This article was downloaded by: [Academy of Management ] On: 21 June 2012, At: 19:27 Publisher: Routledge Informa

Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

The Academy of Management Annals

Publication details, including instructions for authors and subscription information: [Link]

From Smoke and Mirrors to Walking the Talk: Decoupling in the Contemporary World
Patricia Bromley & Walter W. Powell
a b a b

Department of Political Science, University of Utah School of Education, Stanford University

Available online: 08 May 2012

To cite this article: Patricia Bromley & Walter W. Powell (2012): From Smoke and Mirrors to Walking the Talk: Decoupling in the Contemporary World, The Academy of Management Annals, DOI:10.1080/19416520.2012.684462 To link to this article: [Link]

PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: [Link] This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should be independently verified with primary sources. The publisher shall not be liable

for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

Downloaded by [Academy of Management ] at 19:27 21 June 2012

The Academy of Management Annals 2012, 1 48, iFirst article

From Smoke and Mirrors to Walking the Talk:

Decoupling in the Contemporary World
Downloaded by [Academy of Management ] at 19:27 21 June 2012

PATRICIA BROMLEY
Department of Political Science, University of Utah

WALTER W. POWELL
School of Education, Stanford University

Abstract The pervasive spread of rationalizing trends in society, such as the growing inuence of managerial sciences and increasing emphases on accountability and transparency, has created signicant changes in organizations external environments. As a result, there is growing pressure on organizations to align their policies and practices, and to conform to pressures in an expanding array of domains, from protecting the natural world to promoting employee morale. In this context, we reconsider the concept of decoupling as it applies to organizations. Through a review and critique of existing research, we argue that the common understanding of decouplingas a gap between policy and practiceobscures the rise of a more prevalent and consequential

Corresponding author. Email: [Link]@[Link]

ISSN 1941-6520 print/ISSN 1941-6067 online # 2012 Academy of Management [Link] [Link]

The Academy of Management Annals

form of decouplinga gap between means and ends. We describe when to expect both policy practice and means ends decoupling, and we indicate promising areas for research. The major consequences of this overlooked form of decoupling are that in an effort to monitor and evaluate activities where the relationship between means and ends is opaque, (a) internal organizational structures become increasingly complex, (b) organizations persist in a state of perpetual reform, and (c) resources are often diverted away from core goals. Introduction In the late 1960s and early 1970s, a number of organizational scholars adopted an open systems view of organizations. They argued that external inuences both directly in the form of legislation, public policy, and the professionalization of management, and more diffusely through public opinion and social activismled organizations to be more attentive and responsive to their external environments. One line of argument suggested that organizational responses to such pressures often led to buffering of internal practices from outside inspection (Pfeffer & Salancik, 1978; Thompson, 1967). Such efforts were expected to create a decoupling of formal policies from daily practices in an organizations internal technical core (Meyer & Rowan, 1977). Accordingly, organizations adopted policies to conform to external expectations regarding formally stated goals and operational procedures, but in practice did not markedly change their behavior (Scott, 2008). The reasons for this lack of alignment were varied. In some cases, lofty hopes were dashed by the difculty and demands of the tasks (Pressman & Wildavsky, 1973); in others, decoupling was an unintended consequence (Selznick, 1949). And in some circumstances, internal buffering was a straightforward effort to protect internal activities from external monitoring. In recent decades, external inuences on organizations have intensied, perhaps nowhere more notably than in potent movements toward accountability, assessment, and transparency. Scholars trace this fundamental change in societies all around the world, using terms such as audit society (Power, 1994) and audit culture (Strathern, 2000) to describe increasing emphases on monitoring and tracking organizational activities. For instance, in the domain of higher education Espeland and Sauder (2007) richly detail how the inclination to evaluate, count, and rank is deeply reshaping law schools. These regulatory challenges and external pressures, driven by legal and proto-legal inuences, push organizations to align their policies and practices more closely and to conform to external evaluative criteria. The ascendency of neo-liberal ideology, and its attending reication of market principles, has reinforced this focus on measurement. At the same time, organizations are regarded more and more as corporate citizens endowed

Downloaded by [Academy of Management ] at 19:27 21 June 2012

From Smoke and Mirrors to Walking the Talk

with interests and actorhood (Meyer & Bromley, 2012). Their activities are expected to display an increasingly wide array of the proper characteristics of members of society, particularly in such spheres as the environment and equality that may be far removed from (or even contradictory to) the core goal of production (Brunsson, 1989). This general trend has been referred to as the rationalization of the institutional environment (Boli, 2006; Zucker, 1987; see Frank & Meyer, 2002, for an explanation of how growing individualism is linked to broader rationalization). These marked changes in the context in which organizations are embedded call for a reexamination of classic arguments about decoupling, and a richer conceptual understanding of how, when, and why decoupling occurs, as well as its consequences for organizations and society. We review and critique important studies to develop a richer conception of decoupling as it relates to the contemporary world. Most research has focused on the gap between policy and practice. This emphasis interprets the conceptual underpinnings of decoupling too narrowly and mistakenly leads to the criticism that most institutional changes are mere window dressing. In modern organizations, both the type of decoupling and the outcomes may differ from the earlier focus on unimplemented policies. Indeed, gaps between formal policy and implementation may be less common than analysts initially expected (Bromley, Hwang, & Powell, 2009; Coburn, 2004). Moreover, even ceremonial adoption can unintentionally trigger a chain of reactions that have real organizational effects. We develop a typology of decoupling that provides greater insight into the features of modern organizations. Specically, we argue that decoupling occurs at two levels, policy practice and means ends. It is valuable to distinguish between these two forms because of their varied consequences for organizations. Studies that focus on policy practice gaps often focus on why and how managers fail to, or avoid, implementing formal rules. And often the emphasis is on aligning policy and practice over time. We argue that decoupling has conceptual value beyond this common depiction, as it includes instances when means and ends are decoupled. In means ends decoupling, policies are thoroughly implemented but have a weak relationship to the core tasks of an organization. In our expanded conception, means ends decoupling is both a key to understanding the increasingly elaborate internal structures of organizations and it is a main source of heterogeneity in institutional processes. Further, in an increasingly managerial world that emphasizes evaluation, standardization, and benchmarking, the policy practice form of decoupling may become less common, whereas means ends decoupling is likely on the rise. In what follows, we review the evolution of decoupling research and theory, then develop a typology that builds on this literature. Our view reveals under-examined aspects of decoupling, particularly extensive efforts to implement unproven policies as well as buffering between

Downloaded by [Academy of Management ] at 19:27 21 June 2012

The Academy of Management Annals

subunits. Our arguments develop a more comprehensive view of decoupling that incorporates not only the oft-described gap between policy and practice, but also reects the complexity of contemporary organizational practices and goals. We articulate a future research agenda throughout the paper and conclude by proposing consequences for the theory and management of organizations. Theoretical Foundations of Decoupling
Downloaded by [Academy of Management ] at 19:27 21 June 2012

Classical organization research took a straightforwardly functional view of the formal purposes and policies of organizations. Stemming from Webers (1930 [1992]) account of bureaucracy as the outcome of relations between politics and markets, formal organizational elements were expected to be tightly integrated with actual work activities and to play a direct role in controlling and directing day-to-day practices.1 In this view, organizational practices should follow from the goals of their formal policies. Scholars began to challenge this approach with numerous observations of instances where relations among formal structures were loosely coupled (Weick, 1976); they also noted weak linkages between the problems faced by organizations and solutions they came up with (March & Olsen, 1976). One of the earliest applications of the language of loose coupling to organizational research was developed by Karl Weick. In the case of an educational organization, he wrote: it may be the case that the counselors ofce is loosely coupled to the principals ofce. The image is that the principal and the counselor are somehow attached, but that each retains some identity and separateness and that their attachment may be circumscribed, infrequent, weak in its mutual affects, unimportant and/or slow to respond. (Weick, 1976, p. 3) For Weick, the cause of loose coupling was that formal policies were an ideal theory of organizational action, whereas concrete decisions and actions depend on specic contexts and resources. Weick went on to observe a benet of loosely coupled systems. New units could be added to organizations with little disruption to existing operations, and that can be advantageous to organizational survival.2 Weicks observations continue to inuence organizational research today. For example, in a recent book on managing modern organizations, Roberts (2004, p. 68) describes the degree of coupling as a tradeoff between optimal efciency and resilience: A system that is tightly coupled, so that changing any aspect of the design or the environment will compromise performance severely unless numerous other aspects are also adjusted, may work very well if all goes according to plan. He advises organizational designers to recognize the inherent uncertainty of environments and then decide how tightly to link organizational elements.

From Smoke and Mirrors to Walking the Talk

Downloaded by [Academy of Management ] at 19:27 21 June 2012

Meyer and Rowan (1977) extended these initial studies by conceptualizing a more institutional account of decoupling. Consistent with earlier work (Cohen, March, & Olsen, 1972; March & Olsen, 1976; Thompson, 1967; Weick, 1976), they argued that organizations are deeply interpenetrated by their external environment. But in a departure from prior decoupling theory, they offered a phenomenological explanation for the observed gap between formal policy and actual practices. Drawing on Berger and Luckmans (1966) description of the social construction of reality, their institutional view conceptualized modern organizations as the enactment of societal beliefs rather than as rational agents of instrumental exchange. Institutional decoupling was a consequence of organizations becoming isomorphic with widely shared understandings of the social reality. Formal elements of organizations are thus a reection of the environment; internal and external boundaries cannot be sharply distinguished. Structural elements, they observed, 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 efciency, and evaluation and inspection systems are subverted or rendered so vague as to provide little coordination. (Meyer & Rowan, 1977, p. 343) In this view, formal aspects of organization also emerge as reections of rationalizing pressures or myths in the external environment, rather than purely as a result of technical and administrative needs of production.3 Rationalizing pressures from the environment come in two forms. First, organizations seek to secure legitimacy. They are pressed to incorporate elements that are externally legitimated by socially constructed beliefs in the external environment, regardless of their effects on technical tasks at hand. Second, organizations adopt formal policies in order to avoid legal sanctions and the glare of public opinion. In subsequent work, Meyer and colleagues documented the modern institutionalized rules that inuence organizations. These rules are intricately connected to ideas of science, the professions, and the education system (see, for example, the collections in Drori, Meyer, & Hwang, 2006; Krueken & Drori, 2009). Organizations are assumed to reect theories of production that are ingrained in the public imagination and legitimated by the educational system, the legal and political system, and professional bodies. Consequently, policies are often necessarily decoupled from daily practices and activities, which are concerned more directly with production. Later work showed that decoupling also arises due to inertia or to protect local power relations (Dobbin, 1994). Such a view contrasts sharply with the idea that formal organization emerges as an effective functional response to problems of coordination and control.4 In these early conceptualizations, gaps between policy and practice were attributed to the fact that intention and action are driven by different processes

The Academy of Management Annals

(March & Olsen, 1976). The garbage can model of decision-making in organizations emphasizes the loose connections between problems and choices. Cohen et al. (1972, p. 16) note that although decision-making is ideally thought of as a process for solving problems, in reality a major feature of the garbage can process is the uncoupling of problems and choices. March and his colleagues have repeatedly demonstrated the persistent decoupling between information gathering and action (Cohen & March, 1974; Feldman & March, 1982). Weick (1976, p. 4) built on these ideas to develop his view of decoupling. He said that intentions are a poor guide for action, intentions often follow rather than precede action, and that intentions and action are loosely coupled. Here, the argument is that regardless of initial managerial intentions, organizational consequences are unforeseeable in many circumstances. This view is important for understanding some instances of the failure of public or organizational policies to achieve desired outcomes. Similarly, Brunsson (1985, 1989) contends that much irrationality and hypocrisy in organizations can be attributed to the fact that decision-making is based on limited information and a biased set of possible solutions, creating a gap between ideas and action that is difcult to reduce. These early scholars consistently emphasize, however, that decoupling should neither be understood in terms of managerial control, nor as an organizational failure. Decoupling could well be the most functionally effective path in the face of constraints, and it may nonetheless confer legitimacy. Early research highlighted decoupling in such diverse settings as schools (Meyer & Rowan, 1977; Weick, 1976), hospitals (Covaleski & Dirsmith, 1983), courts (Hagan, Hewitt, & Alwin, 1979), and prisons (Thomas, 1984). These public and non-prot sector settings were rife with competing demands, overworked staff, and limited resourceshence, the desire to protect internal operations was understandable. As more domains of the external environment become rationalized, a wider range of organizations face multiple and competing institutional pressures to expand their formal structures in new ways, notably with respect to issues such as equal opportunity, safety, diversity, justice, the environment, transparency, accountability, and consumer relations more generally. The operating environment for organizations becomes more fragmented as the numbers and types of uncoordinated actors on whom organizations depend for material resources and legitimacy increase. Importantly, the environmental fragmentation that shapes modern organizations is often of this highly rationalized form.5 The most inuential external pressures come, for example, in the form of rules and law, in areas such as accounting and auditing standards, consumer safety, labor regulations, or protection for the natural environment. But pressures also stem from soft laws, including numerous forms of standards, ratings, rankings, and the rights-based claims of social movements, as well as general social or professional norms (Jacobsson & Sahlin-Andersson, 2006;

Downloaded by [Academy of Management ] at 19:27 21 June 2012

From Smoke and Mirrors to Walking the Talk

Moerth, 2006). Such competing, fragmented external environments generate conicting demands within organizations. This enlarged operating context now affects much of contemporary organizational life; consequently, there is growing research attention to how organizations respond to a myriad of institutional inuences (Lounsbury, 2002, 2007; Schneiberg & Soule, 2005; Scott, Ruef, Mendel, & Caronna, 2000; Thornton, 2004).

