employer “promised that [subject matter of grievance] would ‘be handled differently’ by the company’s ‘new personnel.’ ”
How later courts described this case
- employer “promised that [subject matter of grievance] would ‘be handled differently’ by the company’s ‘new personnel.’ ”
- company precluded from asserting objection to bargaining order issues sua sponte by Board since it did not move for reconsideration below
- employer "would never sign a contract with Union"
Written by the judges who cited it.
The opinion
ROSENN, Circuit Judge,
dissenting.
The majority opinion, under the rubric of judicial deference, approves of the imposition of a bargaining order without a shred of evidence that a rerun election is impractical. Because I believe that the Board’s order lacks a reasonable basis in law and is unsupported by substantial evidence, I respectfully dissent. 1
The Board’s explanation for the extraordinary remedy of a bargaining order is unacceptable because it is not predicated upon factual determinations but is structured upon speculation built, on speculation. For example, the Board concludes, without any factual evidence to support the conclusion, that the total effect of threats of plant closure during the campaign “was to instill in employees a strong fear of loss of employment that would continue to be operative in the event of a second election.” This is speculation of the highest rank because (1) there is no real evidence that a substantial portion of the work force was affected by “strong fears of the loss of employment” engendered by management threats, and (2) the record does not show the composition of the work force four to five years later. How many of the present work force knew of the previous threats? What is there in this record to show that the employees retain any fears that would affect their ability to freely cast a ballot in a second election conducted by the Board?
The Board apparently believed it was properly punishing the company for engaging in unfair labor practices. Such a simplistic approach, however, ignores the fact that the real victims of the bargaining order are the employees. The employees, many of whom were not in the work force during the election, are being punished for they are being deprived of their right to vote and they are being represented by a bargaining agent they may not desire. I believe the majority errs when it “defers” to alleged Board “expertise” under the facts of this case.
I.
Over six years ago, on March 28, 1974, a representation election was conducted for the employees of Hedstrom. Following the election which the Union lost by a vote of 125 to 113, unfair labor practice charges were lodged against the employer. After an NLRB administrative law judge (ALJ) held hearings on the campaign and election, he concluded that the company had committed unfair labor practices and that the election should be set aside. Although noting that the company had threatened to close the plant if the Union won, the ALJ evidently viewed Hedstrom’s transgressions as “numerous” but not major. The ALJ rejected the contention that the unfair labor practices warranted a bargaining order. Having heard the testimony of the witnesses, he concluded that the employer’s unfair labor practices could be remedied by an appropriate cease-and-desist order and were *320 not so serious as to preclude a board supervised rerun election.
The Board cursorily reversed the ALJ’s decision and issued a bargaining order. In granting that order, the Board relied on NLRB v. Gissel Packing Co., 395 U.S. 575 , 89 S.Ct. 1918 , 23 L.Ed.2d 547 (1969). According to that decision, once a Union enjoys the support of a majority of employees as indicated by authorization cards, and unfair labor practices by the employer undermine this majority and leave only a slight chance for a fair rerun of the election, the Board may order the employer to bargain with the Union. Id at 614, 89 S.Ct. at 1940 .
When we first reviewed the bargaining order issued to Hedstrom, we remanded for an explanation of the specific reasons why a new election could not be held. Hedstrom Co. v. NLRB, 558 F.2d 1137 (3d Cir. 1977). See NLRB v. Armcor Industries, Inc., 535 F.2d 239 (3d Cir. 1976). Further proceedings before the ALJ and the Board followed. However, the Board did not take additional testimony; it did not request position statements or briefs from counsel on remand. In April 1978 the Board found that Hedstrom had committed additional unfair labor practices since the first bargaining order. Referring to the original violations and the subsequent unfair labor practices, the Board proceeded to draft an explanation for its second bargaining order. The principal issue now before this court is the sufficiency of that explanation.
II.
