AMERICAN FEDERATION OF GOVERNMENT EMPLOYEES, National Border Patrol Council, AFL-CIO, Petitioner v. FEDERAL LABOR RELATIONS AUTHORITY, Respondent, 446 F.3d 162

Opinion for the Court filed by Circuit Judge SENTELLE. SENTELLE, Circuit Judge. The National Border Patrol Council (“the Union”) petitions for review of an order of the Federal Labor Relations Authority (“FLRA” or “the Authority”). The FLRA found that a firearms training policy change by the Bureau of Customs and Border Protection (“the Bureau”) did not have a greater-than-de-mmmis effect on the working conditions of bargaining-unit employees. Because the FLRA unreasonably deemed the change de minimis in its effect, we grant the petition. I.

The Union exclusively represents nonsu-pervisory, border patrol employees within the Bureau, a division of the Department of Homeland Security. Before Congress shifted them to the recently created Bureau, the employees worked for the Immigration and Naturalization Service (“INS”). The Bureau succeeded to INS’s obligations under a 1995 collective bargaining agreement, by which the Union and the Bureau continue to abide even though it is now expired. The facts of this case took place during each agency’s tenure as employer.

The bargaining unit includes Basic Trainee Officers (“BTOs”), essentially first-year probationary employees. BTOs must meet proficiency standards in firearms skill, physical fitness, and foreign language capability, among other areas. The Bureau may terminate BTOs for deficiency in any of these areas during the probationary period. The Bureau’s firearms policy sets out the structure of its firearms training program and the qualifications for trainees’ firearms proficiency. In 1996, the Union and INS bargained over revisions to the firearms policy, including aspects of the policy dealing with training.

The revised policy provided for an initial eight-hour training period followed by proficiency testing. To bring deficient BTOs into compliance, the revised policy authorized up to eighty hours of remedial training. During the same period, the Union and INS agreed to a Memorandum of Understanding (“MOU”), which requires the agency to give the Union notice and an opportunity to bargain over changes to the firearms policy. A separate bargaining unit of Bureau employees, represented by a different union, follows the same firearms policy and training regimen. INS again revised its firearms policy six years later.

In pertinent part, it reduced the number of authorized remedial hours for firearms-deficient BTOs from eighty to eight. While mulling over the changes, INS did not notify, bargain with, or otherwise consult with the Union. Upon finalization the agency gave the Union a copy of the revisions. INS claims that by that time, it had already implemented the reduced remedial training hours prior to making the policy change official. Indeed, the Bureau now insists that neither it nor INS ever offered or gave more than eight hours of remedial firearms training to any BTO since 1996.

Claiming that INS committed unfair labor practices while revising the policy, the Union filed a charge with the FLRA in 2002. In the only claim germane to this petition, it asserted that the Bureau violated both statutory and contractual duties to provide notice and an opportunity to bargain over the reduction in remedial training hours. The FLRA General Counsel subsequently issued a complaint against the Bureau alleging violations of 5 U.S.C. § 7116(a)(1) and (5).

In a hearing before an Administrative Law Judge (“ALJ”), the Bureau argued that the reduction in remedial training hours would not have a greater-than-de-minimis effect on working conditions. To prove greater-than-de-mmimis effect, the General Counsel called the Union’s president to testify. See Social Security Admin. Office of Hearings & Appeals, Charleston, S.C., 59 FLRA 646, 655 (2004) (placing burden of proof on General Counsel). According to the witness, the Bureau fired at least one BTO for firearms deficiency without providing the employee more than eight hours of remedial training.

The witness also testified that the Bureau fired at least one nonbargaining-unit employee — under the same firearms policy — without granting eighty remedial training hours. On this evidence, the ALJ found the effect of the hours reduction “somewhat speculative.” Nevertheless, the ALJ concluded that evidence showing the Bureau fired even one BTO after granting only eight remedial training hours sufficed to exceed the de minimis standard.