Types of Decoupling: Policy Practice and Means Ends

Downloaded by [Academy of Management ] at 19:27 21 June 2012

Through a review of relevant literature, we have identied two distinctive forms of decoupling that help us gain greater purchase on the relevance and consequences of decoupling in the contemporary world.6 We distinguish, in particular, between policy practice and means ends decoupling. At the policy practice level, decoupling occurs when rules are unimplemented or routinely violated. Decoupling also occurs in the relationship between means and ends, when policies are implemented but the link between formal policies and the intended outcome is opaque.7 This latter form is, we argue, likely to be increasingly prevalent in contemporary society and to have signicant consequences for organizations, particularly in terms of internal complexity, endemic reform, and diversion of resources. Table 1 provides a summary of each type. Policy Practice A gap between policy and practice has become virtually synonymous with decoupling, although it is just one form of the phenomenon. This type includes policies adopted purely as ceremonial window dressing or implemented, evaluated, and monitored so weakly that they do little to alter daily work routines. Figure 1 is a simple conceptual diagram of policy practice decoupling relative to an ideal organization. The causal chain on the left depicts a model in which formal elements of organization are linked to daily activities and these activities create an intended outcome. Studies of policy practice decoupling make the observation that policies are rarely a strong predictor of daily activities, as depicted in the process on the right. This institutional line of argument developed by Meyer and Rowan triggered a wide body of subsequent research related to policy practice decoupling. As an example, in one of the earliest relevant empirical studies, Bussigel, Barzansky, and Grenholm (1986) examine medical schools and nd the presence of incompatible and inconsistent goals whose relationships to each other are only vaguely articulated. Policy practice decoupling allows an organization to adopt multiple, even conicting, policies in response to external pressures, without unduly disrupting daily operations by trying to implement inconsistent strategies. More recently, in an impressive series of

The Academy of Management Annals

Table 1 Overview of Types of Decoupling Type Policy practice Description Formal rules systematically violated and unimplemented Evaluation/ inspection is not present or is intended as a symbolic act. It is so vague that it provides little relevant information Rules and policies implemented, but with uncertain relationship to outcomes May be multiple unrelated goals Key insight Explains why organizations routinely adopt policies and do not implement them Key consequences Legitimacy or resources or both; adoption may be ceremonial Buffering of core activities; may promote efciency or protect interests of internal constituents

Downloaded by [Academy of Management ] at 19:27 21 June 2012

Means ends

Explains why organizations dedicate resources to practices that have little known relationship to intended goals

Legitimacy Implementation contributes to organizational heterogeneity and complexity If evaluation and inspection are present, generates continual periods of reform when inconsistencies are revealed Arational allocation of resources from purely instrumental perspective; can direct time and attention away from core goals

studies, Westphal, Zajac, and colleagues show that corporations adopt elaborate and purportedly instrumental initiatives such as total quality management programs, long-term incentive plans (LTIP) for CEOs, or stock repurchase programs, but that the implementation of these models varies considerably (Fiss & Zajac, 2004, 2006; Westphal, Gulati, & Shortell, 1997; Westphal & Zajac, 1994, 2001; Zajac & Westphal, 2004). Similarly, Weber, Davis, and Lounsbury (2009) examine how the mechanism of diffusion affects the creation and implementation of national stock exchanges, nding that ceremonial

From Smoke and Mirrors to Walking the Talk

Downloaded by [Academy of Management ] at 19:27 21 June 2012

Figure 1

Ideal Organization versus Policy Practice Decoupling.

adoption is closely associated with international coercion. This research provides ample evidence of policy practice decoupling and extends the initial concept by illustrating that policies may not be implemented because of such factors as the interests of organizational leaders and the internalization of environmental pressures. Although conceptually helpful, the insight that policies are routinely adopted to increase legitimacy and avoid punishment can lead to a criticism that institutional adoption is largely symbolic and inconsequential (Anderson-Levitt, 2003; Hannan & Freeman, 1989, p. 94). The central response to this smoke and mirrors criticism has been to argue that legitimacy can be directly related to success and survival through the ability to gain resources or protect an organizations technical core from pressures in the environment, and thus has real effects even if formal policies are not implemented. Earlier studies showed that one benet of decoupling is protection of the technical core of an organization from pressures in the environment. Covaleski and Dirsmith (1983) found that nursing administrators could avoid cost controls by decoupling their communications with hospital administrators from their communications with other nurses. Just as classroom teachers can operate largely independently of formal policies in their classrooms, nurses could exercise a fair amount of autonomy in treating their patients despite formal regulations. In one of their early studies, Westphal and Zajac (1998) showed that stock markets react favorably to the

10

The Academy of Management Annals

adoption of certain governance features, regardless of implementation. At the individual level, Zott and Nguyen Huy (2007) demonstrated that managers who take symbolic steps to signal their personal credibility, professional organizing skills, organizational achievement, and the quality of their external relationships gain more resources for new ventures than those who do not. These examples notwithstanding, only a limited number of studies have directly tested this argument by empirically comparing resource increases or survival among organizations that adopt a program versus those that do not. The idea of preserving a technical core or maintaining symbolic legitimacy is commonly interpreted negatively when environmental pressures are perceived as morally superior to routine technical operations. In one of the earliest studies to emphasize negative consequences of decoupling, Vaughan (1982) argued that fraudulent transactions are more likely when the activities of subunits are buffered from top management. Orton and Weick (1990, p. 211) described literature along these lines as using the voice of compensations, meaning that loose coupling is an unsatisfactory condition that should be reversed. They point to numerous examples in the eld of educational administration (Firestone, 1985; Murphy & Hallinger, 1984; Weick, 1982). Recent work analyzes this type of decoupling in the corporate setting. Lyon and Maxwell (2011) describe non-governmental organizations accusations that BP has attempted to divert attention away from its environmentally harmful petroleum exploration activities and toward its investments in renewable energy. BP is depicted as symbolically adopting policies that signal attention to environmental protection and environmental philanthropy, but for the purpose of directing attention away from its harmful actions. Also in the realm of environmental policies, Short and Toffel (2010) show that industrial facilities are more likely to comply with their symbolic commitments to selfregulation under the Clean Air Act standards if they are subject to heavy regulatory surveillance and do not have a history of poor compliance. A number of other studies have illustrated the ability of some organizations to resist institutional pressures by framing non-compliant practices in a way that avoids penalties (Deephouse & Suchman, 2008; Elsbach & Sutton, 1992; MacLean & Behnam 2010). In their review of organizational misconduct, for example, Greve, Palmer, and Pozner (2010, p. 77) describe the non-implementation of stock buyback plans described in Westphal and Zajacs (2001) study as unethical behavior that approached violation of SEC reporting regulations. In this view, morally devious and power-hungry managers consciously decide not to implement valuable programs in order to gain resources and avoid sanctions. Many of these accounts identify the actions and interests of individuals as a central determinant of decoupling between policy and procedure. These studies seek to integrate ideas of agency into neo-institutional research by

Downloaded by [Academy of Management ] at 19:27 21 June 2012

From Smoke and Mirrors to Walking the Talk

11

illustrating how external pressures are interpreted and carried out by individuals. For instance, in his study of recycling programs at colleges and universities, Lounsbury (2001) shows that social movement activity by student groups inuenced schools to adopt full-time recycling manager positions staffed by committed environmentalists, whereas schools without student activism created part-time positions staffed by less dedicated managers. Westphal and Zajacs studies show that decoupling policy from practice occurs more often when top executives have inuence over their boards, when a rm has network ties to other rms that engage in similar decoupling, and when a rm has prior experience with decoupling (Westphal & Zajac, 1994, 2001; Zajac & Westphal, 2004). In other words, policy practice decoupling is more common when it serves the interests of powerful leaders. These results suggest a circuit of decoupling, in which networks of top managers and organizations become skilled at symbolic manipulation. Similarly, Tilcsik (2010) describes how a policy that was purely symbolic can, over time, become fully implemented when internal lines of authority shift. In an earlier study, Covaleski, Dirsmith, and Michelman (1993) showed that these internal adoption processes feed back into broader notions of legitimacy, thereby playing a role in establishing and perpetuating power structures in organizational elds. In addition to adding to individual-level interest- and power-based explanations of decoupling, Tilcsiks study contributes to a growing body of work that draws attention to coupling processes. He found that a government agencys adoption of a new budgeting system to appease external stakeholders was intended to be purely symbolic. But over time, internal power dynamics changed and eventually led to its full implementation. Espeland (1998) was among the rst to examine recoupling processes through which formal policies and daily activities move from being disconnected to being closely linked. In her analysis of the Orme Dam in Arizona, she shows that Old Guard engineers became invested in the project for its own sake, although it was not the most optimal way to generate electricity and water. Eventually, a new group of engineers implemented a cost-benet decision-making process, and plans for building the dam were abandoned in favor of a more effective approach, thereby re-aligning practice with technical rationality. Similarly, renewed attention to schools has resulted in accounts of the recoupling of school practices to formal policies due to the heightened salience of external accountability measures (Spillane, Parise, & Sherer, 2011). In a rich ethnographic study, Hallett (2010) emphasized that the process of recoupling in schools is fraught with much individual-level turmoil and conict. Kelly and Dobbin (1998) describe a comparable phenomenon in the implementation of anti-discrimination laws over the period 1961 1996. As federal oversight increased, rms began implementing policies that were generally ceremonial but intended to ward off external sanctions. Nevertheless, in so doing, corporations created staff positions and units that took on a life of their own, and

Downloaded by [Academy of Management ] at 19:27 21 June 2012

12

The Academy of Management Annals

committed professionals created new policies inside their organizations. Such studies overcome a common criticism of decoupling research; namely that formal elements of organization adopted in response to institutional pressures are largely ceremonial, with little disruption of daily routines. Parallel to this organizational-level work, a host of cross-national research further supports the argument that, over time, ceremonial efforts often become binding, coupling external demands to policy commitments. Hafner-Burton and Tsutsui (2005, p. 1378) report that when nation-states make formal legal commitments to symbolize human rights compliance even while they are in violation, this process of empty institutional commitment to a weak regime paradoxically empowers non-state advocates with the tools to pressure governments toward compliance. Their analysis of this transformation is based on human rights treaty ratication and violations in 153 countries over the period 1976 1999. They build on a similar argument made in a qualitative study by Thomas (2001), which examined the effects of the Helsinki Final Act in 1975 on political developments in socialist states. Political scientists have posited a boomerang effect where the adoption of national human rights policies, which are not strongly supported domestically, nonetheless give local activists the ammunition to enlist the support of international non-governmental organizations in efforts at reform (Keck & Sikkink, 1998; Risse, Ropp, & Sikkink, 1999). In a study of environmental policies and outcomes, Schofer and Hironaka (2005) show that domestic environmental outcomes improve as support and mobilization for the environment becomes more institutionalized at the world level. In a related vein, Cole (2005) shows that countries may experience an apparent initial widening of the discrepancy between human rights practices and policies due to increased reporting and monitoring (rather than an actual increase in violations), but this is followed by gradual alignment. Further, he advocates making a distinction between decoupling due to a lack of capacity versus decoupling due to lack of will (Cole, 2012), as do Lim and Tsutsui (2012) in their global analysis of corporate social responsibility (CSR) adoption patterns. Similarly, in the realm of national science policy, Drori, Meyer, Ramirez, and Schofer (2003) nd that policies are less likely to be implemented in developing countries. Related to these ndings, John Meyer and colleagues (Meyer, Boli, Thomas, & Ramirez, 1997, p. 155) had warned earlier that it is a conceptual mistake to be too cynical about decoupling. Just as managers of organizations may vary in the resources available to them to implement policies, government authorities differ in their ability to exercise effective control within the borders of their own polity (Krasner, 1999, p. 4). Clearly, longitudinal studies of decoupling are invaluable for providing greater insight into when and why some policies might remain as window dressing for long periods of time, while others evolve into a

