NOTICE: Federal Circuit Local Rule 47.8(b) states that opinions and orders which are designated as not citable as precedent shall not be employed or cited as precedent. This does not preclude assertion of issues of claim preclusion, issue preclusion, judicial estoppel, law of the case or the like based on a decision of the Court rendered in a nonprecedential opinion or order.
Robert E. LONG, Petitioner,
v.
UNITED STATES POSTAL SERVICE, Respondent.

No. 91-3369.

United States Court of Appeals, Federal Circuit.

May 6, 1992.

Before NIES, Chief Judge, and PLAGER and LOURIE, Circuit Judges.

DECISION

LOURIE, Circuit Judge.

Robert E. Long appeals from the May 1, 1991, decision of the Merit Systems Protection Board, No. DA07529010467, sustaining his removal by the United States Postal Service (the "agency"). The Board found that Long fabricated a story that he had been attacked and it sustained his removal for disrupting postal operations at the Melear Station on April 19, 1990. We hold that the Board's decision is unsupported by substantial evidence. Accordingly, we reverse.

BACKGROUND

Long was employed as a Supervisor of Mails and Delivery at the Melear Station in Arlington, Texas. On April 19, 1990, Long reported to his managers that at approximately 6:55 a.m., while he was working in the vault room, he was attacked by an unidentified assailant armed with a knife. He alleged that he went into the vault room, without turning on a light, to rearrange stock in preparation for a delivery later that day. He said that from the hallway light he could see well enough to open one of the safes in the room. Allegedly, while kneeling in front of the open safe, someone pushed him toward the safe. When Long tried to stand, the alleged attacker shut the safe door on his head. Long claims that as he was falling backwards, his attacker grabbed him and came toward him with a knife. The knife apparently missed Long and was found stuck in the wall behind him. While Long claims he did not see his assailant, he identified the person's right hand as that of a large, black male with a silver ring. Long's injuries were confirmed by a doctor as contusions of his head and temple area. Long alleges that someone was trying to kill him, since he had previously received several notes on postal service forms threatening his life.

An investigation was conducted by the Postal Inspection Service and postal security personnel which resulted in closing window operations at the Melear Station for approximately four hours and thereby disrupting postal operations. The postal inspectors were unable to discover any evidence to confirm or to deny Long's story. For example, no assailant could be found, no fingerprints recovered, and no witnesses were discovered. Also, the inspectors were unable to reposition the knife in the wall according to Long's description of the event. In total, the inspectors found thirteen "inconsistencies" between Long's story and the results of their investigation.

Finding that a preponderance of the evidence supported a conclusion that Long fabricated the story, the agency determined that the story was untrue. In the Board's initial decision, the administrative judge (AJ) agreed with the agency and found that Long had made a false report of being attacked. It concluded that Long's removal from his position for disrupting postal service operations was warranted. Long's report had caused the facility to close its service windows for several hours. The Board denied Long's petition for review and Long brought this appeal.

DISCUSSION

I. Standard of Review

We review the Board's decision under a narrow standard, affirming the judgment 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) (1988).

To sustain the agency's contention that Long filed a false report of his alleged attack, the agency was required to prove by a preponderance of the evidence that Long knowingly supplied wrong information, and that he did so with the intention of deceiving or misleading the agency. Kumferman v. Department of the Navy, 785 F.2d 286, 290 (Fed.Cir.1986). The Board, without making specific findings on intent, concluded that Long knowingly supplied wrong information. Because the Board's finding on the charge of knowingly supplying wrong information was unsupported by substantial evidence, we will reverse. We therefore need not reach the issue of intent.

In order to show that the report Long filed contained wrong information, the agency needed to prove by a preponderance of the evidence that the attack did not occur. See 5 U.S.C. § 7701(c)(1)(B) (1988). A preponderance of the evidence is defined as:

That degree of relevant evidence which a reasonable mind, considering the record as a whole, might accept as sufficient to support a conclusion that the matter asserted is more likely to be true than not true.

