dissenting.
I respectfully dissent from the majority opinion because there is no showing that Brown’s private consensual conduct, improper though it was, had an adverse impact on the efficiency of the service. The Board, and the majority opinion, has stretched the nexus requirement so thin that Brown has become subject to the moral judgments of his employer. As one of our sister circuits has repeatedly pointed out, that is an improper role for the government. See Norton v. Macy, 417 F.2d 1161, 1165 (D.C.Cir.1969) ([T]he notion that it could be an appropriate function for the federal bureaucracy to enforce the majority’s conventional codes of conduct in the private lives of its employees is at war with elementary concepts of liberty, privacy, and diversity.); Doe v. Hampton, 566 F.2d 265, 273 n. 20 (D.C.Cir.1977) (The nexus requirement ... serves to minimize unjustified governmental intrusions into the private activities of federal employees.); Gloster v. General Servs. Admin., 720 F.2d 700, 703 (D.C.Cir.1983) (The nexus requirement is particularly strict in cases involving non-work-related misconduct.).
BACKGROUND
Preliminarily, I wish to clarify two points in' the majority’s presentation of the factual background of this case. First, in my view the misconduct in this case has not been clearly framed. Brown’s relationship with a married woman predated his position at Camp Lejeune and had no causal connection with his job. He was a neighbor of the woman and met her before he started working at Camp Lejeune. There was no evidence in the record that his contacts with the woman occurred at, were facilitated in any way by, or were related to any events or activities sponsored by the MWR Department. Even Colonel Stewart, the official who removed Brown, stated that there was no allegation that Brown used his position to further his misconduct.
Second, the impact of Brown’s misconduct on his job has not been clearly framed. The record reveals no evidence at all of any negative impact. The record indicates that Brown received the highest possible performance appraisal, and a Certificate of Commendation, for the period covering virtually the entire duration of his misconduct. Even Colonel Helland, testifying against Brown at the Board’s eviden-tiary hearing, could not criticize Brown’s support of the Colonel’s unit. Colonel Helland stated that Brown’s performance was adequate and that the Colonel had no complaints.
With that framework in mind, I now address the Board’s finding, which the majority affirms, that Brown’s misconduct somehow had adversely affected the efficiency of the service. As all parties recognize, an employee can only be removed for such cause as will promote the efficiency of the service. 5 U.S.C. § 7513(a) (1994); Crofoot v. United States Gov’t Printing Office, 761 F.2d 661, 664 (Fed.Cir.1985); see also Pararas-Carayannis v. Department of Commerce, 9 F.3d 955, 957 (Fed.Cir.1993). This requires that there be a nexus between the employee’s misconduct and the efficiency of the service. Sanders v. United States Postal Serv., 801 F.2d 1328, 1332 (Fed.Cir.1986); see also Pararas-Carayannis, 9 F.3d at 957. The Board *1365summarized the ways in which such a nexus can be shown as consisting of: (1) existing egregious circumstances that give rise to a rebuttable presumption of nexus based on the nature and gravity of the misconduct; (2) preponderant evidence showing that the misconduct affects the employee’s or his co-worker’s job performance; (3) preponderant evidence showing that the misconduct affects management’s trust and confidence in the employee’s job performance; and (4) preponderant evidence showing that the misconduct interfered with or adversely affected the agency’s mission. See Brown v. Department of Navy, No. DC-0752-97-1018-1-1, slip op. at 7 (M.S.P.B. Jan.2, 1998) (the AJ’s opinion) (Board Opinion). The Board did not find that Brown’s misconduct had affected his job performance. The Board relied only on the last two of these four possible avenues to a nexus, and I address below the lack of substantial evidence to support each.
ANALYSIS
1. Trust and Confidence
The AJ found that the agency has ... proven nexus by showing by preponderant evidence that [Brown’s] misconduct affected management’s trust and confidence in his job performance. Board Opinion, slip op. at 11. As the AJ correctly pointed out, and our own case law requires, management’s loss of trust and confidence must be directly related to the employee’s job performance. See Pararas-Carayannis, 9 F.3d at 956-58 (Employee used both government time and money to launder prostitution proceeds.); Sanders, 801 F.2d at 1330 (Employee used agency premises and on-duty time, on two occasions, to engage in the sale of cocaine.); Brown v. Department of Transp., 735 F.2d 543, 545, 548 (Fed.Cir.1984) (Air traffic controller supervisor spoke at an off-duty rally of striking air traffic controllers, thus implicating his own loyalty to management.).