The justification advanced by the Board must be examined in light of the paramount principle that the preferable way to measure employees’ support for a union is by secret election. Otherwise a union, unwanted by either the employer or the employees, may be foisted upon a company. Although Gissel approves the Board’s resort to authorization cards as a gauge under limited circumstances, the opinion leaves no doubt that, when possible, an election should be held. 395 U.S. at 603-04 , 89 S.Ct. at 1934-1935 . This preference for the elections is a cornerstone of labor relations. Elections permit employees to express their personal preferences by secret ballot in the highly critical determination as to who, if anyone, will exclusively represent them in collective bargaining. The secrecy of the ballot most effectively wards off the pressures created by the advocates for and the opponents of the union; it insulates employees from intimidation by either union organizers or the company. The relative formality of an election, supervised by NLRB agents, as contrasted with the informal solicitation of authorization cards, and whatever unknown representations are made in the process, may impress employees as to the seriousness of their choice and the necessity for deliberation. After a campaign in which both the union and the company debate the merits of unionization, employees can make a more informed choice than when they informally sign authorization cards based solely on union representation. Authorization cards are not foolproof and their solicitations are not always free from misrepresentation or pressure. It is the independent, deliberate, secret, and informed choice of employees — the choice made in an election — to which the Board should, whenever possible, give effect.
[Democracy in industry must be based upon the same principles as democracy in government. Majority rule, with all its imperfections, is the protection of workers’ rights, just as it is the surest guaranty of political liberty that mankind has yet discovered.
Senator Robert Wagner paraphrasing Lloyd K. Garrison, former chairman, National Labor Relations Board, 79 Cong.Rec. 7571 (1935).
Judged in light of the preference for elections, the Board’s bargaining order in this case is a misapplication of Gissel and is unsupported by substantial evidence. Under Gissel, as interpreted by this court in NLRB v. Armcor Industries, Inc., supra, the Board may issue a bargaining order here only if unfair labor practices have so corrupted conditions at the Hedstrom plant that, even with the passage of time, a fair election, could not be held. The Board at *321 tempts to satisfy this standard by the following argument: in 1974 Hedstrom committed unfair labor practices that precluded a fair election then, and because of subsequent unfair labor practices “the passage of time has not increased the likelihood that a fair second election can be held.” As the Board wrote in its Supplemental Decision and Order,
[T]he effect of [Hedstrom’s] threats of plant closure and its extensive unfair labor practices during the course of the campaign was to instill in employees the fear that they would lose their jobs through plant closure if the Union won the election. There is nothing to indicate that this fear has been erased during the past 3 years. To the contrary, recent events indicate [Hedstrom’s] vigorous opposition to the Union continues to result in unfair labor practices, thereby sustaining, reinforcing, or increasing the fear that the Company might still carry out its earlier threat to close operations if the Union prevails.
As I have already indicated, this is sheer speculation. The facts do not substantiate this argument. I doubt that the record in this case shows that a new election could not have been conducted in 1974; it certainly does not show that a fair election was precluded when the bargaining order was issued in 1978.
As the ALJ recognized in his decision concerning the 1974 violations, Hedstrom’s unfair labor practices were numerous. With the exception of the threat to close the plant, however, the violations were not, in his view, “extensive” in terms of their weight and effect. He concluded that even the threat of closure did not preclude a fair election:
Here, the Union obtained 133 authorization cards. It polled 113 votes. Thus, even if the assumptions are made that all 133 persons who placed their names on the Union’s authorization cards did so because they wanted representation and that all 20 of those who changed their minds were coerced into doing so by [Hedstrom’s] unfair labor practices, [Hedstrom] dissipated only 15 percent of the Union’s majority. The impact of unfair labor practices of such limited effectiveness is not sufficiently lingering to justify a finding “that the possibility of erasing the effects . . . and ensuring a fair . . rerun ... by the use of traditional remedies, though present, is slight and that employee sentiment once expressed through cards would, on balance, be better protected by a bargaining order.”
(Footnote omitted quoting NLRB v. Gissel Packing Co., supra, 395 U.S. at 614-15 , 89 S.Ct. at 1940 .)