5 C.F.R. § 1201.56(c)(2) (1990). This court has stated that preponderance of the evidence "means the greater weight of evidence, evidence which is more convincing than the evidence which is offered in opposition to it." Hale v. Department of Transp., FAA, 772 F.2d 882, 885 (Fed.Cir.1985).

We review the Board's findings to ascertain whether substantial evidence supports a conclusion that Long's report was not true, i.e., that the attack did not occur. Substantial evidence is such relevant evidence, in the record as a whole, as might be accepted by a reasonable mind as adequate to support the verdict under review. Ellermets v. Department of the Army, 916 F.2d 702, 705 (Fed.Cir.1990). Our consideration of the evidence of record indicates that the Board's conclusion that the attack did not occur is not supported by substantial evidence.1

II. The Burdens--Long's Credibility

Long alleged and the agency agreed that Long was the only witness to the attack. The Board's determination that Long fabricated the story thus depended on his credibility alone. The AJ relied on Hillen v. Department of the Army, 35 M.S.P.R. 453 (1987), in determining credibility. Hillen instructs an AJ how to resolve credibility disputes.

[A]n administrative judge must first identify the factual questions in dispute; second, summarize all of the evidence on each disputed question of fact; third, state which version he or she believes; and, fourth, explain in detail why the chosen version was more credible than the other version or versions of the event.

Id. at 458. Pursuant to Hillen, several factors must be considered in making a credibility determination. These factors include the following:

(1) The witness's opportunity and capacity to observe the event or act in question; (2) the witness's character; (3) any prior inconsistent statement by the witness; (4) a witness's bias, or lack of bias; (5) the contradiction of the witness's version of events by other evidence; (6) the inherent improbability of the witness's version of events; and (7) the witness's demeanor.

Id.

We endorse Hillen's instructions for resolving credibility determinations by administrative judges. We also agree that the factors listed in Hillen are reasonable and a valid starting point in making credibility determinations. However, in the present case, the AJ erred in the way she applied the Hillen test. She agreed with the agency, even though it failed to refute Long's story of what did happen on April 19, 1990 with preponderant evidence. The AJ incorrectly relied primarily on the finding that Long's allegations were inherently improbable and inconsistent. These findings lack substantial evidence.

Overall, the AJ found that Long's version of the story was not credible, while noting that Long "presented a convincing demeanor during his testimony and a number of witnesses, including his father and his minister, testified as to [Long's] truthfulness." The AJ agreed with the agency that Long's story was more than likely not true, but substantial evidence does not support this finding. The substantial evidence requirement means that there must be a sufficient amount of evidence that a reasonable fact-finder could conclude the story was not true, and here there was none.

We acknowledge that credibility determinations are virtually unreviewable on appeal, Hambsch v. Department of the Treasury, 796 F.2d 430, 436 (Fed.Cir.1986), and that this court will not "second-guess a presiding official's credibility determination, based as it was on demeanor...." DeSarno v. Department of Commerce, 761 F.2d 657, 661 (Fed.Cir.1985) (citing Griessenauer v. Department of Energy, 754 F.2d 361, 364 (Fed.Cir.1985)).2

Overall, the Hillen factors support Long's credibility and do not provide substantial evidence to the contrary. Long was the only person with the opportunity and capacity to observe the event (the first factor). Long's character (the second factor) was good; he had a good work record with no prior offenses. The fourth and fifth factors, Long's possible bias or lack of bias and the possible contradiction of Long's version of events by other evidence, are not sufficiently probative here to add or detract from Long's credibility. It has not been suggested that Long would have benefited by filing a report which was fabricated or indeed that he had any motive to do so.

