In ancient Egypt, the Instruction of Ptahhotep (c. 2300 B.C.) declared, “To resist him that is put in authority is evil” (Lichtheim, 1973). Pericles, in Athens during the fifth century b.c., said that a spirit of reverence and respect for authority and the laws constrained public acts and inhibited wrongdoing (Thucydides, 404 b.c./1910). In the sixth century, Saint Benedict’s rule for the authority and responsibilities of the abbots, priors, deacons, and monks were essential to the survival of the monasteries he founded (Chittister, 1992). John Locke (1960), writing at the end of the seventeenth century, held that authority was needed to make a liberal society work and survive. He noted that for the creation, maintenance, and progress of a liberal society, authority that would be more than leadership had to be established (Weaver, 1991). Edmund Burke (1790/1967) believed that those appointed to exercise authority were assigned a holy function. Recognized authority gives rise to power, and power brings responsibility. Thus Friedrich Nietzsche (1888/ 1935) declared that responsibility increases toward the summit. Abraham Lincoln, in his second annual message to Congress, observed that those who have the power to save the Union also bore the responsibility to do so.
The legitimate power of individuals in groups, organizations, and societies derives from their authority and responsibility. Whether they are viziers, company presidents, military officers, school principals, or first-line supervisors, their responsibility should be commensurate with their authority so as to minimize their role conflict. Often, it is not. Variations in and mismatchings of authority and responsibility have been reported for people who hold the same kinds of positions. Surveys by Bass and Valenzi (1974) discovered many managers who thought that they were delegated responsibility without the necessary authority to go along with it. Interviews with 32 college presidents by Birnbaum (1988a) disclosed that responsibility without authority made the presidents’ leadership role most difficult. The presidents coped by accepting the ambiguities of their situation, by incrementally pursuing limited objectives, and by practicing management by exception.
A study by Munson (1981) of social work supervisors and their subordinates revealed differences between preferred and actual authority and in the structure of relations that were relevant to supervisory practices and control. Aiken and Bacharach (1985) studied 44 local governmental administrative agencies in Belgium and found considerable discrepancy between subordinates’ self-reported authority to make decisions and the authority their superiors said these subordinates had. The discrepancy was increased if the local government was controlled by a coalition, the organization was open to outside influences, and the structure of control was fragmented.
The allocation of responsibility and authority to managers should reflect what is required to meet organizational objectives. However, responsibility and authority may, in fact, be allocated as rewards or punishments, as political symbols, or as political gestures to imply that improvements are occurring in the organization (Benze, 1985). The managers’ accountability implies their acceptance of the responsibilities of their leadership position, the expectation that they will be linked publicly to their words and actions, and the expectation that they may be called on to explain them (Wood & Winston, 2005).
Formally, authority is the legitimate right to exercise power. According to Freud (1939), the need for authority figures develops during childhood from the yearning for a father. Zander, Cohen, and Stotland (1957) observed that the felt possession and exercise of authority and the exercise of authority are highly correlated. Traditionally, legitimacy comes from norms dealing with society, organizations, and membership within them (House, 1991). It includes the right to command and to induce compliance as well as the ability to control subordinates and to make decisions by oneself (Katzenbach & Smith, 1992). Authority is a central feature of the structure of formal organizations. It prescribes “expectations that certain individuals should exert control and direction over others within defined areas of competence.” Authority is derived from implicit or explicit contracts concerning the individual’s position or knowledge. The potential conflicts arising from incongruities between authority and responsibilities and between authority and knowledge are the same as those for imbalances in status and competence. Loss of authority because of a restructuring of the organization will be reflected in feelings of deprivation (C. S. George, 1972).
Both traditional and behavioral definitions are phrased in terms of value judgments in support of two conflicting moralities or ideologies (Wells, 1963). Peabody (1962) found that executives in a city welfare agency emphasized legitimacy and position as the basis of authority, whereas policemen in the same city stressed knowledge and competence as the basis. In a later study, Peabody (1964) observed that supervisors in a police department, a welfare agency, and a public school thought of authority in terms of internal (superior-subordinate) relations, whereas the people supervised thought of it in terms of external (worker-client) relations. Supervisors in the police department unquestioningly accepted authoritative instructions that produced conflict more frequently than did those in the welfare agency or the school. The unacceptable use of authority was more common in the welfare agency than in the police department or the school.
Influenced by Barnard (1938), behaviorists maintained that leaders have authority only to the extent that followers are willing to accept their commands. Authority was defined operationally in terms of the areas in which members of an organization are to carry out their responsibilities (Petersen, Plowman, & Trickett, 1962). Authority in an organization specifies a member’s perceived area of freedom of action and interaction, along with the formally delegated or informally recognized right to initiate action. Members act in accordance with their perception of: (1) the degree of freedom that is allowed to them, and (2) the initiative that they feel they can safely exercise. Their perception may or may not coincide with the expectations of their supervisors, peers, or subordinates. Authority is also connected with the affirmation of certain values, as illustrated by the inaugural oath of the president of the United States to preserve, protect, and defend the Constitution (Paige, 1977).
In general, group leaders perceive themselves as having a higher degree of authority than do the leaders of subgroups; and leaders of subgroups, in turn, perceive their authority to be higher than the perceived authority of individual members of the subgroups who are responsible for performing individual tasks. Authority gives legitimate power, which depends on the norms and expectations held by a group that are appropriate to a given role—the accorded rights, duties, and privileges that go along with appointment or election to a position.
