[thongidr]

Institutional Theory

The origins of "new institutional theory," also known as neo-institutionalism, lie in part in research on prisons, where it was found that informal procedures dependent on inmates were often more important than guard-enforced formal procedures. This was generalized to the notion that explanation of human behavior within institutions had to encompass not only the old institutional focus on formal rules ("by the book" procedures) and value-maximizing rationality, but also a new focus on the role of norms, symbols, myths, belief systems, and informal arrangements forming the culture of the organization. Campbell & Pedesen (2001: ch1), among others, argue that academic interest in institutionalism grows more broadly out of post-WWII upheavals, such as the collapse of Communism, which made questions about the relation of institutions to political and economic change critical. Others trace the rise of institutional theory to a reaction against the rise of social science behavioralism after WWII (Hall and Taylor 1996: 936). In any event, institutional theory is perhaps currently the single most popular theoretical approach within public administration, endorsed by such figures as H. George Frederickson (1999), a leading figure in PA theory. Because of its prevalence, this theory is termed simply "institutional theory" below, but note that some authors wish to reserve that label for traditional theories or public choice theory, and instead use "neo-institutional theory" as the label for what follows.

Overview
Since any writing that deals with governments, corporations, churches, or even families is dealing with an institution, "institutional theory" can be construed to include an extremely wide variety of writings by very different authors. One widely used typology seeking to give order to a very disparate field is that by Hall & Taylor (1996), who differentiated three traditions within institutionalism:

1. Rational choice refers to theories such as public choice theory and other forms of analysis based on the assumption that decision-makers are rational, self-interested optimizers, or social choice theories based on rational voters, principal-agent theories focusing on contracts and incentives linking superiors to subordinates or agents, or market theories in general, which also focus on optimization by individuals in markets. That is, in rational choice theories, social phenomena are interpreted in terms of preference-maximizing strategies pursued by individuals. Because of preference-maximization, Hall & Taylor refer to the rational choice tradition as the "calculus approach" to institutionalism. In the public arena, preference-maximizing strategies revolve around exchanging compliance for preferences in a social bargain which, if not kept, may lead to the breakdown of existing institutional patterns of power (Levi, 1990). While rational choice theories do seek to analyze decisions affecting institutions, this tradition is often seen as a contrast to rather than part of institutional theory for reasons given below.

2. Historical institutionalism grows out of group theories of politics and structural functional theories in sociology, both popular in the 1960s and 1970s. Group theory in political science reacted against postwar behavioralism's attempt to generate class-wide generalizations, instead preferring to study each case as a unique constellation of cooperating and competing actors shaped by unique historical backgrounds. Concepts such as incrementalism and "muddling through", rather than rational planning or even rational choice, dominated historical institutionalism. Structural-functional theory in sociology reacted against economic determinism, instead emphasizing the power of culture, values, and norms in institutional change. Pierson's (2000) work on path dependence fits in this tradition, emphasizing the inertia of institutions as self-reinforcing systems, as does the earlier work by North (1990) on the obstacles to change posed by organizational norms designed to make existing patterns self-perpetuating. Historical institutionalism in general emphasizes the constraints institutional factors place on decision-making and individual discretion.

3. Sociological institutionalism is the type most today equate with "institutional theory" and is largely reflected in other sections below. The sociological tradition is in explicit contrast to rational choice theory, and has overlap with historical institutionalism insofar as it also emphasizes the socio-political-cultural embeddedness of institutional decisions. Hall & Taylor (1996) thus contrast the "calculus approach" of the rational choice tradition with the "cultureal approach" of the historical and sociological traditions. Sociological institutionalism is differentiated from historical institutionalism by less emphasis on power (as in group theory) or even norms (as in structural-functionalism) and more emphasis on the sociocultural construction of "realms of meaning" which in turn guide individual behavior within institutions. Sociological institutionalism does not view social patterns as tending toward equilibrium, as do rational choice theories, nor does it view change as being as tightly constrained by norms and traditions as does historical institutionalism. Themes in the sociological tradition include the symbolic functions of institutional forms, which are not necessarily optimized in functional terms; the cultural specificity of different institutional enactments of the same technology; and the quest for organizational legitimacy as the dynamic behind the diffusion of innovation. Change is seen as a necessary institutional process in pursuit of legitimacy (Levi, 1990).

