The second step in the model of organizational change specifies that one outcome of the distribution of power within the organization is the, selection and tenure of individuals in major administrative positions. Since power conveys the ability to influence organizational decisions (e.g., Pfeifer and Salancik, 1974), it is likely that power will be used to influence the choice of top administrative personnel. Indeed, the: designation of formal leaders is one of the most obvious ways of in- stitutionalizing power in organizations.
We assume that the problems and difficulties facing the organization are likely to be attributed to the administrator, regardless of whether the administrator is the cause of such problems (e.g., Gamson and Scotch, 1964). Visible failure calls into question the competence of those in control, and the administrators become visible and, at times, arbitrary targets. One empirical implication of this assumption is that administrators will tend to be removed when organizational perform mance is below acceptable levels. However, performance and organizational problems are not the critical variables in administrator removal. The administrator’s own perceived capacity or incapacity for dealing with the problems is important. Another critical factor is the importance of the problems facing the organization. The problems* themselves must be important. And finally, another factor affecting the removal or tenure of administrators is the institutionalization of power; those who have institutionalized their positions of control should last longer given the same level of performance problems. In brief, the tenure and removal of administrators should be a function of the, existence of critical difficulties, frequendy deriving from the organization’s context, and the ability of the administrator to cope with these problems or to maintain control independent of his coping capability, The selection of new administrators should be similarly related to the environmental context of the organization. Those in power should tend to select individuals who are capable of coping with the critical problems facing the organization. Such selection may occur as a consequence of planning or it may occur unintentionally. Organizational members in power may intentionally select administrators who have characteristics that would be useful in coping with the organizations context and contingencies. However, a similar result would arise unintentionally from the decision situation. There is a tendency, under certain conditions, for decision makers to favor candidates who are, similar to themselves (e.g., Byme, 1969; Berscheid and Walster, 1969). If some organizational subunits have more influence than others, it is likely that the leader selected will tend to have characteristics similar to those with power in the organization. There will be a tendency for accountants to think that an accountant should head the organization, and for marketing personnel to favor a leader with a marketing background. The resolution of such differences in opinion would be that those with more power would have more influence in the decision, so the person selected would reflect characteristics of those in power. Since subunit power is derived from coping with critical contingencies, the selected administrators will tend to have characteristics appropriate for the organization’s context. Therefore, for either intended or unintended reasons, the characteristics of administrators should be related to the context of the organization, and particularly in the case of recently selected administrators rather than those who have more institutionalized positions.
We view administrative succession as a political process of contested capability, where the contest is resolved by subunit power. Surprisingly. there is very little literature relevant to this position. Succession and tenure have most frequently been related to organizational size (Grusky, 1961; Gordon and Becker, 1964), with inconsistent results. Size itself may be confounded with political aspects of organizations. Large size may be correlated with a larger number of departments or subunits, which provide more bases of power and therefore more potential contests for control. Organizations differ in the extent to which they have clear-cut performance criteria, as well as in their actual performance. Size may be complexly related to both of these factors. The typical approach to examining administrator tenure does not provide information about the process or causes of change, but it is not necessarily inconsistent with the approach we are developing.
A notable exception to the size-turnover theme was the study by Zald (1965) of succession in a large social welfare organization. Zald argued that succession to the position of chief administrator was the outcome of political processes within the organization. Similarly, Perrow (1961) has noted that control in hospitals passed from the trustees to the doctors and then to the professional hospital administrators. There is some suggestion in Perrow’s analysis that control has changed as environmental contingencies facing hospitals have changed. While hospitals first faced problems of obtaining funding and later problems of accreditation and quality, today hospitals face problems of cost controls that require professionally trained administrators.
Pfeifer and Salancik (1977) directly examined the relationship between an organization’s context and the tenure and selection of administrators in 57 midwestem hospitals. This study related context directly to administrator succession, without examining the intervening causal linkage of the relationship of context to power and power to succession. Though not explicating this intervening step, the study represents the only empirical examination relating succession to organizational context.
Pfeiffer and Salancik argued that the selection and tenure of chief executives in organizations are consequences of the organization s context and the ability of administrators to cope with the uncertainties and contingencies deriving from that context. Three hypotheses guided the study. First, the frequency of executive turnover should be a function of the stability and problems confronted by the organization (Gouldner, 1954). Second, the characteristics of the chief executive should be systematically related to the organization s requirements for dealing with critical contingencies (Thompson, 1967). And third, to the extent that power is institutionalized, the relationship between context and selection will be attenuated. To examine these hypotheses, questionnaires were given to the chief administrators of hospitals. Data were obtained on the hospital’s sources of funding, ownership, size, and competitive conditions. Professional backgrounds of current , administrators were obtained, and tenure was assessed by measuring, the length of time the administator had served in that position.