Even the sixth factor, the inherent improbability of Long's version of events, is no worse than neutral. While the alleged events were uncommon and certainly a bit bizarre, they were not inherently improbable considering the circumstances. No one disputes that something did happen to Long on the morning of April 19, 1990. Long suffered contusions to his head, confirmed by a doctor. A knife was recovered from the wall in the vault room where the alleged incident took place. Long had documented proof that he had been previously threatened. Given these facts, and the fact that no contrary story was alleged by the agency, such as that he inflicted the injury on himself, Long's version of the alleged attack was not inherently improbable. Finally, and importantly, the AJ found Long's demeanor (the seventh factor) to be credible.

The inspectors determined some of Long's statements to be inconsistent with prior statements (the third factor), but we do not agree that these were inconsistencies. The inspectors found thirteen inconsistencies in Long's story which the AJ relied on in making her determination that Long falsified his report. In reviewing the evidence, we determine that these inconsistencies were not in fact inconsistencies. At most, the majority of these determinations were indications of improbability or speculations, and most are not supported by substantial evidence.

For example, the first "inconsistency" noted is that there were "[n]o witnesses, when numerous employees were in a position to have seen or heard something. No employee claimed to have heard or seen anything unusual the entire morning of April 19, 1990." This is not an inconsistency. Long notes that the two employees most likely to have witnessed the event were not even questioned during the investigation. The agency presented no evidence to the contrary. Thus, there is no inconsistency in the fact that no employee heard or saw the alleged event.

Additionally, the inspectors determined that it was "[h]ighly unlikely that a non-employee could have entered the building." Clearly, this determination is not an inconsistency. Long alleged that people can enter the post office from open back and side doors at all hours of the day, and the agency presented no evidence to the contrary.

The inspectors also found Long's not turning on the light when entering the safe room to be an inconsistency. Once again, this is hardly an inconsistency, but merely an undisputed fact. Additionally, both parties agree that light from the hallway at least partially illuminated the vault room.

Another inconsistency found by the inspectors was the injury received compared with the description of the attack. The inspectors stated:

There were no bruises to Long's arms, shoulder or side which would have been expected if the safe door [were] slammed on him. How could his head be caught between the safe and the safe's door without the door bruising his shoulders or arms?

At the hearing, Long presented evidence on how the safe door could have hit only his head without striking his shoulder or arm. However, the AJ found "the injuries which he sustained to be completely incompatible with his story." While we cannot judge the question of physical injury, it is certainly not farfetched that Long's head could suffer contusions without the rest of his body being harmed. The examining doctor, according to Long, stated that Long was lucky his skull was not broken given the weight of the door.

The fact that no one inside the post office fit the description of the alleged assailant is an "inconsistency" cited by the AJ. Long initially thought that this person was a postal employee, but later stated that it may have been someone from outside. The agency does argue that no one in the post office fit the description given by Long. However, this is not an inconsistency, but a lack of evidence as to the identity of the attacker.

Other asserted inconsistencies noted by the inspectors include that it was "[h]ighly unlikely that an employee would attempt the alleged assault without expecting to be identified and therefore, jeopardizing his job." Further, they said it was "[h]ighly unlikely that, as Long claims, the Hurst burglar has a contact within Melear Station who did the attack."3 The inspectors regarded Long's inability to see his assailant to be another inconsistency. Also, the inspectors found it inconsistent that Long was "not seriously injured when the assailant had him defenseless on the floor." The fact that the knife was left at the scene was determined to be an inconsistency as well as the lack of fingerprints obtained from the scene. Additionally, the inspectors found Long's demeanor after the attack to be inconsistent with his description of the event. We do not find any of these determinations to be "inconsistencies," but rather speculations or additional noncontradicting facts. They certainly are not evidence sufficient to refute Long's story.

Finally, the inspectors found the location and angle of the knife in the wall to be inconsistent with Long's description of the alleged attack. The agency alleges that the knife could not be inserted back into the wall when simulating the attack as Long described it. This is actually the only potential inconsistency in Long's story. However, we will not speculate about the attack or about the accuracy of Long's description of it; this fact alone does not disprove Long's story. The fact of the matter is that there was a knife retrieved from the wall and there was no contrary evidence, such as Long's fingerprints, as to how it got there.

