This is a petition for review and a cross-petition for enforcement of an order of the National Labor Relations Board adopting the findings of an administrative law judge (Kleinman, J.), who concluded that New York University Medical Center ("NYU Medical Center") violated Sections 8(a)(1) and 8(a)(3) of the National Labor Relations Act, 29 U.S.C. §§ 158(a)(1) and (a)(3), by laying off ten psychiatrists at Bellevue Hospital and by threatening adverse consequences if they continued their participation in a labor organization. The ALJ found that NYU Medical Center's unit chiefs were not "supervisors" under the Act and ordered (inter alia ) that NYU Medical Center reinstate the unit chiefs and Staff Psychiatrists with back pay and full benefits. On appeal, NYU Medical Center (1) argues that each of the doctors was terminated for demonstrated administrative and clinical deficiencies, and that there is no substantial evidence to support the NLRB's finding that NYU Medical Center's actions were motivated by animus against the labor organization; and (2) that in any event the four unit chiefs were supervisors within the definition of the Act, and therefore outside the Act's protection. NYU Medical Center also contends that the statements characterized as threats by the NLRB were no more than predictions of likely consequences flowing from the Association's recalcitrance on certain issues. For the reasons set forth below, each petition is granted in part and denied in part.

NYU Medical Center, a subdivision of New York University, provides physicians' services and other medical services to Bellevue Hospital under an Affiliation Agreement with New York City Health and Hospitals Corporation ("HHC"), which owns and operates Bellevue. Before the layoffs at issue, Bellevue's Inpatient Department was divided into 13 separate psychiatric wards, called "Units," each with a unit chief, two to six doctors, and an average of 25 beds. Each unit was also served by nurses, social workers, therapists, and maintenance people, but these classes of employees had their own, in some ways more clearly delineated, lines of authority.

What follows are brief descriptions of the key participants in the case. Dr. Trujillo is the Director of Psychiatry at Bellevue. Dr. Castaneda, the Director of Inpatient Services, reported to Dr. Trujillo. Dr. Castaneda headed a staff of approximately 50 physicians, both psychiatrists and other medical specialists, including unit chiefs. The complainants in this action are six of the ten doctors laid off on March 25, 1995: Drs. Kermani, Geller, Graham, and Portnow were unit chiefs, while Drs. Mahon and Steiner were attending psychiatrists.

Dr. Kermani had been employed at Bellevue for 25 years, and was board certified in psychiatry and neurology, child psychiatry, and forensic psychiatry. He was unit chief of a combined medical-psychiatric unit and had substantial expertise in the treatment of patients with AIDS.

Dr. Portnow had been employed at Bellevue for 26 years, 21 years as a unit chief. Widely published and well-known as a lecturer, he had expertise in several subspecialties of psychiatry, and was board certified in psychiatry, forensic psychiatry, and neurology. Prior to the layoffs, he was unit chief of an inpatient adult unit.

Dr. Geller was employed at Bellevue for 27 years, 21 years as a unit chief. Dr. Geller was chief of the teaching unit.

Dr. Graham was employed by NYU Medical Center for 27 years, 19 as a unit chief. He was board certified in psychiatry and neurology.

Dr. Mahon worked in a unit that was closed as a result of the budget cuts that precipitated this case. Shortly before her termination, she received a favorable evaluation.

Dr. Steiner worked with Dr. Kermani. Dr. Wallach testified that Steiner had been transferred there after experiencing problems in another unit.

In 1973, the psychiatrists at Bellevue formed the Association of Staff Psychiatrists, Bellevue Hospital, for the purpose of addressing such matters as working hours, working conditions, salaries, and member grievances. At the time of the layoffs, the Association's membership included 45 psychiatrists, and the Executive Board consisted of Dr. Graham, president; Dr. Kermani, vice president; Dr. Geller, faculty representative; Dr. Steiner, treasurer; and Drs. Portnow and Mahon, members at large.

Soon after Dr. Trujillo became the director of psychiatry in October 1991, he began talking to Dr. Geller about the Association. Trujillo told Geller that he "had heard very good things about [Geller] and about the unit [Geller] worked on," that the unit "was one of the finer units at the hospital, [and] that [Geller] was one of the finest physicians." Trujillo then added that he was distressed that the doctors would put themselves in a position "adversarial" to Bellevue management and (in particular) himself. When Geller indicated that Association members regarded their association as being somewhat of a union, Trujillo stated that there could be no such thing as a union of physicians at a hospital and that "such a thing should cease to be."