In neither its original decision nor its opinion after remand has the Board answered these persuasive findings. 2 On the basis of the Board’s failure to repudiate the findings of the ALJ, there is serious doubt that Hedstrom’s violations would have prevented a fair rerun election, even in 1974.
In view of this doubt, what is for me decisive is the Board’s failure to point to actual evidence rather than mere speculation that the effects from the 1974 violations continued to afflict Hedstrom’s employees when the Board issued its second opinion in 1978. The Board rested its conclusion on additional unfair labor practices which it found Hedstrom had committed since 1974. I turn to these alleged violations.
The Board specifically pointed to four unfair labor practices which it said had taken place since 1974 and had continued the effects of the 1974 violations: (1) the discharge of Rena Ritchey; (2) the interrogation of Ritchey; (3) the incident involving Clark Ferguson and Delores Casteel; and (4) the threat that Erma England would be assigned more burdensome condi *322 tions for work. 3 I agree with Judge Adams that we should sustain the finding concerning the discharge of Ritchey, ordering reinstatement with back pay. However, there is nothing whatever in her discharge which has the slightest evidentiary value to support the Board’s inability to conduct a fair election. Her discharge grew out of a failure on her part to respond to a recall to work following a work stoppage. I disagree with the majority’s enforcement of the Board’s findings as to the incident involving Ferguson and Casteel and believe that the Board’s finding about a threat to England is also insupportable.
Two nights after the strike was called off, a minor company supervisor, Clark Ferguson, encountered Delores Casteel, a Union employee, in a bar. Ferguson, who obviously had been drinking heavily, called Casteel “a scab,” told her the Union would never get a contract and concluded by saying that she “had no job.” Taken alone, this language could well be viewed as threatening an employee with discharge and tending to emphasize futility in Union activity. NLRB v. Varo, Inc., 425 F.2d 293, 299 (5th Cir. 1970). However, in context, it is clear that Casteel did not labor under any threat of discharge from Ferguson’s drunken remarks and that Ferguson acted in excess of his authority. It is conceded that Casteel knew that Ferguson was drunk; that he was not her supervisor or in any way connected with company employment policy; that she did have a job, because that same afternoon she had been called back to work by a company official; and, finally, that Ferguson was so drunk he did not know what he was talking about. 4
Under the circumstances, I can find no substantial evidence supporting a conclusion that Casteel was in any way coerced, nor in any way threatened in a manner tending to discourage Union activity. I view the Ferguson-Casteel incident as an isolated, personal barroom exchange between company employees with sharply different views. The company was completely divorced from complicity for this barroom incident and should not be held responsible. Accordingly, I would reverse the finding of the Board in this respect.
As for the alleged threat to England, the ALJ found that the conversation was not coercive. Both section 10(c) of the Labor-Management Relations Act and the Board’s own regulations dictate that unless an exception to a finding of the ALJ is filed, the finding is final and cannot be reviewed by the Board. 29 U.S.C. § 160 (c) (1976); 29 C.F.R. §§ 102.46 (h), 102.48(a) (1978). Although the General Counsel lodged no exception to the ALJ’s finding about the alleged threat to England, the Board overturned that finding and determined the conversation violated section 8(a)(1). In taking this action, the Board disregarded both the statute and its own regulations, which vouchsafe to parties appearing before it a full opportunity to be heard on questions that the Board is deciding. Even if the statute were not implicated, the rules afford procedural rights, which the Board may not ignore. See Morton v. Ruiz, 415 U.S. 199, 235 , 94 S.Ct. 1055, 1074 , 39 L.Ed.2d 270 (1974). I would therefore hold that the alleged threat to England was not within the authority of the Board to consider. See NLRB v. International Union of Operating Engineers, Local 66, 357 F.2d 841, 845-46 (3d Cir. 1966).