Authority depends on the relationship between leaders and followers. Leaders can restrict the authority of subordinates by withholding their rights to act and decide. They can increase their subordinates’ authority by delegating to them the right to act. Followers can reduce the leader’s authority by failing or refusing to accept the leader’s decisions. They can increase their leader’s authority by referring matters to the leader for decision. The authority of leaders and followers also depends on other sources of power. The authority of professionals, for instance, stems from a wide variety of sources, such as their exclusive information, their control of resources, and their assigned responsibilities (Filley & Grimes, 1967).
Conflict results if colleagues think that leaders have misperceived or exceeded the limits of their authority (Pondy, 1967). For example, U.S. workers would find it unacceptable if their supervisors gave them orders about where they should live. According to Haas, Porat, and Vaughan (1969), an almost universal source of conflict is the tendency of individuals at any designated level to see themselves as having more authority and responsibility than their superiors believe they have. Lennerlöf (1965a) also noted a tendency for supervisors to rate themselves as having more authority than their superiors reported delegating to them. George W. Bush has argued that as president during wartime he has the authority to ignore congressional statutes and the Constitution, as did Abraham Lincoln during the Civil War.
Legitimacy for actions may be accorded by higher officials, but it still depends on an acceptance by subordinates. Followers give a form of consent to legitimacy, which they can grant or withhold, sometimes at considerable cost (Hollander, 1978). According to Barnard (1938), this means that authority is delegated upward and is granted by one’s subordinates. Furthermore, the perception that a superior has the right to give orders must be widely held and shared by other subordinates. Nor does one gain such support of one’s authority from subordinates passively. Bendix (1974) thought that the compliance of subordinates with authority depends on the leader’s active cultivation of the legitimacy of authority. Whitson (1980) suggested that there is a tense equilibrium between authority and the confidence followers have in the authority structure. Ironically, as confidence wanes, more formal authority and more coercion are applied, escalating the loss of confidence. Presumably, renewed confidence, accompanied by less coercion, is required before authority can be restored.
Zones of Indifference and Acceptance. Barnard expanded his argument about the extent to which authority derives from subordinates’ acceptance of orders by noting conditions that increase or decrease such compliance. Orders will be complied with to the extent that they are understood, are consistent with the purpose of the organization, and are compatible with the personal interests of the subordinate. Also, orders will be followed to the extent that the subordinate is physically and mentally able to comply with them. Each subordinate has a “zone of indifference” within which orders are acceptable with no conscious questioning of authority. These zones of indifference are maintained by the interests of the group. In bureaucracies, they give the higher authority some latitude and autonomy in its dealings with organizational members (Downs, 1967).
A leader’s authority, legitimated by his or her position, can bring about a wide or narrow zone of acceptance in subordinates by instilling in them the belief that the benefits of compliance will exceed the costs involved in remaining members in good standing in the organization. Using the Professional Zone of Acceptance Inventory (Hoy, Tarter, & Forsyth, 1978), Johnston (undated, b) surveyed 490 teachers in 55 elementary and secondary schools. In the 26 secondary schools, the teachers’ zones of acceptance correlated .71 with their loyalty to the principals. However, the more authoritarian the principal, the lower was the teacher’s zone of acceptance. The more the principal enforced the rules rationally and legitimately in the eyes of the teacher, the larger was the teacher’s zone of acceptance. This finding was consistent with Gouldner’s (1954) observation that according to the factory workers he interviewed, rational discipline was necessary for efficient functioning of the work process. Rational discipline maintained workers’ loyalty to their supervisor. Workers thought that it required being lenient and “not too strict” (p. 46): “[When] there’s work to be done they expect you to do it. … Otherwise they leave you alone” (p. 47). However, with professionals, Johnston found that the zone of acceptance was increased more by the extent to which the principals granted the teachers professional autonomy than by rational discipline and rule enforcement.
Authority is not power. “No amount of legal authority over the grizzly bears of British Columbia would enable you to get yourself obeyed by them out in the woods” (National Research Council, 1943). In supervisor-subordinate relationships, coercive and reward power, as well as legitimate power, are expected in the supervisor’s position in addition to authority, but subordinates have learned to associate reward and punishment with the supervisory position and its symbols of status and authority. An analysis of authority and power requires an examination of attitudes and expectations about authority figures and symbols, as well as of the rules and regulations that describe what should be expected of holders of positions of authority (Bass, 1960).
