with whom Mr. Justice White, Mr. Justice Marshall, and Mr. Justice Blackmun join, dissenting.
In holding that the full-time faculty members of Yeshiva University are not covered employees under the National Labor Relations Act, but instead fall within the exclusion for *692supervisors and managerial employees, the Court disagrees with the determination of the National Labor Relations Board. Because I believe that the Board’s decision was neither irrational nor inconsistent with the Act, I respectfully dissent.
I
Ten years ago the Board first asserted jurisdiction over private nonprofit institutions of higher education. Cornell University, 183 N. L. R. B. 329 (1970). Since then, the Board has often struggled with the Procrustean task of attempting to implement in the altogether different environment of the academic community the broad directives of a statutory scheme designed for the bureaucratic industrial workplace. See, e. g., Adelphi University, 195 N. L. R. B. 639, 648 (1972). Resolution of the particular issue presented in this case — whether full-time faculty members are covered “employees” under the Act — is but one of several challenges confronting the Board in this “unchartered area.” C. W. Post Center, 189 N. L. R. B. 904, 905 (1971).
Because at the time of the Act’s passage Congress did not contemplate its application to private universities, it is not surprising that the terms of the Act itself provide no answer to the question before us. Indeed, the statute evidences significant tension as to congressional intent in this respect by its explicit inclusion, on the one hand, of “professional employees” under § 2 (12), 29 U. S. C. § 152 (12), and its exclusion, on the other, of “supervisors” under § 2 (11), 29 U. S. C. § 152 (11). Similarly, when transplanted to the academic arena, the Act’s extension of coverage to professionals under §2 (12) cannot easily be squared with the Board-created exclusion of “managerial employees” in the • industrial context. See generally NLRB v. Bell Aerospace Co., 416 U. S. 267 (1974).
Primary authority to resolve these conflicts and to adapt the Act to the changing patterns of industrial relations was *693entrusted to the Board, not to the judiciary. NLRB v. Weingarten, Inc., 420 U. S. 251, 266 (1975). The Court has often admonished that “[t]he ultimate problem is the balancing of the conflicting legitimate interests. The function of striking that balance to effectuate national labor policy is often a difficult and delicate responsibility, which the Congress committed primarily to the National Labor Relations Board, subject to limited judicial review.” NLRB v. Truck Drivers, 353 U. S. 87, 96 (1957). Accord, Beth Israel Hospital v. NLRB, 437 U. S. 483, 501 (1978); NLRB v. Erie Resistor Corp., 373 U. S. 221, 235-236 (1963). Through its cumulative experience in dealing with labor-management relations in a variety of industrial and nonindustrial settings, it is the Board that has developed the expertise to determine whether coverage of a particular category of employees would further the objectives of the Act.1 And through its continuous oversight of industrial conditions, it is the Board that is best able to formulate and adjust national labor policy to conform to the realities of industrial life. Accordingly, the judicial role is limited; a court may not substitute its own judgment for that of the Board. The Board’s decision may be reviewed for its rationality and its consistency with the *694Act, but once these criteria are satisfied, the order must be enforced. See Beth Israel Hospital v. NLRB, supra, at 501.
II
In any event, I believe the Board reached the correct result in determining that Yeshiva’s full-time faculty is covered under the NLRA. The Court does not dispute that the faculty members are “professional employees” for the purposes of collective bargaining under § 2 (12), but nevertheless finds them excluded from coverage under the implied exclusion for “managerial employees.” 2 The Court explains that “[t]he controlling consideration in this case is that the faculty of Yeshiva University exercise authority which in any other context unquestionably would be managerial.” Ante, at 686. But the academic community is simply not “any other context.” The Court purports to recognize that there are fundamental differences between the authority structures of the typical industrial and academic institutions which preclude the blind transplanting of principles developed in one arena onto the other; yet it nevertheless ignores those very differences in concluding that Yeshiva’s faculty is excluded from the Act’s coverage.