In May 1993, Bellevue Medical Director Dr. David Cohen phoned Dr. Kermani at home after one of Kermani's patients committed suicide by jumping out of a window at Bellevue. Dr. Kermani, under the mistaken impression that the caller was a resident on duty, told him that dealing with the family was an administrator's job, and hung up. Dr. Kermani did drive to the hospital almost immediately, however, and later apologized to Dr. Cohen. Pamela Brier, Executive Director of Bellevue, and Dr. Cohen reacted initially by demanding that Kermani be fired. They simmered down, however, and a compromise was reached whereby Kermani would be counselled by Trujillo, but remain.

In the counselling sessions, Dr. Trujillo told Dr. Kermani that the Association had given him "nothing but a headache," and denounced its president, Dr. Graham as a "son-of-a-bitch," who had written "idiotic" letters. Trujillo told Kermani that the Association made too many demands and that Kermani should "disassociate" himself from Drs. Graham and Kirshbaum (then president and secretary of the Association, respectively).

Dr. Kermani reported Dr. Trujillo's statements to Dr. Graham, who called an emergency meeting of the Association. Shortly after the meeting, Dr. Trujillo reprimanded Kermani for disseminating confidential information arising out of a counselling meeting, and placed him on a six-month probation.

On June 17, 1993, the Association filed an unfair labor practices charge on Dr. Kermani's behalf, alleging that NYU Medical Center was retaliating against him for his Association activities when he was placed on probation. Based on that charge the NLRB's Regional Director filed a complaint. An informal settlement agreement between the parties provided for, among other things, revocation of Dr. Kermani's probation.

Shortly after this disposition, Dr. Trujillo told Dr. Kermani that he had always believed that Kermani was a "conscientious and good doctor," but warned "it is better to disassociate yourself from Kirshbaum and Graham." When Kermani demurred, Trujillo (according to Kermani's version) became angry and stated "Then I will have to shoot them through you ... next time there will be no association to go to for cover."

When Dr. Trujillo learned in October 1994 that Dr. Graham was seeking reelection as the Association's president, Trujillo advised Dr. Portnow, "[y]ou guys don't know what's in your best interest." Trujillo called Graham "incompetent" and the Association "useless," then asked Portnow to use his influence to defeat Graham and elect Dr. Kirshbaum. Following Graham's reelection, Trujillo accused Portnow of "fail[ing]" him because he did not convince Kirshbaum to challenge Graham.

In late 1994, Trujillo told Dr. Geller that he had "high esteem" for him, but "that if he had his way, he would be glad to get rid of people like Kermani, Graham, Mahon and Kirshbaum." On March 9, 1995, Trujillo told Portnow that "the Association interfered with [my] ability to run the department and the department would be better off without the Association."

For years, HHC had expressed dissatisfaction with a practice under which certain psychiatrists employed at Bellevue worked on-site only from 9:00 a.m. to 3:00 p.m. In a memo to Dr. Trujillo dated September 30, 1994, Pamela Brier, who as Executive Director of Bellevue was an HHC employee, warned that because of a budgetary crisis, "[e]ffective November 1, 1994, all full-time staff must fulfill the obligation explicitly delineated in the affiliation contract to work at least 35 hours per week."

NYU Medical Center convened several meetings with staff psychiatrists to discuss the 9-to-3 issue. At the September 30, 1994 meeting, Trujillo announced that because of budget problems, all psychiatrists had to work from 9:00 until 4:30 or take a cut in pay. Members of the Association's Executive Committee spoke out against the change in hours. Dr. Mahon waspishly suggested that, as a cost-cutter, clinical work be assigned to administrators such as Drs. Wallack (the Medical Director of Bellevue who was employed by HHC), Castaneda, Levy, and Varas. Dr. Kermani suggested that the psychiatrists should not be "victimized" but instead everyone, including Dean Scotch of the medical school and Trujillo, should take a cut in salary. Dr. Kermani's comments provoked Dean Scotch, who said that Kermani "was the whole problem, that if Kermani wasn't leading the opposition, or leading the riot, ... there wouldn't be any problem." Dr. Graham commented that he thought the 9-3 schedule was protected by the provision in the affiliation contract that on-call time or preparation time be counted towards the psychiatrists' hourly obligations.