Authority is power that is legitimatized by tradition, law, agreements, religion, and the rights of succession; it is distinguished from force and coercion (J. M. Burns, 1978). In bureaucracies, superiors in the hierarchy have authority over subordinates. This provides formal organizational coordination of members’ behavior based on officially defined structure. In traditional societies, authority is the legitimate support for the father, the priest, and the noble. It comes from God, the will of heaven, or nature. In modern societies, formal authority derives from man-made constitutions, compacts, charters, legislation, judges’ rulings, and due process. The state has the authority to charter the board. The board has the authority to appoint the organization’s president. The president has the authority to hire a staff, and so on. Decisions can be reversed in the same way. Those who appoint the leaders to a given position usually have the authority to revoke functions previously assigned, to remove units of the organization from the leaders’ jurisdiction, and to dismiss leaders from office. Even when supported by their superiors, appointed leaders may find themselves frustrated and powerless if subordinates (like grizzly bears) refuse to obey their commands. In volunteer organizations, nongovernmental organizations (NGOs), and political protest groups, formerly powerful elected or appointed leaders may find that they no longer have influence after their followers and subordinates withdraw support or give it to opponents. Thus the concept of role legitimatization calls attention to the fact that the power and influence of leaders are dependent on the acknowledgment of their authority by followers and, in some circumstances, by superiors or peers. Needless to say, the acceptance and exercise of authority will also depend on the interpretation by the individuals in authority of their rights, duties, obligations, privileges, and powers. Thus, for instance, “President Eisenhower simply did not believe that he should be leading crusades of a moral, humanitarian, or civil rights nature. He believed that his job was to operate exclusively within the governmental powers of his office” (Larson, 1968, p. 21).
Some people argue that authority and the concomitant differentiation of power are essential for cooperative efforts in organizations; others advocate equalization, leveling, and power sharing. At one extreme, Carney (1982) saw authority as the organization’s universal panacea for keeping power in the hands of a few. The absence of such authority made improbable the spontaneous and comprehensive cooperation of the organization’s members to work together toward common goals. Authority provided managers with the right to control rewards and punishments; to maintain discipline, order, and security; and to avoid confusion and frustration. However, at the other extreme, Baker (1982) noted that when members of an organization are committed to equality in the distribution of power, as was seen in a radical feminist community, the maintenance of an encrusted bureaucracy was prevented, and a shifting cadre of leaders could emerge, both informally and formally, to generate policy and cooperation. Authority could be shared.
School principals were seen as administering rules in three ways. In 31 secondary and 39 elementary schools, Johnston (undated) observed that when some principals enforced the rules with explanations and understanding, compliance by the staff was high. When some other principals enforced rules with discipline and punishment, the staff complied grudgingly. When still other principals did not enforce the rules, the teachers did not obey the rules. The principal’s influence and the felt loyalty of the teachers to their principal correlated .39 and .51, respectively, when the principal used explanation and understanding to enforce the rules. The correlations when punishment was used to maintain the rules were .01 and –.03. The correlations with influence and loyalty were negative (–.31 and –.20) when the rules were neither enforced by the principal nor obeyed by the teachers. Venable (1983) corroborated the finding that the principals’ exercise of nonpunitive legitimate authority (use of “rule-administration behavior”) correlated with the teachers’ loyalty to the principals.
Authority is pervasive in its effects, often accounting for unquestioned influence. Thus in Milgram’s (1965b) famous experiment, participants accepted without much consideration an experimenter’s authority and delivered supposedly dangerous shocks to other student participants. Scheffler and Winslow (1950) found that people in low-status positions did not reject authority any more than did those in high-status positions. W. E. Scott’s (1965) study of the attitudes of workers in a professional organization that allowed little autonomy on the job found that workers tended to accept the system of restrictive supervision. However, those who were professionally (externally) oriented were more critical of the authority structure than were those who identified themselves as internally oriented. Self-confidence about one’s interpersonal relationships also appeared to make a difference in one’s responsiveness to authority. Thus Berkowitz and Lundy (1957) observed that individuals strong in interpersonal confidence tended to be influenced more readily by authority figures than by peers, whereas the reverse was true for persons weak in interpersonal confidence.
Authority is not confined to the leadership structure. All members of an organization possess or believe they possess some degree of authority to perform their respective jobs. In a study of more than 1,700 individuals in formal organizations, Stogdill (1957a) found that only about 1 in 500 checked the statement, “I have no authority whatsoever.” Unskilled mechanical workers rated themselves as having more than zero responsibility and authority.
Motivational Effects on Acceptance of Authority. Control over what others want establishes power over them (Bass, 1960). For leaders, such power, rather than empty formal authority, depends on the extent to which they can activate the needs and motives of their subordinates (J. M. Burns, 1978). The stronger the motivational base the leader can tap, the more authority, power, and control he or she can exercise. Thus Hollander and Bair (1954) reported that compared with members whose motivation was low, highly motivated members of a group identified more strongly with authority figures. According to the German field marshal Erwin Rommel, “The commander must try, above all, to establish personal and comradely contact with his men, but without giving away an inch of his authority” (as quoted in Mack & Konetzni, 1982, p. 3). Taylor (1983) suggested that leader-subordinate relationships in the U.S. Army were systematically undermined during the 1960s and the 1970s by the decrease in personal contact because of the introduction of information technology. More detached, self-centered subordinates and institutional policies also contributed. Lower-level commanders were given more responsibility but less authority. An effort to restore personal contact and authority at each level was initiated in 1979. In the same vein, H. I. Bowman (1964) revealed that school principals who described their superiors as high in consideration on the Leader Behavior Description Questionnaire tended to rate themselves as high in responsibility and authority. In Shakespeare’s plays Richard III, Macbeth, and Coriolanus, fear was used to maintain authority, and ultimately loyalty to the leaders was lost (Corrigan, 1999).
