Organizational Discretion in Responding to Institutional Practices: Hospitals and Cesarean Births

Article excerpt

During the past 15 years, organizational theorists have sought to understand why organizations adhere to dominant practices in their interorganizational fields. Neoinstitutional theory has provided a common framework for answering this question. It suggests that organizations seek legitimacy and attain it by conforming to prevailing institutional norms for practice. Theorists have viewed institutional expectations as agreements about the correct way to do things and have investigated their impact on structural conformity and isomorphism, through mechanisms such as societal norms, professional training and accreditation, and state regulation (Meyer and Rowan, 1977; Scott and Meyer, 1983; DiMaggio and Powell, 1983).

An important aspect of these early views was that organizations and their interests were underemphasized, and in some cases discounted, as relevant to understanding institutionalized practices. More recently, theorists have sought to renew attention to interests and agency (e.g., Mezias, 1990; Oliver, 1991; DiMaggio, 1991; Brint and Karbel, 1991; Goodstein, 1994). They argue that complete and uncontested institutionalization is rare and that interests and agency play a role in determining how organizations adapt to their institutional environments. Oliver (1991), for example, noted that because institutional environments are not always unitary and organizations are not always passive, an organization may respond to institutional pressures according to its resource dependencies. Goodstein (1994) suggested that organizations respond strategically to institutional pressures, depending on their idiosyncratic constraints and incentives.

While these recent views return attention to organizational interests, they also risk discounting the social-fact quality of institutions, much as earlier theorists discounted the role of agents. In this paper, we extend arguments about interest and agency in institutional environments while preserving the essential features of an institutional perspective. We propose that institutional standards are insufficient for constraining practice when they are uncertain. As a result, practices are indeterminant, and actors will need to impose additional constraints to define a practice to follow. Because discretion is created by the resulting uncertainty (Pfeffer and Salancik, 1978), actors may use their own particularistic interests to guide their further definition of appropriate action. This discretion, however, is bounded by the institutions that gave rise to it, as will be the choices organizations make when pursuing their own interests.

We suggest that there may be a core set of institutions or institutional standards for which agreement exists and others at the margins for which it does not. By implication, organizational influences on practice will be greatest when institutional standards are most uncertain, and organizational strategic interests therefore influence practices at the margins more than at the core. Organizations thus generate variation in practice while conforming to their institutions by pursuing their strategic interests within the limits of the discretion permitted by the institutions generating it. We examine these arguments within a context of medical decisions to deliver a baby surgically rather than vaginally.


Neoinstitutional theory developed in response to empirical anomalies in organizational studies - certain practices, procedures, and structures could not easily be explained by prevailing rational-actor theories (DiMaggio and Powell, 1991). A common thread of much early work was that the environment consists of taken-for-granted beliefs and rules that penetrate organizations, creating the lenses through which actors view and construct the world. Early work sought to explain structural conformity and isomorphism in areas in which technical uncertainty was high and technical rationality proved insufficient, such as in education (e. …