As reflected in the legislative history of the Taft-Hartley Amendments of 1947, the concern behind the exclusion of supervisors under § 2 (11) of the Act is twofold. On the one hand, Congress sought to protect the rank-and-file employees from being unduly influenced in their selection of leaders by the presence of management representatives in their union. “If supervisors were members of and active in the union which represented the employees they supervised it could be pos*695sible for the supervisors to obtain and retain positions of power in the union by reason of their authority over their fellow union members while working on the job.” NLRB v. Metropolitan Life Ins. Co., 405 F. 2d 1169, 1178 (CA2 1968). In addition, Congress wanted to ensure that employers would not be deprived of the undivided loyalty of their supervisory foremen. Congress was concerned that if supervisors were allowed to affiliate with labor organizations that represented the rank and file, they might become accountable to the workers, thus interfering with the supervisors’ ability to discipline and control the employees in the interest of the employer.3
Identical considerations underlie the exclusion of managerial employees. See ante, at 682. Although a variety of verbal formulations have received judicial approval over the years, see Retail Clerks International Assn. v. NLRB, 125 U. S. App. D. C. 63, 65-66, 366 F. 2d 642, 644-645 (1966), this Court has recently sanctioned a definition of “managerial employee” that comprises those who “ 'formulate and effectuate management policies by expressing and making operative the decisions of their employer.’ ” See NLRB v. Bell Aerospace Co., 416 U. S., at 288. The touchstone of managerial status is thus an alliance with management, and the pivotal inquiry is whether the employee in performing his *696duties represents his own interests or those of his employer.4 If his actions are undertaken for the purpose of implementing the employer’s policies, then he is accountable to management and may be subject to conflicting loyalties. But if the employee is acting only on his own behalf and in his own interest, he is covered under the Act and is entitled to the benefits of collective bargaining.
After examining the voluminous record in this case,5 the Board determined that the faculty at Yeshiva exercised its decisionmaking authority in its own interest rather than “in the interest of the employer.” 221 N. L. R. B. 1053, 1054 (1975). The Court, in contrast, can perceive “no justification for this distinction” and concludes that the faculty’s interests “cannot be separated from those of the institution.” Ante, at 688.6 But the Court’s vision is clouded by its failure fully to discern and comprehend the nature of the faculty’s role in university governance.
Unlike the purely hierarchical decisionmaking structure that prevails in the typical industrial organization, the bureaucratic foundation of most “mature” universities is characterized by dual authority systems. The primary decisional *697network is hierarchical in nature: Authority is lodged in the administration, and a formal chain of command runs from a lay governing board down through university officers to individual faculty members and students. At the same time, there exists a parallel professional network, in which formal mechanisms have been created to bring the expertise of the faculty into the decisionmaking process. See J. Baldridge, Power and Conflict in the University 114 (1971); Finkin, The NLRB in Higher Education, 5 U. Toledo L. Rev. 608, 614-618 (1974).
What the Board realized — and what the Court fails to apprehend — is that whatever influence the faculty wields in university decisionmaking is attributable solely to its collective expertise as professional educators, and not to any managerial or supervisory prerogatives. Although the administration may look to the faculty for advice on matters of professional and academic concern, the faculty offers its recommendations in order to serve its own independent interest in creating the most effective environment for learning, teaching, and scholarship.7 And while the administration may attempt to defer to the faculty’s competence whenever possible, it must and does apply its own distinct perspective to those recommendations, a perspective that is based on fiscal *698and other managerial policies which the faculty has no part in developing. The University always retains the ultimate decisionmaking authority, see ante, at 675-676, and the administration gives what weight and import to the faculty’s collective judgment as it chooses and deems consistent with its own perception of the institution’s needs and objectives.8
The premise of a finding of managerial status is a determination that the excluded employee is acting on behalf of management and is answerable to a higher authority in the exercise of his responsibilities. The Board has consistently implemented this requirement — both for professional and nonprofessional employees — by conferring managerial status only upon those employees “whose interests are closely aligned with management as true representatives of management.” (Emphasis added.) E. g., Sutter Community Hospitals of Sacramento, 227 N. L. R. B. 181, 193 (1976); Bell Aero*699space, 219 N. L. R. B. 384, 385 (1975); General Dynamics Corp., 213 N. L. R. B. 851, 857 (1974).9 Only if the employee is expected to conform to management policies and is judged by his effectiveness in executing those policies does the danger of divided loyalties exist.