Effects of the Scope of Authority. Scope of authority depends on the influence needed to accomplish legitimate and recognized role requirements and organizational objectives (Barnard, 1952). Even though the scope may be expressed in writing, the boundaries may be blurred, and conflicts in understanding and actions may result (Reitz, 1977). Bachand (1981) contrasted Canadian crown corporations headed by government-appointed boards of directors with Canadian private-sector corporations headed by boards of directors who were elected by stockholders. The boards differed in their scope of authority, which affected their actions and those of the executives who reported to them. The members of the crown boards of directors, compared with those of the boards of directors of private corporations, believed they had the authority to do more than just give advice and counsel. The members of the crown boards faced more potential for direct conflict between political and economic objectives and had to be more active than members of the private-sector boards. As appointees of the government, the members of the crown boards of directors believed themselves to be more independent than did the members of the private-sector boards. In turn, the chief executives of the crown corporations sought their support for specific policies and decisions more often than the chief executives of private-sector corporations sought the support of their boards.
D. E. Tannenbaum (1959) observed that boards of directors in five agencies enlarged or restricted the chief executives’ role by their willingness or refusal to delegate authority to the executives to act and decide. S. Epstein (1956) studied the effects of this enlargement or restriction on supervisors in experimental groups who were given different degrees of freedom of action. Severe restriction induced the supervisors to restrict the behavior of their subordinates, to supervise them more closely, and to supervise in a management-oriented manner. The absence of such restrictions induced the supervisors to give their subordinates more freedom and to supervise them less closely.
Dalton, Barnes, and Zaleznik (1968) studied the effects of changes in the authority structure of some departments of a research organization when other departments were left unchanged. In those departments where authority was transferred from department heads to scientists and engineers, the scientists and engineers favored the change, but the department heads did not. In the unchanged departments, expectations were heightened by the shifts in authority in the changed departments. Subordinates in the unchanged departments became dissatisfied with their superiors, and there was a greater tendency of those whose authority remained unchanged to seek positions elsewhere.
Parallel Authority Structures. Throughout the former Soviet Union, the scope of authority of the formal organizational leadership was limited by a parallel Communist Party structure. The dualism was seen in the parallel state and party organizational structures, factory management and party representatives, military commanders and political commissars. The recognition of the costliness of this dualism resulted in its elimination in the military (Mack & Konetzni, 1982) and industry (Gorbachev, 1988). Heifitz (1994) clarified the role of authority relative to leadership by considering the physician-patient relationship. In a type 1 situation, when faced with a purely definable technical problem with a clear solution about which the physician is expert and the patient lacks any knowledge, the physician should use his or her authority to recommend the appropriate course of action. In a type 2 situation, the problem is definable but the solution is not clear. Here, adaptive work is needed. The physician and patient need to pursue the solution together. In a type 3 situation, the problem is not clearly definable and there are no recognizable technical solutions. Again, adaptive work must be done in which leadership is required to promote learning to understand the problem and create solutions.
Traditionally, leadership was fully determined by authority (Barker, 1994). But Hunt, Osborn, and Schuler (1978) conceptualized leadership as consisting of a discretionary and a nondiscretionary component. According to this concept, discretionary leadership is under the control of the leader, and nondiscretionary leadership is invoked by the organizational setting in which the leader operates. With civilian control and with concern about avoiding incidents that might provoke war, U.S. military officers, in particular, are limited in their discretionary activities. Although they are fully accountable and responsible for the actions of their subordinates, they are held on a tight leash by the higher command (Mack & Konetzni, 1982). Interviews with 32 college presidents revealed that their discretionary opportunities to lead are increasingly being constrained by governmental intervention, demographic trends, fiscal constraints, unrealistic public expectations, divergent interest groups on campus, and confusing patterns of authority. Beyond the presidents’ nondiscretionary responsibility to balance their budgets, they still have discretion in how much they try to locate new sources of support (Birnbaum, 1988b).
One way of measuring discretionary and nondiscretionary leadership is to use subordinates’ responses to the Leader Behavior Description Questionnaire (LBDQ), asking them on selected items if their supervisor “can and does,” “could but doesn’t,” “can’t and doesn’t,” or “can’t but tries anyway” (Martin & Hunt, 1981). Van de Ven and Ferry (1980) asked supervisors and employees to judge the amount of authority they had in making decisions in four areas of their jobs: (1) determining what tasks the employee will do; (2) setting quotas; (3) establishing rules and procedures; (4) determining how exceptions to the usual work were to be handled. In a large engineering division of a public utility, Hunt, Osborn, and Schuler (1978) derived a leader’s nondiscretionary score from the results of a leadership survey. Nondiscretionary leadership was that portion of the original score predicted by organizational practices. The discretionary score was equal to the difference between predicted and original scores. Table 14.1 shows the relative extent to which each of four original behaviors by leaders toward subordinates (approval, consideration, disapproval, and ego deflation) were accounted for as discretionary and nondiscretionary. It can be seen that in the public utility, the rewarding leadership behavior of managers had more of a nondiscretionary component than did disapproval or ego deflation.
Table 14.1 Percentage of Variance That Accounts for the Discretionary and Nondiscretionary Leadership Components of a Manager’s Behavior
| Manager’s Leader Behavior | ||||
| Approval | Consideration | Disapproval | Ego Deflation | |
| Discretionary | 77 | 77 | 90 | 81 |
| Nondiscretionary | 20 | 22 | 9 | 1 |
| Error | 3 | 1 | 1 | 18 |
SOURCE: Adapted from Hunt, Osborn, and Schuler (1978).