In spite of different emphases, Immergut (1998) argues that all three neo-institutional schools share the belief that preferences are not expressed by atomistic individuals but are institutionally aggregated. That is, individual decisions do not sum to collective decisions because of the influence of organizational aggregation. The institutions doing the aggregating are biased in particular directions. Rational choice theory points in the direction of an equilibrium of preferences. Historical institutionalism points in the direction of maintenance of the status quo. And sociological institutionalism points in the direction of consolidation of institutional legitimacy in a changing environment.

[thongidr]

Key Concepts and Terms
o Institutions are governance structures based on rules, norms, values, and systems of cultural meaning. Institutions are defined by Elinor Ostrom as "the set of working rules that are used to determine who is eligible to make decisions in some arena, what actions are allowed or contrained, what aggregation rules will be used, what procedures must be followed, what information must or must not be provided, and what payoffs will be assigned to idividuals dependent on their actions" (Ostrom, 1990: 51). Jepperson (1991), however, defines institutions more broadly to be any social pattern characterized by standard sequences of interactions. The definitional focus tends to reflect the social science discipline. In economics, institutions serve to minimize market transaction costs (North, 1990). In sociology, institutions are social structures which come to be sanctioned by the norms and values of the society. In the public sphere, institutions create checks and balances, facilitate political cooperation, and reduce political uncertainties.

o Governance as networking. Institutional theory embraces institutions in all sectors and part of its contribution is addressing public administration issues in a world characterized by greater inter-sector partnering, cooperation, and interchange. Frederickson (1999) notes that whereas traditional public administration focused on government as an actor amid clashing interests each seeking to "win," institutionalism refocuses public administration on government actors embedded in networks of cooperation which often transcend the public sector. This refocusing is more appropriate for what he sees as the emergence of a "fragmented and disarticulated state," quite in contrast with traditional Weberian and Wilsonian conceptions of a hierarchical, command-driven state sector separate from other sectors. Institutionalism is closely allied with network theory as developed by authors like O'Toole (1997a, 1997b, 1997c).

+ Transformational leadership. Institutional analysis suggests new roles for agency leaders in governance through networks, and also in helping to construct the symbols and meanings by which those in the network (both internal to the organization and external in its environment) perceive what it is the organization does and what its choices are (March & Olsen, 1984; Frederickson, 1999: 704).

o Embeddedness and legitimation. The behavior of individuals in organizations cannot be simplified into a small set of motivations, as public choice theory attempts to do with its assumptions about individuals being rational, benefit-maximizing actors. Instead, behavior is deeply rooted in and reflective of multiple contexts, including culture, the legal framework, agency interest, and so on. Behavior must be explained on a situational basis, with each context different (Scott, 2001). The drive for resource stability leads organizations to seek legitimation, which is achieved by the agency becoming embedded in political, legal, organizational, and cultural relationships which confer legitimacy.

+ Organizational legitimation. DiMaggio & Powell (1983), among other institutionalist writers, emphasize the primacy of the motive of organizational legitimation over the motive of organizational efficiency. When forced to choose, organizations will select options which preserve and enhance organizational legitimation. A corollary belief is that legitimation leads to continuation of the resource stream upon which the organization depends, more so than does organizational effectiveness in achieving the agency mission. Reliance on legitimating external institutions rather than internally-generated efficiencies is a strategy by which organizations reduce turbulence and maintain stability (Meyer and Rowan, 1977:348-349).