At the second staff meeting on the 9-3 issue, a few days later, Brier flatly stated that because of budgetary constraints, psychiatrists would no longer be allowed to work 9-to-3. Dr. Graham complained that the 9-3 doctors were being discriminated against and that the Association was going to look into the matter. Dr. Mahon said that she (and others) had been specifically hired to work the shorter hours at a commensurately reduced salary and questioned how NYU Medical Center could require increased hours or a further reduction in salary. Brier responded that the 9-3 issue was no longer negotiable. Mahon continued to try to attract Brier's attention by raising her hand; Brier said to her, "You're trying to cause me trouble, aren't you?"

At the third meeting on the 9-3 question, Trujillo made the now-familiar announcement that the psychiatrists would have to work longer hours. Dr. Mahon repeated her suggestion that the administrators pitch in and perform some clinical work. When Dr. Wallack began to address the suggestion, Dr. Trujillo hushed him and said "don't even respond to her."

After the third meeting, Trujillo described Mahon's conduct to Portnow as "disgraceful, disloyal, undesirable" and predicted that it would "lead to trouble" for the psychiatrists. Trujillo also told Portnow following this meeting that "the Association would be punished for its position on the 9-3 issue." Trujillo told another member of the Association's executive committee, Dr. Weiss, that during a budget shortfall "it was not a good idea ... to have an open issue" like the 9-3 issue because "it makes you very visible for the chopping block."

After all these discussions, Dr. Geller asked Dr. Trujillo if he could continue to work his usual 7:30 to 3:00 schedule, and Trujillo said that because of his esteem for Geller, he would try to arrange it. But when Trujillo learned that Geller had repeated this conversation to Kermani, Trujillo exploded, declined to "do what I promised," and responded to Geller's remonstrances: "You ... people don't give a damn about the future of the department. All you care about is yourselves."

On March 27, 1995 (i.e., a day or two after terminating many of the physicians who vocally objected to the change in hours), NYU Medical Center issued a memorandum announcing that as of April 3, all psychiatrists would work 9:00 to 4:30 p.m. or resign.

In the fall of 1994, HHC notified the NYU Medical Center Administrator for Affiliation Contracts that the mental health portion of the affiliation contract budget at Bellevue had to be reduced by approximately $2 million annually. NYU Medical Center administrators, including Dr. Trujillo, strenuously argued against the reduction; but in the end, HHC insisted on the budget cut and determined that it could be achieved only by laying off psychiatrists.

In late 1994 and early 1995, NYU Medical Center administrators developed a reorganization program to achieve the inevitable reduction in staff via a "treatment plan" to rapidly stabilize and release patients. Under this plan, staff cuts were mainly in the inpatient units; one of them was to be closed and others were to be combined, reorganized or redirected.

The number of psychiatrists slated for lay-off was ten. Drs. Castaneda and Trujillo, Deputy Director of the Department of Psychiatry Wallack, and Senior Clinical Coordinator Levy, along with other administrators, participated in several meetings to determine who would go; the final list included Drs. Graham, Kermani, Geller, Portnow, Mahon and Steiner. On March 22, 1995, each of them was summoned to a meeting with Seltzer, Trujillo, and Wallack, and given a letter terminating his or her employment, for budgetary reasons.

This Court reviews the decisions of the NLRB for substantial evidence. However, "[t]he 'substantial evidence' standard does not leave factual questions wholly to the NLRB; to the contrary, it requires [courts] to take account of the evidence that undermines the NLRB's conclusions." Caremore, Inc. v. NLRB, 129 F.3d 365, 369 (6th Cir.1997). This Court has chastised the Board for its "disregard" of probative evidence and criticized the "practice followed all too often by the Board of rejecting evidence that does not support the Board's preferred result." Spentonbush/Red Star Cos. v. NLRB, 106 F.3d 484, 490 (2d Cir.1997). The Supreme Court has held that "[w]hen the Board purports to be engaged in simple factfinding, ... it is not free to prescribe what inferences from the evidence it will accept and reject, but must draw all those inferences that the evidence fairly demands." Allentown Mack Sales and Service, Inc. v. NLRB, 522 U.S. 359, ----, 118 S. Ct. 818, 829, 139 L. Ed. 2d 797 (1998). Our independent review of the evidence before the Board in the present case leads to a confirmation of most of the Board's conclusions.