Effective leaders in order to clarify what is required of their subordinates use discretionary leadership (Gast, 1984), but the subordinates may also display discretionary leadership behavior. Goodacre (1953) studied the performance of good and poor combat units on simulated problems. There was a greater tendency in good units than in poorly performing units for the men without authority to take the initiative in giving orders during the problem and to be better satisfied with their leaders’ management of the problem. Hambrick and Abrahamson (1995) obtained academics’ and security analysts’ ratings of managerial discretion in selected industries. Discretion was rated highest in R & D–intensive industries such as computers, motion picture production, and scientific instruments; discretion was rated lowest in steel mills, petroleum and natural gas production, and gold and silver mining.
Heller (1985, p. 488) summed up the decline in authority in Britain and elsewhere during the past several decades, at the macro level and the microlevel, as the basis of relationships between the subordinate and the superior: “The loss of authority is … evidenced by a decline in public confidence in institutions and institutional leaders, a loss of loyalty and commitment of organizational members, and a trend toward identification with multiple organizations. … The loss is [also] evidenced in a decline in willingness to be bossed as well as a loss of desire to be the boss, and in a trend toward rating oneself as ‘better than’ the boss on desirable traits.” However, Heller noted that the loss of authority, brings on a challenge for new bases for inducing compliance. One possibility is a return to strong, personal, charismatic leaders. Other alternatives include industrial democracy, power sharing, participative management, and the development of commitment to common goals. The social network may be another substitute for authority. Personal initiative may also substitute for authority. One night, when the outpost of guards responsible for patrolling for North Korean submarine infiltrators refused to take action because an incident was outside their area, a Korean taxi driver took it on himself to arouse a sleeping South Korean army garrison and lead the soldiers to a group of suspicious men he had noticed. They turned out to be North Korean infiltrators from an offshore submarine (Kristoff, 1996).
In addition to cultural changes that have diminished authority as a basis for successful leadership, Zaleznik (1980) presented a number of cases of executives whose neuroses got in the way of their obtaining compliance from subordinates. Sometimes, for example, the superior isolated himself, became tyrannical, and caused a palace revolution against his authority. In other cases, executives could not exert their authority because of their obsessive fear of hurting their subordinates. Generally, managers are becoming less dependent on their authority for their leadership and influence. Peter Drucker sees a marked change from the command and control structure where rank provides authority. This was copied from the Prussian army in the 1870s by the emerging big businesses system. “We are now evolving toward structures in which rank means responsibility but not authority … and in which your job is not to command but to persuade” (Cohen, 1996, p. 16). Boccialetti (1996) argued that a more reciprocal approach to authority is needed, particularly in times of change in internal and external organizational conditions. Those lower in the organizational hierarchy need to become more proactive in their relations with higher-ups, stay better informed about big issues, and accept more responsibility.
Organizational leaders are given responsibility to carry out an assignment and are held accountable for it. They are expected to carry out their responsibilities in conformance with the norms and rules of the organization. The leaders should have the authority (legitimate power, right, or permission) to do so. Authority in organizations is meant to be used to fulfill assigned responsibilities (Hollander, 1978). The responsibilities are the members’ perceptions of expectations by the organization that they will perform on its behalf. Generally, leaders perceive their responsibilities to be broader and more far-reaching than other group members perceive their own responsibilities. Leaders consider it their responsibility to make policies and to initiate action for themselves and for those members for whom they are responsible. The other members perceive themselves to be responsible for initiating action in specific subgroups or for executing individual assignments. In any event, neither leaders nor followers can learn to assume responsibilities until they are given them (Deci, 1972). Supervisors in naval organizations who are more conscious of their active leader-ship are perceived to be more responsible naval officers by their subordinates (Stogdill, Scott, & Jaynes, 1956). In Senge’s (1990) learning organization, mind-sets are shifted so that members assume responsibility for actions connected to outcomes. Negative outcomes may result in rejection of responsibility. Thus failure of group decisions may result in a diffusion of individuals’ sense of responsibility (Whyte, 1991). Sense of responsibility may be a matter of: (1) moral and legal standards of right and wrong, or legal and illegal; (2) obligation or sense of duty; and (3) concern for the consequences (Winter, 1991). Alternatively, having a sense of responsibility may be a matter of how one evaluates one’s own character.
Like authority, responsibility also depends on leader-follower relations. The leader can reduce the responsibility of other group members by failing or refusing to relinquish duties that others could perform, by overly close supervision, and by requiring consultation before others can perform. Other group members can reduce the leader’s responsibility by performing tasks that the leader would be expected to perform. They can increase the leader’s responsibility by failing to carry out assigned and expected duties. Even when responsibilities are closely defined, the actions of their leaders tend to condition the perceived responsibilities of the subordinates, and the performance of the subordinates tends to condition the responsibilities of the leader. To be effective, anyone desiring influence in an organization needs to accept responsibility for knowing about values, strengths, performance, and communications with other members (Drucker, 1999).
H. H. Meyer (1959) found that supervisors and their bosses did not differ significantly on 77 items in their evaluations of the amount of the supervisors’ responsibilities. However, more effective supervisors rated themselves significantly higher in responsibility than ineffective supervisors did. In a later study in 21 plants, Meyer (1970b) found that the most effective supervisors assumed that they had full responsibility when there was any ambiguity about who was in charge. Ziller (1959) studied group leaders’ decision making regarding a difficult problem that risked the safety of their group. The leaders could base their actions on a throw of dice or by making a rational decision. Leaders who accepted responsibility for their group’s action tended to be nonconformists. They possessed personal resources that enabled them to take risks without undue stress and strain.