Downloaded by [Academy of Management ] at 19:27 21 June 2012

From Smoke and Mirrors to Walking the Talk

13

meaningful aspect of regular activities. The incorporation of a long-term perspective is an important step forward for studies of policy practice decoupling. But even as longitudinal research reveals that in some cases policies and practices can become more aligned over time, it is crucial to recognize that this phenomenon can be driven by diffuse institutional changes rather than top down implementation of policies. Meyer and Rowan (1978) showed that new educational ideas are taken up by advocates at all levels, not implemented solely through formal policies. As a result, new practices may sometimes precede the existence of a formal policy, and policies are debated in the absence of well-understood methods for achieving desired results (Kelly & Dobbin, 1998; Meyer & Rowan, 1978). March and Olsen (1976) also discussed the misalignment between intention (such as formal policy) and action (including daily practices), which suggests that external inuences act directly on each element. Going a step further, we contend that not only are an organizations policies (or strategies) and practices (or activities) each directly inuenced by environmental pressures, but so are organizational outcomes. For instance, Zajac and Westphal (2004, p. 433) posit that institutionalization processes might increase the market value of a policy as more rms adopt it, despite growing evidence of decoupling. In the eld of education, Soltow (1981) illustrates how a shifting cultural view of education led to both the expansion of enrollments and the diffusion of compulsory education laws (often subsequent to dramatic increases in enrollment) in the US. Snellman (2011) labels this reverse decoupling and illuminates it in a study of the embrace of shareholder value practices in Finnish corporate governance. Global nancial pressures played only a limited role in the adoption of Wall Street practices by Finnish corporations; internal organizational competition among various elite groups and domestic politics mattered more. Indeed, some Finnish companies adopted the walk of shareholder value without talking the talk in order to keep their embrace of the favored American practices out of the public eye. More broadly, we stress that all levels of an organization are subject to pressures from the environment; therefore, organizational and institutional changes are often uneven and irregular. This view contrasts with the idea that change proceeds from the top down, with policies rst policies created, then implemented, leading to the intended outcomes. What can we learn from this active strand of research on decoupling and the relationship between policy and practice? Taken together, this research presents a number of arguments regarding when policies may be decoupled from (or recoupled to) both formal and informal practices. Policy practice decoupling is more likely if:
.

Downloaded by [Academy of Management ] at 19:27 21 June 2012

adoption is motivated by legitimacy rather than technical demands;

14
. . .

The Academy of Management Annals

it is early in the adoption process; there is weak capacity to implement policies; internal constituents do not reinforce external institutional pressures. Weak internal reinforcement may result when adopted policies conict with or are not central to the interests or beliefs of internal parties; organizational leaders learn of non-compliance practices in other rms or have a history of non-compliance; there is a lack of external enforcement, especially legal; there is a poor t between imposed goals and existing identities and practices; or internal constituents have relatively more power than external constituents and can resist institutional pressures.

Downloaded by [Academy of Management ] at 19:27 21 June 2012

The majority of management research has focused on the gap between formal procedure and actual practice. But several authors have observed that this form of decoupling is empirically less common than the literature suggests (Bromley et al., 2009; Coburn, 2004). Others conclude that decoupling is only a temporary phenomenon (Boxenbaum & Jonsson, 2008; Haack, Schoeneborn, & Wickert, forthcoming). We believe that these interpretations emerge in large part because contemporary studies, including our own prior work, have often dened decoupling too narrowly, in particular neglecting a critical insight from earlier theoretical discussions: Decoupling can also occur at the level of the means ends relationship if policies or practices have an uncertain link to outcomes. A broad view of decoupling that includes internal differentiation highlights that policies and structures can serve as adaptive symbols rather than instances of ceremonial adoption, a point we now turn to. Means Ends Decoupling Decoupling between means and ends occurs in settings where formal structures have real organizational consequences, work activities are altered, and policies are implemented and evaluated, but where scant evidence exists to show that these activities are linked to organizational effectiveness or outcomes. Often, extensive rhetoric or cultural beliefs posit a causal link between activities and outcomes, but organizational participants may also consciously recognize that the actions are of limited utility but pursue these practices because they perceive larger structural pressures. In other words, organizational activities are integrated not through a direct connection to production, but rather through operational systems such as accounting, personnel management, evaluation, or monitoring. Operating systems and techniques for evaluation and monitoring may themselves have an uncertain relationship with intended outcomes. Means ends decoupling, illustrated in Figure 2, examines why organizations direct substantial resources toward implementing policies and evaluating practices that have a tenuous link to core goals. In contrast, policy

From Smoke and Mirrors to Walking the Talk

15

Downloaded by [Academy of Management ] at 19:27 21 June 2012

Figure 2

Means Ends Decoupling.

practice decoupling focuses on why formal policies are not implemented. Whereas policy practice decoupling can be thought of as symbolic adoption, means ends decoupling is better characterized as symbolic implementation. Means ends decoupling helps to explain why organizations implement a range of practices (and associated ways of evaluating practices) that have an opaque relationship to outcomes. Many organizational activities reect larger social and cultural meta-narratives rather than technical requirements (Dobbin, 1994). The external environment generates means ends decoupling by creating the myths of causality, control, and coherence that organizations are structured around. These rationalizing accounts (rather than direct observation or instrumentality) provide organizations with appropriate means and ends, particularly in highly rationalized and fragmented environments. Thus, means ends decoupling includes cases where formal organization and evaluation techniques are applied to obscure goals (as in a minister trying to measure the effectiveness of a sermon). It also includes instances where organizations adopt new ends that are not directly related to core goals (as in CSR practices). Conceptually, the connection between these practices is that the link between their enactment and an organizations core goal is unclear. In this form of decoupling, the relationship between means and ends stem from institutionalized conceptions of the appropriate way to achieve goals that emerge in the environment.

16

The Academy of Management Annals

Practices that we now identify as instances of means ends decoupling did not exist in an earlier era because specic means and ends were not seen as coupled in broader cultural-cognitive models. In particular, in our contemporary, post-industrial society, the evaluation and measurement of effectiveness and the efcacy of work activities take on a seemingly moral imperative. A great deal of effort goes into developing, implementing, monitoring, and evaluating policies, even though their ultimate effects are unclear. Further, as rationalization proceeds, more domains become subject to formal organizing, creating a highly fragmented environment. When the external environment is fragmented into multiple rationalized domains, organizations are shaped by the exogenously constructed rationalities of a variety of audiences or stakeholders. Means ends decoupling can, in some ways, be conceived of as the simultaneous appearance of conicting forms of rationality within an organization. Weber (1921[1968], 1930[1992]) identied the tensions between formal and substantive rationalities. Formal rationality involves means ends calculation based on the widespread application of rules and regulations that reect inherent self-interest, as in the classic bureaucracy. The goal is to calculate the most precise and efcient means for the resolution of problems by ordering them under universal and abstract regulations (Kalberg, 1980, p. 1158). In contrast, actions driven by substantive rationality reect deeply held values such as love or friendship. In cases where the trappings of formal rationality take on substantive value over and above their known direct contribution to achieving desired ends (e.g. in education testing), we observe means ends decoupling. Similarly, when substantive values in the environment (e.g. protecting the natural world or human rights) emerge in formally rational organizations, the clear link between means and ends becomes blurred. Below, we esh out our conception of means ends decoupling and then discuss three consequences for organizationsgreater internal complexity, endemic reform, and the diversion of resources. We believe that instances of means ends decoupling are on the rise in the contemporary world. Conversely, instances of policy practice decoupling may become less common. In a rationalizing world, with heightened emphasis on transparency and accountability, policy practice decoupling is increasingly likely to be seen as a moral and operational failure, in contrast to early conceptual depictions that emphasized the legitimacy benets of decoupling. Although it is largely overlooked, Meyer and Rowans (1977) classic theorization was not limited just to policy practice decoupling. It also described decoupling as an internal buffering of organizational units from one another to protect an organizations technical core from practices that are imposed by wider institutional demands.8 Hence, the buffering of internal practices is similar to policy practice decoupling in that it protects core activities, but it

Downloaded by [Academy of Management ] at 19:27 21 June 2012

From Smoke and Mirrors to Walking the Talk

17

differs in that within each subunit policies and practices may be largely implemented. This conception of decoupling, reecting subunits or practices that are largely separate silos within organizations, has been less explored.9 There are just a handful of exceptions to the general dearth of research on buffering between internal practices. Early on, Boynton and Zmud (1987) showed that information and technology departments are commonly buffered from the rest of an organization. Within hospitals, Leatt and Schneck (1984) observed that subunits containing nurses, physicians, or paramedical staff are often distinct entities. In the strategy domain, scholars have long recognized the value of having business units operate relatively independently (Gupta, 1984; Horwitch & Thietart, 1987). Recently, however, a surge in attention to organizational complexity and institutional pluralism (Greenwood, Raynard, Kodeih, Micelotta, & Lounsbury, 2011) and multiple institutional logics (Lounsbury, 2007; Thorton & Ocascio, 1999) indicates that scholars increasingly observe incongruent and buffered elements within organizations, many instances of which would t our characterization of means ends decoupling. In 2006, nearly 30 years after his seminal article with Meyer, Rowan revisited the issue of decoupling in the American school system. He shows that education in the U.S. has changed dramatically in the decades since institutional theory was initially developed. A relatively technical theory of schooling has emerged in the form of large testing agencies, increasing credentialing for education professionals, and stronger policy instruments designed to support particular curricular reforms and hold teachers and schools accountable for student performance. Rowan (2006) argues that a new institutional analysis of schooling would provide more insight into contemporary education if it focused on the social construction of these technical theories of schooling and on the role of institutions in promoting conformity to new educational models. He says, institutional governance. . . can, in fact, result in more than ceremonial conformity and loose coupling, even in regulated organizations such as schools (Rowan, 2006, p. 24). He goes on to describe how tight conformity to governance of education can occur, but the controlling inuence of disciplinary cultures seems to result from taken-for-granted assumptions (p. 25). In other words, although government regulation of education can have real consequences, the regulations themselves are based on a set of institutional beliefs located in society at large, rather than in well-vetted measures of effective schooling. Put differently, new myths have emerged that celebrate testing and measurement, even though what is assessed may have little relationship to learning. In settings like education, where the means of achieving goals are often derived from institutionalized beliefs rather than observable effects, there can be a persistent gap between means and ends.10

Downloaded by [Academy of Management ] at 19:27 21 June 2012

18

The Academy of Management Annals

In their review paper, Orton and Weick (1990) described the circumstance of an unclear relationship between means and ends as causal indeterminacy. They argue that decoupling due to an uncertain means ends relationship may be particularly prevalent in the production of complex social or public goods. Organizations whose ultimate goal is to produce a complex good face the challenge of developing meaningful metrics of accomplishment and clearly establishing the causal efcacy of particular programs or activities. Such causal indeterminacy characterizes decoupling at the level of means ends, rather than policy practice. Not surprisingly, these challenges loom large in the social sector. Clark (1983), for example, describes universities as organizations fated to be loosely coupled systems because their primary product is knowledge. Universities generate rich theories of knowledge production and sometimes measure the processes they implement, but the true impact of the knowledge they have produced is never known, particularly in the short run. The means ends gap is perhaps even more profound when we consider religious organizations. We can never know how many souls are saved, but churches increasingly try to improve their management and outcomes (Monahan, 1999). Putnam and Campbell (2010, p. 45) describe the tension one priest feels between his calling and the daily duties of running a corporate parish with an increasing emphasis on planning non-worship activities for the community using a large professional staff: The priest says, I didnt get into this to become an events coordinator. In another striking example, the non-prot Room to Read, founded by former Microsoft executive John Wood, is widely lauded for its results-oriented approach to improving education in developing countries. Its mission is to transform the lives of millions of children in developing countries by focusing on literacy and gender equality in education (Room to Read website 2011: [Link]). It does this by providing books and libraries to children in developing countries. Room to Read regularly reports on its impact by citing the number of books it has distributed, libraries built, and participants in its programs. But the organization eventually realized that many recipients lacked the minimum language skills required to use the books. Sometimes, the books were considered too precious to risk being lost or damaged by students, so they were locked up in libraries away from classrooms and students. Consequently, the organization came to recognize that counting numbers of participants or books distributeda common strategy to measure outcomes for non-prot organizationssays little about whether lives are transformed or students become more literate. This widespread challenge is faced by a great many organizations. In the current climate, most are working feverishly to improve measures of outcomes. Room to Read itself has launched a new pilot program to try to improve its measures. Such concerns are reected in its recent strategic plan:

Downloaded by [Academy of Management ] at 19:27 21 June 2012

From Smoke and Mirrors to Walking the Talk

19

Over the past several years, we have monitored our efforts closely and uncovered concerns about whether or not we have maximized the utilization of the programs. For example, although we know that in 2008 a total of 2,341,941 local language and/or English-language books were provided to the 4,478 libraries established between 2006 and 2008, and that on average 88% of our libraries have check-out systems in place, we still had concerns about just how many books were being checked out and read by children. (Room to Read website 2011, p. 4) Also in the charitable sector, in a case study of the Edna McConnell Clark Foundation (EMCF), the evaluation director comments, One of the key qualities of effective organizations is that they are systematic in gathering, analyzing, and using informationin other words, they have the capacity to measure whatever must be tracked in order to keep them on course for maintaining or even improving their effectiveness (Grossman & Curran, 2002, p. 14). EMCF is notable among US foundations in being especially focused on whether its grantees are making genuine progress toward their stated goals. It is also unusually generous in providing infrastructure support toward realizing the goals. But seen from a different angle, the foundation infers whether an organization is effective by seeing whether it gathers, analyzes, and collects information. A risk is that the focus can be more on the process rather than the intended social outputs a potential grantee is thought to be producing. It is easy to imagine that an organization could collect information and still remain ineffective, or that extensive efforts to gather and analyze information could actually detract from program activities. Feldman and March (1982) elaborate on the many ways in which information gathering can be a form of signaling and symbolism in organizations; they posit that most of the information gathered remains unused. In our rationalizing world, internal and external stakeholders value good faith efforts to monitor and evaluate, even in the absence of a clear link between policies and outcomes. Such activities become, to use Selznicks (1957, p. 17) oftquoted phrase, infused with value beyond the technical requirements of the task at hand. Similarly, Pache and Santos (2010, p. 460) describe a common conict between organizational goals and means for achieving those goals, when technical prescriptions are so institutionalized that they become ends in themselves. Information collection can, in some cases, be a rst step to uncovering practices that best achieve outcomes; the question is whether data collection becomes an end in itself or is used as a diagnostic part of a forward-looking process. Although means ends decoupling is common in organizations with a social goal, it is pervasive far beyond the non-prot sector. Thomas (1983) describes the emergence of loose coupling between the ideological basis of due process and judicial practice, which he attributes to competition for

Downloaded by [Academy of Management ] at 19:27 21 June 2012

20

The Academy of Management Annals

Downloaded by [Academy of Management ] at 19:27 21 June 2012

control over trial proceedings. In work on nancial auditing, Power (1997, p. 9) develops the concept of an expectations gap, where there is loose coupling in the auditing eld between accounts of potential and operational capability. The meta-narrative surrounding auditing is that its ultimate goal is to prevent fraud. Its technical aspects, however, necessarily rely on a small sample of transactions (so errors are easily missed), the opinions of experts (so decisions are not entirely transparent), and the cooperation of people inside the target organization to assess their own control systems (so the ability to have a purely external audit is limited). The gap between the belief or rhetoric of what audits should accomplish and what they actually measure generally goes unchallenged, and the practice is governed instead by a logic of condence in accounting as a profession. Periodically, however, malfeasance or auditing failures are revealed, generating waves of reform in auditing practices. The potential for reform and change is just one possible consequence of means ends decoupling. In what follows, we discuss these outcomes in more detail.

Internal complexity. Powers work on auditing aptly demonstrates that means ends decoupling is often linked to greater internal complexity: one prima facie sign of decoupling is the creation or enhancement of organizational sub-units explicitly to manage the external audit process (audit committees, internal auditors, audit ofcers, etc.) (Power, 1997, p. 96). Unlike policy practice decoupling, which may be largely ceremonial, means ends decoupling has extensive consequences, albeit often unintended ones. In fragmented and rationalized environments, organizations seek to respond to a greater number of external stakeholders; adapting to these external pressures can generate both internal complexity and inconsistency. To give a more detailed example in a setting familiar to many readers, most universities are made up of numerous departments and units (academic and non-academic) that operate relatively independently. Subunits may have their own budgets and sources of funding, mission statements, and annual reports. The website of one unit at Stanford University, the Land Use and Environmental Planning (LUEP) Ofce, states: LUEP is made up of regional, campus and resource planners and analysts with backgrounds and training in the design, planning and science professions. [Our ofce] advises the senior administration and Board of Trustees about long-term land use and the protection of land-based resources (LUEP Website, 2011). Very indirectly, the ofce is linked to teaching and learning because academic activities take place on university land, but the department is clearly populated with its own professionals and well-articulated goals. Further, it is easy to envision situations where the LUEP Ofces goals conict with academic goals. Students may be best served through the construction of new classroom buildings, but such construction could be detrimental to either protecting land resources or

From Smoke and Mirrors to Walking the Talk

21

maintaining the balance of political and environmental regulations with Santa Clara County, which prefers to see more on-campus housing and less commuting. Perhaps energy conservation would be best achieved by devoting extensive resources to environmental upgrades to the university, such as the installation of solar panels and energy-efcient lighting; although students might benet more if resources were devoted to hiring faculty and limiting class size. Although LUEP works closely with the Ofce of Development, Finance, and the top administrative ofcers of the university, the decisions that have a vital and lasting effect on the universitys size, shape, and future appearance are only weakly linked to faculty and students immediate concerns. The state of environmental science may be such that the LUEP ofce is able to measure the effect of its policies on land use, so environmental standards are rigorously pursued. Indeed, the ofce has pressed to have all recently constructed buildings obtain environmental certication, a sign of the rationalities of energy awareness. The most energy inefcient buildings, however, may well be those that are most actively used by researchers. But such energy-saving efforts are, at best, only moderately tied to academic departments goals of teaching and scholarship. Buffered activities and relationships like these are common in all sorts of organizations. The peculiar observation that organizations rigorously pursue many activities that have unknown effects and relationships to core goals is not new, although it has seldom been studied systematically. Only a handful of scholars have documented instances where extensive organizational work is directed toward activities with uncertain outcomes. Edelman (1992) says that Equal Employment Opportunity/Afrmative Action (EEO/AA) legislation did lead rms to make changes as they attempted to comply with the law, but the legal mandate emphasized broad procedural adaptations rather than specic outcomes. Consequently, scant effort was made to ascertain whether these changes were causally linked to improvements in the status of minorities in the workplace. Edelman (1992, p. 1534) states that studies of work-force demographics show improvements in the economic and occupational status of minorities and women, albeit not necessarily in response to the law. Edelman (1992, p. 1539) remarks that decoupling implementation from outcomes is not unique to employment law; a similar process occurs in the realm of anti-pollution law. A parallel case could be made with human rights treaty ratication and human rights practices; the causal mechanism linking the means of treaties with the ends of human rights practices is unclear. Edelman (1992, pp. 1543 1544) goes on to describe how EEO/AA ofces may be given autonomy and authority so that they may monitor organizational practices in other departments, but if that monitoring can affect other departments, the ofce should be considered coupled. In this case, more visible and public organizations (measured by size and proximity to the public sphere) were more accountable to outside stakeholders and thus more likely to

Downloaded by [Academy of Management ] at 19:27 21 June 2012

22

The Academy of Management Annals

develop elaborate internal structures. Procedurally, EEO/AA ofces may be tightly coupled, with their monitoring and implementation affecting other units. But when considering means and ends, the implementation of the law (means) is decoupled from the intended outcome (ends), and diversity activities (means) are but weakly tied to the core goals of most corporations (ends). This study and related works are among the rst to document the pervasiveness of internal differentiation as a feature of modern organizations (Edelman, 1990, 1992; Edelman, Uggen, & Erlanger, 1999). Similarly, Dobbin and colleagues (Dobbin, 2009; Dobbin & Kelly, 2007; Kalev, Dobbin, & Kelly, 2006; Kelly & Dobbin, 1998) show that diversity programs have assumed prominence inside large corporations. These studies describe a remarkable transformation inside contemporary rms. Large-scale efforts to create equal opportunity and justice are ubiquitous, but the quest for equality has been transformed into the pursuit of diversity as a business goal, championed by professional human resource managers. Kalev et al. (2006) note that there is little systematic evidence of the effectiveness of many diversity programs despite their costs and prevalence. They nd clear, but highly contingent patterns. Some programs are effective (notably, establishing responsibility, especially through mentoring and company-wide task forces), whereas others are less so (networking), and some not at all (mandated training and evaluations). Some underrepresented groups benet modestly (white women, followed by black women), whereas others do not (black men and Hispanics). The authors speculate that rms implement these programs either to protect themselves from litigation or to boost morale. Their extensive research indicates that, at the level of the program itself, means ends decoupling can be mitigated when the organization creates structural responsibility for ensuring outcomes, such as with a full-time diversity management position or a diversity taskforce. Still, these types of policies are less common than those carried out under the direct purview of human resource departments. Moreover, companies that need diversity programs (i.e. those with low diversity) are the least likely to do so (Dobbin, Kim, & Kalev, 2011). And in a broad sense, when the programs are examined in light of a corporations nancial goals, it is hard to see why companies dedicate substantial resources to creating and implementing programs that go beyond the scope of the law and are difcult to justify in terms of effectiveness in creating a diverse workforce or contribution to productivity. Here, again we see the importance of a wider environment in which the values of opportunity, fairness, and public procedure and accountability have become widely accepted and imported into formal organizational structures. One response to multiple, conicting institutional logics is compartmentalization (Greenwood et al., 2011; Pratt & Foreman, 2000). One can read many existing studies of the adoption of practices as instances where specic activities are isolated from the rest of the organization. Studies of corporate

Downloaded by [Academy of Management ] at 19:27 21 June 2012

From Smoke and Mirrors to Walking the Talk

23

philanthropy and environmentalism (Lyon & Maxwell, 2011) and CSR (Basu & Palazzo, 2007) have analyzed these practices as symbolic actions tied to managing external pressures, largely disconnected from an organizations core work. Similarly, Binder (2007) shows extensive buffering between departments of a housing organization and attributes the gap to a tension between competing logics of passion and nance, which are linked to different external funding sources. Taken together, these studies suggest that greater complexity in the external environment promotes internal buffering of subunits, one form of decoupling, as a mechanism for incorporating inconsistent or contending inuences into organizational practices. Environmental complexity, however, takes a particular form. General uncoordinated chaos is unlikely to contain the types of legitimated and orderly stakeholders that inuence contemporary organizations. Pache and Santos (2010) argue that conicting institutional demands are likely to be felt most strongly by organizations in fragmented elds that are moderately centralized, as opposed to highly decentralized, because in such a domain no logics are sufciently dominant to exert potent inuence. Moreover, conicting institutional expectations, and accompanying internal differentiation, are more likely to arise in fragmented organizational elds (Aurini, 2006; Binder, 2007; George, Chattopadhyay, Sitkin, & Barden, 2006; Lok, 2010). The existence of multiple logics in the environment may also be a source of micro-level strategic action, providing wiggle room to carve out distinctive responses (Clemens & Cook, 1999; Friedland & Alford, 1991; Powell & Colyvas, 2008; Seo & Creed, 2002). Most relevant to our purposes here, micro-level processes within organizations can inuence the extent and nature of decoupling. Greenwood and Hinings (1996) say that if internal groups can supply a cogent alternative, organizations are more likely to resist institutional pressures for change. Internally, professionals often act through informal channels to promote the goals of their respective subunits. Such efforts are most common when professionals are brought inside the organization because they carry expert knowledge of rules and regulations from different elds (DiMaggio & Powell, 1983). In Binders (2007) account, some of the distance between organizational subunits was due to the fact that staff were steeped in outside professional logics and not as cognizant of existing internal practices: The Housing director had a degree in nance and none of the staff had a background in housing, unlike the Discovery Center and Family Support departments. Similarly, Edelman (1992) nds that the structural elaboration of organizations to include EEO/AA ofces is more likely if human resources personnel are already present, and Lounsbury (2001) nds that recycling programs are more extensive when people with relevant professional backgrounds are hired. These professionals act as internal champions to promote units with their own logics.

Downloaded by [Academy of Management ] at 19:27 21 June 2012

24

The Academy of Management Annals

Downloaded by [Academy of Management ] at 19:27 21 June 2012

Most recently, Greenwood et al. (2011) suggest that more visible and consumer-conscious organizations may be particularly targeted by certain stakeholder groups. Large corporations like Walmart and Starbucks create huge public displays of their purported benets to the community, their environmental protection efforts, and their CSR activities. And, again, such efforts can, in turn, transform internal practices: Walmart has become a leader in environmentally sustainable practices and dedicates extensive resources to support its corporate philanthropy programs. Note that we are not assuming there are positive social outcomes of such programs; instead we emphasize that organizations pursue many activities despite the lack of a clear connection to outcomes (for the community and the company) and despite a clear integration with core goals (in the case of Walmart, the emphasis on prots through retail).

