Georgetown Hotel v. National Labor Relations Board

SILBERMAN, Circuit Judge,

dissenting:

This case, which involves an unusual set of facts, presents a very narrow question of substantive labor law but, it seems to me, a much more important issue as to the appropriate scope of judicial review.

It is perfectly true, as the majority emphasizes, that an employer is not obliged to accept and validate a union’s proffered authorization cards to determine whether they constitute evidence of majority status. Linden Lumber Division v. NLRB, 419 U.S. 301 , 95 S.Ct. 429 , 42 L.Ed.2d 465 (1974). It has, instead, every right to insist on an election. Id. On the other hand, if an employer does agree to recognize a union, subject to a card check, and subsequently satisfies that condition, it has bound itself to recognize the union al *1473 though the employer, at least under present Board law, may withdraw from the agreement at any time prior to its execution. United Buckingham Freight Lines, 168 N.L.R.B. 684 (1967). Here, as found by the Administrative Law Judge (“ALJ”) and the Board, the employer’s expressed condition subsequent was more limited than the normal validation of cards; he asked not to determine whether the employees voluntarily and knowingly signed the union authorization cards or whether the signatures were genuine, cf. Harding Glass Indus., 216 N.L.R.B. 331 (1975), but rather, as a new manager, only to determine whether the signatories were employees on his hotel’s payroll (and not floater employees actually on the payroll of two other nearby affiliated hotels), and if so, whether they were a majority. The ALJ concluded as a finding of fact (adopted by the Board) 1 that the employer had satisfied the condition — that it had “checked the names on the signed cards with names on its payroll records.” ALJ op. at 15, J.A. 25. It is that finding of fact that the majority concludes is unsupported by substantial evidence and therefore declines to accept.

The finding rests, as the AU explicitly stated, not on any direct testimony that the employer had compared the cards with the payroll list, but rather on inferences the ALJ drew from the employer’s subsequent conduct. Specifically, the AU inferred that when the employer’s counsel sent the Union representative (preparatory to the private election arranged to save the manager, Rosenbloom, embarrassment) the very payroll list against which the manager had agreed to compare the names on the cards, either the manager or his counsel had compared the one to the other. 2 Of course, it is possible that neither the attorney nor the manager compared the two. The majority emphasizes that counsel testi-fled he advised the manager not to check the cards. Majority op. at 1472. If Chati-lovicz had testified to that effect, the AU, of course, was not obliged to credit his testimony. 3 But Chatilovicz did not so testify. Instead, Chatilovicz said that he told “Mr. Rosenbloom that he could simply ignore the visit and do nothing. I said he could take a look at the cards further and try to determine through a payroll list or some other method to see if the employees did represent a majority_” J.A. 306. Later on, when asked whether Rosenbloom “at any time indicate[d] he was going to do that, check his records,” Chatilovicz responded “No, to the contrary. I had advised him to put the cards in an envelope and to send them over to me.” J.A. 309. Chatilovicz never said he advised Rosen-bloom not to look at the cards, and he never said that Rosenbloom denied looking at the cards. Moreover, Chatilovicz never denied having checked the cards himself. Nor did Rosenbloom testify that he was advised not to check the cards, or that he did not do so. What is more, one of the two options Chatilovicz suggested to Ro-senbloom was to check the cards against the payroll. This is exactly what the AU found was done. What the majority does is even more problematic than relying on non-credited testimony — it relies on testimony that, as far as I can determine, is not in the record.

The Board’s inference that the employer compared the cards to a payroll list is further strengthened because once a quick informal private election was agreed upon in lieu of the payroll check, there was no longer any disincentive to examine the payroll list; so long as the private election agreement was in force that comparison had no legal consequences. It was only after the employer discharged Chatilovicz, retained other counsel, and disavowed the *1474 private election agreement that — as the Board held, Board op. at 2 n. 3, J.A. 33— the prior existing agreement was resuscitated. And at that point the prior agreement had, by virtue of the preparations for the private election, been completely executed. The AU thought that since Rosen-bloom’s only condition upon recognition was mere assurance, by examination of the payroll, that the employees who signed the cards were employed by the hotel, it was reasonable to infer that the very preparation of the payroll list to be sent to the union more likely than not satisfied the condition — that the manager or his counsel had at one point looked at both the cards and the list.

Surely if, after the original meeting between Richardson and Rosenbloom, the latter had merely sent the payroll list to Richardson, and no other evidence one way or the other indicated whether Rosenbloom had compared the list against the cards, a finding that Rosenbloom had in fact compared the two would be unassailable. I think this case is not significantly different. The agreement to a private election is relevant only because it created the circumstances whereby the payroll list was compiled and transmitted.

The majority, it seems to me, asks much too much when it demands that the circumstantial evidence upon which the AU relies must necessarily establish the fact found. Administrative agencies are entitled to find facts based on inferences drawn from circumstantial evidence that are not, as a matter of human experience, inevitable. Determination of what turned out to be the critical fact in this case — whether or not Rosenbloom or Chatilovicz compared the payroll list to the names on the cards — in the absence of direct testimony must rest on inferences “and it has long been settled that an agency’s conclusions based upon such inferences should not be set aside by a reviewing court unless they transgress the bounds of reason.” Catholic Medical Center v. NLRB, 620 F.2d 20 , 22 (2d Cir.1980). Indeed, this seems precisely the point at which an agency's expertise is most useful and should be most respected by a reviewing court. For “it is the Board to which Congress has delegated administration of the Act. The Board, therefore, is viewed as particularly capable of drawing inferences from the facts of a labor dispute. Accordingly, it has been said that a Court of Appeals must abide by the Board’s derivative inferences ... unless those inferences are ‘irrational,’ ‘tenuous’ or ‘unwarranted.’ ” Penasquitos Village, Inc. v. NLRB, 565 F.2d 1074, 1079 (9th Cir.1977) (citations omitted).