* Under section 8(a)(1) of the National Labor Relations Act, it is an unfair labor practice for an employer "to interfere with, restrain, or coerce employees in the exercise of" their rights to form, join, or assist labor organizations.1 29 U.S.C. § 158(a)(1). An employer's conduct violates section 8(a)(1) if, under all the existing circumstances, the conduct has a reasonable tendency to coerce or intimidate employees, regardless of whether they are actually coerced. See Irving Air Chute Co. v. NLRB, 350 F.2d 176, 178-79 (2d Cir.1965). Any assessment of statements made by an employer to its employees "must take into account the economic dependence of the employees on their employers, and the necessary tendency of the former, because of that relationship, to pick up intended implications of the latter that might be more readily dismissed by a more disinterested ear." NLRB v. Gissel Packing Co., 395 U.S. 575, 617, 89 S. Ct. 1918, 23 L. Ed. 2d 547 (1969). It is well-settled that threats of layoffs or other adverse economic consequences violate section 8(a)(1) if they are motivated by or conditioned upon an employee's participation in a labor organization. See HarperCollins San Francisco v. NLRB, 79 F.3d 1324, 1329 (2d Cir.1996); NLRB v. Pace Oldsmobile, Inc., 681 F.2d 99, 100 (2d Cir.1982) (per curiam ).

Section 8(a)(3) of the Act makes it an unfair labor practice for an employer to discriminate "in regard to hire or tenure of employment or any term or condition of employment to encourage or discourage membership in any labor organization.... " An employer violates section 8(a)(3) by firing an employee for engaging in union activity. See NLRB v. Transportation Management Corp., 462 U.S. 393, 397, 401-03, 103 S. Ct. 2469, 76 L. Ed. 2d 667 (1983). Similarly, an employer violates section 8(a)(3) by taking action against employees who engage in "concerted" activity for their mutual aid and protection. See Metropolitan Edison Co. v. NLRB, 460 U.S. 693, 703, 103 S. Ct. 1467, 75 L. Ed. 2d 387 (1983).

We conclude that, except as to Dr. Steiner (discussed in Part II of this opinion), substantial evidence supports the NLRB's conclusion that NYU Medical Center both (a) threatened Portnow, Geller, Graham, Kermani and Mahon with adverse employment action as a result of their protected union activity and (b) terminated these physicians' employment on the basis of their involvement in protected activities. Of course, a competing inference is available; NYU Medical Center vigorously urges that the layoffs were motivated by budgetary and performance considerations. However, in light of the administrators' marked, explicit hostility toward the Association, evidenced by numerous comments and veiled threats by Dr. Trujillo over a period of time, the NLRB's conclusion does rest on substantial evidence. This conclusion is reinforced in part by credibility determinations, which this Court is generally reluctant to disturb: the ALJ credited the accounts of Drs. Mahon, Kermani, Geller, Graham and Portnow, while discrediting much of the testimony of Drs. Trujillo and Castaneda, and the NLRB endorsed those credibility determinations.

NYU Medical Center argues that the statements construed as threats by the NLRB were merely predictions, protected by the First Amendment and section 8(c) of the Act, as to what would happen if, to everyone's regret, the Association's abusive demands provoked HHC to make deeper budget cuts. Section 8(c) provides in part:

The expressing of any views, argument, or opinion ... shall not constitute or be evidence of an unfair labor practice ... if such expression contains no threat of reprisal or force or promise of benefit.

29 U.S.C. § 158(c). Moreover, an employer is free to make a "prediction" as to the "demonstrably probable consequences" of protected activity. See Gissel Packing Co., 395 U.S. at 618, 89 S. Ct. 1918.

The statements in question (as detailed in the Background section of this opinion) are outside this safe harbor if the complaining doctors' version of those statements is credited, as it was. Permissible predictions need to be phrased so that the speaker's prognostications clearly rest on objective facts outside the employer's control. Here, the perceived threats were not couched in terms of what HHC might do. Moreover, HHC did not select the doctors for lay-off; that selection was made entirely by Drs. Trujillo and Castaneda, among other administrators, acting on behalf of NYU Medical Center. Trujillo's statements, as recounted by the complainants, did "threaten a reprisal" in the event that the Association did not toe the line. While Dr. Trujillo's version of the statements renders them innocuous, the NLRB did not credit Dr. Trujillo's account of his remarks.