Pared of its insupportable findings of fact, the Board’s conclusion of lingering effects rests on the questioning of one employee, Ritchey, and her later discharge under circumstances indicating her own par *323 tial fault. I do not discern any warrant for the Board’s conclusion that these two events have sustained since 1974 the taint of the original unfair labor practices in such a degree that even now an election cannot be held. As the ALJ demonstrated, the 1974 unfair labor practices eroded, at the most, 15 percent of the Union’s majority. Without more, I fail to perceive how the interrogation and discharge of this employee, herself partially at fault, provides substantial evidence for the view that the effects of Hedstrom’s violations had not sufficiently abated to allow a fair election by 1978.
This case is not unlike L’Eggs Products Inc. v. NLRB, 619 F.2d 1337 (9th Cir. 1980), where the employer allegedly engaged in threatening activities prior to a Union election. The Union lost the election and filed charges of unfair labor practices. The ALJ found that the company’s “campaign against the Union included coercive interrogations of employees [and] threats of more onerous working conditions.” The Board affirmed the findings and imposed a bargaining order. The Ninth Circuit, in ordering the Board to reconsider the bargaining order, concluded that the discharge of two employees could not provide the sole basis for the Board’s order.
In this case the Board has not attempted to assess how the interrogation and discharge of Ritchey — even the other alleged unfair practices — could possibly interfere with the freedom of choice of employees in a Board-supervised, secret election. Indeed, no possible effect could be perceived. The ALJ made an assessment and, reached a negative conclusion. The Board’s conclusion on the necessity for a bargaining order is not grounded on the evidence but on sheer unsupported conjecture. In criticizing the reliance on assumptions underlying the Board’s regulation of campaign tactics, Judge Skelly Wright, writing for the D.C. Circuit, observed that the Board was suffering
from the obvious self-justifying tendency of an institution which in over 30 years has itself never engaged in the kind of much needed systematic empirical effort to determine the dynamics of an election campaign or the type of conduct which actually has a coercive impact. The public interest need for such an empirical investigation into the assumptions underlying the Board’s regulation of campaign tactics has for some time been recognized by labor law scholars.
Getman v. NLRB, 450 F.2d 670, 675-76 (D.C.Cir.1971) (footnotes omitted).
The majority finds support for the imposition of the bargaining order in Electrical Products Division of Midland-Ross v. NLRB, 617 F.2d 977 (3d Cir. 1980). I believe Midland-Ross involved a much different factual context and is easily distinguished from this case. In Midland-Ross the Company announced to all employees only two days before the election that it was closing one of its plants because of unprofitability. The Company then stated its belief that unionization leads to lower profitability. The clear implication of the Company’s statement was that a Union victory could mean an eventual plant shutdown. The shutdown of the other plant lent credence to management threats. In this case there was no other plant shutdown. Unlike the record in Midland-Ross, supra, id. at 987, there is no evidence whatsoever that Hedstrom’s workers presently fear any closedown because of unionization.
The crucial facts of this case bear considerable similarity to those in Rapid Manufacturing Co. v. NLRB, 612 F.2d 144 (3d Cir. 1979). In Rapid Manufacturing the Board imposed a bargaining order on a company which had committed two unfair labor practices during an election campaign. This court denied enforcement, stating:
As we have previously observed, by no means can we regard two unfair labor practices, such as these, committed during the course of a two month election campaign, as being “pervasive.” Nor does the evidence reveal that this was an “extraordinary” ease by any standard.
612 F.2d at 149 . Similarly, in this case, the Board relied on several unfair labor prac *324 tices which, by no stretch of the imagination, can be labeled “pervasive.”
In the instant case, the Board again has mechanistically imposed a bargaining order, the effect of which is to mandate upon non-consenting employees 5 a labor organization as the exclusive bargaining agent. Such a draconian remedy should be justified by something more substantial than the traditional deference to the Board’s claim of expertise in determining whether campaign tactics are likely to interfere with an employee’s freedom of choice. This claimed expertise of the Board, however, is more mythical than real and has caused labor scholars to question the ability of the Board to determine what actually influences the voters’ decision. See Getman and Goldberg, The Myth of Labor Board Expertise, 39 U.Chi.L.Rev. 681 (1972); Samoff, NLRB Elections: Uncertainty and Certainty, 117 U.Pa.L.Rev. 228 (1968).