However, all of the evidence adduced at the Board’s hearing was conjectural, speculative, or unrelated to Brown’s actual job responsibilities. Bear in mind that Brown’s principal role was to organize activities for Marines in the units he supported; Brown did not occupy a position, such as a counselor or therapist, that would require that the Marines he supported repose confidence and trust in him. Colonel Helland, the commanding officer of an expeditionary marine unit, stated that Brown’s job was to be a facilitator, making it easier to obtain resources and facilities, and Mr. Mayberry, a family readiness officer, described Brown’s job as a party and recreation event planner.
Representative of this evidence is the testimony of Colonel Stewart, Assistant Chief of Staff of the MWR Department, commonly known as the head of the MWR. Colonel Stewart speculated in his testimony, emphasis added, that Brown will' not be efficient at- all and that he did not believe that people will turn to him for [ ] support. But Colonel Stewart did not connect this perceived blackballing of Brown by others to Brown’s actual job responsibilities of planning events. Mr. Blare, Director of the Operations Division of the MWR Department and Operations Officer of the MWR Department, also testified in a conclusory fashion. He stated that anything short of removal would not solve the problem, but failed to identify what-aspect of Brown’s job performance was in question. Mrs. Leferbre, a key volunteer coordinator and the wife of a commanding officer, stated that the wives of deployed Marines may be vulnerable. She also expressed her opinion that Brown’s indiscretion would threaten the reputation of her program, the key volunteer program. However, she did not provide any evidence that Brown’s job performance had been impaired or how it would be impaired because of his indiscretion. Colonel Helland testified to the generality that family problems can have considerable impact on a unit’s perspective, but failed to provide any evidence that Brown’s job performance had been compromised or might be im*1366paired because of his particular misconduct. Mr. Mayberry opined generally that the Marines Brown served needed to be able to trust Brown, but Mr. Mayberry offered no evidence that any Marines had lost trust or that Brown’s actual job performance would be impaired because of Brown’s indiscretion. The testimony identified by the majority opinion fares no better. See ante, at 1359-60 (majority opinion) (repeatedly using the conditional would). In my opinion, there is not substantial evidence of record showing that management’s loss of confidence is related to Brown’s actual job performance.
Rather than requiring a connection to an employee’s job performance, the majority opinion risks being viewed as sanctioning the removal of an employee simply because of a difference in values between that employee and his superiors. Colonel Stewart testified to this effect: I can’t imagine a marine commander wanting to deal with somebody who has ... a track record of this kind of thing. Brown’s superiors, including his immediate boss and two Marine Colonels, essentially testified that they disapproved of Brown’s behavior. If this evidence is enough to warrant removal, then no employee is safe from the threat of removal. Mr. Blare testified that it would have been different if Brown had held some other position, specifically, a janitor, in the MWR Department. However, Mr. Blare fails to provide any connection between Brown’s personal indiscretion and his actual job performance, without which a charge of loss of trust and confidence in that performance rings hollow. Again, the only evidence of Brown’s performance is positive.
The situation might be different if the record contained testimony from those Marines and their families who were not in a position of authority over Brown and whom Brown was charged with supporting. See Brown v. Department of Navy, No. DC-0752-97-1018-I-1, slip op. at 7, 83 M.S.P.R. 230 (M.S.P.B. Aug.10, 1999) (Sla-vet, Vice Chair, dissenting) (raising the same issue). The AJ recognized the utility of such evidence, asserting that Brown had lost his credibility, integrity and ability to function with [the Marines and their families] as a result of his misconduct, and that this directly interfered with his ability to accomplish his assigned mission. Board Opinion, slip op. at 9. However, this finding is totally unsupported by any testimony from those Marines and their families. The Board and the majority opinions include only testimony, such as that discussed above, that is premised on the surmised disapproval of the Marines, or the families of the Marines, in the units supported by Brown.
Testimony from these supposedly disapproving individuals would have served to establish the necessary connection with Brown’s job performance. These individuals might have testified that they no longer felt comfortable attending the functions that Brown organized or having Brown attend those functions as part of his assigned duties. The deployed Marines might have testified that they no longer felt comfortable having their families attend those functions while they were deployed. Key volunteers that were also the wives of Marines that Brown was charged with supporting might have testified that they no longer felt comfortable working with Brown in organizing events. Such testimony could have demonstrated that Brown’s job performance and effectiveness would indeed have been impaired, in that he would have been unable to attend the events he planned, to get others to attend them, or even to plan them in the first place. Many possibilities were open to the Government to establish a connection between Brown’s personal conduct and his actual job responsibilities and performance, but the Government did not avail itself of them.