Dr. Steiner did not testify at the hearing before the ALJ. The NLRB's finding of a violation as to him rests solely on his membership on the Association's Executive Board and the single comment by Dr. Trujillo coupling Steiner with the more vocal psychiatrists. However, not all the Board members were terminated. Board members Kirshbaum and Weiss, for example, were retained, and Dr. Trujillo's comments about them were more numerous and more hostile than anything he said about Dr. Steiner. Further, Dr. Steiner does not appear to have been an activist or a gadfly, and there is no substantial evidence that he even heard the threats that were supposedly intended to intimidate him.

Moreover, Dr. Steiner's license to practice medicine has been revoked so that reinstatement and back pay are inappropriate in his case. See In Matter of Steiner v. De Buono, 239 A.D.2d 708, 657 N.Y.S.2d 485, 486-87 (3d Dep't 1997) (upholding revocation of Steiner's license based on patient's testimony that Steiner performed fellatio on him as part of his therapy, met with him in public restaurants and "resort locations" and had induced him to lend Steiner money on three occasions).

NYU Medical Center's best hope on this appeal is that, whatever its conduct, the sections of the Act cited by the NLRB do not protect "supervisors," and that Drs. Kermani, Geller, Graham, and Portnow were supervisors by virtue of being unit chiefs. This Court and several others have noticed the NLRB's tendency to narrow artificially the category of supervisors, and have concluded therefore that more strict review is required in cases where the Board must determine whether complainants are supervisors. See Spentonbush, 106 F.3d at 492 ("the Board's biased mishandling of cases involving supervisors increasingly has called into question our obeisance to the Board's decisions in this area."); Schnuck Markets, Inc. v. NLRB, 961 F.2d 700, 704 (8th Cir.1992)("[S]crutiny is particularly appropriate in cases where the Board determines supervisory status .... [because] our review necessarily becomes more probing when the Board has exhibited a pattern of applying the statute inconsistently."); Children's Habilitation Ctr., Inc. v. NLRB, 887 F.2d 130, 132 (7th Cir.1989) ("More important than verbal niceties in the standard of review is judicial impatience with the Board's well-attested manipulativeness in the interpretation of the statutory test for 'supervisor.' ").

Nonetheless, as a nondeferential review confirms, there is substantial evidence that the role of a Bellevue unit chief in recruitment, hiring, scheduling, and the other categories of supervisory responsibility was negligible. According to the psychiatrists' testimony, which the ALJ could credit (and did), the attending physicians who staffed the Units generally functioned as colleagues of the unit chief rather than as their subordinates. Other medical professionals (such as nurses) and other service personnel (such as social workers) functioned within their own, entirely separate, lines of authority.2

NYU Medical Center argues that the burden is on the NLRB to prove all elements of a section 8 claim, and that the NLRB therefore should have to prove that the unit chiefs are employees rather than supervisors. See Health Care & Retirement Corp. of America v. NLRB, 987 F.2d 1256 (6th Cir.1993) (vacating order where NLRB failed to meet its burden of proving substantial evidence of non-supervisory status), aff'd on diff. grounds, 511 U.S. 571, 114 S. Ct. 1778, 128 L. Ed. 2d 586 (1994); NLRB v. Beacon Light Christian Nursing Home, 825 F.2d 1076, 1080 (6th Cir.1987). It is evident, however, that supervisory status is a statutory exception, separately stated, to the provisions of the Act that extend its coverage to all employees. The applicable rule of statutory construction is that the party claiming the benefit of such an exception must demonstrate its applicability. See FTC v. Morton Salt Co., 334 U.S. 37, 44-45, 68 S. Ct. 822, 92 L. Ed. 1196 (1948). In other words, the unit chiefs are employees, and they may or may not also be supervisors, which would take them outside the ambit of the Act. The NLRB bears--and has carried--its burden of demonstrating that the unit chiefs are employees of NYU Medical Center; the burden of establishing that they are supervisors is on the respondent. See NLRB v. Bakers of Paris, Inc., 929 F.2d 1427, 1445 (9th Cir.1991) (noting that "burden of proving supervisory status rests upon the party asserting it"). This conclusion is reinforced by the employer's natural advantage in adducing proof as to how it organizes its operations and personnel.