Accountability of an organizational leader may be defined as: (1) acceptance of responsibility for one’s own and associates’ behavior and effects; (2) answerability—providing explanation and justification of actions, commitments, decisions, and opinions; and (3) liability for them (Dubnick, 2003). In the governance of public and private institutions, accountability is a formal means of feedback and control common in bureaucracies (Gruber, 1987). Wood and Winston (2005) expanded the definition so that accountability implied: (4) willingness to accept the responsibility to serve the well-being of the organization; (5) the expectation that position holders would be linked to their actions or words; and (6) the expectation that they would be called on to explain their beliefs, decisions, commitments, and actions to their constituents.
Along with a leader’s authority and responsibility must come accountability. The leader is held answerable for using authority legally and meeting responsibilities (Brooks, 1995). Leaders are more likely than others to be held accountable. A leader cannot evade the consequences of having more influence over others; more control over events; greater visibility and recognition; and greater responsibility for failures, misplaced efforts, or inaction in the face of an evident threat to the group’s well-being (Hollander, 1978). Mack and Konetzni (1982, p. 5) noted: “In navies in general, and in the United States Navy in particular, strict accountability is an integral part of command. Not even the profession of medicine embraces the absolute relationship found at sea. A doctor may lose a patient under trying circumstances and continue to practice; but a naval officer seldom has the opportunity to hazard a second ship.” More safeguards are imposed as a leader’s reputation for trustworthiness declines (Hall, Blass, Ferris, et al., 2004).
In democracies, leaders need power to get things done. But accountability must go along with that power. Many others, such as bureaucrats, elected officials, educators, military leaders, land developers, marketers, and bankers, are granted powers that have important consequences to the public and need to be held accountable for their actions (Gardner, 2003). Formal accountability mechanisms are placed on leaders in the public and private sectors as safeguards against violations of the public trust. Fry, Scott, and Mitchell (1987) obtained survey results about accountability of deans, chairpersons, and faculty members at 369 business schools. Sources of information were the president, the dean, and two department heads. Faculty members were much more likely than deans to be evaluated formally and more frequently by multiple raters from below. Deans and department heads were more likely to be evaluated informally from above, and less frequently.
Accountability builds and sustains a climate of trust and credibility in leadership. The accountability of leaders to their group has important effects when they represent their group in negotiations (Lamm, 1973). Studies have found that being held accountable increases resolution of ethical issues (Brief, Dukerich, & Doran, 1991) and leads to more thoughtful decision processes (McAllister, Mitchell, & Beach, 1979). Also, making corporate boards, executives, and managers more accountable for their decisions and actions reduces illegal behavior, reduces lawsuits (T. M. Jones, 1986), and reduces the use of unapproved influence tactics (Baucus & Near, 1991). In a study of 210 respondents from a variety of occupations, Hall, Hochwarter, Ferris, et al. (2003) reported, as expected, that increases in accountability were correlated with increased “organizational citizenship.” But an unexpected finding was that for respondents high in job efficacy (their ability to perform their jobs), political behavior increased with accountability, whereas for those low in job efficacy the ability decreased.
Demands for accountability have increased with increased reports about violated trust and scandals in the clergy, business, politics, and government. Accountability reminds leaders of their need to comply with prevailing norms (Wood & Winston, 2005). However, leaders may try to avoid accountability or displace blame by resorting to collective responsibility vested in a committee, a board, or shared authority.
A leader’s feelings of accountability depend on perceptions of the applicability and enforcement of external laws, rules, and regulations to which organizations add their own norms, standards, and constraints on actions and outcomes for which managers should feel accountable. Accountability will also be a matter of what we think others expect of us and how they will treat us for our actions (Tetlock, 1985). In the Fry, Scott, Mitchell study, accountability appeared to be greater with the use of multiple evaluators, frequency and formality of evaluation, and accessibility of results. We may feel accountable as individuals. Also, we may be socialized into feeling accountable if we and others in a group are interdependent (Roberts, 2001). Accenture and Williams (2000) surveyed 104 respondents from a midwestern marketing firm and found that they felt personally accountable for their own development to the degree that that: they could do their work well (r = .47); they were high in need for achievement (r = .34); they were in a supportive work situation (r = .25); they had a supportive supervisor (r = .55); they intended to engage and actually engaged in developmental behavior (r = .80). Their time and resources for development mediated how much they participated.
Time Span. As a measure of responsibility, Jaques (1956) introduced the time span of responsibility—the length of time during which a manager or employee can be held accountable for decisions. Time span was found to correlate between .86 and .92 with the compensation that the job occupant judged to be fair (Richardson, 1971). It also was a good index of the location of a position in an organizational hierarchy. Thus a supervisor may have a time span of one month; a department head, one year; a general manager, three years; and a chief executive officer of a corporation, 20 years. Similarly, in an army, the company commander may have a time span of three months; the battalion commander, one year; the brigade commander, two years; and a full general, up to 20 years.
Ethical Accountability. Berman and Van Wart (1999) suggested holding managers, particularly public service administrators, ethically accountable to the public. Cost efficiency and profitability are relatively less important to evaluate than effectiveness of services. Public service managers may be held accountable for providing services working with partners in the private sector that are timely, responsive, reliable, accurate, error-free, customer-friendly, and convenient.