Endemic reform. As should be clear, organizations of many types routinely pursue activities that may not result in the achievement of goals. By denition, rationalization includes increasing efforts to clarify these means ends relationships. Thus, in a rationalizing environment, we see greater attempts to measure and evaluate activities, particularly in realms where the link between activities and outcomes is not directly observable. Especially in opaque settings, an emphasis on measurement can become a valued activity in itself, a taken-for-granted means of achieving ends.11 Monitoring efforts, however, can also expose inconsistencies, triggering waves of reform in policies, implementation, and evaluation inside organizations or even at the eld level. In some instances, reforms may bring us closer to achieving some intended outcomes, but in many others the ultimate effects will never be known (e.g. the extent to which religious organizations save souls, or museums shape cultural experience and awareness, or an environmental organization slows global warming). In general, our tools are likely best at capturing immediate, direct, and observable consequences. They are less able to adequately measure externalities, spillovers, and unobservable, long-term, or second-order effects of various policies or practices. The existing literature suggests two main forms of perpetual reform. In the rst, periods of calm are interrupted by revelations of misconduct or failure, followed by waves of reform or transformation. Power (1997) characterized eld-level auditing reforms as driven by dramatic and recurrent revelations of failure, and this pattern is likely common to many systems of monitoring and evaluating. Alternatively, ongoing debates and reforms in education are rarely accompanied by massive, system-wide changes. Instead the history of education reform is characterized by incremental, gradual changes, a process Tyack and Cuban (1995) described as tinkering towards utopia. Irrespective of whether changes are wholesale or piecemeal, when the environment is rationalized and means and ends are decoupled, organizations are unlikely to be stable over long periods of time.

From Smoke and Mirrors to Walking the Talk

25

Diverting resources. A nal consequence of meansends decoupling can be a diversion of resources and attention away from core activities. Universities provide an apt example of how the internalization of various environmental pressures can also involve extensive reshaping of core activities. Focusing on law schools, Sauder and Espeland (2009, p. 68) observe that rankings have prompted broad changes in legal education, affecting how resources are distributed, decisions are made, and status is dened. US News and World Report rankings inuence a very large number of constituents, including current and prospective students and their parents, alumni, employers, trustees, and the faculty. Consequently, university administrators direct very signicant resources toward strategies for managing the rankings and, in some cases, twist traditional rules and priorities in order to do better. Clearly, in an implementation sense, university actions and rankings are tightly coupled. At the same time, administrators view the law school rankings with disdain and deride them as too stupid to be taken seriously (p. 68). Sauder and Espeland (2009, p. 68) recount the dean at Yales description of the rankings as an idiot poll, and Harvards dean used the phrases Mickey Mouse, just plain wacky, and totally bonkers. The rankings encourage behaviors that many consider counterproductive, and create distinctions among schools where no meaningful contrasts exist. Thus, in a more fundamental sense, the rankings are highly disconnected from a law schools actual merits, a situation we place rmly in the category of decoupled means and ends. To be sure, university and college rankings serve useful purposes for some audiences. Moreover, they have rapidly come to have signicant organizational effects in higher education. But the methodology of rankings remains highly disputed. The component parts, including measures of reputation, selectivity, post-graduation placement, research resources for faculty, faculty education, faculty publication records, and class size, are only modestly correlated. Reputation, for example, is assessed with a bevy of weak proxies. In some instances, if the rankings produce unexpected results that do not t common intuition, the criteria are readjusted to t existing perceptions. And at the general level of university rankings, there are scores of competing evaluations, measuring things in different ways or using different categories. The rankings game has become an expansive (and expensive) business. But rankings are only loosely connected to the quality of academic work at a university, despite their extensive impact. Efforts aimed at scoring higher on rankings are not necessarily linked to efforts to improve teaching and learning. And, as we have seen in recent history, university staff can become so obsessed with rankings that they engage in deception and malfeasance, eroding trust in all universities and colleges. This diversion of attention away from core goals, a feature of means ends decoupling, is pervasive in the modern world. In many ways, traditional universities around the world have become highly managed organizations, and core activities are not always insulated from

Downloaded by [Academy of Management ] at 19:27 21 June 2012

26

The Academy of Management Annals

other goals stemming from various pressures of hard and soft law, accounting, and evaluation (Kruecken & Maier, 2006). A similar phenomenon has occurred in hospitals and medical practice (Scott et al., 2000). Sauder and Espeland (2009) argue that participation in law school rankings, despite common criticisms and concerns about their value, becomes widespread through both external enforcement and internalization of the pressures.12 Activities become deeply shaped by external criteria that may have little to do with an organizations core goals. The pervasiveness of rankings is a relatively new phenomenon, emerging in the late 1980s. Educational rankings have burgeoned in tandem with a wide array of evaluations and standards. For example, Rao (1998) analyzes the development of consumer product research, Chatterji and Toffel (2010) describe the increasing array of CSR ratings, and Bartley (2007) assesses the effects of the creation of voluntary standards in labor and forestry certication on businesses. This wider movement of standards and certication reects the growing rationalization of the environments in which most organizations operate. Increasingly, all organizations, regardless of their core goals, must also conform at some level to the law and normative pressures. In some cases, meeting external demands requires devoting large amounts of resources to develop extensive internal practices that comply, in some fashion, with external criteria. Organizations are expected to be particularly responsive to aspects of the environment when they are institutionalized in various forms of hard and soft law, and when claims are more rationalized. Accordingly, environmental and human rights regimes operate through a combination of formal legal channels, social movement advocacy, and the creation of professional champions for these causes. Further, in the highly litigious and decentralized U.S., the ensuing complex elaboration of organizational practices may be greater than in countries where legal intervention is either less common or more directly controlled by the state, rather than through organizational interpretations of ambiguous law and public involvement in legal and regulatory affairs (Dobbin, 2009). The central questions associated with meansends decoupling are Why and when do organizations pursue activities that are weakly linked to their intended goals? and How and to what extent do these new activities reshape existing policies and practices? We do not have denitive answers, but our conception of meansends decoupling draws attention to the widespread existence of organizational practices that are aligned with formal policies but have unproven utility. A focus on information and procedure rather than directly on ultimate goals is at the core of meansends decoupling. On the basis of these studies and observations, we suggest that meansends decoupling will be more prevalent:
. .

Downloaded by [Academy of Management ] at 19:27 21 June 2012

in contexts where the effects of actions are difcult to measure; over time, as rationalizing pressures expand into arenas where outcomes are hard to measure;

From Smoke and Mirrors to Walking the Talk

.

27

. .

when conicting pressures are more institutionalized in hard or soft law, especially when law emphasizes procedure over outcomes; when internal constituents champion an external cause; and when the rationalized environment is fragmented.

Environmental fragmentation may be due to:

.

Downloaded by [Academy of Management ] at 19:27 21 June 2012

direct accountability to a greater number of stakeholders through, for example, diverse funding streams; greater societal pressure due to visibility because of size, status, or the perception of a public interest. Fragmentation may also be higher in contexts where traditional forms of authority are weak (such as the radically decentralized U.S. versus more centralized countries like Germany, or in heavily regulated industries versus those with less government oversight, or in organizational elds and time periods where older forms of religious or charismatic authority such as the legal or medical professions are fragile); and may increase over time with the worldwide adoption of New Public Management and neo-liberal ideologies.

Table 2 provides an overview of selected empirical studies relevant to our arguments. Alternatives to Institutional Decoupling We have made predictions throughout the paper regarding when each type of decoupling is more likely to appear. There are, of course, instances when decoupling will not be the response to institutional pressures. Oliver (1991) describes a spectrum of responses: acquiescence, compromise, avoidance, deance, and manipulation. Of these, avoidance is most closely linked to policypractice decoupling because it involves symbolic conformity, including concealing non-compliance and buffering technical structures. Acquiescence and compromise may incorporate a great deal of meansends decoupling.13 Deance and manipulation, however, should not be viewed as institutional decoupling. Deance involves rejecting institutional demands, and manipulation entails attempting to control the source and content of the institutional pressure.

Technical versus Institutional Decoupling: Normal Accidents and Deviance An underemphasized distinction in research on decoupling is that these causal processes outlined in early theory are relevant to institutionalized organizations. To some degree, all organizations must be understood as institutionalized, because many elements, such as legal incorporation, are dictated by law. Other arenas, such as pressures to conform to labor and environmental standards, have also become highly institutionalized, legitimated by courts and in

28

Table 2 Summary of Empirical Studies

Downloaded by [Academy of Management ] at 19:27 21 June 2012

Author(s), publication year Binder (2007)

The Academy of Management Annals

Empirical setting Case study primarily based on 89 interviews from a non-prot transitional housing organization in a large western city of the U.S., 2003 2005 Ethnography of reading instruction practices primarily based on three teachers in two schools including 35 interviews of administrative personnel and 14 additional teacher observations in California, 19831999 Examines human rights practices and ratication of International Human Rights Covenants for more than 130 countries, 1966 1999 Cross-national, time-series analysis examining human rights practices and treaty ratication, 1981 2007 Use of case-mix accounting systems based on a diagnosticrelated group framework in hospitals

Relevant contribution(s) Shows internal buffering between subunits, driven by links to external nancial ties and divergent professional logics Complete decoupling of policy from practice is rarely the response; responses mediated by messages about instruction in environment, teachers preexisting beliefs and practices, and nature of message itself Gap between policy and practice narrows over time; decoupling is due to both lack of capacity and lack of will

Coburn (1994)

Cole (2005)

Cole (2012) Covaleski et al. (1993) Dobbin (2009)

Edelman (1990)

Gap between policy and practice narrows over time; decoupling is due to both lack of capacity and lack of will Adoption plays a role in establishing and perpetuating the legitimacy of a policy or practice, thus adoption is related to power structures Examines creation of regulations and practices related to equal Widespread internal structuration in rms in response to opportunity in U.S. rms ambiguous laws; inuence of professional personnel managers in perpetuating and implementing diversity and equal opportunity practices Adoption more common in larger organizations closer to Emergence of formal grievance procedures in 52 U.S. government and with internal personnel professionals organizations around the San Francisco Bay area including even-history analysis, 18501983

Edelman (1992)

Elsbach and Sutton (1992) Espeland (1998)

Fiss and Zajac (2004) Fiss and Zajac (2006)

Haack et al. (forthcoming)

Hafner-Burton and Tsutsui (2005) Hallett (2010)

National survey of 346 U.S. organizations conducted in 1989 Legitimacy gained through visible compliance of creating internal structures, but internal elaboration is not necessarily and event-history analysis regarding compliance with EEO/ linked to changed outcomes AA practices, 1964 1989 Legitimacy gained through symbolic conformity to norms; Examines illegitimate actions by members of two social decoupled formal structures from illegitimate actions when movement organizations primarily based on meetings, brought to light events, and interviews in 1990 Examines contention surrounding government proposal to Recoupling of policies to decision-making processes; build the Orme Dam in southwestern U.S. contestation over denition of rationality between multiple stakeholders Adoption of shareholder value orientation with a sample of Presence of powerful or committed key actors reduces 112 German companies operating in 1990, 1990 2000 decoupling Framing of strategic change toward shareholder value and real Decoupling is not binary, can involve multiple ways of presenting and justifying organizational choices; forces returns to shareholders using a sample of 112 German toward aligning policy and practice may come from within companies operating in 1990, 1990 2000 rm to correct cognitive inconsistencies of rm identity Decoupling between signing the principles and enacting their Examines adoption and implementation processes of the Equator Principles for environmentally responsible investing contents decreases over time using 42 interviews and approximately 750 documents over period 2009 2011 Policies and practices aligned over time; changed human Cross-national, time-series study of human rights treaty rights practices attributed to direct effects of global human ratication and government repression in 153 countries, rights regime, not necessarily to policies themselves 19761999 Two-year ethnographic study in urban elementary school in Cultural pressures toward accountability create recoupling the U.S., 1999 2001 of policies and practices; policies and practices may be closely aligned despite loose relationship to and disruption of core goals

Downloaded by [Academy of Management ] at 19:27 21 June 2012

From Smoke and Mirrors to Walking the Talk

29

30

Table 2 Summary of Empirical Studies (Continued)

Downloaded by [Academy of Management ] at 19:27 21 June 2012

Author(s), publication year Kalev et al. (2006)

The Academy of Management Annals

Empirical setting

Relevant contribution(s)

Kelly and Dobbin (1998)

Lim and Tsutsui (2012) Lounsbury (2001)

Lounsbury (2002) Lounsbury (2007)

Lyon and Maxwell (2011)

Workforce data for 708 U.S. companies combined with survey Firms dedicate extensive resources to practices that have little data on employment practices, 1971 2002 known effectiveness; programs that establish managerial responsibility have strongest outcomes, rms subject to government regulation also have better outcomes Historical analysis of EEO/AA in the U.S., including trends in Firm EEO/AA measures have not straightforwardly followed enforcement efforts; hiring of internal specialists and organizational practices, related laws, and professional professionals meant internalization of diversity values in discourse, 1961 1996 rms regardless of regulation Decoupling is due to both lack of capacity and lack of will; Cross-national, time-series analysis examining patterns of civil society linkages encourage CSR adoption, but adoption of the Global Compact and Global Reporting developed countries resist substantive adoption Initiative through analysis of 99 countries, 2000 2007 Archival data analysis and survey data gathered in 1996 for Civil society pressures shape more extensive adoption of higher education recycling programs in the Great Lakes area recycling program through creation of new, full-time professional role Professionalization of nance occupations for U.S. nance Changing logics in the environment shape ideas and practices associations, 1945 1993 of professionalization Competing logics in environment leads to practice variation Mutual fund practices in establishing contracts with money managers specically observing Boston and New York and time-series analysis for all mutual funds, 1944 1985 Examines positive and negative disclosures of environmental Activist pressure deters greenwashing (i.e. disclosing positive performance in rms using formal economic modeling environmental performance but not disclosing negative performance), but also decreases overall disclosure

Downloaded by [Academy of Management ] at 19:27 21 June 2012

MacLean and Behnam (2010) Power (1997)

Case study of deceptive sales practices at a large nancial services rm using various forms of archival data Auditing practices in the U.K.