The majority, by relying on “common sense,” majority op. at 1471-72, to reject the inference drawn by the AU is either crediting a version of Chatilovicz’ and Rosenbloom’s testimony that neither actually advanced or it is preferring its own expertise to that of the Board: I do not believe either is permissible. It is true, as the majority points out, that Rosenbloom or Chatilovicz could have prepared the eligibility list without checking the names on the cards. Id. at 1472. But it is also reasonable to infer, as did the AU and the Board, that they did check the names; this finding is hardly “irrational.” “[T]he possibility of drawing two inconsistent conclusions from the evidence does not prevent an administrative agency’s finding from being supported by substantial evidence.” Consolo v. Federal Maritime Comm’n, 383 U.S. 607, 620 , 86 S.Ct. 1018, 1026 , 16 L.Ed.2d 131 (1966); Public Citizen Health Research Group v. Tyson, 796 F.2d 1479, 1485 (D.C.Cir.1986). Our role as a reviewing court is not to decide which of the two possible conclusions we would choose, but whether the Board’s choice is supported by “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Universal Camera Corp. v. NLRB, 340 U.S. 474, 477, 488 , 71 S.Ct. 456, 459, 464 , 95 L.Ed. 456 (1951); American Federation of Labor v. Marshall, 617 F.2d 636 , 649 n. 44 (D.C.Cir.1979). If this standard is applied, then the Board’s ruling must stand.

It is illustrative to consider whether if we were reviewing a jury verdict in favor of the Board, rather than an agency determination, we would grant a judgment n.o.v. For substantial evidence has been defined *1475 as “enough to justify, if the trial were to a jury, a refusal to direct a verdict when the conclusion sought to be drawn from it is one of fact for the jury.” Consolo v. Federal Maritime Comm’n, 383 U.S. at 620 , 86 S.Ct. at 1026 . It is clear that “a verdict based on circumstantial evidence is not infirm simply because the evidence supports an equally probable inference to the contrary. It is the jury that chooses among allowable inferences. The standard for determining whether an inference is allowable is ... whether it is one that ‘reasonable and fair-minded men in the exercise of impartial judgment’ might draw from the evidence. An inference is not unreasonable simply because it is based in part on conjecture, for an inference by definition is at least partially conjectural.” Daniels v. Twin Oaks Nursing Home, 692 F.2d 1321, 1326 (11th Cir.1982) (citation omitted). See Williams v. Steuart Motor Co., 494 F.2d 1074, 1082 (D.C.Cir.1974). I think it obvious that on these facts we would not grant judgment n.o.v. to the petitioner and yet the majority does not hesitate to reverse the expert finding of the agency.

To be sure we are not here faced — as we often are — with a highly technical or scientific finding of an administrative agency to which deference seems more natural. This case involves only an unusual twist on a familiar, easily understandable event. But it is no less objectionable, in my view, for the court to prefer its judgment as to the weight of the circumstantial evidence to that of the Board’s than it would be if the question were more complex or unfamiliar. Restrictions on judicial review of agency fact findings — like limitations on review of jury verdicts — are based not just on notions of relative institutional competence, but rather on our understanding of the federal judiciary’s appropriately limited role.

Nor do I agree with the majority that affirming the AU and the Board’s finding would result in the “rigid rule” the majority fears — that whenever an employer accepts cards from a union, verification will be assumed. Majority op. at 1472. An employer has an absolute right to reject authorization cards and to seek a Board election. Even when an employer accepts cards subject to verification that normally refers to a good deal more than the limited inquiry this employer imposed as a condition subsequent to the recognition agreement, and therefore a finding that verification had taken place would usually be more difficult. See, e.g., Cam Indus., 251 N.L.R.B. 11 (1980), enforced, 666 F.2d 411 (9th Cir.1982); Harding Glass Indus., 216 N.L.R.B. 331 (1975). The majority takes the Board to task for not explaining why this court should adopt the “rigid rule” it claims has been adopted here. But it seems to me this finding of fact did not a rigid rule make, and the Board has not characterized its decision as the adoption of a rule of any sort. It is inappropriate to require an explanation for that which has not been done.

In sum, this is, as I said at the outset, an unusual case but since I believe the majority’s scope of review to be outside the bounds of the National Labor Relations Act, I respectfully dissent.

.The Board adopted the ALJ's findings and conclusions. The footnote authored by two Board members did not, as the majority suggests, majority op. at 1470, reject the ALJ’s finding that the names were actually checked against the payroll list. Board op. at 2 n. 3, J.A. 33.

. At oral argument, counsel for the Board said nothing inconsistent with this. But cf. majority op. at 1471.

. Indeed, the ALJ specifically declined to credit other important portions of Chatilovicz’ testimony. ALJ op. at 13, J.A. 23.