Judith MISKILL, AFGE Local 1923, Petitioners v. SOCIAL SECURITY ADMINISTRATION, Respondent
Judith Miskill challenges an Arbitrator's award sustaining her removal by the Social Security Administration for violations of the Agency's time and attendance policy. Because the Arbitrator's decision is not in accordance with law, we vacate and remand.
Ms. Miskill was employed as an Information Technology Specialist with the Social Security Administration (the Agency) for more than 14 years. On August 16, 2013, Larry Schwab, Ms. Miskill's first-line supervisor, proposed to remove Ms. Miskill for violations of the Agency's time and attendance policy between July 1, 2011 and June 28, 2013. Mr. Schwab charged Ms. Miskill with: 1) Extending Lunch Period and Break Periods; 2) Failure to Accurately Record Arrival Times; 3) Failure to Accurately Record Departure Times; and 4) Receiving Overtime Pay or Compensatory Time Off When Ineligible. On September 25, 2013, Assistant Associate Commissioner Dan Parry sustained all four charges against Ms. Miskill and removed her from Federal Service.
On November 6, 2013, the American Federation of Government Employees (the Union) submitted a grievance on behalf of Ms. Miskill asserting that the removal “violated the just cause, progressive discipline and warning and counseling provisions contained in Article 23, Section 1, the timeliness provisions of Article 23, Section 2 and the requirements of Douglas v. Veterans Admin., 5 MSPR 313 (1981)․” J.A. 302. On December 3, 2013, after Executive Officer Don Henry denied the grievance, the Union invoked arbitration.
In preparation for the arbitration hearing, Ms. Miskill requested the turnstile records and time reports of the eight other individuals within her component at the Division of Network Engineering (DNE) between July 1, 2011 and June 28, 2013. The Agency provided the records to her in August and November 2014. Those records were analyzed by Candace Dayton, a Certified Public Accountant, Certified Product Examiner, and Certified Information Technology professional. J.A. 132. Based on those records, Ms. Dayton concluded that the eight other DNE employees had committed the same or similar violations as Ms. Miskill, yet not one of these eight employees was investigated or charged with misconduct. Specifically, Ms. Dayton found that: (1) six DNE employees had more error minutes than Ms. Miskill for inaccurately recording lunch and break periods; (2) four DNE employees had more error minutes than Ms. Miskill for inaccurately recording their arrival time; (3) four DNE employees had more error minutes than Ms. Miskill for inaccurately recording their exit time; and (4) one DNE employee had more error minutes than Ms. Miskill for inaccurately recording overtime. Ms. Dayton concluded that five employees had more overall error minutes than Ms. Miskill.
Ms. Dayton's analysis was provided to the Agency on June 17, 2015. The Arbitrator held a hearing on the merits on September 2, 2015. The Arbitrator received into evidence Ms. Dayton's analysis showing the time and attendance discrepancies of the eight other DNE employees. Based on this evidence, Ms. Miskill argued that the penalty of removal was too harsh when “eight other comparators, employees working within DNE, also engaged in similar misconduct ․ but had not been disciplined.” J.A. 4. At the end of the hearing, the parties stipulated that the comparator employees were under investigation for potential violations of the Agency's time and attendance policy, but had not yet been charged with any misconduct. J.A. 196–97. On December 19, 2015, the Arbitrator sustained Ms. Miskill's removal after finding that “the eight comparators were [not] similarly situated to [Ms. Miskill] because, based on a stipulation of the parties, the possible disciplinary action regarding these other employees is still pending an investigation.” J.A. 4.
Ms. Miskill appeals. We have jurisdiction under 5 U.S.C. § 7121(f) and § 7703(a).
Federal employees who are also union members may challenge removal either by direct appeal to the Board or through arbitration (with exceptions not applicable here). 5 U.S.C. § 7121(e)(1). We review an arbitrator's decision under the same standard of review that is applied to decisions from the Merit Systems Protection Board. Id. § 7121(f); Johnson v. Dep't. of Veterans Affairs, 625 F.3d 1373, 1376 (Fed. Cir. 2010). Thus we must affirm the decision of the arbitrator unless it is: “(1) arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law; (2) obtained without procedures required by law, rule, or regulation having been followed; or (3) unsupported by substantial evidence.” 5 U.S.C. § 7703(c)(1)–(3). This standard of review “contemplates de novo review of questions of law.” Johnson, 625 F.3d at 1376.
The government argues that the Arbitrator violated the Collective Bargaining Agreement (CBA) by considering Ms. Miskill's evidence regarding the time and attendance violations of the eight DNE employees.