Sauder and Espeland (2009)

Qualitative case study of U.S. News rankings effects on law schools primarily using data collected through interviews and analysis of secondary sources Cross-national and world-level longitudinal analyzes of links between CO2 and CFC emissions and CO2 and deforestation, 19801996 Sample of 7274 U.S. industrial facilities subject to the Clean Air Act, 1993 2003 Case study of four schools administrative responses to regulative emphases on instruction, 1999 2003 case study of how a post-Communist government agency allocated public funds to subsidiaries using interviews from 20042008 and archival data from 19922008 Cross-national, time-series study of creation of national stock exchanges in developing countries, 19802005 Adoption of LTIPs for CEO compensation in 570 large U.S. rms, 1972 1990

Schofer and Hironaka (2005) Short and Toffel (2010) Spillane et al. (2011) Tilcsik (2010)

Decoupling of compliance from core activities contributes to appearance of legitimacy but allows for misconduct Practice of auditing decoupled from rhetoric that a core goal is fraud prevention; rise of auditing practices accompanied by internal structuration in rms Internalization of external pressures reduces decoupling; policies and practices may be closely aligned despite loose relationship to and disruption of organizations core goals; over-auditing takes on pathological character Decoupling is less common if institutions are highly structured, span multiple levels, and persist over time Compliance with self-regulation measures more likely in presence of heavy legal regulation, in absence of direct threat of sanctions; less likely among prior offenders Recoupling of regulation with teaching reinforced by leaders creating new organization routines and formal structure ideology of powerful organization leaders inuences whether decoupling is and whether it persists; can have alignment over time if leadership changes Decoupling prevalent if adoption was coerced; implementation associated with mimesis of peer groups and high status countries Adoption decoupled from implementation; decoupling more prevalent in rms with powerful CEOs, or poor prior performance, and among later adopters

From Smoke and Mirrors to Walking the Talk

Weber et al. (2009)

Westphal and Zajac (1994)

31

32

Downloaded by [Academy of Management ] at 19:27 21 June 2012

The Academy of Management Annals

Table 2 Summary of Empirical Studies (Continued) Author(s), publication year Westphal and Zajac (1998) Westphal and Zajac (2001) Empirical setting Data from 408 corporations examining stock market reactions to adoption of governance mechanisms, 1982 1992 Implementation of stock repurchase programs in 412 midsized industrial and service rms in U.S., 1985 1991 Relevant contribution(s) Symbolic adoption of one governance policy deters other governance reforms; symbolic adoptions can have positive nancial outcomes Decoupling of policy and practice more likely when top executives have power over boards and when social structural or experiential factors enhance awareness of potential for decoupling Can be consequential outcome of adoption (positive market reaction) over time as policy becomes institutionalized in environment; forces toward aligning policy and practice may come from within rm to correct cognitive inconsistencies of rm identity Entrepreneurs gain more resources if they demonstrate symbolic attention to their personal credibility, professional organizing, organizational achievement, and quality of stakeholder relationships

Zajac and Westphal (2004)

Adoption of 860 stock repurchase plans among 463 rms listed in Forbes 500 or Fortune 500 in 1980, 1980 1994

Zott and Nguyen Huy (2007)

Two-year eld study of 26 British entrepreneurial ventures, 20022004

Note: This summary table is limited to empirical studies published after the last comprehensive review of decoupling by Orton and Weick in 1990.

From Smoke and Mirrors to Walking the Talk

33

public opinion, and have wide-reaching effects inside organizations. But the observation that institutional rules affect all organizations to some extent does not mean that all organizational decoupling is similarly institutional. Greve et al. (2010) discuss a number of cases where actions and outcomes do not follow from policy, focusing on organizational accidents and misconduct. Building on Perrow (1999) and Mezias (1994), they describe complex technical systems where interactions are rapid and closely linked, such as nancial markets, petrochemical reneries, nuclear power plants, or airliners, and a simple accident or series of oversights can lead to unexpected and even catastrophic outcomes. Similarly, deviance from established organizational rules can be a one-time occurrence or it can become normalized and informally accepted as the way things are done. Building on Vaughans (1996) analysis of the Challenger tragedy, Barnett and Finnemore (1999) note that normalization of deviance is a common pathology in international organizations. Individuals make small exceptions to formal rules, perhaps even with the intention of being more efcient in a specic case, under the assumption that a minor change from established rules will not cause a major bureaucratic failure. Over time these small deviations become routinized and accepted as standard practice for getting things done. They are no longer seen as rule-breaking. It is plausible that many instances of decoupling occur either through such normal accidents or intentional skirting of rules for both moral and immoral reasons. These forms of decoupling are unlikely to be the product of institutional forces, unless they occur routinely in a large number of organizations. Such ubiquity would suggest that broader social inuences are at play. These examples of normal accidents and deviance are instances of decoupling, but their causes tend to be rooted in either technical malfunctions or intentional efforts to bypass otherwise implemented rules. To be sure, pressures for productivity or performance can stem from external parties. But it is useful to distinguish between competitive and institutional inuences, as they can lead to distinctive forms of decoupling. Meyer and Rowan (1977, p. 347) observed that the two explanations of environmental isomorphism [competitive and institutional] are not entirely inconsistent. Organizations both deal with their environments at their boundaries and imitate environmental elements in their structures (Meyer and Rowan, 1977, p. 347). Some elements of organization arise for more functional reasons, whereas others respond to institutionalized myths in the wider environment. In both cases, external pressures shape organizations, but the source of external inuence carries different implications. Organizations operating in sectors where there are consistent demands, and where the relation between formal structure and production is clear cut, will maintain closer alignment between formal policy and daily practices than organizations in environments with multiple, competing demands. Such execution is

Downloaded by [Academy of Management ] at 19:27 21 June 2012

34

The Academy of Management Annals

possible because conformity to regulations can be easily observed and enforced through direct inspection, the denition of conformity to a policy is unambiguous and easily understood to have a direct link to production, and the various units of an organization have coordinated goals. But even in a highly transparent production chain, there are times where policies and practices fall out of alignmentthat is, they become decoupled, for reasons unrelated to institutional pressures. Simple human error, intentional or unintentional, can easily derail an efcient system, as can technological or mechanical failures. Or, a new production system may require a period of time to become fully operational. Nevertheless, given the ability to control, coordinate, and measure outputs in a technical system, instances of decoupling are likely to be short-lived. For example, the cashier systems at McDonalds restaurants are designed to operate with maximum efciency, with registers that are so intuitive they require almost no training to operate. The restaurants goal is to keep lines moving as quickly as possible in order to sell more burgers and fries. Nonetheless, cash registers can break; an employee can be particularly unmotivated or distracted and move slowly or make errors. In these cases, the formal policies of how the system should be working would differ from their actual implementation, a form of decoupling. But the source of dysfunction is identiable and can be corrected. Machines can be xed and employees can be encouraged to work harder or be replaced. Any decoupling is likely to be temporary and local, and it will not provide general legitimacy benets. Few aspects of modern organizational life are as simple as a fast-food counter. In practice, most situations fall somewhere along a spectrum between technical and institutional. Espelands (1998) case of dam-building, for example, is clearly a more technical setting than more social arenas such as schools, welfare services, or courts. But even in the construction of a dam, competing denitions of land use give rise to contentious debates over what a desirable outcome would be. Careful analysis of the level of institutionalization of a setting should be taken into account when one is examining decoupling. Technical decoupling is likely to be episodic, have an observable negative effect on efciency, and be correctible in a straightforward way. Similarly, if decoupling is created unintentionally, as in the normalization of deviance, it is likely to be remedied after the discrepancy is shown to create problems. Conversely, institutional decoupling is likely to persist as long as activities are aligned with prevailing myths in the environment, and have legitimating benets. Our main point is that not every instance of decoupling should be understood in purely institutional terms. Conclusions: Beyond Window Dressing Our view of decoupling provides insight into a number of macro-social trends in contemporary organizations. Two developments are most notable. Broadly,

Downloaded by [Academy of Management ] at 19:27 21 June 2012

From Smoke and Mirrors to Walking the Talk

35

in a rationalizing environment, we should expect widespread means ends decoupling as elements of formal organizations, such as standards of evaluation and measurement, are increasingly applied to complex social and cultural phenomena where means and ends are opaque. Consequently, in many elds, we observe waves of reform as one set of tools is revealed to have a poor relationship to intended outcomes and a new set of tools is ushered in. Hence, in a society that is both rationalizing and increasingly fragmented, the number of conicting external pressures is increasing, generating heterogeneity and complex organizational structures. The policies, practices, and goals of organizations are often developed in response to institutional pressures. If these externally driven practices conict with core aspects of production, a rational response is to avoid implementing them as much as possible. But as rationalization proceeds, the alignment of policy and practice becomes a moral obligation, and it is often monitored through the channels of hard and soft law combined with various counting and accounting tools. A number of studies show that, over time, even policies intended to be symbolic can become integrated into organizations in unexpected ways (Edelman, 1992; Hallett, 2010; Sahlin & Wedlin, 2008; Sauder & Espeland, 2009). Future research on policy practice decoupling should seek to understand when and why some policies remain symbolic while others become implemented. For example, the source or type of institutional pressure, characteristics of the organization itself, or features of the policy being adopted may be linked to the degree or persistence of policy practice decoupling. We speculate that policy practice decoupling may decline because: (a) more emphasis is placed on implementing policies, (b) policies are selected in part because they can be implemented and measured, and (c) pressures in the environment that drive the creation of policies may independently be changing practices and outcomes, regardless of policy implementation. And as the environment becomes more rationalized, more social domains where ultimate goals are hard to measure become subject to formal organization and the environment becomes more fragmented. Thus, means ends decoupling increases over time as organizations create policies in complex and ambiguous settings and work to incorporate rationalizing aspects of the environment internally. These propositions about the rise and fall of each type of decoupling, however, should be tested empirically. We have put forth a view that policies, practices, and outcomes are independently inuenced by the environment. This implies that a great deal of decoupling is unrelated to deliberate managerial strategy (Greenwood et al., 2011; Weaver, Trevino, & Cochran, 1999). Hironaka and Schofer (2002, pp. 215 216) similarly argue: scholars increasingly equate decoupling (and failed implementation) with cynical action by organizations that desire the legitimacy of

