The final link in the model of environment-organizational effects suggested that the administrator can affect decisions that direct the structure and activities of the organization. Presumably, such direction is to make the organization more consistent with the requirements of its environment. Thus, hospital administrators bring ideas for change, establishing elaborate inventory and cost accounting systems and investing in modern management systems and advanced equipment reduce costs. They make decisions that redesign the organization in line with their views of what is needed. As in the case of the other;’ linkages, the link between the administrator and organizational action is imperfect.
Child (1972) has noted that many researchers have posited various forms of technological or environmental determinism. Environmental 01 technological factors are presumed to place overwhelming constraints or requirements on the organization. The role of administrators can be safely ignored since the design and activities of organizations are dictated by these constraints. Although the few studies that have attempted to estimate the amount of variance in organizational performance attributable to administrators (Lieberson and O’Connor, 1972; Salancik and Pfeifer, 1977) have found that the administrator accounts for relatively little variation, there is some administrator effect. Complete determinism seems unlikely. Some change in organizational activities and performance is associated with administrators.
Although we are sympathetic with the view that the environment constrains the organization’s design and activities—indeed, this is one of the points we have tried to emphasize throughout the book—we believe a more sophisticated and complex understanding of the process is required. There is some empirical evidence that organiza- tional administrators can affect decisions, particularly for their own benefit. Stagner (1969), for instance, in a survey of business executives, reported that the executives claimed that powerful departments could get their way without Regard to the welfare of the entire organization. Pfeffer and Salancik (1974) found that power of subunits was an important predictor of budget allocations in a university, and that even after accounting for subunit power, administrative strategies had statistically significant effects on resource allocations (Pfeffer and Salancik, in press). The various case studies of business firm decision making tend to indicate the importance of who is in control for predicting organizational actions. Cyert et al. (1956) found that forecasts were developed to justify the desired decisions, rather than having the decisions based on the forecasts. Baldridge’s (1971) case study of an eastern university also suggests the applicability of a political model of organizational functioning, stressing the importance of who is in control.
A more realistic perspective on organizational action would recognize that organizational actors mold organizational activities, but do so within constraints which limit their discretion to take action. To explicate such a perspective it is necessary to develop a model that suggests how much constraint an administrator faces in formulating action. Both the omnipotent administrator and the impotent administrator are equally inaccurate representations of reality. It is necessary to develop a model specifying how much effect administrators can, or should, have and under what conditions.
Perhaps the most important thing to recognize about discretion is that it is rarely unilateral. Discretion, as well as power, is shared in organizations. The manager, though a leader, is also a follower who responds to the demands of those with whom he deals and upon whom he depends for support to accomplish his own activities. The subordinate, although sometimes thought of as dependent, can be seen as powerful, since the subordinate frequently controls resources 1 or performances critical to the activities of the manager (e.g., Mechanic, 1962). Pfeffer and Salancik (1975) attempted to illustrate the external constraint on leader behavior in a study of first-line supervisors in a housing office in a large state university. The supervisors were asked to describe how they behaved with respect to a variety of activities and also how they thought their subordinates, bosses, and peers wanted them to act. Analyses indicated that supervisors were more likely to behave as their subordinates wanted when they were more similar to their subordinates (in sex and work activities) and more integrated with subordinates in the social network. The supervisors reported acting in conformity with the demands of their bosses to the extent that they were more similar to the boss and that they faced pressure for performance.
In another study conducted in an insurance company, Salancik et al. (1975) asked subordinates how they wanted their bosses to act and compared these demands with the managers’ own reports of activities. The behaviors consisted of those commonly associated with work and supervision. Supervisors were classified according to how much their organizational position required them to coordinate their department’s activities with other departments. The hypothesis stated that the more a given manager had to coordinate with other departments, the more constrained his behavior would be and the less likely he would be able to construct behaviors corresponding to the demands of subordinates. As predicted, an inverse relationship was observed between the extent of coordination with others and the supervisor’s satisfaction of subordinate demands (r = -.91).
In another attempt to understand how administrators formulate actions in a political and social context, Pfeffer and Salancik (1976) examined the relative discretion of mayors on city budget items for a sample of 30 United States cities. They argued that the mayor should have more or less discretion depending upon the level of organized interests operating in the city. Measuring mayor effect as the amount of variation in budget categories attributable to the mayors from an analysis of variance procedure (e.g., Salancik and Pfeffer, 1977), Pfeffer and Salancik found that mayors confronting conditions associ- ated with less organized interests had more effect on budget categories. Mayors with more discretion were in cities with a higher proportion of nonwhites, lower median income, and with a higher proportion of persons employed in government employment and construe- tion. Mayors with less discretion were in cities with higher median income, more persons employed in manufacturing (a surrogate measure of unionization), and with a higher proportion of professional and managerial persons. These results indicate that administrator effects are reduced by organized, potent interests, and that factors associated with these interests therefore can be used to partially predict administrator effects.
Although there has been remarkably little attention to organizational decision making and managerial action, the information presented above suggests that a model of managerial behavior and its consequences can be developed. It will be necessary to take into account the social and political realities of organizations and to consider that actions are formulated under constraints mediated by others in the organization who have their own power and discretion. This model of decision making within organizations, it should be evident, is similar to the argument we have developed with regard to organizations operating in externally controlled environments. Power is shared and actions are consequences of power. Power, in turn, results from dependencies and interdependencies between actors attempting to achieve their own objectives.
Child (1972) has argue„cl that a number of organizational structures and actions may be consistent with environmental requirements. His emphasis on strategic choice reintroduces the importance of control and power to the analysis of organizations. Of course, it is possible that the strategic variations resulting from choice are only short-term effects, and that over longer periods of time, environmental determinism such as that resulting from selection may predict the organization- environment relationship. Although this possibility warrants exploration, the available evidence indicates that organizational decisions differ depending on who controls the organization.
Source: Pfeffer Jeffrey, Salancik Gerald (2003), The External Control of Organizations: A Resource Dependence Perspective, Stanford Business Books; 1st edition