Yeshiva’s faculty, however, is not accountable to the administration in its governance function, nor is any individual faculty member subject to personal sanction or control based on the administration’s assessment of the worth of his recommendations. When the faculty, through the schools’ advisory committees, participates in university decisionmaking on subjects of academic policy, it does not serve as the “representative of management.” 10 Unlike industrial supervisors *700and managers, university professors are not hired to “make operative” the policies and decisions of their employer. Nor are they retained on the condition that their interests will correspond to those of the university administration. Indeed, the notion that a faculty member’s professional competence could depend on his undivided loyalty to management is antithetical to the whole concept of academic freedom. Faculty members are judged by their employer on the quality of their teaching and scholarship, not on the compatibility of their advice with administration policy. Board Member Kennedy aptly concluded in his concurring opinion in Northeastern University, 218 N. L. R. B. 247, 257 (1975) (footnote omitted):
“[T]he influence which the faculty exercises in many areas of academic governance is insufficient to inake them 'managerial’ employees. Such influence is not exercised 'for management’ or 'in the interest of the employer,’ but rather is exercised in their own professional interest. The best evidence of this fact is that faculty members are generally not held accountable by or to the administration for their faculty governance functions. Faculty criticism of administration policies, for example, is viewed not as a breach of loyalty, but as an exercise in academic freedom. So, too, intervention by the university administration in faculty deliberations would most likely be considered an infringement upon academic freedoms. Conversely, university administrations rarely consider themselves bound by faculty recommendations.”
It is no answer to say, as does the Court, that Yeshiva’s faculty and administration are one and the same because their interests tend to coincide. In the first place, the National Labor Relations Act does not condition its coverage on an antagonism of interests between the employer and the em*701ployee.11 The mere coincidence of interests on many issues has never been thought to abrogate the right to collective bargaining on those topics as to which that coincidence is absent. Ultimately, the performance of an employee’s duties will always further the interests of the employer, for in no institution do the interests of labor and management totally diverge. Both desire to maintain stable and profitable operations, and both are committed to creating the best possible product within existing financial constraints. Differences of opinion and emphasis may develop, however, on exactly how to devote the institution’s resources to achieve those goals. When these disagreements surface, the national labor laws contemplate their resolution through the peaceful process of collective bargaining. And in this regard, Yeshiva University stands on the same footing as any other employer.
Moreover, the congruence of interests in this case ought not to be exaggerated. The university administration has certain economic and fiduciary responsibilities that are not shared by the faculty, whose primary concerns are academic and relate solely to its own professional reputation. The record evinces numerous instances in which the faculty’s recommendations have been rejected by the administration on account of fiscal constraints or other managerial policies. Disputes have arisen between Yeshiva’s faculty and administration on such fundamental issues as the hiring, tenure, promotion, retirement, and dismissal of faculty members, *702academic standards and credits, departmental budgets, and even the faculty's choice of its own departmental representative.12 The very fact that Yeshiva’s faculty has voted for the Union to serve as its representative in future negotiations with the administration indicates that the faculty does not perceive its interests to be aligned with those of management. Indeed, on the precise topics which are specified as mandatory subjects of collective bargaining — wages, hours, and other terms and conditions of employment13 — the interests of teacher and administrator are often diametrically opposed.
Finally, the Court’s perception of the Yeshiva faculty’s status is distorted by the rose-colored lens through which it views the governance structure of the modern-day university. The Court’s conclusion that the faculty’s professional interests are indistinguishable from those of the administration is bottomed on an idealized model of collegial decisionmaking that is a vestige of the great medieval university. But the university of today bears little resemblance to the “community of scholars” of yesteryear.14 Education has become *703“big business,” and the task of operating the university enterprise has been transferred from the faculty to an autonomous administration, which faces the same pressures to cut costs and increase efficiencies that confront any large industrial organization.15 The past decade of budgetary cutbacks, declining enrollments, reductions in faculty appointments, curtailment of academic programs, and increasing calls for accountability to alumni and other special interest groups has only added to the erosion of the faculty’s role in the institution’s decisionmaking process.16
*704These economic exigencies have also exacerbated the tensions in university labor relations, as the faculty and administration more and more frequently find themselves advocating conflicting positions not only on issues of compensation, job security, and working conditions, but even on subjects formerly thought to be the faculty's prerogative. In response to this friction, and in an attempt to avoid the strikes and work stoppages that have disrupted several major universities in recent years, many faculties have entered into collective-bargaining relationships with their administrations and governing boards.17 An even greater number of schools — Yeshiva among them — have endeavored to negotiate and compromise their differences informally, by establishing avenues for faculty input into university decisions on matters of professional concern.