Downloaded by [Academy of Management ] at 19:27 21 June 2012

36

The Academy of Management Annals

conformity but do not wish to bear costs of proper implementation (Krasner, 1999). Although some examples t this pattern. . .we resist this narrow meaning of decoupling and the actor-centric strategic presumptions that it implies. Given that much decoupling is unintended, we also see a need for research that moves beyond the study of policy practice gaps. A useful direction would be to develop a clearer specication of the relationships between organizations and their environments. This research should examine the mechanisms through which policies, practices, and outcomes become aligned or maintain distinct trajectories, with an eye to analyzing the role of external pressures at all stages. To provide a specic example, we posit that in a rationalized environment, the revelation that a policy or practice is failing to meet its intended objectives feeds back into the environment, potentially generating new models and contributing to further fragmentation of the environment as competing models emerge. Over time, understandings of how to achieve and measure complex social goals may improve, reducing the gap between specied means and ends. Or competing models may further elaborate, creating pidgin languages rather than a lingua franca that is easily implemented. Such dynamics at present are poorly understood. An added theoretical insight that we gain is the identication of meansends decoupling as a distinct form, which affords insight into how larger institutional forces generate heterogeneity both within and between organizations. In contrast to earlier arguments that stressed conformity, the increasing complexity of organizational structures and the associated meansends decoupling is driven by the greater fragmentation of a highly rationalizing environment. As organizations respond to diverse inuences with appropriate steps, their internal structures become more elaborated. Further, within an organization various practices may be decoupled in different ways, and across a eld a common practice may be decoupled in divergent ways by organizations. Future research should document and analyze these trends empirically by collecting detailed data on the internal structures and activities of organizations, as well as developing more sophisticated measures of environmental fragmentation. In a curious irony, a consequence of decoupling is that the organizations most inuenced by external rationalizing pressures will devote extensive resources to enacting and evaluating practices that are only loosely related to their core goals. In other words, the most highly rationalized organizations may also be the least instrumental, judged by purposive action focused on avowed goals. Particularly when the link between means and ends is unclear, technical procedures of accounting, personnel production and selection, information gathering, and evaluating can become ends in themselves, establishing an organization as responsible, rational, and successful. But the link between these rationalized practices and an organizations mission or intended

Downloaded by [Academy of Management ] at 19:27 21 June 2012

From Smoke and Mirrors to Walking the Talk

37

Downloaded by [Academy of Management ] at 19:27 21 June 2012

production is unspecied, uncertain, and poorly measurable by existing metrics. At the organizational level, extensive resources are dedicated to performing these rituals to demonstrate organizational rationality and maintain dominant theories of progress, efciency, and effectiveness. This observation should not be interpreted as a critique of contemporary organizations, but rather as a description. Both greater good and more activity that is unrelated to primary tasks are simultaneously possible. Indeed, future studies could examine these normative implications in particular settings. One broad managerial implication of our work is that there is a need for more reective and proactive responses to external pressures. We would encourage leaders to focus thoughtfully on shaping tools, such as systems of reporting, monitoring, and evaluation, in ways that are more directly linked to their organizations core activities, and to search for ways to inuence the nature of external standards in the environment. We also suggest that applied studies of how organizations can mediate environmental pressures and how they can shape their environment would be useful. Research on how bottom up or bootstrapping practices, in which organizations relate external demands more directly to their daily activities, rather than external conformity, would be welcome. Extending beyond the organizations, societywide consequences emerge from the creation of new roles for monitoring and evaluation, and shifting lines of authority as those responsible for creating measures gain inuence over substantive professionals.

Acknowledgements Thanks to Wade Cole, Gili Drori, Royston Greenwood, Alexandra Kalev, Valeska Korff, Mike Lounsbury, John W. Meyer, and Charles Perrow for their comments on earlier drafts, and to Nick Romriell for research assistance.

Endnotes
1. Perhaps the central theme in Webers writing is the increasing control in social and material life. He was emphatic that this enhanced mastery, and its associated calculability and predictability, did not entail progress. He recognized an everpresent tension between modernity and the process of modernization, and saw rationality as a multi-faceted enterprise, with practical, substantive, theoretical, and formal components (Kalberg, 1980). Indeed, procedural efciency may well, he argued, come at the expense of substantive rationality (Brubaker, 1992). Quoting Goethe, he argued that rationalization was producing specialists without spirit, sensualists without heart (Weber, 1930[1992], p. xix). Orton and Weick distinguish between the concepts of loose coupling and decoupling in the following way: If there is neither responsiveness nor distinctiveness,

2.

38

The Academy of Management Annals

the system is not really a system, and it can be dened as a noncoupled system. If there is responsiveness without distinctiveness, the system is tightly coupled. If there is distinctiveness without responsiveness, the system is decoupled. If there is both distinctiveness and responsiveness, the system is loosely coupled (Orton and Weick, 1990, p. 205). For our purposes, and as is often the case in neo-institutional research, we treat the terms loose coupling and decoupling interchangeably, referring to a spectrum of connectedness rather than distinct concepts (Beekum and Ginn, 1993). Further, institutional studies describe decoupling and loose coupling as both a state and a process (which can, thus, lead to recoupling). These widely shared beliefs or myths, in the environment are rationalized because they take a particular form. They are impersonal prescriptions that identify various social purposes as technical ones and specify in a rulelike way the appropriate means to pursue these technical purposes rationally (Meyer and Rowan, 1977, pp. 343 344). We describe organizations as having a technical core, or main goal of production, but we view this as an institutionalized property of the organization rather than as a functional characteristic. That is, organizations are structured around the idea of production, although notions of how and what to produce are institutionalized. We do not, thus, distinguish sharply between more and less rational parts of the organization; but we do assume that organizations have self-proclaimed purposesthat their existence is based on the idea of doing something. The classic example of the QWERTY keyboard illustrates how even technical environments are strongly shaped by institutional pressures (David, 1985). There are many examples where even the core goals of production are rooted in broad institutional shifts, such as the rise of green products, fair-trade labor, or socially responsible investing. Following Meyer and Jepperson (2000, p. 102), by rationalization we refer to the cultural accounting of society and its environments in terms of articulated, unied, integrated, universalized, and causally and logically structured schemes. Our discussion focuses on decoupling between elements within an organization, but decoupling also occurs at other levels. For instance, Orton and Weick (1990) observe that decoupling can also occur between and within individuals. Studies of decoupling between individuals in an organization have included analyses of design principles in a steel mill (Hedberg, 1984), relationships within a rms investment division (Schall, 1983), baseball teams (Keidel, 1984), and the judicial system (Hagan et al., 1979). More recently, Pager and Quillian (2005) demonstrate a gap between employers reported attitudes toward hiring ex-offenders and their actual hiring behavior. This study builds on social psychological work indicating that at the individual level there is much decoupling between self-reports, or hypothetical intentions, and actual behavior (see Fiske, 2004, for a meta-analysis). Decoupling can also occur between organizations that are members of larger associations or involved in collective relationships, and research on these metaorganizations is an understudied area (see Ahrne and Brunsson, 2008, for more

Downloaded by [Academy of Management ] at 19:27 21 June 2012

3.

4.

5.

6.

From Smoke and Mirrors to Walking the Talk

39

7.

8.

9.

10.

11.

discussion). Provan (1983) identied ve types of relationships between organizations; he ranked them, in order, from loosely coupled to tightly coupled: coalitions, participatory federations, independent federations, mandated federations, and intra-organizational systems. Extra-organizational decoupling has also been studied in investment coalitions (Bygrave, 1988), networks of state agencies (Covaleski, Dirsmith, & Jablonsky, 1985; Kaplan, 1982), and among quasi rms in the hospital industry (Luke, Begun & Pointer, 1989). Naturally, individuals and organizations champion particular visions of means ends relationships and give instrumental accounts or rationales for their actions, but these beliefs reect externally constructed causal stories, what Mills (1940) called vocabularies of motive, rather than clearly observable or irrefutably provable links. In contrast to this institutional explanation for the buffering of internal organizational structures, Weick (1976) speculated that the existence of redundant systems aimed at producing the same outcome was benecial for organizations because it reduced the chance of complete organizational failure. If one system broke down, other processes serving the same function would continue to operate, making an organization more resilient. Similarly, Perrow (1981) argued that a certain level of redundancy promoted responsiveness in the face of error or failure. In a study of environmental organizations, Hironaka and Schofer (2002, pp. 215 216) make a similar observation. They state: In its original conception, loose coupling is very broad. Weak or disconnected links may be horizontal or vertical and may take many forms, ranging from simple autonomy of organizational units to matters such as the disconnection between plan and implementation. . . However, the common usage of decoupling in organizational research in now decidedly more specic. Decoupling is most frequently used to describe the disconnection between high-level organizational decisions, plans, and policies with concrete implementation and efcacious organizational outcome on the ground. Notably, almost no one would suggest that the way to improve education is to design policies to reect what schools could reasonably achieve. Instead, schools are envisioned as simultaneously being sites for the formation of human capital; building good citizens; instilling proper morality; teaching life skills as diverse as banking and interpersonal relations; providing adequate food; sex education; opportunities for physical tness and ne arts; and more. The focus is on xing the practices to help schools achieve goals dened in the environment rather than changing the goals themselvesan emphasis likely observed in many organizational settings. The idea of measurement should be interpreted broadly to include all sorts of accounting, counting, evaluation, monitoring, ranking, and reporting activities. For example, Gaertner and Ramnayayan (1983) discuss decoupling in four different types of accounts used by organizations that t this view: enacting frameworks for external audiences (auditing) and internal audiences (implementation

Downloaded by [Academy of Management ] at 19:27 21 June 2012

40

The Academy of Management Annals

plans), and setting frameworks for external audiences (legitimation) and internal audiences (integration). The pervasive inuence of rankings on legal education is likely due to the fact that one source, US News and World Report, dominates the realm of evaluation; in contrast, universities and colleges, business schools, and other academic disciplines are subject to multiple ranking systems. Consequently, there is greater leeway in adjusting to, and manipulating, multiple evaluative criteria in elds outside legal education. Applying Olivers categories to study decoupling between policy and practice in German rms, Fiss and Zajac (2006, p. 1188) nd that organizations that fervently proclaim their conformity to demands for strategic change are in fact less likely to be the ones that actually implement structural changes, while those that do implement changes may often feel compelled to downplay their conformity. Other studies suggest that acquiescence is more likely if organizations depend on institutional referents for resources, including legitimacy (DiMaggio & Powell, 1983; Oliver, 1991; Pfeffer & Salancik, 1978).

12.

13.

Downloaded by [Academy of Management ] at 19:27 21 June 2012

References
Ahrne, G., & Brunsson, N. (2008). Meta-organizations. Cheltenham, UK: Edward Elgar. Anderson-Levitt, K. (Ed.). (2003). Local meanings, global schooling: Anthropology and world culture theory. New York: Palgrave McMillan. Aurini, J. (2006). Crafting legitimation projects: An institutional analysis of private education businesses. Sociological Forum, 21(1), 83 111. Barnett, M., & Finnemore, M. (1999). The politics, power, and pathologies of international organizations. International Organization, 53, 699 732. Bartley, T. (2007). Institutional emergence in an era of globalization: The rise of transnational private regulation of labor and environmental conditions. American Journal of Sociology, 113(2), 297 351. Basu, K., & Palazzo, G. (2007). Corporate social responsibility: A process model of sensemaking. The Academy of Management Review, 33(1), 122 136. Beekum, R.I., & Ginn, G.O. (1993). Business strategy and interorganizational linkages within the acute care hospitals industry. Human Relations, 46, 1291 1317. Berger, P., & Luckman, T. (1966). The social construction of reality: A treatise in the sociology of knowledge. New York: Doubleday. Binder, A. (2007). For love and money: One organizations creative and multiple responses to a new funding environment. Theory and Society, 36, 547 571. Boli, J. (2006). The rationalization of virtue and virtuosity in world society. In M.L. Djelic & K. Sahlin-Andersson (Eds.), Transnational governance: Institutional dynamics of regulation (pp. 95 118). Cambridge, UK: Cambridge University Press. Boxenbaum, E., & Jonsson, S. (2008). Isomorphism, diffusion, and decoupling. In R. Greenwood, C. Oliver, K. Sahlin, & R. Suddaby (Eds.), The Sage handbook of organizational institutionalism (pp. 299 323). London: Sage. Boynton, A.C., & Zmud, R.W. (1987). Information technology planning in the 1990s: Directions for practice and research. MIS Quarterly, 11, 59 71.