Such testimony would also have served as a hedge against an abuse of power by Brown’s superiors. Before the Government enforce[s][its] conventional codes of conduct in the private lives of its employ*1367ees, Norton, 417 F.2d at 1165, it has the responsibility to establish the necessary connection between the employee’s off duty misconduct and the employee’s job-related responsibilities, White v. United States Postal Serv., 768 F.2d 334, 336 (Fed.Cir.1985).
2. Agency Mission
The AJ also found that the agency has proven nexus ... by showing by a preponderance of the evidence ... that [Brown’s] misconduct ... interfered with and adversely affected the agency’s mission. Board Opinion, slip op. at 8. As this court has recognized, such a finding provides a nexus without a direct showing of an impact on the employee’s job performance. See Allred v. Department of Health and Human Servs., 786 F.2d 1128, 1131 (Fed.Cir.1986) (Courts have repeatedly held that where an employee’s misconduct is contrary to the agency’s mission, the agency need not present proof of a direct effect [sic — affect] on the employee’s job'performance.) (citing Masino v. United States, 218 Ct.Cl. 531, 589 F.2d 1048 (1978) and Giles v. United States, 213 Ct.Cl. 602, 553 F.2d 647 (1977)); see also Masino, 589 F.2d at 1056; Giles, 553 F.2d at 650.
However, the Board does not discuss the mission of the agency, which in this case is the Department of the Navy. See Board Opinion, slip op. at 1. Instead, the Board, and the majority opinion, focus on the mission of the MWR Department and the purpose of Brown’s position:
The agency has established ... that the purpose of the appellant’s position, and the MWR DepaHment, was to provide support for the Marines, and the families of those Marines while they are deployed. Since Major “B” was a Marine Officer who was assigned to one of the very MEU’s [sic — MEUs] that the MWR Department, and the appellant’s position, existed to support, and that the affair took place during the time Major “B” was deployed overseas with his unit, I find that the agency has sufficiently established that that [sic] the appellant’s misconduct was antithetical to the agency’s very mission.
Board Opinion, slip op. at 10-11 (emphasis added). In measuring employee conduct, it is the agency’s mission at large that must be considered and not merely a division or lesser group within the agency. I have found no case from this, or any other, court, in which the mission of a unit smaller than the agency was considered. See Allred, 786 F.2d at 1129 (focusing on the mission of the Department of Health and Human Services); Stump v. Federal Aviation Admin., 761 F.2d 680, 681 (Fed.Cir.1985) (focusing on the mission of the Federal Aviation Administration); Giles, 553 F.2d at 650 (focusing on the mission of the Internal Revenue Service); Masino, 589 F.2d at 1050, 1055 (focusing on the mission of the United States Customs Service). Further, the cases consistently focus on: (1) broad purposes, as opposed to more narrower purposes of a division or other lesser group within the agency; and (2) the public’s perception, which is not likely to be concerned with or even aware of the divisions or lesser groups within an agency. See Allred, 786 F.2d at 1131 (focusing on whether the misconduct might undermine public confidence in the agency); Stump, 761 F.2d at 681-82 (focusing on the broad purpose of the FAA and public confidence in that agency); see also Borsari v. Federal Aviation Admin., 699 F.2d 106, 110 (2nd Cir.) (addressing the goals of the FAA, the agency, to establish a mission connection), cert. denied, 464 U.S. 833, 104 S.Ct. 115, 78 L.Ed.2d 115 (1983); Wild v. United States Dep’t of Hous. and Urban Dev., 692 F.2d 1129, 1133 (7th Cir.1982) (listing numerous examples of, and focusing exclusively on, broad purposes of various agencies). This would be a different case if the Board’s decision and the evidence of record had established that Brown’s misconduct had implicated this country’s national defense, arguably the mission of the Department of the Navy. But the mission of the Navy was *1368not discussed and I must conclude that the Board committed legal error by finding a nexus on this ground.
By focusing on the MWR Department, and then even on Brown’s particular job, the Board and the majority are simply restating their belief that Brown’s job performance would be impacted. However, again, there is no evidence showing any connection between Brown’s misconduct and his job performance.
3. Off-Duty Conduct
The majority opinion cites the three cases of Masino, Giles, and Allred, all of which are binding on this court, for the proposition that off-duty conduct can give rise to a nexus. First, having cited Brown v. Department of Transp. earlier, I certainly do not quarrel with the position that off-duty conduct can give rise to a nexus. However, neither that case, nor any of the three raised by the majority opinion, reverses the long-standing rule that a trust and confidence nexus must contain a connection to the employee’s actual job responsibilities and performance. Indeed, Brown v. Department of Transp. supports this rule, and the other three cases rely principally on the egregious nature of the misconduct or its impact on the agency’s mission to establish a nexus.
