National Labor Relations Board v. American Medical Services, Inc.

POSNER, Circuit Judge. This is a companion case to NLRB v. Res-Care, Inc., 705 F.2d 1461 (7th Cir.1983), decided today, and familiarity with our opinion in that case is assumed. The Board ordered this respondent, which operates a 240-bed nursing home in Wisconsin, to bargain collectively with a union representing a unit consisting of 17 registered nurses employed at the home. The respondent refused to bargain, on the ground that these nurses are supervisors within the meaning of section 2(11) of the National Labor Relations Act, 29 U.S.C. § 152(11). The Board ordered the respondent to bargain and asks us to enforce its order. The nursing home in this case is much larger than the one in Res-Care. Not only does it have more than three times as many beds, but it has 235 employees, compared to 52 in Res-Care. Of the 235, 165 are in the nursing department, and a detailed breakdown is available only for them. The 165 include 6 registered nurses who are conceded to be supervisors, the 17 other registered nurses whose status is in dispute (and whom we shall call “the registered nurses,” ignoring the conceded supervisors), and between 140 and 145 licensed practical nurses and nurse’s aides — all of whom are conceded in this case (though not in Res-Care ) to be nonsupervisory employees. An evident difference between the cases is in the ratio of supervisors to employees if the Board’s position is upheld: 1 to 8 in Res-Care, only 1 to 27 here (1 to 6 if the respondent prevails). Since, as we explained in Res-Care, the purpose of excluding supervisors from the protections of the National Labor Relations Act is to give the employer a solid cadre of employees who owe undivided loyalty to him, this discrepancy in ratios in two facilities that are similar except in size raises a warning flag. Another factual difference between the cases involves the amount of supervision exercised during the evening and night shifts by the nurses whose status is in question. In Res-Care those nurses, who were licensed practical nurses rather than registered nurses, were as here the highest-ranking employees on the premises during the evening and night shifts. But the number of other employees present was small (3 to 5 nurse’s aides, and an unknown but presumably small number of nonmedical employees), and there was no evidence that the licensed practical nurses exercised any significant supervisory authority — their supervisors being only a phone call away. In this case, the highest-ranking employees on the premises during the evening and night shifts are the registered nurses, and although their supervisors carry beepers and can thus be reached at any time, the evidence, for which there is no counterpart in the record of the Res-Care case, is that the registered nurses on the premises really are in charge — months may pass before any of them gets in touch with their supervisor during the evening or night shift — and of a large professional staff: 21 to 22 licensed practical nurses and nurses’ aides on the evening shift, 9 to 12 on the night shift. Of course, as we emphasized in Res-Care, supervision is a technical concept in section 2(11). It does not mean directing the work of other employees; there must be some control over their employment prospects or conditions. However, the record in this case shows that a registered nurse once discharged another employee on the spot for leaving a patient unattended on the [1474] toilet for several hours, and although it is true that the discharge was confirmed the next day by her supervisor, that it is the only discharge by a registered nurse in the record, and that the provocation was extreme, there is no doubt she did fire this employee — not merely report her to a supervisor as in Res-Care. This was an unequivocal act of supervision within the meaning of section 2(11), and the fact that the nurse acted out of concern for the patient did not make her act professional rather than supervisory. Simply ordering the employee from the premises for the sake of the patients might have been a professional, nonsupervisory decision. But firing an employee is inherently supervisory. And the power to fire is significant to the purposes of section 2(11) even if rarely exercised. An employer who delegates to an employee the power to fire other employees has a right to demand that employee’s undivided loyalty. He would not have it if the employee could join a union, with all the protections that the Act gives employees engaged in concerted activity for mutual aid or protection, for he might then have a difficult choice between carrying out his obligation to the employer and maintaining solidarity with his fellow union members. The record also contains an instance where a registered nurse suspended an employee without — as in Res-Care — consulting her superior. Moreover, the registered nurses fill out employee evaluation forms that according to undisputed evidence can lead to terminating probationary employees; in Res-Care the forms were new and their significance uncertain. The Regional Director’s statements that “it is uncontested that RN’s do not by the use of the evaluation forms recommended pay changes or discipline” and that “the record does not establish that the RN is involved in any discussion concerning the decision to refrain from or to issue discipline” are not supported by substantial evidence. The nurse who was the principal union witness contradicted the second statement, and did not support the first; the employer introduced substantial testimony contradicting both statements. The record also shows that the registered nurses have power over work assignments that goes beyond the ministerial responsibilities of the licensed practical nurses in Res-Care. Although the employer sets minimum staffing levels, the registered nurses are authorized to operate at lower levels when patient care would not be compromised and to call in additional staff when the needs of patient care dictate. The licensed practical nurses in Res-Care authorized overtime but had no authority to decide who would get the overtime work; they just went down a list of employees that was supplied them by their superiors. Although one of the registered nurses who testified for the union in this case said she “just basically” goes down the list, and “often” goes right down the list, other registered nurses — also union witnesses — testified without equivocation that the registered nurses decide “who is called” and have discretion “to figure out who can come in and wants to pick up some time.” The Regional Director’s unqualified statement that “RN’s will call employees on a preestablished list to inquire whether they will come in to work” thus is not supported by substantial evidence either. The potential for conflicts of interest in allowing employees who control staffing levels to engage in protected activity would be substantial. Such an employee would be caught between his obligation to the employer to minimize staffing levels consistently with the needs of the patients and his obligation to his union to see that his fellow union members were fully employed. The registered nurses in this case thus wield the carrot (overtime for favored employees) and the stick (discharge, suspension, adverse employee evaluations). The extremely low ratio of supervisors to employees that would result if the Board’s order were enforced, the authority of the registered nurses to discharge and otherwise discipline other employees, and their discretion over who works when — all factors not present in Res-Care, in Misericordia Hospital Medical Center v. NLRB, 623 F.2d 808, 816-18 (2d Cir.1980), or in any other case we have found upholding the [1475] Board’s certification of a nurses’ collective bargaining unit (Beverly Enterprises v. NLRB, 661 F.2d 1095, 1098-1100 (6th Cir.1981), is close, but the nurses alleged to be supervisors there had no power of discharge) — convince us that the Board has improperly certified a unit of supervisors. Paying attention as we must to the purpose behind section 2(11) of preventing serious conflicts of interest, we conclude that the Board’s order is not supported by substantial evidence on the record as a whole and that enforcement must therefore be Denied.