*705Today's decision, however, threatens to eliminate much of the administration’s incentive to resolve its disputes with the faculty through open discussion and mutual agreement. By its overbroad and unwarranted interpretation of the managerial exclusion, the Court denies the faculty the protections of the NLRA and, in so doing, removes whatever deterrent value the Act’s availability may offer against unreasonable administrative conduct.18 Rather than promoting the Act’s objective of funneling dissension between employers and employees into collective bargaining, the Court’s decision undermines that goal and contributes to the possibility that “recurring disputes [will] fester outside the negotiation process until strikes or other forms of economic warfare occur.” Ford Motor Co. v. NLRB, 441 U. S. 488, 499 (1979).
Ill
In sum, the Board analyzed both the essential purposes underlying the supervisory and managerial exclusions and the nature of the governance structure at Yeshiva University. Relying on three factors that attempt to encapsulate the fine distinction between those professional employees who are entitled to the NLRA’s protections and those whose managerial responsibilities require their exclusion,19 the Board concluded *706that Yeshiva’s full-time faculty qualify as the former rather than the latter. I believe the Board made the correct determination. But even were I to have reservations about the specific result reached by the Board on the facts of this case, I would certainly have to conclude that the Board applied a proper mode of analysis to arrive at a decision well within the zone of reasonableness. Accordingly, in light of the deference due the Board’s determination in this complex area, I would reverse the judgment of the Court of Appeals.
“It is not necessary in this case to make a completely definitive limitation around the term 'employee.' That task has been assigned primarily to the agency created by Congress to administer the Act. Determination of ‘where all the conditions of the relation require protection’ involves inquiries for the Board charged with this duty. Everyday experience in the administration of the statute gives it familiarity with the circumstances and backgrounds of employment relationships in various industries, with the abilities and needs of the workers for self-organization and collective action, and with the adaptability of collective bargaining for the peaceful settlement of their disputes with their employers. The experience thus acquired must be brought frequently to bear on the question who is an employee under the Act. Resolving that question . . . ‘belongs to the usual administrative routine’ of the Board.” NLRB v. Hearst Publications, Inc., 322 U. S. 111, 130 (1944). Accord, NLRB v. Seven-Up Bottling Co., 344 U. S. 344, 349 (1953).
Because the Court concludes that Yeshiva’s full-time faculty are managerial employees, it finds it unnecessary to reach the University’s contention that the faculty are also excluded as “supervisors” under §2 (11). Ante, at 682. My discussion therefore focuses on the question of the faculty’s managerial status, but I would resolve the issue of their supervisory status in a similar fashion.
See H. R. Rep. No. 245, 80th Cong., 1st Sess., 14 (1947):
“The evidence before the committee shows clearly that unionizing supervisors under the Labor Act is inconsistent with the purpose of the act ... . It is inconsistent with the policy of Congress to assure to workers freedom from domination or control by their supervisors in their organizing and bargaining activities. It is inconsistent with our policy to protect the rights of employers; they, as well as workers, are entitled to loyal representatives in the plants, but when the foremen unionize, even in a union that claims to be 'independent' of the union of the rank and file, they are subject to influence and control by the rank and file union, and, instead of their bossing the rank and file, the rank and file bosses them.”
See also S. Rep. No. 105, 80th Cong., 1st Sess., 3-5 (1947).
Section 2 (11) of the Act requires, as a condition of supervisory status, that authority be exercised “in the interest of the employer.” 29 U. S. C. §152 (11). See also NLRB v. Master Stevedores Assn., 418 F. 2d 140 (CA5 1969); International Union of United Brewery Workers v. NLRB, 111 U. S. App. D. C. 383, 298 F. 2d 297 (1961).
The Board held hearings over a 5-month period and compiled a record containing more than 4,600 pages of testimony and 200 exhibits.
The Court thus determines that all of Yeshiva’s full-time 'faculty members are managerial employees, even though their role in university decisionmaking is limited to the professional recommendations of the faculty acting as a collective body, and even though they supervise and manage no personnel other than themselves. The anomaly of such a result demonstrates the error in extending the managerial exclusion to a class of essentially rank-and-file employees who do not represent the interests of management and who are not subject to the danger of conflicting loyalties which motivated the adoption of that exemption.
As the Board has recognized, due to the unique nature of their work, professional employees will often make recommendations on matters that are of great importance to management. But their desire to exert influence in these areas stems from the need to maintain their own professional standards, and this factor — common to all professionals — should not, by itself, preclude their inclusion in a bargaining unit. See Westinghouse Electric Corp., 113 N. L. R. B. 337, 339-340 (1955). In fact, Congress clearly recognized both that professional employees consistently exercise independent judgment and "discretion in the performance of their duties, see 29 U. S. C. § 152 (12), and that they have a significant interest in maintaining certain professional standards, see S. Rep. No. 105, 80th Cong., 1st Sess., 11 (1947). Yet Congress specifically included professionals within the Act’s coverage. See NLRB v. Bell Aerospace Co., 416 U. S. 267, 298 (1974) (White, J., dissenting in part).