The Board itself has based its findings concerning the impact of various campaign tactics largely on its own speculation and has never conducted an empirical study to determine the impact of various campaign techniques on voting behavior. Recently, however, some empirical research has been conducted. The preliminary results of this research indicate that campaigns have little impact on voting behavior and that the impact they may have is often just the opposite of that which the Board has assumed they would have.
Harlan # 4 Coal Co. v. NLRB, 490 F.2d 117 , 123 n.5 (6th Cir.), cert. denied 416 U.S. 986 , 94 S.Ct. 2390 , 40 L.Ed. 763 (1974).
I therefore conclude that the Board’s bargaining order lacks the support of substantial evidence and is a misapplication of Gissel. I would not enforce it. I would also deny enforcement of the Board’s order as to all other unfair labor practice violations except for an order reinstating Ritchey with back pay. As to her, I would affirm the Board’s order.
WEIS and GARTH, Circuit Judges, join in this dissent.
. As for the subsidiary issue of the reinstatement of Ritchey with back pay, I concur with the majority. I disagree with the majority’s enforcement of the remaining unfair labor practices.
. Although our standard of review is not altered when the Board reaches a conclusion different from that reached by the ALJ, the Supreme Court in Universal Camera Corp. v. NLRB, 340 U.S. 474, 496 , 71 S.Ct. 456, 468 , 95 L.Ed. 456 (1951), has noted that evidence supporting the Board’s conclusion may be viewed as being “less substantial” in such circumstances. NLRB v. PPG Industries, 579 F.2d 1057, 1058 (7th Cir. 1978).
. The majority also alludes to a fifth violation found by the Board: Hedstrom had allegedly promulgated new work rules without consulting the Union. But the Board did not cite that violation as support for the bargaining order. Furthermore, in the absence of a collective bargaining agreement or valid bargaining order at the time, there was nothing to preclude the company from adopting such work rules— rules which were substantially similar to those which had been in existence for many years.
. I think I (Casteel) said “You’re drunk and you don’t know what you’re saying.” “ * * * because if he wouldn’t have been drunk, I don’t think he would have been saying it.’’ (Emphasis added.)
. The number of employees in the bargaining unit in the first election was 249. Hedstrom Co. v. NLRB, 558 F.2d at 1141 n.6. At the time of the hearing in these proceedings, there were 312 employees in the bargaining unit, each of whom would be directly affected by the bargaining order.
The majority suggests that the Board need not consider the impact of employee turnover because to do so would “put a premium upon continued litigation by the employer.” Maj. op. at 312 quoting NLRB v. L. B. Foster Co., 418 F.2d 1 (9th Cir. 1969). Whatever justification this rationale may have in cases such as L. B. Foster, where the delay is solely the result of proceedings before the Board, is lacking in cases such as this where a major cause of delay was an error by the Board which necessitated a remand by the court of appeals. Requiring the Board to consider subsequent events in cases such as this not only promotes fairness to the employees but places a “premium” upon correct original decisions by the Board; it therefore does not unjustly reward an employer’s protraction of the administrative proceedings. In fact, Chromalloy Mining and Minerals, Etc. v. NLRB, 620 F.2d 1120 (5th Cir. 1980), which is heavily relied upon by the majority, see Maj. op. at 312, 312 n.8, recognizes that the Board should consider subsequent events, such as employee turnover, after a case has been remanded to it by the court of appeals. Chromalloy, at 1131 citing NLRB v. American Cable Systems, 427 F.2d 446 (5th Cir.), cert. denied, 400 U.S. 957 , 91 S.Ct. 356 , 27 L.Ed.2d 266 (1970); see “After All, Tomorrow is Another Day”: Should Subsequent Events Affect the Validity of Bargaining Orders?, 31 Stan.L.Rev. 505 (1979). Thus, since a major cause of the delay in this case was the Board’s own failure to meet our Armcor standards, see Hedstrom I, supra, we should not condone the Board’s failure to examine the impact of events which transpired during the course of that delay.