The CBA states:
Only issues identified in the written grievance will be considered by the grievance deciding official. Neither party may consider issues that were not raised at the last step of the grievance process. An arbitrator may only consider issues that were raised at the last step of the grievance process.
J.A. 246 (emphasis added).
Under the government's interpretation, the CBA prohibits the Arbitrator from considering evidence related to a disparate treatment defense unless that specific defense—the “issue” under the CBA—has been presented to the Agency during the grievance process.
“Interpretation of a collective-bargaining agreement is a question of law we review de novo.” Garcia v. Dep't of Homeland Sec., 780 F.3d 1145, 1147 (Fed. Cir. 2015). The Court “begin[s] with the plain language of the agreement” and then “give[s] the words in the agreement their ordinary meaning unless the parties mutually intended and agreed to an alternative meaning.” Id. (citing Harris v. Dep't of Veterans Affairs, 142 F.3d 1463, 1467 (Fed. Cir. 1998)).
Here, we conclude that the CBA does not require that all evidence supporting an “issue” must be raised before the Agency in order for the Arbitrator to properly consider it. Thus, we decline to interpret “issue” as narrowly as the government insists. Ms. Miskill's grievance stated that her removal was, inter alia, not in compliance with the requirements of Douglas v. Veterans Administration, 5 M.S.P.R. 280 (1981). J.A. 302. One of the Douglas factors requires agencies, when determining a penalty, to consider the consistency of the penalty with those imposed upon other employees for the same or similar offences. 5 M.S.P.R. at 305. Accordingly, by arguing that the Agency failed to properly consider the Douglas factors, Ms. Miskill sufficiently raised the issue of disparate treatment. That she did not present evidence supporting the disparate treatment argument until arbitration is of no consequence as the CBA does not impose such a requirement. Therefore, the Arbitrator did not err by considering the comparator employee evidence.
The arbitrator did err, however, in its treatment of that comparator evidence. Specifically, the arbitrator's categorical conclusion that the eight DNE employees could not be comparator employees because they were under investigation is an incorrect statement of the law. Although the fact that a comparator employee is under investigation is a factor to be considered in determining whether that comparator is similarly situated, it is not a complete bar.
“Arbitrators, like the Board, must review de novo the merits of an agency's decision to take adverse action against an employee.” Norris v. Sec. & Exch. Comm'n, 675 F.3d 1349, 1355 (Fed. Cir. 2012). Thus, when reviewing an agency's decision to take adverse action against an employee, the arbitrator must “independently assess[ ] the relevant Douglas factors to determine whether the penalty imposed was reasonable.” Id. at 1357.
The Douglas factor at issue here is the “consistency of the penalty with those imposed upon other employees for the same or similar offences.” 5 M.S.P.R. at 305. To establish disparate penalties, the employee must show “that the charges and the circumstances surrounding the charged behavior are substantially similar,” which includes “proof that the proffered comparator was in the same work unit, with the same supervisor, and was subjected to the same standards governing discipline.” Lewis v. Dep't of Veterans Affairs, 113 M.S.P.R. 657, 660 (2010) (Lewis II). “Where an employee raises an allegation of disparate penalties in comparison to specified employees, the agency must prove a legitimate reason for the difference in treatment by a preponderance of the evidence.” Lewis v. Dep't of Veterans Affairs, 111 M.S.P.R. 388, 391 (2009) (Lewis I). Moreover, “evidence regarding similarly-situated employees who received no discipline after committing similar misconduct would also support [a] disparate penalty claim.” Fearon v. Dep't of Labor, 99 M.S.P.R. 428, 434 (2005).
Here, Ms. Miskill was removed in August 2013 for violations of the time and attendance policy from 2011–2013. At her arbitration hearing, Ms. Miskill presented evidence that her entire department committed the same or similar offenses during the same time period, yet no other employee had been disciplined. The government did not directly confront that evidence but merely argued that, because the eight other employees in the DNE were currently under investigation and had not yet been charged, they could not be considered comparator employees. J.A. 234. The Arbitrator agreed and sustained Ms. Miskill's removal after finding that “the eight comparators were [not] similarly situated to [Ms. Miskill] because ․ the possible disciplinary action regarding these other employees is still pending an investigation.” J.A. 4. As we have noted, this was in error.