Formally, in a hierarchy in the private sector, accountability is ensured when a leader’s mandated actions and decisions are evaluated by his or her higher authority, with consequential reward or punishment (Kearns, 1996). For administrators in the public sector, authority is vested in the elected president and his appointees, the legislature, and the judiciary. Elected officials are held accountable through the election process, rules, regulations, and traditions. But leaders create arrangements to protect their positions and avoid openness, transparency, and accountability. As evidenced by numerous financial scandals, the CEO is often not held accountable by the corporate board of directors—a signal flaw in corporate governance in the private sector and NGOs. Monitoring systems are required. In the private sector, these include effective legal regulation and uncorrupted accountants, dedicated lawyers, investigative media, and protected whistle-blowers. In the public sector, they include a free press, citizen advocacy groups, open political parties, rule of law, minimum secrecy in agencies, and an uncorrupted electoral process and judiciary (Gardner, 1988). Unfortunately, T. R. Mitchell (1993) found no assured accountability in either the public or the private sector. Neither government administrators nor business managers can be recalled by democratic voting procedures. More often than not, administrators and managers are protected by a concentration of economic and governmental power and by immunity from criminal and civil liability. Unconstrained power affects accountability by biasing appraisals of one’s own and others’ performance (Kipnis, 1987). Nevertheless, “The requirement that one be answerable for one’s decisions and actions is an implicit, if not explicit, assumption of organizational systems. Accountability is an integral part … of understanding … organizational theory” (T. R. Mitchell, 1993, pp. 116–117).
Delegation implies that one has been empowered by one’s superior to take responsibility for certain activities. The degree of delegation is associated with the trust the superior has for the subordinate to whom responsibilities have been delegated.1 When a group is the repository of authority and power, it likewise may delegate responsibilities to its individual members.
The delegation of responsibilities should not be confused with laissez-faire leadership2 or abdication. A leader who delegates is still responsible for following up whether the delegation has been accepted and the requisite activities have been carried out. The leader remains accountable (Anonymous, 1989).
The delegation of decision making implies that the decision is lowered to a hierarchical level that is closer to where it will be implemented. Such delegation is consistent with the encouragement of self-planning, self-direction, and self-control.
Yukl (1998) suggests several reasons why leaders delegate: (1) the task can be done better by a subordinate; (2) the task is low-priority and not urgent; (3) the task is a career development experience; (4) the task is not central to the leader’s role. Furthermore, the leader can use the time for other necessary activities. The leader’s capabilities and influence can be multiplied. Better use can be made of the subordinate’s time and abilities. The leader can demonstrate trust in the subordinate. Delegation can develop a greater sense of challenge, initiative, responsibility, authority, and autonomy in the subordinate (Avolio & Bass, 1991).3
Kuhnert (1994) offered several reasons why leaders may avoid delegating. A transactional leader cannot suspend an agenda or coordinate an agenda with others. A transformational leader may place too much weight on current role expectations and may be unable to make difficult decisions that might entail possible loss of respect.
Stogdill and Shartle (1948, 1955, 1975) developed the RAD scales to measure organizational responsibility (R), authority (A), and delegation (D). Responsibility and authority, when combined, were said to approximate accountability. The selfrated responses by individuals describing their own jobs formed scales that had high reliability. Correlational results were obtained in 10 organizations engaged in the production of chemicals, the manufacturing of metal products, and governmental work. As expected, the responsibility and authority of job occupants were correlated positively, ranging from .13 to .63 with a mean of .45. But the relationship of delegation to authority and responsibility was not as high. Delegation averaged .17 in correlation with responsibility in the 10 organizations, with a range from –.27 to .38. Delegation averaged only .23 with authority, with a range from organization to organization from –.42 to .49. (These surveys are over half a century old and would be worth replicating in the twenty-first century.)
A complaint of managers is that they are often delegated a great deal of responsibility by their superiors without the associated authority. This is a commonplace occurrence where leaders are delegated to lead teams of professionals (McKenna & Maister, 2002). The leaders need to depend on their personal expertise, esteem, and persuasiveness. Yet satisfaction and productivity are greater when delegation, responsibility, and authority are matched. In one of the 10 organizations that Stogdill and Shartle studied, there was a correlation of .42 between the managers’ self-rated delegation and their actual authority. Dissatisfaction and ineffectiveness were rampant. In the eight organizations in which delegation correlated highly, authority, satisfaction, and effectiveness were much higher than in the two organizations in which delegation and authority were mismatched. Similar inferences could be suggested about the wide range of correlations, from .13 to .63 obtained between responsibility and authority. In the organization in which responsibility and authority correlated .63, operations went much more smoothly than in the organization in which the correlation was only .13.