One must be careful not to overvalue the significance of the faculty’s influence on academic affairs. As one commentator has noted, “it is not extraordinary for employees to seek to exert influence over matters embedded in an employment relationship for which they share a concern, or that management would be responsive to their strongly held desires.” Finkin, The NLRB in Higher Education, 5 U. Toledo L. Rev. 608, 616 (1974). Who, after all, is better suited than the faculty to decide what courses should be offered, how they should be taught, and by what standards their students should be graded? Employers will often attempt to defer to their employees’ suggestions, particularly where — as here — those recommendations relate to matters within the unique competence of the employees.
Moreover, insofar as faculty members are given some say in more traditional managerial decisions such as the hiring and promotion of other personnel, such discretion does not constitute an adequate basis for the conferral of managerial or supervisory status. Indeed, in the typical industrial context, it is not uncommon for the employees’ union to be given the exclusive right to recommend personnel to the employer, and these hiring-hall agreements have been upheld even where the union requires a worker to pass a union-administered skills test as a condition of referral. See, e. g., Local 42 (Catalytic Constr. Co.), 164 N. L. R. B. 916 (1967); see generally Teamsters v. NLRB, 366 U. S. 667 (1961).
The Board has also explained that the ability of the typical professional employee to influence company policy does not bestow managerial authority:
“Work which is based on professional competence necessarily involves a consistent exercise of discretion and judgment, else professionalism would not be involved. Nevertheless, professional employees plainly are not the same as management employees either by definition or in authority, and managerial authority is not vested in professional employees merely by virtue of their professional status, or because work performed in that status may have a bearing on company direction.” General Dynamics Corp., 213 N. L. R. B., at 857-858.
Where faculty members actually do serve as management’s representatives, the Board has not hesitated to exclude them from the Act's coverage as managerial or supervisory personnel. Compare University of Vermont, 223 N. L. R. B. 423 (1976) (excluding department chairmen as supervisors), and University of Miami, 213 N. L. R. B. 634 (1974) (excluding deans as supervisors), with Northeastern University, 218 N. L. R. B. 247 (1975) (department chairmen included within bargaining unit because they act primarily as instruments of the faculty), and Fordham University, 193 N. L. R. B. 134 (1971) (including department chairmen because they are considered to be representatives of the faculty rather than of the administration). In fact, the bargaining unit approved by the Board in the present case excluded deans, acting deans, directors, and principal investigators of research and training grants, all of whom were deemed to exercise supervisory or managerial authority. See ante, at 678, n. 7.
Nor does the frequency with which an employer acquiesces in the recommendations of its employees convert them into managers or supervisors. See Stop & Shop Cos., Inc. v. NLRB, 548 F. 2d 17, 19 (CA1 1977). Rather, the pertinent inquiries are who retains the ultimate decisionmaking authority and in whose interest the suggestions are offered. A different test could permit an employer to deny its employees the benefits of collective bargaining on important issues of wages, hours, and other conditions of employment merely by consulting with them on a host of less significant matters and accepting their advice when it is consistent with management's own objectives.
See, e. g., App. 740-742 (faculty hiring); id., at 232-233, 632, 667 (tenure); id., at 194, 620, 742-743 (promotion); id., at 713, 1463-1464 (retirement); id., at 241 (dismissal); id., at 362 (academic credits); id., at 723-724, 1469-1470 (cutback in departmental budget leading to loss of accreditation); id., at 410, 726-727 (election of department chairman and representative).
See 29 U. S. C. §158 (d).