+ Technology enactment. Fountain's (2001) "theory of technology enactment" is an example of institutional theory emphasizing embeddedness, for the context of adoption of information technology. Fountain shows that contrary to technological determinism, information technology is implemented ("enacted") by agency leaders and staff in ways that reflect how the agency is embedded in legal, political, social, and its own organizational culture, leading to outcomes different from what might be expected on the basis of sheer technological logic and efficiency.

o Cultural persistence. Zucker (1977, 1983, 1991) described a three step sequential process of development of institutional culture:
1. Habitualization. New problems lead to patterned problem-solving within the organization.
2. Objectification. A shared understanding of problem-solving patterns associated with a particular problem develops.
3. Sedimentation. The shared understanding becomes a patterned typification and behavior reflects this typification rather than the original problem. "Path dependence" emerges in which following the pattern is imbued with a myth of effectiveness, not requiring justification by objective evidence (March & Olsen, 1989; North, 1990; Pierson, 2000).
The further in the three step process of cultural persistence, the less the degree of critical evaluation and modification of behavior.

+ Scripts. DiMaggio and Powell (1991: 15) developed the related concept of scripts. Organizations respond to stimuli through application of rules which apply a pre-established classifications to the stimulus and then apply taken-for-granted procedures. Whereas stimulus response in traditional institutionalist analysis focused on applying internalized organizational values and culture, the concept of scripts was more external, not assuming that the individual employee embraced and internalized organizational culture and acted based on these internalized values; rather the employee was only assumed to apply organizational scripts without necessarily having internalized values or having been socialized into commitment of organizational culture.

+ Bounded rationality. In terms of its intellectual history, the concept of cultural persistence emerged from traditional institutional theory, particularly Herbert Simon's concept of bounded rationality, meaning that institutional cultures develop boundaries restricting the range of solutions which are legitimate to examine (Simon, 1947). Decision-makers do not search all alternatives or truly optimize. Rather, as Simon (1947), observed, they "satisfice" by selecting among choices restricted by organizational rules and customs. The inability of decision-makers to absorb more than limited information facilitates acceptance of such restriction (March & Simon, 1958; March, 1994; March & Olsen, 1999). March & Olsen (1989) refer to such acceptance as the "logic of appropriateness."

[thongidr]

o Cultural self-perpetuation. Cultural persistance may become extended to cultural self-perpetuation as agencies devise rules, scripts, patterns, and arrangements which reinforce core values and structures. In this light, Pierson's (2000) concept of "institutional stickiness" focuses on how institutions develop rules and practices which make policy arrangements resistant to reversal. Path dependence reinforces such rules and practices.

o Formal vs. actual organizational procedures: Stories and rationales. Formal organizational procedures may serve symbolic purposes of legitimation rather than substantive purposes of performance (Meyer & Rowan, 1977). Meyer and Rowan noted the divergence of the formal and the actual was supported by belief systems and associated symbolism they termed "rationalized myths." Rationalized myths were stories constructed by the organization to conform to what the organization was supposed to be doing, even when stories diverged sharply from actual processes. Stories served symbolic purposes of reassuring influential actors and publics whose support the organization needed.

+ Constructivist theory holds that human behavior within organization reflects the social construction of common meanings attached to political, economic, and social decisions. Where constructivism sees the definition of social reality emerging on a post-hoc basis as rationalization of organizational choices, in contrast previous rational theories of institutions believed that objective problems and solutions predictably determined later organizational policy explanations and justifications. See March and Olsen (1989).

+ Formal structure as a ceremonial function. Jepperson (2001: 7) described formal organizational structure as ceremonial in function, serving to signal to and reassure external groups as well as organizational constituents that the organization is legitimate. Such "signal rationality" is a strategy for organizational survival which can be contrasted with the economic rationality which traditional institutional theory emphasized as a strategy toward the same end. For instance, establishing an agency website may do little to achieve actual efficiencies but a great deal to signal to legitimating organizations that the agency is a legitimate modern organization (Clerkin, 2006).