1. Organizational Context and Administrator Jenure
There is some evidence that long tenure in office is negatively associated with organizational problems. Grusky (1963) studied turnover of baseball managers and found turnover to be negatively related to team performance, which Gamson and Scotch (1964) argued represented scapegoating. Pfeifer and Leblebici (1973) found that the average, length of tenure of chief executives in business organizations de-, creased as the organization’s debt to equity ratio increased- McEachem (1975) observed that turnover of chief executives is more likely when the firms managed suffered declines in profits four or more years in a row. In a university, Salancik, Staw, and Pondy (1975) observed that turnover was more likely in academic departments that needed resources and had difficulty in acquiring them. Departments characterized by high resource interdependence among membeis also had higher turnover in the leadership position.
The study by Salancik et al. (1975) attempted to show that turnover was not only associated with the presence of organizational diffiW culties, but that a critical variable was the ability of the administrator to cope with these difficulties. In addition to characterizing departments by their requirements for resources, departments were also characterized by their degree of paradigm development. It was assumed that departments in high paradigm disciplines should find it easier to resolve issues of resource allocation since such departments have well-developed frameworks within which to make such decisions (Lodahl and Gordon, 1972). Salancik and his colleagues found that over a twenty-year period, turnover was greater in departments characterized by poorly developed paradigms and high resource interdependence. Departments that had difficulties in obtaining requisite re- sources and that were also unable to resolve conflict because of the absence of shared judgments experienced twice the rate of administrator turnover of other departments. Administrators, removed when problems confronting the subunit could not be resolved, were scapegoated, since the lack of consensus is a function of the department’s discipline and not the administrator.
In the study of hospital administrators, Pfeffer and Salancik (1976) attempted to relate tenure to hospital problems by examining three sources of external contingencies: competition for funding and staff; relationships with business and general community interests; and the condition of the hospital’s operating budget, specifically whether surpluses or deficits were being run.
Correlations between administrator tenure and hospital context are displayed in Table 9.1. Tenure was longer in hospitals with less competition for funding and staff. Tenure was also longer for administrators in hospitals with better;relationships with the business and local community. On the other hand, there was no simple relationship be-tween tenure in office and the state of the hospital s operating budget.
The lack of a relationship between tenure and the operating budget might at first appear to be inconsistent with the idea that executive tenure is reduced by organizational problems. A more sophisticated analysis, however*, suggests that the operating budget is more or less critical for different hospitals. Some hospitals rely on operating surpluses for financing capital projects. Other hospitals, however, rely more on private contributions for capital expenditures. Clearly, private donations and operating funds represent alternative sources of financing, and the importance of operating funds for the hospital is contingent on the importance of private donations. Therefore, for hospitals which do not rely on private donations, tenure should be related to the status of the hospital’s operating budget. For hospitals relying on private donations, tenure and the operating budget should be relatively uncorrelated.
More important to hospitals relying on private donations should be the relationship with sources providing donations, such as the business community.
To examine the effect of funding sources on hospital administrator tenure, the 57 hospitals were divided into those with larger and those with smaller proportions of their capital budgets obtained from private donations. For hospitals relying more on donations, administrator tenure was highly correlated with the hospital s relations with the business community (r — .43) and was virtually independent of the status of the operating budget (T— .01). For hospitals relying less on donations and more on operating funds, administrator tenure was correlated with the operating budget condition (r — .23), and while also correlated with the hospital’s relations with the business community, this correlation was smaller than for the other set of hospitals. The evidence suggests that it is not the presence of problems, but the criticality of the problems to the organization that affect administrator tenure.
Another factor we have suggested that might affect administrator, tenure is the institutionalization of power in the organization. To the extent power becomes institutionalized, the administrator can continue even when confronted by unresolvable contingencies. This source of indeterminacy in the environmental-context- administrator-tenure rela- . tionship can be indirectly examined by considering the moderating * effects of variables which might be associated with power institutional- . ization. One such variable is the ownership of the organization. In the study of hospitals, there was some relationship between ownership and tenure, with a positive association for administrators in private, non-profit hospitals and a negative association between tenure and administering a hospital affiliated with a religious denomination. There are a viariety of ad hoc explanations that could be mustered for these relation-ships, including the fact that administrators of private, nonprofit hos-pitals are not embedded in a larger organization with a hierarchy of authority, as is the case for both religious and government hospitals.