As we have said, our review of the Board's adoption of the ALJ's finding on this point is searching; and the employer's statutory burden is light. Section 2(11) of the NLRA defines a supervisor as:

any individual having the authority, in the interests of the employer, to hire, transfer, suspend, lay off, recall, promote, discharge, assign, reward, or discipline other employees, or responsibly to direct them, or to adjust their grievances, or effectively to recommend such action, if in connection with the foregoing the exercise of such authority is not of a merely routine or clerical nature, but requires the use of independent judgment.

29 U.S.C. § 152(11). It is well settled that section 2(11) is to be read in the disjunctive, and "an employee who in the course of employment uses independent judgment to engage in one of the 12 listed activities, including responsible direction of other employees, is a supervisor." Health Care & Retirement, 511 U.S. at 579, 114 S. Ct. 1778 (emphasis added).

There was evidence that the unit chiefs exercised some degree of authority over attending physicians within their individual units. For example, Dr. Graham admitted that he told the attendings when their work needed improvement. Dr. Kermani wrote a series of warning letters to one of his attending physicians. Similarly, Dr. Geller admitted that he evaluated the attending physicians on his unit. Dr. Portnow indicated "I have spent a huge amount of time supervising Dr. Gray and seeing over his shoulder." The unit chiefs admitted that their signatures were required on the attendings' vacation request forms. The unit chiefs expressed their staffing needs and, on occasion, participated to some degree in the selection of attending physicians. Dr. Graham testified that he often mediated disputes, although he would refer a problem nurse or social worker to the authority within those disciplines if his mediation efforts failed.

This showing by NYU Medical Center is offset however, by evidence that the unit chiefs actually exercised surprisingly little control over such matters. The unit chiefs consistently denied that they had authority to hire or fire employees. The credited evidence indicates that the unit chiefs did not usually meet with psychiatrists assigned to their unit until after the hiring. On one occasion, Dr. Graham met with a candidate as a courtesy, but he had seen none of the applicant's documentation, so the interview was not highly substantive.

At one time, the unit chiefs gave performance evaluations to the attending physicians; but the record allows the finding that they exercised no independent judgment in carrying out this function. Their testimony, which the NLRB credited, was that Dr. Castaneda instructed them as to the range of grades to be given, or discarded the evaluations and redid them himself. Evaluations that do not affect job status of the evaluated person are inadequate to establish supervisory status. Highland Superstores, Inc. v. NLRB, 927 F.2d 918, 922-23 (6th Cir.1991); NLRB v. Res-Care, Inc., 705 F.2d 1461, 1467 (7th Cir.1983). There was no evidence that any of the evaluations written by the unit chiefs had any effect on the employment of the attending physicians.

Dr. Kermani sent some warning letters to attending physicians, but his testimony, which was credited, was that these letters were dictated by Castaneda. Kermani testified that he exercised no independent judgment in performing this task. Similarly, the evidence accepted by the NLRB demonstrated that the unit chiefs had no control over assignment of patients, which was done by a nurse or clerk pursuant to a system agreed upon by all of the psychiatrists in the unit.

NYU Medical Center's strongest evidence of supervisory authority is the position description of the unit chief position: the first duty listed is to "[s]upervise[ ] unit psychiatrists and other physicians .... [e]valuate[ ] their performance, [and] recommend[ ] hiring and termination." However, the document is undated, and none of the unit chiefs had ever seen it. Theoretical or paper power does not a supervisor make. See Food Store Employees Union Local 347 v. NLRB, 422 F.2d 685, 690 (D.C.Cir.1969).

In sum, substantial evidence supports the Board's conclusion that the unit chiefs, their title notwithstanding, did not function as supervisors. We agree with NYU Medical Center that competing inferences and opposing evidence are fairly present in the record, but the choice among permissible competing inferences is left to the Board.

The NLRB is affirmed and the cross-petition for enforcement granted as to Drs. Graham, Geller, Kermani, Portnow, and Mahon. The NLRB is reversed and the cross-petition for enforcement is denied as to Dr. Steiner.