+ In the field of sociology, new institutional theory was associated with compatible intellectual trends in ethnography, phenomenology, symbolic interactionism, and other schools which emphasized intersubjectivity and the creating of meaning based on differing contexts for human interaction. These approaches contrasted purported behavior under the formal structure of institutions with actual behavior created in context through informal as well as formal human interaction.

o Isomorphism. In any field, organizations tend to become homogenous in both process and structure over time. Though institutional innovations may spread at first for performance reasons and organizational desire to be seen as being in the vanguard, later in the diffusion process innovations are apt to be adopted for reasons of legitimacy and reducing uncertainty rather than reasons of promoting actual performance. Early differentiation by leading edge innovating agencies gives way over time to homogeneity of approaches and even of organizational structures across agencies in later stages of the diffusion process. Meyer and Rowan (1977: 340) set forth the mechanics of isomorphism: "[O]rganizations are driven to incorporate the practices and procedures defined by prevailing rationalized concepts of organizational work and institutionalized society. Organizations that do so increase their legitimacy and their survival prospects, independent of the immediate efficacy of the acquired practices and procedures." The tendency toward homogeneity was called "isomorphism" by DiMaggio and Powell (1983, 1991), who noted two types: competitive isomorphism arising from market forces and institutional isomorphism arising from competition for political and organizational legitimacy. DiMaggio and Powell focused on institutional isomorphism, identifying its three major mechanisms (see Paradis & Cummings, 1986, on isomorphism in hospice care; Lai, Wong, & Cheng, 2006 on isomorphism in IT adoption).

1. Coercive: Pressures to make organizational procedures and/or structure conform to best practices, arising from the demands of actors on whom the organization is dependent for resources (the "resource dependence model" associated with Pfeffer and Salancik, 1978) or even from outright regulation and mandates (ex., Paradis and Cummings, 1986) .
2. Mimetic: Pressures arising from the drive to reduce uncertainty. Under uncertainty, imitating successful peers is seen as a safe strategy. Haunschild & Miner (1997), for instance, found uncertainty increases mimesis. Note Mizruchi & Fein(1999) argue that subsequent authors utilizing the work of DiMaggio and Powell (1983) have selectively emphasized mimesis over other forms of isomorphism and have used measures of mimesis which are confounded with the constructs of coercive and normative isomorphism.
3. Normative: Pressures arising from professionalization, which socializes personnel within the organization to view certain types of structure and process as legitimate. Socialization occurs not only through formal education but also through professional associations, trade associations, and professional media.

+ Social actor theory holds that human behavior within organizations reflects interactions with important political, economic, and social actors; with peers who are role models defining individual and collective identity; and with networks of professional and organizational groups. These three influences parallel the three bases of isomorphism developed by Dimaggio and Powell. An example, applied to adoption and implementation of information technology within an organization, is found in the work of Lamb and Kling (2003). The social actor metaphor contrasts, of course, with "rational actor" and "economic man" metaphors found in public choice and social choice theories.

o Loose coupling. In spite of cultural persistence and isomorphism, organizations may implement any given innovation, regulation, or policy in different ways depending on the contexts in which the organization is embedded. That is, institutionalism is not a deterministic theory. Meyer and Rowan (1977: 343) wrote, "Structural elements are only loosely linked to each other and to activities, rules are often violated, decisions are often unimplemented, or if implemented have uncertain consequences, technologies are of problematic efficiency, and evaluation and inspection systems are subverted or rendered so vague as to provide little coordination." Loose coupling accounts for the diversity of ways in which agencies implement OMB technology mandates (Fountain, 2001 ) or drug treatment programs (Peyrot, 1991), for instance. Loose coupling can even be used to account for the findings opposed to the bulk of institutional theory, such as the finding of Kraatz and Zajac (1996) that technical and economic pressures led American liberal arts colleges to adopt practices for their professional programs which directly opposed long-held institutional values and the culture within which such colleges are embedded.

+ Chamelon strategies take advantage of loose coupling to present the agency to different audiences in different ways (Peyrot, 1991, discusses this with regard to drug abuse programs).

+ Decoupling. Under certain circumstances, an agency may go so far as to decouple its implementation from the original policy. Decoupling serves to buffer the agency from unwanted associations as it seeks greater legitimation through other associations (Bastedo, 2004, discusses strategic decoupling in educational policy, for ex.).