Having found a greater-than-de-TOimmis effect, the ALJ held that the Bureau had violated both its statutory duty and its contractual duty under the MOU to give notice and bargain over the changes. The Bureau filed exceptions to the ALJ’s ruling, and the FLRA reversed. In the Authority’s view, the General Counsel failed to prove that the Bureau had fired any BTO solely for firearms deficiency without providing eighty remedial training hours. In addition, the Authority rejected all evidence related to nonbargaining-unit employees.

Accordingly, the Authority held that the General Counsel did not prove greater-than-de-TOmimis effect, and it therefore concluded that the Bureau had no statutory duty to bargain over the policy revisions. The Authority also held that the absence of a statutory duty to bargain precluded a contractual duty under the MOU. The Union timely petitions this court for review of the FLRA’s order, challenging the Authority’s application of the de min-imis exception.

The Union also argues that even if a change is de minimis, an agency may still lawfully choose to bargain over such a change and that the Bureau was required to do so here in light of a provision in the MOU requiring the Bureau to bargain “to the fullest extent allowable under law.” It has since conceded, however, that it waived this issue by failing to raise it before the agency. See 5 U.S.C. § 7123(c). Accordingly, we only address the reasonableness of the FLRA’s application of the de minimis exception to the firearms policy revisions. II.

Federal law requires an agency to “negotiate in good faith” with its employees’ chosen representative. 5 U.S.C. § 7116(a)(5). The employer’s duty to bargain, however, extends only to bargainable issues, not including a range of matters labeled “management rights.” 5 U.S.C. § 7106; see also Nat’l Treasury Employees Union v. FLRA 437 F.3d 1248, 1249 (D.C.Cir.2006). In addition to these explicit exceptions, the FLRA has interpreted the statute to include an unwritten de minimis exception, and we have deferred to its

interpretation. See Ass’n of Admin. Law Judges v. FLRA, 397 F.3d 957, 959 (D.C.Cir.2005) (“AAL/”). Our deference, however, is not without limits: The FLRA has the burden before this court to “show[ ] that any particular application of the de minimis exception is reasonable.” Id. at 963. Because the Authority has not carried that burden, we grant the petition for review. As we stated in AALJ, “[a] de minimis change is not a proper subject of bargaining not because management has a ‘right’ to make it but because it has no appreciable effect upon working conditions.” Id. at 962.

Accordingly, any policy change having an appreciable effect on working conditions cannot find shelter in the de minimis exception. Appreciable effects may surface not only through actual past effects but also through likely future effects.

The FLRA recognizes these two avenues of inquiry in its de minimis exception formula: “In assessing whether the effect of a decision on conditions of employment is more than de minimis, the Authority looks to the nature and extent of either the effect, or the reasonably foreseeable effect, of the change on bargaining unit employees’ conditions of employment.” United States Dep’t of the Treasury, IRS, 56 FLRA 906, 913 (2000) (citation omitted). Following this methodology, the Authority found that the firearms policy change had a de minimis effect on working conditions.

The Authority noted “conflicting” testimony regarding whether the INS first, and the Bureau later, ever offered more than eight hours of remedial firearms training to any BTO under either the original or revised policy. The Authority suggested, however, that since 1996 the agency may have in practice used the same training policy that it adopted officially in 2002. Placing great weight on these years of practice, the Authority reasoned that any adverse effects from the eight-hour cap on remedial training should have manifested themselves during that span.

The Authority found no actual adverse effects during the period after 1996 and therefore concluded that the official policy change in 2002 entailed no reasonably foreseeable effects either. But the Union contends that the policy change has caused actual adverse effects. According to unrefuted testimony, the Bureau terminated at least one BTO for firearms deficiency without offering the eighty remedial hours. The witness further testified that some nonbargaining-unit employees, subject to the same firearms policy, had also been terminated without receiving more than eight hours of remedial training.