Age, Rank, and R, A, and D. Some other patterns of interest emerged in Stogdill and Shartle data that were a useful way to describe likely differences among organizations in leadership and management. For instance, age and seniority varied from one organization to another in their relation to responsibility, authority, and delegation. In one organization, authority was invested in older members, but in another organization it was invested in younger ones. The correlation between age and authority within the organizations ranged from –.26 to .45. It does not take much imagination to contrast the organization in which self-estimated authority correlated .45 with age and the one in which the correlation was –.26. The organization with the negative correlation was a rapidly expanding technology firm; the organization with the high positive correlation was a stable agency where age and promotion were in lockstep. More often than not, the educational level of an organization’s managers correlated positively with their responsibilities and authority. The managers’ own perceived responsibility and authority were unrelated to their subordinates’ job satisfaction, but a mean correlation of .19 for the 10 organizations was found between the managers’ perceived authority and the extent to which their superiors were seen as considerate in leadership behavior. Evidently, the higher members rise in authority in an organization, the more they are treated with consideration by their superiors. H. J. Bowman (1964) found, similarly, that school principals who rated themselves higher in responsibility, authority, and delegation described their superiors as being more considerate of them.
Rank, Salary, and R, A, and D. Kenan (1948) used the RAD scales to study large governmental organizations and found, as expected, that executives in higher-level positions described themselves as being higher in responsibility and authority than did those in lower-level positions. Correspondingly, Browne (1949) noted that executives’ salaries related positively to self-estimates of responsibility and authority. D. T. Campbell (1956) reported that authority and delegation but not responsibility were positively and significantly related to one’s level in various organizations, military rank, time in the position, and recognition for being in a position of leadership. Strong positive correlations should be expected between Jaques’s (1956) time span of accountability and authority and responsibility as measured by the RAD scales.
Hierarchical Effects. In four large naval organizations, Stogdill and Scott (1957) found that the higher the responsibility and authority of their superiors, the less the subordinates tended to delegate. Yet, as might be expected, those superiors who delegated the most had subordinates who rated themselves highest in responsibility, authority, and delegation.
When superiors rated themselves as having a high level of responsibility, their subordinates rated themselves as having high levels of responsibility and authority, but these subordinates’ self-ratings were unrelated to the authority of their superiors.
Stogdill and Scott (1957) analyzed the responsibility, authority, and delegation scores of commanding officers and executive officers and the average RAD scores of their junior officers on submarines and on landing ships. The executive officers tended to delegate more freely to those below them on both types of ships when the commanding officers exercised a wider scope of responsibility and authority and delegated more freely. But the relationship between commanding officers and executive officers on the submarines and landing ships differed substantially. Executive officers reported more responsibility and authority when their commanding officers were high in responsibility, authority, and delegation on the submarines, but the reverse was true on the landing ships.
The RAD scores of the junior officers were systematically related to the RAD scores of their executive officers and to those of their commanding officers. As the authority of their executive officer increased, junior officers’ RAD scores tended to decrease. When their commanding officer delegated more freely, the junior officers reported an increase in responsibility on landing ships but a decrease in responsibility on submarines. Stogdill and Scott concluded that the commanding officers could increase or decrease the workloads and freedom of action of their executive officers. In the same way, junior officers tended to tighten their controls as superiors increased their own sense of responsibility and freedom of action. However, responsibility and authority did not flow without interruption down the chain of command. The responsibility, authority, and delegation of subordinates were more highly influenced by the subordinates’ immediate supervisors than by their higher-level commanding officers and executive officers.
According to Stogdill, Scott, and Jaynes (1956), when U.S. Navy supervisors were perceived as delegating freely, their subordinates not only rated themselves higher in responsibility and authority but thought that they should have a high degree of responsibility and authority. However, the subordinates believed that they themselves spent too much time on inspections, preparing procedures, training, and consulting peers and not enough time on coordination and interpretation.
The responsibility of subordinates appeared to be more highly related than their authority to the responsibility and authority of their superiors. An increase in the responsibility of superiors had the effect of increasing the responsibility of their subordinates, but it did not necessarily increase the subordinates’ authority. This was a latent cause of dissatisfaction. An increase in the superiors’ responsibility, authority, and delegation also appeared to require that the subordinates increase their coordination activities.
Authority, responsibility, delegation, and accountability have been subjects of commentary and research in leadership and management since early in the twentieth century. Authority is the legitimate right to exercise power, but it depends on the willingness of others to accept it. It is less often the main source of leadership than it used to be. Authority is both allocated from above and acknowledged from below before it converts into power and leadership and becomes strong in its effects on leadership and wide in scope. Leaders are assigned responsibility and held accountable for using it commensurately with their authority. This situation can provide a check on leaders. Discretionary leadership is needed beyond that required by authority and responsibility, particularly as a consequence of changing patterns of authority. Authority and responsibility are correlated, and both contribute to accountability. Leaders’ responsibilities are broader than those of their subordinates but are dependent on their subordinates. Responsibilities but not accountability can be delegated. Delegation should be to levels in the organization closest to the decisions that need to be made.
Supervisors with more responsibility generate more responsibility among their subordinates, but the same is not true for the downward flow of authority. The responsibilities of superiors tend to influence the performance of subordinates, but their authority has a stronger impact on the subordinates’ expectations. When superiors have a great deal of responsibility and authority, their subordinates believe that the demands made on them for coordination increase. However, when superiors delegate a great deal to their subordinates, the subordinates report that they are overburdened with responsibilities and need more authority than they possess. Subordinates do not view delegation as an unmixed blessing when they have critical or burdensome duties to perform.
Organizational effectiveness depends on the congruence of responsibility, authority, and delegation and their interrelationships at different levels of the organization. Their combination provides a gauge of their accountability. The link between the authority and responsibility of superiors and subordinates is embedded in the larger framework of superior-subordinate interdependence and mutual reinforcements.