Business organizations provide a better setting for observing the effects of ownership on tenure. In an insightful reexamination of the ; separation of ownership and control issue (e.g., Berle and Means, 1932), McEachem (1975) categorized firms as being: (1) owner-managed, in which the firm was both controlled and managed by the : owners; (2) externally controlled, in which the firm was controlled by owners but managed by other managers; and (3) manager-controlled, in which the firm was both controlled and managed by nonowning managers. In the first case, dominant shareholders are involved in the management of the firm. For the owners to fire the managers would involve firing themselves. In the second case, there are dominant t shareholders who control the firm but leave its management to others. 4 case, the manager is readily fired since there is some concentra-tion of power to accomplish the removal. It might be expected that tenure would be longer for executives in firms of the third type, in which the managers both controlled and ran the organization.
McEachem related the three types of ownership to executive turnover. Estimated equations for a sample of 96 firms and an industry sample of 48 firms are presented in Table 9.2. As expected, there are some dramatic differences between types of ownership and tenure. Tenure is about twice as long in the owner-managed firms than in either of the other two categories. While the coefficients in the two equations are in a direction indicating some effect of management control to lengthen tenure, neither coefficient is statistically significant.
Neither of the analyses reported by Pfeffer and Salancik ( 1977) or by McEachem (1975) test the more reasonable proposition that ownership has effects on tenure depending on the level of organizational performance. One possible reason why neither study found stronger results is that it would be expected that ownership would interact with organizational performance to determine tenure. If the administrator were handling the job brilliantly, he would remain in office regardless of ownership. It is only when the organization confronts difficulties that the institutionalization of power, represented by differences in ownership, would have a differential effect on tenure.
The length of an administrator’s tenure in office is likely to be less when there is visible evidence of his inability to cope with critical contingencies and he is unable to institutionalize control. While there has been limited research directed to this issue, the available evidence is at least not inconsistent with this proposition.
2. Organizational Context and the Characteristics of Administrators
We have suggested that the organizational context tends to encourage the selection of administrators appropriate for coping with that context. Anecdotal evidence seems consistent with this position. At one time corporate executives tended to be drawn from the ranks of line production executives, who could cope with the problems of. production. As production became increasingly routinized and mechanized, the problems changed to selling the products, and marketing executives became more prominent. Then, in the 1960s, organizational contingency. Financial executives came into power in organizations. Most recently, corporations have faced regulatory and legal problems, and lawyers have become increasingly powerful and prominent as chief executives. Pacific Gas and Electric, a public utility, formerly was dominated by engineers. Today, most of the engineers , who populated the high administrative positions are gone, replaced by lawyers who can deal with the increasingly complex legal and regulatory environment.
If our argument is correct, and hospitals also select administrators who can cope with critical contingencies, then we should find systematic relationships between the hospital’s context and the characteristics’1, of its chief administrator. Pfeffer and Salancik (1977) examined th relationship between the hospital administrator s formal training ffi, hospital administration and dimensions of the organizational conteste The relevant correlations are displayed in Table 9.3. Training in hosv pital administration was related, as expected, to the context of the funding sources. Hospital administrators tended to have more training to the extent that the organizations received insurance funds and less training when private donations were more important.
The correlations in Table 9.3 are not very large, however. We have suggested that the extent to which administrators’ capabilities will be aligned with environmental contingencies will vary with the institutionalization of control by the administrator. More recently appointed administrators should be more aligned with the environmental contingencies facing the hospital, for the longer-tenured administrators may, in part, reflect stable institutionalized structures of control.
The sample of 57 hospitals was divided according to the length of the administrator’s tenure. The subsample of recently appointed administrators, constituting about one-half the sample, held office for four years or less. The correlations between context and training in hospital administration for the two subsamples of long-tenured and recently appointed administrators are displayed in Table 9.4. As expected, the relationships displayed in that table are stronger for more recently appointed administrators. This suggests that not only are administrator characteristics predicted by the organization’s context, but that new appointments are even more strongly affected by the organizational environment. The study of hospital administrators is generally consistent with our argument that one manifestation of the environment’s impact on organizations is through the selection of administrators.
Source: Pfeffer Jeffrey, Salancik Gerald (2003), The External Control of Organizations: A Resource Dependence Perspective, Stanford Business Books; 1st edition