material fact regarding whether the manager impermissibly relied on the supervisor‘s biased recommendation. See id. at 1313. This prompted the district court to effectively inquire whether Ms. Parker—the bearer of the burden of persuasion on the cat‘s pаw theory—had sought to gain answers to the some of these questions it identified regarding the manager‘s decisionmaking by deposing the manager. ‍​​‌​​​​‌​‌​‌‌‌​​​‌​​​​​‌​​​‌‌​‌​‌​‌​​​‌‌‌‌‌‌‌‌‌‌‍Notably, Ms. Parker responded in the negative, stating that the decision not to depose the manager “was not a strategic decision, [it] was merely a decision based on dollars and cents” and that deposing the manager “just wasn‘t in [their] litigation war chest.” Id. at 1314–15. There is no indication from Ms. Parker‘s arguments on appeal that she deposed United‘s appellate reviewer either—perhaps based on a similar financial calculation. In any event, as with the manager that presided over the IRM, Ms. Parker bears the burden of showing that the appellate reviewer‘s decision was impermissibly infected by the supervisor‘s assumedly prohibited bias, and she cannot rely on speculation and conjecture to do that.
3
The district court did not address this argument. But we can rely on this argument because United raised it in district court, the parties fully briefed it there, and United reurges the argument on appeal. See Havens v. Johnson, 783 F.3d 776, 782 (10th Cir. 2015).
4
Ms. Parker also argues that her supervisor acted with retaliatory intent. See Appellant‘s Opening Br. at 42-53. Because United showed that it had fired Ms. Parker for reasons unrelated to the supervisor‘s allegedly retaliatory motives, we need not address this evidenсe. See Singh v. Cordle, 936 F.3d 1022, 1039 (10th Cir. 2019) (discussing evidence presented to a grievance committee and stating that this evidence had “conclusively broke[n] the causal chain between [a supervisor‘s] alleged animus and Plaintiff‘s nonrenewal“).
5
The court clerk instructed Ms. Parker to publicly file all of the previously sealed exhibits that United no longer seeks to keep confidential (H, L, M, Q, R, U, V, and W).
Read the detailed case summary