The court in Brown v. Department of Transp. first examined the specific requirements of Brown’s supervisory position and observed that [cooperation, loyalty, and trust are particularly important among those managing the operation of a complex, sophisticated transportation system where the lives of hundred of innocent members of the public may ... depend upon split-second judgment. Such need for trust was even more acute at the moment Brown uttered his remarks.... Id. at 547. The court later concluded that it was common sense that Brown’s actions could easily turn into a situation where his loyalty to management could be cast in doubt — as indeed it was. Id. at 548. Thus, management’s loss of trust and confidence in Brown was directly connected to his job performance.
In Masino, the Court of Claims upheld the removal of a Customs Inspector for the off-duty smoking and transporting of marijuana. The Masino court upheld the nexus finding based principally on the egregious nature of the conduct, a ground specifically not found in the present case. See Masino v. United States, 218 Ct.Cl. 531, 589 F.2d 1048, 1056, 1057 (1978) (stating in its analysis that the transportation and use of the very contraband which a law enforcement officer is sworn to interdict, is clearly misconduct which “speaks for itself,” and summarizing in its conclusion that [t]he misconduct speaks for itself.); Board Opinion, slip op. at 8 (I do not find that [Brown’s] misconduct in this instance was so egregious that “it speaks for itself,” and that a “presumption of nexus” is applicable.).
Further, although the Masino court presented a subsidiary justification for finding a nexus, that justification is best characterized as being based on an adverse impact with the mission of the United States Custom Service, the agency in Masino. The Masino court stated that Masino ... was specifically charged with enforcing the laws concerning contraband, including marijuana. Since possession and/or use of marijuana is a violation of federal criminal statues, he was clearly not conducting himself in a manner to be expected of a Government employee engaged in law enforcement duties. Masino, 589 F.2d at 1056. The Masino court thus contrasted the broad purpose of enforcing the laws concerning contraband, common to all Customs Inspectors, with Masino’s obvious contravention of those same laws. Id. The implication is clear that the mission of the United States Customs Service was adversely impacted by Masino’s behavior. See Allred, 786 F.2d at 1131 (identifying Masino as a case where the nexus was based on an adverse impact on the mission of the agency).
*1369In Giles, the Court of Claims upheld the removal of an Internal Revenue Service (IRS) revenue officer for his own willful, flagrant violation of tax laws in not filing or not timely filing his state and federal tax returns over a three year period. Giles, 553 F.2d at 648, 649. The Giles court based its nexus finding on the impact that Giles’ behavior had on the overall effectiveness and credibility of the agency, the IRS, stating that:
If ... taxpayers are encouraged to believe they can unilaterally determine whether or not to file ..., the need for investigations increases and the efficiency of the Service is impaired. [Giles’] example undercuts the Service’s efforts to encourage voluntary compliance and, if condoned, could impair the credibility of IRS with tax officers and the public generally.
Id. at 650; see also Allred, 786 F.2d at 1131 (identifying Giles as a case where the nexus was based on an adverse impact on the mission of the agency). Thus, the Giles court upheld a finding of nexus based on an impact to the IRS’ mission, not to Giles’ job performance.
In Allred, this court upheld the removal of an employee of the Department of Health and Human Services (HHS) based on his conviction, for child molestation. See Allred, 786 F.2d at 1129-30. The misconduct was considered so egregious that a nexus was presumed, thus no connection to Allred’s job performance needed to be shown. See id. at 1131.
The egregiousness of the misconduct also led the Allred court to uphold the finding of nexus on two secondary grounds, finding both an adverse impact on the mission of the agency, HHS, and a loss of management’s trust and confidence in Allred’s job performance. However, the Allred court tied the loss of trust and confidence to Allred’s specific duties, which included representing] the agency before state and local governments, universities, hospitals, and various other grantees in preparing cost allocation plans to ensure the plans conformed with the agency’s regulations. Id. at 1129. The Allred court found that management had lost trust and confidence in Allred’s ability to represent HHS with these groups that were outside of the agency. See id. at 1131. Although the Allred court does not appear to have required any testimony from these clients of Allred’s, as I would require in the present case, the egregiousness of Allred’s misconduct, and the attending presumption of nexus, precluded the need for doing so.
CONCLUSION
Although I cannot condone Brown’s conduct, I also cannot concur in the majority opinion. It grants virtually unbridled discretion to management, allowing it to remove an employee based on disapproval of that employee’s off-duty behavior that does not implicate either job performance or the mission of the agency, in this case the Navy. Accordingly, I respectfully dissent and would reverse the decision of the Board.