See generally J. Brubacher & W. Rudy, Higher Education in Transition: A History of American Colleges and Universities, 1636-1976 (3d ed. 1976). In one of its earliest decisions in this area, the Board recognized that the governance structure of the typical modem university does not fit the mold of true collegiality in which authority rests with a peer group of scholars. Adelphi University, 195 N. L. R. B. 639, 648 (1972). Accord, New York University, 205 N. L. R. B. 4, 5 (1973). Even the concept of “shared authority,” in which university decisionmaking is seen as the joint responsibility of both faculty and administration, with each exerting a dominant influence in its respective sphere of expertise, has *703been found to be “an ideal rather than a widely adopted practice.” K. Mortimer & T. McConnell, Sharing Authority Effectively 4 (1978). The authors conclude:
“Higher education is in the throes of a shift from informal and consensual judgments to authority based on formal criteria. . . . There have been changes in societal and legislative expectations about higher education, an increase in external regulation of colleges and universities, an increase in emphasis on managerial skills and the technocratic features of modem management, and a greater codification of internal decision-making procedures. These changes raise the question whether existing statements of shared authority provide adequate guidelines for internal governance.” Id., at 269.
In 1976-1977, the total expenditures of institutions of higher education in the United States exceeded $42 billion. National Center for Education Statistics, Digest of Education Statistics 137 (Table 133) (1979). In the same year, Yeshiva University, a private institution, received over $34 million in revenues from the Federal Government. Id., at 132 (Table 127).
University faculty members have been particularly hard hit by the current financial squeeze. Because of inflation, the purchasing power of the faculty’s salary has declined an average of 2.9% every year since 1972. Real salaries are thus 13.6% below the 1972 levels. Hansen, An Era of Continuing Decline: Annual Report on the Economic Status of the Profession, 1978-1979, 65 Academe: Bulletin of the American Association of University Professors 319, 323-324 (1979). Moreover, the faculty at Yeshiva has fared even worse than most. Whereas the average salary of a full professor at a comparable institution is $31,100, a full professor at Yeshiva averages only $27,100. Id., at 334, 348. In fact, a severe financial crisis at the University in 1971-1972 forced the president to order a freeze on all faculty promotions and pay increases. App. 1459.
As of January 1979, 80 private and 302 public institutions of higher education had engaged in collective bargaining with their faculties, and over 130,000 academic personnel had been unionized. National Center for the Study of Collective Bargaining in Higher Education, Directory of Faculty Contracts and Bargaining Agents in Institutions of Higher Education i-ii (1979). Although the NLRA is not applicable to any public employer, see 29 U. S. C. §152 (2), as of 1976, 22 States had enacted legislation granting faculties at public institutions the right to unionize and requiring public employers to bargain with duly constituted bargaining agents. Mortimer & McConnell, supra n. 14, at 53. See also Livingston & Christensen, State and Federal Regulation of Collective Negotiations in Higher Education, 1971 Wis. L. Rev. 91, 102.
The upsurge in the incidence of collective bargaining has generally been attributed to the faculty’s desire to use the process as a countervailing force against increased administrative power and to ensure that the ideals of the academic community are actually practiced. As the Carnegie Commission found, “[unionization for [faculty] is more a protective than an aggressive act, more an effort to preserve the status quo than to achieve a new position of influence and affluence. . . .” Carnegie Commission on Higher Education, Governance of Higher Education 40 (1973). See also Mortimer & McConnell, supra n. 14, at 56; Lindeman, The Five Most Cited Reasons for Faculty Unionization, 102 Intellect 85 (1973); Nielsen & Polishook, Collective Bargaining and Beyond, The Chronicle of Higher Education 7 (May 21, 1979).
The Carnegie Commission, in concluding that “faculty members should have the right to organize and to bargain collectively, if they so desire,” Carnegie Commission on Higher Education, supra, at 43, observed:
“We may be involved in a long-term period of greater social conflict in society and greater tension on campus. If so, it may be better to institutionalize this conflict through collective bargaining than to have it manifest itself with less restraint. Collective bargaining does provide agreed-upon rules of behavior, contractual understandings, and mechanisms for dispute settlement and grievance handling that help to manage conflict.” Id., at 51.
Contrary to the Court's assertion, see ante, at 685, the Board has not abandoned the “collective authority” and “ultimate authority” branches of its analysis. See Reply Brief for Petitioner in No. 78-857, pp. 11-12, n. 8. Although the “interest/alignment analysis” rationale goes to the *706heart of the basis for the managerial and supervisory exclusions and therefore provides the strongest support for the Board’s determination, the other two rationales are significant because they highlight two aspects of the university decisionmaking process relevant to the Board’s decision: That the faculty’s influence is exercised collectively — and only collectively— indicates that the faculty’s recommendations embody the views of the rank and file rather than those of a select group of persons charged with formulating and implementing management policies. Similarly, that the administration retains ultimate authority merely indicates that a true system of collegiality is simply not the mode of governance at Yeshiva University.