JUSTICE O'CONNOR, with whom JUSTICE POWELL joins, dissenting.
I join THE CHIEF JUSTICE in the introductory comments and Part I of his dissent. However, I write separately because I believe labor peace would be advanced by avoiding the absolute positions adopted both by the majority and by the dissent of THE CHIEF JUSTICE. Because I am convinced that the Board should examine the circumstances surrounding and following an impasse to determine whether an unusual circumstance sufficient to justify withdrawal has occurred, and because I cannot accept the Court's conclusory statements concerning the effects of all interim agreements, I respectfully dissent.
I
The Court agrees with the Board that an impasse is not an unusual circumstance "sufficiently destructive of group bargaining to justify unilateral withdrawal." The Board adopted that position after identifying an impasse as (1) simply a "temporary deadlock or hiatus in negotiations" (2) which may be brought about intentionally by one of the parties and (3) which in almost all cases is "eventually broken, either through a change of mind or the application of economic force." Charles D. Bonanno Linen Service, Inc., 243 N.L.R.B. 1093, 1093-1094 (1979). There are, of course, impasses that fit this description. Others do not. Unfortunately, having developed its premise, the Board has chosen to ignore the reasons which justified it, and now "reasons" that an impasse, regardless of duration, does not justify employer withdrawal. The problem with the Board's approach is that it reasons by definition. That is, while an impasse may be a temporary deadlock, a deadlock cannot be made temporary simply by calling it an impasse. Thus, while the rule may be efficient, it does not contribute to principled decisionmaking. This case provides an excellent example of the result which obtains when the Board applies a general rule without analysis of the particular factual situation.
More than a temporary lull in negotiations developed here. When Bonanno withdrew from the bargaining unit in November, the parties had been deadlocked for more than six months. Nevertheless, although the Board defines an impasse as a "temporary" deadlock, it inexplicably views the passage of time as irrelevant to the question of whether something more than a "hiatus" in negotiations was involved. [ Footnote 3/1 ]
Closely related to the Board's view of an impasse as temporary is its view that an impasse is not an unusual circumstance, because it is broken in almost all cases either through a change of mind or by the application of economic forces. However, in this case, the Board did not determine whether, when Bonanno withdrew, the parties were likely to have broken the impasse. The union and association had unsuccessfully utilized the most common economic weapons: the union had called a selective strike; association members had locked out their drivers; and Bonanno had hired replacement drivers. [ Footnote 3/2 ] The Board made no finding that the arsenal of this union or these employers contained additional economic weapons which, if used, might have ended the impasse.
Moreover, neither the Board nor any party to the negotiations suggested that Bonanno precipitated the May 15 impasse as a means to excuse its withdrawal from the association. The Board's third identifying feature of an impasse is thus also missing from this situation.
The impasse which all parties agree existed on May 15 may fit the Board's definition of impasse; the situation which existed on November 21 does not. The point is not that this Court should substitute its judgment as to the "significance of impasse and the dynamics of collective bargaining" for that of the Board. The point is that the Board should be required to analyze, not simply label, a deadlock in negotiations. If the Board had utilized its expertise to examine the facts of this case, it well might have found a complete breakdown in negotiations, not a temporary impasse. When such a complete breakdown occurs, I would afford an employer a right to withdraw.
II
Neither can I agree with the Court's conclusion that employers who execute interim agreements invariably maintain an equivalent stake in securing a final agreement, and that interim agreements always deter fragmentation of the employer unit. That conclusion, like the "impasse rule" adopted today, sweeps too broadly. The conclusion is least likely to be accurate when applied to a highly competitive industry which relies upon skilled workers and counts heavily upon repetitive patronage. If one member of a struck employer association in such an industry reaches an interim agreement, he will gain a competitive advantage sufficient to produce a natural and powerful interest in prolonging the deadlock. In fact, the Board has found that an employer with such an advantage is less likely to push for prompt settlement of the labor dispute. See, e.g., Connell Typesetting Co., 212 N.L.R.B. 918 (1974). Moreover, as we recognized in NLRB v. Brown, 380 U. S. 278 (1965), the notion that allowing a practice which unfairly advantages one employer would "succeed in breaking up the employer association was not at all fanciful." Id. at 380 U. S. 284. Likewise, when an interim agreement affords one employer a competitive advantage, the notion that allowing the agreement will promote fragmentation of the bargaining unit is not at all fanciful.
Other factors could also affect the impact of an interim agreement. For instance, an agreement between a union and the employer of 40 percent of a workforce could shatter a bargaining unit. Such an agreement with an employer of two percent of the workforce might have little effect. No magic inheres in the word "interim," as none inheres in "impasse." Identification of an agreement as "interim," rather than "final," is the beginning, not the end, of the required analysis of the agreement's effect on the bargaining unit. Yet today the Court gives blanket approval of any interim agreement. [ Footnote 3/3 ] Here too, I would require the Board to apply its expertise to determine the effect of such an agreement in a particular instance, rather than approve the Board's practice of decision by label. If an agreement, interim or final, operates to fragment a bargaining unit, I would allow withdrawal by an employer.
III
The goal of multiemployer bargaining is to promote "labor peace through strengthened collective bargaining." NLRB v. Truck Drivers, 353 U. S. 87, 353 U. S. 95 (1957). Neither a complete breakdown in negotiations nor a fragmented bargaining unit furthers that goal. Because today's decision allows both, I dissent.
Notes
[ Footnote 3/1 ]
As THE CHIEF JUSTICE notes in his dissent, the Board conceded during oral argument that it will not regard an impasse as an unusual circumstance even if it continues for two years.
[ Footnote 3/2 ]
The parties submitted this dispute on stipulated facts which do not indicate whether other association members hired temporary workers after the strike and lockout, a step which they had the right to take. NLRB v. Brown, 380 U. S. 278 (1965).
[ Footnote 3/3 ]
By adopting a per se rule that interim agreements never fragment the bargaining unit, the Court takes a position more extreme than that urged by the Board. At least until today, the Board has allowed employer withdrawal when interim agreements resulted in unit fragmentation. See, e.g., Typographic Service Co., 238 N.L.R.B. 1565 (1978); Connell Typesetting Co., 212 N.L.R.B. 918 (1974). Approval of a rule so favorable to one party to negotiations is hardly the way to encourage multiemployer bargaining or industrial peace.