Assumptions
o Bidirectionality. Culture influences human behavior in organizations, but the reverse is also true. Weick (1995) describes "sensemaking" processes within organizations which lead to evolution of cultural understandings.

[thongidr]

o Culture over power. Though institutional theorists do discuss power, the central thrust to date has emphasized cultural factors rather than interest and power as had the 'old institutionalism.' Some institutional theorists as well as critics of institutional theory believe that a viable institutional theory must resurrect the focus on interest and power, not just culture (see DiMaggio 1988; Brint and Karabel 1991).

o Embeddedness over change. Kim (2005) critiqued neo-institutional theory for overemphasizing institutional embeddedness, inertia, and the constraints of organizational culture, failing thereby to provide adequate explanation for institutional change, which is common. In contrast, Kim noted, systems theory would see embeddedness more as a cause of change rather than of inertia.

o Situationality. Institutionalism holds that behavior must be predicted on the basis of the particular situation in which the actors are embedded, not by generalized explanations such as those found in rational actor theories Taken to an extreme, institutional theory can become a form of particularism, lacking generalized hypotheses and explaining every event by unique elements in the particular setting. For instance, in the study of policy diffusion, homogenization could be explained by isomorphism and lack of homogenization explained by loose coupling - rendering institutional theory something post-hoc rather than predictive in nature. Frederickson (1999: 704) observes, "It could be said of contemporary public administration that we are all now institutionalists." However, a theory that is all things to all people is not a theory with testable hypotheses but rather a generalized vocabulary better described as a jargon than a theory. Hira & Hira (2000) critiqued institutional theory along the same lines, noting that by explaining phenomena on so many bases (technological, cultural, ideological), institutional theory becomes ambiguous and self-contradictory, and as a result may be of less analytic value than if the researcher were to pursue component theories (ex., rational action, cultural determinism) separately since the components lead to more unambiguous hypotheses. To apply institutional theory as a theory, not a jargon, the researcher must limit the importance given to situationalism and embeddedness and instead draw out generalized hypotheses from other key concepts embraced by institutionalism.

A similar point is made by Gorges (2001), who found little explanatory power in new institutionalist hypotheses about institutional change. "The new institutionalism fails to provide an adequate explanation of institutional change because, by relying on variables such as critical junctures, path dependency, leadership or the role of ideas, it leaves institutions behind and employs a grab-bag of explanations that proponents of almost any theoretical perspective could use. The conditions under which these variables matter are unspecified and the causal relevance of institutions themselves is unclear. New institutionalists should specify more rigorously the factors that change institutions and explicate the links between these factors and institutional change. Doing so, however, could mean abandoning their emphasis on the primacy of institutions in developing explanations for political phenomena." (Gorges, 2001: 137).



Illustrative Hypotheses
Hypotheses below are illustrative and not all authors associated with this theory would subscribe to all hypotheses listed.

o Organizations have informal as well as formal structure and procedures, and the two diverge to a lesser or greater extent.
o The greater the divergence of formal and actual, the more prevalent the rationalization of processes in stories providing organizational legitimation.
o The more prevalent rationalization, the more the content of communications with influential actors relies on stories.
o Individual behavior in organizations is motivated by the culture associated with informal as well as formal structure, not just by the formal forms of the organization.
o When the legitimating structures and procedures of an organization conflict with rational efficiency and effectiveness, the organization will opt for the former over the latter.
o Organizations which opt for external legitimation over internal efficiencies will be more stable/experience less turbulence.
o Individual behavior cannot be predicted on the basis of rational-technical considerations alone (ex., technological optimization, interest optimization). It must be predicted on a situational basis, taking into account culture and interests as well as the legal-formal setting.
o The more long-standing an agency's dealing with a problem, the more the objectification and sedimentation (problem-solving patterns become typified and behavior reflects this typification, not the problem itself).
o The more long-standing an agency's dealing with a problem, the less the degree of evaluation and modification of behavior.
o Early diffusion of innovations is more performance based, while late adoption is motivated by needs for legitimacy and desire to reduce uncertainty.
o The more uncertain the environment, the greater the imitation of successful peers.
o The greater the degree of professionalization and professional networking, the greater the isomorphism with regard to an innovation.
o The more long-standing the field, the greater the organizational homogeneity.
o Leadership giving priority to/adept at networking will be more effective than leadership which focuses internally within the organization.
o Institutional theory will better explain any given phenomenon than will public choice theory.