In sum, we find almost all of the thirteen "inconsistencies" to be other facts or non-facts relating to the incident, which, while not supporting Long's case, are not substantial evidence to refute it. In fact, Long states that two of the inspectors testified that, although the assault was unlikely, it could have occurred. We conclude that it was not reasonable to find from the existing facts that the attack on Long did not occur.

We cannot uphold the AJ's conclusion that Long's report of the alleged attack was erroneous without finding substantial evidence to support it. An employer should not have to tolerate a disruption of its operations caused by false reports of an assault. However, an employee should not be discouraged from reporting an attack simply because he or she has no witnesses. For an employer to determine that an employee's report of a criminal attack, although plausible, is erroneous, some evidence is needed to support the employer's position.

The AJ determined that the agency proved by preponderant evidence that the attack could not have taken place. Upon reviewing this decision, we do not discern any evidence of fabrication, but rather a lack of evidence to prove whether Long's assertions were true or not. While the agency was not required to conclusively prove a negative, that the event did not occur, the record must contain substantial evidence that Long fabricated his story. The agency failed to meet its burden.

We hold that the AJ's determination that Long falsely reported to his supervisors that he had been attacked was not supported by substantial evidence. Accordingly, the decision of the Board affirming Long's removal is reversed.

PLAGER, Circuit Judge, dissenting.

I respectfully dissent. Since the opinion in this case is nonprecedential, confusion about the applicable law has been avoided. Nevertheless, because I believe that not only is the result wrong, but the analysis that led to that result is flawed, I write briefly to explain.

In my view, the majority has exceeded its proper role as an appellate court and engaged in impermissible reweighing of the central credibility determination in the case. The key to the entire event was whether or not one believes Long. The agency which investigated disbelieved him; the AJ who held the hearing and took his evidence disbelieved him; the MSPB, in affirming the AJ, determined that the facts supported the agency.

The agency and the MSPB had before it all the evidence there was, and there was more than enough to find Long's story false if, as was the case, he was disbelieved. The panel, on the other hand, carefully reweighs each item of evidence and concludes that the various factual "inconsistencies" were not all that inconsistent. From this it concludes that his story was believable and presumably should have been believed. That is not a function of an appellate court.

The question before this court is not whether we think Long's story is believable, or even whether the agency proved that it was not believable by a preponderance of the evidence. The only question on appeal is whether, on these facts, there is substantial evidence in the record to support the MSPB's determination that a preponderance of the evidence supported the agency's decision to disbelieve Long. 5 U.S.C. § 7703(c) (1988). Substantial evidence is a highly deferential standard. It does not require that we agree with the trier of fact, but only that there be some evidence, more than a mere scintilla, to support the trier's conclusions. "It means such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.... We must determine whether, considering the record as a whole, the agency's evidence is sufficient to be found by a reasonable factfinder to meet the evidentiary burden applicable to the particular case." Bradley v. Veterans Admin., 900 F.2d 233, 234 (Fed.Cir.1990) (citations omitted). The MSPB had all the evidence it could have, and it was enough. There is no basis here for anything other than an affirmance.

Long also argues that the agency violated Section 651.63 of the Employee Relations Manual and thereby committed harmful error. Given that there is no substantial evidence to support the Board's conclusion that Long submitted a false report, we need not reach this issue

The dissent accuses us of reweighing the evidence, which is not the function of an appellate court. However, in order to determine whether the Board based its decision on substantial evidence, it is not only permissible, but essential, that we review the evidence to determine whether it can support the Board's conclusion. Here we conclude that the asserted improbabilities and inconsistencies do not constitute substantial evidence that the attack did not occur

In 1987, Long's home in Hurst, Texas had been burglarized and vandalized by a neighbor's son, who was convicted for the offense. The offender had later threatened Long for pressing charges. In February, 1990, Long's home was again burglarized and vandalized, and Long claims the neighbor's son was again the prime suspect. After the April 19 report, Long alleged that the attacker may have been connected to the Hurst burglary