Without discrediting the witness’s testimony, the FLRA gave it short shrift. Diminishing the importance of the testimony regarding the BTO’s termination, the Authority found that the General Counsel failed to prove that firearms deficiency was the sole cause of the BTO’s termination. The Authority placed even less weight on the nonbargaining-unit evidence, issuing a blanket exclusion: “Evidence of any impact on non-bargaining unit employees does not serve to establish an impact on bargaining unit employees.” The Authority mishandled both pieces of testimony.

Despite accepting the fact that the Bureau likely fired at least one firearms-deficient BTO without providing eighty remedial training hours, the Authority appeared to require a showing that the policy change constituted the sole cause for the termination. Such an evidentiary requirement would fundamentally change the nature of the de minimis exception, which heretofore relieved the employer of any duty to bargain over “trivia.” AALJ, 397 F.3d at 959. But the specter of termination is no trivial matter.

It is nonsensical to say that a termination, if due even in part to the policy change, does not constitute an “appreciable effect upon working conditions.” The Authority’s error runs deeper, however. Even if the Bureau terminated the BTO for other reasons, the evidence demonstrates the existence of reasonably foreseeable effects. The Authority accepted that at least one BTO' — and perhaps some nonbargaining-unit employees — became eligible for termination despite not receiving more than eight remedial hours of firearms training.

The fact that one employee became termination-eligible due to firearms deficiency makes the likelihood of a future termination much greater. The Bureau’s policy change drastically reduced every employee’s ability to remedy a firearms deficiency, thus increasing the likelihood that a deficient BTO will actually be terminated by the Bureau. This increased likelihood of termination, confirmed by a past occurrence, constitutes a reasonably foreseeable effect on working conditions. Accordingly, a greater-than-de-romimis effect exists when an employee has become eligible for termination due to a policy change.

The FLRA also erred in its treatment of the testimony regarding the nonbargain-ing-unit employees. Although we generally do not presume to tell the Authority which evidence it should or should not find persuasive, see 5 U.S.C. § 7123(c) (setting out “substantial evidence” standard), the complete exclusion of this evidence is baffling. The nonbargaining-unit employees worked for the same employer, followed the same firearms policy, and faced the same penalties for firearms deficiency. The policy change therefore would have struck a similar chord in both groups.

The Authority already faced evidence that the Bureau terminated at least one BTO who failed to meet firearms standards. The terminations of similarly situated, nonbargaining-unit employees provide additional evidence of concrete effects and confirm the likelihood of future adverse effects to the bargaining unit. Again, having neither discredited the testimony nor given reasons for finding it unpersuasive, the Authority should not have discarded such evidence. More than one alarm bell should have alerted the Authority to the fact that this policy revision had an appreciable effect on working conditions.

When a policy change increases the likelihood of an employee’s termination, it almost certainly rises above the level of trivia. In addition, the sheer magnitude of the policy change should have given the Authority pause. Because the Bureau trains BTOs initially for only eight hours, eighty remedial hours would have given laggards ten times that in additional training to cure the deficiency. Without consulting the Union, the Bureau unilaterally reduced that number by ninety percent, from eighty to eight.

A change on that order alone — where the penalty may be termination — pushes the bounds of the de minimis exception. In AALJ, we approved the FLRA’s use of the de minimis exception for a nearly seventy-percent reduction in reserved parking spaces. 397 F.3d at 960. But unreserved parking was plentiful and freely available, and improper parking would not lead to dismissal. Id. at 964. The same cannot be said in this case. Here, the Bureau did not replace the eighty remedial hours with an equivalent, and firearms deficiency would result in termination.

This massive change had a reasonably foreseeable, greater-than-cfe-minimis effect on working conditions. Because we find that the reduction in remedial training hours had an appreciable effect on working conditions, we conclude that the FLRA unreasonably applied the de minimis exception. In addition, because the Authority erroneously relied on the de minimis exception in determining that there was no violation of the MOU, we set aside its resolution of the Union’s MOU claim. III. For these reasons, we grant the petition for review.

We remand to the Authority for further proceedings not inconsistent with this opinion.