As further illustration, below are some hypotheses explored by Clerkin (2006).
o Organizations that are involved in informal networks and/or formal collaborations have a higher level of organizational formalization than organizations not involved in informal networks and/or formal collaborations.
o Organizations that are affiliated with other organizations have a higher level of organizational formalization than organizations that are not affiliated with other organizations.
o Environmental uncertainty, measured by number of changes in an organization's community that impact the organization, increases the level of organizational formalization.
o Environmental uncertainty, measured by number of government policy changes that impact the organization, increases the level of organizational formalization.
o Nonprofits reporting a change in their revenue streams have a higher level of organizational formalization than nonprofits not reporting a change in their revenue streams.
o The more types of revenue an organization receives, the more formalized the organization.
o The larger an organization's paid staff, the greater the level of organizational formalization.
o The older an organization, the greater the level of organizational formalization.


Frequently Asked Questions
o How is institutional theory related to principal-agent theory?
In principal-agent theory the central problem is how the principal holds the agent accountable to policies desired by the principal. The concept of loose coupling in institutional theory looks at this same issue, though more from the agency viewpoint, emphasizing how agencies may find it in their interest (ex., in their quest for legitimation) or culture (ex., in applying organizational scripts) to loosely couple or even decouple from policies mandated or recommended from above. See Scheid-Cook (1992) and Scheid and Suchman (2001).


o What is structuration theory and how is it related to institutional theory?
Structuration theory is a variant of institutional theory which grows out of the work of Anthony Giddens (1984). Giddens held that individual actions both shape and are constrained by social structures. This two-way cause-effect relationship Giddens (p. 19) called "the duality of structure." The rules of social life (structure) result from the interactions of individuals with each other and with existing social structures in what Giddens (p. 21) called "the enactment/reproduction of social practices." In studying a particular phenomenon, for example e-democracy (Parvez, 2006; Parvez & Pervaiz, 2006)), structuration theory opposes either the technological determinist view that a technological imperative will override human will, or the opposite human design view that information technology is infinitely malleable to human intentions. Rather, it predicts that human behavior will shape information technology, but that technology will reshape human behavior as well. This prediction is very similar to that made with regard to information technology by institutional theorists such as Fountain (2001) and her "theory of technology enactment."

[thongidr]

In both structuration theory and institutional enactment theory, technology may catalyze behavior change but the exact form of resulting structure (ex., centralized vs. decentralized) is socially as well as technologically determined. Other scholars in structuration theory include Orlikowski & Robey (191) and Orlikowski (1992, 2000) A variant is "adaptive structuration theory" (AST), associated with DeSanctis and Poole (1994). While most structuration scholars place more emphasis on social rather than rational/technological determination of structure, AST emphasizes the concept of "the spirit" of the technology or innovation (DeSanctis & Poole, 1994: 126). Existing social structures combine with the spirit of technology to generate the "structural potential" of a given technology or innovation.


o How may isomorphism be measured?
In their study of liberal arts college professional programs, Kraatz and Zajac (1996) measured isomorphism by constructing a homogeneity indicator based on the standard deviation in the number of professional programs offered. Under the hypothesis of isomorphism, it was predicted that the standard deviation would decline over time until relative homogenization was reached. Clerkin (2006) refined this type of indicator to create a more complex and comprehensive homogeneity index which took account not only of number of programs but also various structural traits. In Clerkin's approach, the homogeneity scale sums the number of times an organization's status on a feature matches that of the mode of the field.