A comparator is an employee that “was in the same work unit, with the same supervisor, and was subjected to the same standards governing discipline.” Lewis II, 113 M.S.P.R. at 660. There is no strict requirement that employees subject to the same standards governing discipline must always receive precisely the same penalty. But similarly situated employees must be subject to the same criteria, and differences in penalties must depend on specific factual differences between those employees. That consideration is, by its nature, a case-dependent, highly factual inquiry and not amenable to bright-line rules.
Here, the Arbitrator erred by imposing a categorical rule of exclusion, i.e., that similarly situated employees under investigation could not be comparators. It is true that, in some circumstances, an ongoing investigation may be a legitimate basis to exclude an employee as a comparator. But that depends on the factual circumstances. Here, the facts strongly suggest that it was improper to exclude the eight employees as comparators simply because they were under investigation at the time of the arbitration. After Ms. Miskill's removal in September 2013, the Agency knew or should have known that there were potential problems with the implementation of time and attendance polices in the DNE unit. Yet the Agency waited to begin its investigation of the eight other DNE employees for almost two years and, in fact, did not initiate an investigation until it discovered that Ms. Miskill was alleging disparate treatment.
The time required for the Agency to perform its investigation once begun also suggests that it was improper to exclude the eight employees as comparators based simply on their investigation status. Ms. Miskill's supervisor received an anonymous tip that Ms. Miskill may have violated the agency's time and attendance policies. Less than two months had elapsed from receipt of that tip until completion of an investigation and the issuance of a Notice of Proposed Removal. But the investigation into the other eight employees had not finished by the time the Arbitrator issued his opinion, which was more than six months after Ms. Miskill submitted her evidence regarding the time and attendance violations of the other employees to the Agency in June 2015. This was true, moreover, despite the fact that the Agency had access to Ms. Dayton's detailed analysis of the time records, which provided a starting point for the Agency's own investigation of those employees. Thus, even putting the initial two-year lag time aside, there is no explanation for such an additional delay.
Although the Agency's timing could be read to suggest that it placed—and kept—the comparators under investigation for the sole purpose of evading Ms. Miskill's disparate treatment allegation, we need not ascribe any particular motivation to the Agency or require any proof of intent to conclude that the Arbitrator erred in categorically excluding these eight employees as comparators. The Agency's conduct here is made even more troubling by its concession at oral argument that upon completion of the investigation, the Agency did not discipline six of the DNE employees and counseled the remaining two. Oral Argument at 13:50–14:18 http://oralarguments.cafc.uscourts.gov/default.aspx?fl=2016-1598.mp3. In short, the Agency determined that removal was the appropriate penalty for Ms. Miskill, but no other similar adverse action was warranted for the eight other employees who committed the same or similar violations. This type of factual evidence is why a categorical rule of exclusion based on an employee's investigatory status is improper.
To be clear, we are not creating an opposite categorical rule—that consideration of an ongoing investigation is never relevant to the question of consistency of punishment. Nor do we implicitly hold that punishment of one employee cannot occur until all possible investigations of all potential employees have been completed. Such rules would be unworkable. Rather, when a particular employee under investigation is raised as a comparator, the arbitrator must evaluate that as one factor, among others, to determine the consistency of the agency punishment.
On remand, if the Arbitrator determines that any or all of the eight DNE employees are comparators, then he must determine if the Agency imposed consistent penalties.1 See Douglas, 5 M.S.P.R. at 305; see also Fearon, 99 M.S.P.R. at 434. Although “penalty decisions have generally looked backward to determine if the penalty selected by the agency is consistent with those which have been historically imposed by the agency,” under certain circumstances, the consideration of post-removal evidence is appropriate. Chavez v. Small Bus. Admin., 121 M.S.P.R. 168, 182 n.7 (2014). Given the Agency's conduct in this case, if the Arbitrator determines that any or all of the eight other employees are comparators, the Arbitrator should reopen the record to compare the penalties imposed upon the comparators with Ms. Miskill's penalty of removal.2
Because the Arbitrator erred in the comparator employee analysis, we vacate and remand. On remand, if the Arbitrator determines that any or all of the eight DNE employees are comparators, the Arbitrator is instructed to reopen the record to determine if the penalties imposed upon the comparators were consistent with the penalty received by Ms. Miskill.
VACATED AND REMANDED
Costs to Petitioners.
1. The evidence regarding the eight comparators may be relevant to other Douglas factors, but we leave this question to the Arbitrator to consider in the first instance.
2. This is not the end of the analysis, however. If the Arbitrator determines that the comparators were treated differently, the burden shifts to the Agency to prove a legitimate reason for the differing treatment. See Lewis I, 111 M.S.P.R. at 391.
Hughes, Circuit Judge.