From Smoke and Mirrors to Walking the Talk

41

Bromley, P., Hwang, H., & Powell, W.W. (2009). Common tendencies and divergent responses: Critical junctures in the institutionalization of strategic planning in nonprots. Paper presented at the 2009 meetings of the American Sociological Association, San Francisco, CA. Brubaker, R (1992). The limits of rationality. London: Routledge. Brunsson, N. (1985). The irrational organization: Irrationality as a basis for organizational action and change. Chichester, UK: John Wiley and Sons. Brunsson, N. (1989). The organization of hypocrisy: Talk, decisions and actions in organizations. Chichester, UK: John Wiley and Sons. Bussigel, M., Barzansky, B., & Grenholm, G. (1986). Goal coupling and innovation in medical schools. The Journal of Applied Behavioral Science, 22, 425 441. Bygrave, W.D. (1988). The structure of the investment networks of venture capital rms. Journal of Business Venturing, 3, 137 157. Chatterji, A.K., & Toffel, M.W. (2010). How rms respond to being rated. Strategic Management Journal, 31, 917 945. Clark, B.R. (1983). The higher education system: Academic organization in crossnational perspective. Berkeley, CA: University of California Press. Clemens, E., & Cook, J. (1999). Politics and institutionalism: Explaining durability and change. Annual Review of Sociology, 25, 441 466. Coburn, C. (2004). Beyond decoupling: Rethinking the relationship between the institutional environment and the classroom. Sociology of Education, 77, 211 244. Cohen, M.D., & March, J.G. (1974). Leadership and ambiguity. New York: McGraw Hill. Cohen, M.D., March, J.G., & Olsen, J.P. (1972). A garbage can model of organizational choice. Administrative Science Quarterly, 17, 1 25. Cole, W.M. (2005). Sovereignty relinquished? Explaining commitment to the international human rights covenants, 19661999. American Sociological Review, 70(3), 472 495. Cole, W.M. (2012). Human rights as myth and ceremony? Reevaluating the effectiveness of human rights treaties, 1981 2007. American Journal of Sociology, 117(4), 1131 1171. Covaleski, M.A., & Dirsmith, M.W. (1983). Budgeting as a means for control and loose coupling. Accounting, Organizations, and Society, 8, 323 330. Covaleski, M.A., Dirsmith, M.W., & Jablonsky, S.F. (1985). Traditional and emergent themes of budgeting: An empirical analysis. Journal of Accounting and Public Policy, 4, 277 300. Covaleski, M.A., Dirsmith, M.W., & Michelman, J.E. (1993). An institutional theory perspective on the DRG framework, case-mix accounting systems and healthcare organizations. Accounting, Organizations and Society, 18(1), 65 80. David, P.A. (1985). Clio and the economics of QWERTY. American Economic Review, 75(2), 332 337. Deephouse, D.L., & Suchman, M. (2008). Legitimacy in organizational institutionalism. In R. Greenwood, C. Oliver, R. Suddaby, & K. Sahlin (Eds.), The Sage handbook of organizational institutionalism (pp. 49 77). Thousand Oaks, CA: Sage. DiMaggio, P., & Powell, W.W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational elds. American Sociological Review, 48(2), 147 160.

Downloaded by [Academy of Management ] at 19:27 21 June 2012

42

The Academy of Management Annals

Dobbin, F. (1994). Cultural models of organization: The social construction of rational organizing principles. In D. Crane (Ed.), The sociology of culture: Emerging theoretical perspectives (pp. 117 141). Oxford, UK: Basil Blackwell. Dobbin, F. (2009). Inventing equal opportunity. Princeton, NJ: Princeton University Press. Dobbin, F., & Kelly, E. (2007). How to stop harassment: The professional construction of legal compliance in organizations. American Journal of Sociology, 112, 1203 1243. Dobbin, F., Kim, S., & Kalev, A. (2011). You cant always get what you need: Organizational determinants of diversity programs. American Sociological Review, 76(3), 386 411. Drori, G., Meyer, J.W., & Hwang, H. (2006). Globalization and organization: World society and organizational change. New York: Oxford University Press. Drori, G., Meyer, J.W., Ramirez, F.O., & Schofer, E. (2003). Loose coupling in national science: Policy versus practice. In G. Drori, J.W. Meyer, F.O. Ramirez, & E. Schofer (Eds.), Science in the modern world polity: Institutionalization and globalization (pp. 155 173). Stanford, CA: Stanford University Press. Edelman, L. (1990). Legal environments and organizational governance: The expansion of due process in the American workplace. American Journal of Sociology, 95, 1401 1440. Edelman, L. (1992). Legal ambiguity and symbolic structures: Organizational mediation of civil rights law. American Journal of Sociology, 97(6), 1531 1576. Edelman, L., Uggen, C., & Erlanger, H.S. (1999). The endogeneity of legal regulation: Grievance procedures as rational myth. American Journal of Sociology, 105, 406 454. Elsbach, K., & Sutton, R. (1992). Acquiring organizational legitimacy through illegitimate actions: A marriage of institutional and impression management theories. Academy of Management Journal, 35(4), 699 738. Espeland, W.N. (1998). The struggle for water. Chicago, IL: University of Chicago Press. Espeland, W.N., & Sauder, M. (2007). Rankings and reactivity: How public measures recreate social worlds. American Journal of Sociology, 113(1), 1 40. Feldman, M.S., & March, J.G. (1982). Information in organizations and signal and symbol. Administrative Science Quarterly, 26, 171 186. Firestone, W.A. (1985). The study of loose coupling: Problems, progress, and prospects. In A. Kerckhoff (Ed.), Research in the sociology of education and socialization (Vol. 5, pp. 3 30). Greenwich, CT: JAI Press. Fiske, S.T. (2004). Social beings. New York: Wiley. Fiss, P., & Zajac, E. (2004). The diffusion of ideas over contested terrain: The (non) adoption of a shareholder value orientation among German rms. Administrative Science Quarterly, 49(4), 501 534. Fiss, P., & Zajac, E. (2006). The symbolic management of strategic change: Sensegiving via framing and decoupling. The Academy of Management Journal, 49(6), 1173 1193. Frank, D.J., & Meyer, J.W. (2002). The profusion of individual roles and identities in the postwar period. Sociological Theory, 20(1), 86 105. Friedland, R., & Alford, R. (1991). Bringing society back in: Symbols, practices, and institutional contradictions. In W.W. Powell & P.J. DiMaggio (Eds.), The new

Downloaded by [Academy of Management ] at 19:27 21 June 2012

From Smoke and Mirrors to Walking the Talk

43

institutionalism in organizational analysis (pp. 232 263). Chicago, IL: University of Chicago Press. Gaertner, G.J., & Ramnayayan, S. (1983). Organizational effectiveness: An alternative perspective. Academy of Management Review, 8, 97 107. George, E., Chattopadhyay, P., Sitkin, S., & Barden, J. (2006). Cognitive underpinnings of institutional persistence and change: A framing perspective. The Academy of Management Review, 31(2), 347 365. Greenwood, R., & Hinings, C.R. (1996). Understanding radical organizational change: Bringing together the old and the new institutionalism. Academy of Management Review, 21(4), 1022 1054. Greenwood, R., Raynard, M., Kodeih, F., Micelotta, E., & Lounsbury, M. (2011). Institutional complexity and organizational responses. Academy of Management Annals, 5(1), 317 371. Greve, H.R., Palmer, D., & Pozner, J. (2010). Organizations gone wild: The causes, processes, and consequences of organizational misconduct. Academy of Management Annals, 4(1), 53 107. Grossman, A., & Curran, D.F. (2002). Edna McConnell Clark Foundation: A New Approach at an Old Foundation. Harvard Business School, Case 302090. Gupta, A.K. (1984). Contingency linkages between strategy and general manager characteristics: A conceptual examination. Academy of Management Review, 9, 399 412. Haack, P., Schoeneborn, D., & Wickert, C. (forthcoming). Talking the talk, moral entrapment, creeping commitment? Exploring narrative dynamics in corporate social responsibility standardization. Organization Studies, 33(5/6). Hafner-Burton, E., & Tsutsui, K. (2005). Human rights in a globalizing world: The paradox of empty promises. American Journal of Sociology, 110(5), 1373 1411. Hagan, J., Hewitt, J.D., & Alwin, D.F. (1979). Ceremonial justice: Crime and punishment in a loosely coupled system. Social Forces, 58, 506528. Hallett, T. (2010). The myth incarnate: Recoupling processes, turmoil, and inhabited institutions in an urban elementary school. American Sociological Review, 75, 52 74. Hannan, M.T., & Freeman, J. (1989). Organizational ecology. Cambridge, MA: Harvard University Press. Hedberg, B. (1984). Career dynamics in a steelworks of the future. Journal of Occupational Behavior, 5, 53 69. Hironaka, A., & Schofer, E. (2002). Decoupling in the environmental arena: The case of environmental impact assessments. In A.J. Hoffman & M.J. Ventresca (Eds.), Organizations, policy, and the natural environment: Institutional and strategic perspectives (pp. 214 234). Stanford, CA: Stanford University Press. Horwitch, M., & Thietart, R.A. (1987). The effect of business interdependencies on product R and D-intensive business performance. Management Science, 33, 178197. Jacobsson, B., & Sahlin-Anderssen, K. (2006). Dynamics of soft regulations. In M.L. Djelic & K. Sahlin-Andersson (Eds.), Transnational governance: Institutional dynamics of regulation (pp. 247 265). Cambridge, UK: Cambridge University Press. Kalberg, S. (1980). Max Webers types of rationality: Cornerstones for the analysis of rationalization processes in history. American Journal of Sociology, 85(5), 1145 1179.

Downloaded by [Academy of Management ] at 19:27 21 June 2012

44

The Academy of Management Annals

Kalev, A., Dobbin, F., & Kelly, E. (2006). Best practices or best guesses? Diversity management and the remediation of inequality. American Sociological Review, 71, 589 617. Kaplan, R.E. (1982). Intervention in a loosely organized system: An encounter with nonbeing. Journal of Applied Behavioral Sciences, 18, 415 432. Keck, M.E., & Sikkink, K. (1998). Activists beyond borders: Advocacy networks in international politics. Ithaca, NY: Cornell University Press. Keidel, R.W. (1984). Baseball, football, and basketball: Models for business. Organizational Dynamics, 12(3), 4 18. Kelly, E., & Dobbin, F. (1998). How afrmative action became diversity management. American Behavioral Scientist, 41, 960 984. Krasner, S. (1999). Sovereignty: Organized hypocrisy. Princeton, NJ: Princeton University Press. Kruecken, G., & Maier, F. (2006). Turning the university into an actor. In G. Drori, J.W. Meyer, & H. Hwang (Eds.), Globalization and organization: World society and organizational change (pp. 241 257). New York: Oxford University Press. Krueken, G., & Drori, G. (Eds.). (2009). World society: The writings of John W. Meyer. Oxford, UK: Oxford University Press. Leatt, P., & Schneck, R. (1984). Criteria for grouping nursing subunits in hospitals. Academy of Management Journal, 27, 150 165. Lim, A., & Tsutsui, K. (2012). Globalization and commitment in corporate social responsibility: Cross-national analyses of institutional and political-economy effects. American Sociological Review, 77(1), 69 98. Lok, J. (2010). Institutional logics as identity projects. Academy of Management Journal, 53(6), 1305 1335. Lounsbury, M. (2001). Institutional sources of practice variation: Stafng college and university recycling programs. Administrative Science Quarterly, 46, 29 56. Lounsbury, M. (2002). Institutional transformation and status mobility: The professionalization of the eld of nance. Academy of Management Journal, 45, 255 266. Lounsbury, M. (2007). A tale of two cities: Competing logics and practice variation in the professionalizing of mutual funds. Academy of Management Journal, 50, 289 307. LUEP Website (2011). Retrieved from [Link] Luke, R., Begun, G.W., & Pointer, D.D. (1989). Quasi rms: Strategic interorganizational forms in the health care industry. Academy of Management Review, 14, 9 19. Lyon, T., & Maxwell, J. (2011). Greenwash: Corporate environmental disclosure under threat of audit. Journal of Economics and Management Strategy, 20(1), 3 41. Maclean, T., & Behnam, M. (2010). The dangers of decoupling: The relationship between compliance programs, legitimacy perceptions, and institutionalized misconduct. Academy of Management Journal, 53(6), 1499 1520. March, J.G., & Olsen, J.P. (1976). Ambiguity and choice in organizations. Bergen: Universitetsforlaget. Meyer, J.W., Boli, J., Thomas, G.M., & Ramirez, F.O. (1997). World society and the nation-state. American Journal of Sociology, 103(1), 144 181. Meyer, J.W., & Bromley, P. (2012). The worldwide expansion of organization. Paper presented at the 8th New Institutionalism Workshop, Barcelona.

Downloaded by [Academy of Management ] at 19:27 21 June 2012

From Smoke and Mirrors to Walking the Talk

45

Meyer, J.W., & Jepperson, R.L. (2000). The actors of modern society: The cultural construction of social agency. Sociological Theory, 18, 100120. Meyer, J.W., & Rowan, B. (1977). Institutionalized organization: Formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340 363. Meyer, J.W., & Rowan, B. (1978). The structure of educational organizations. In M. Meyer (Ed.), Environments and organizations (pp. 78 109). San Francisco, CA: John Wiley and Sons. Mezias, S.J. (1994). Financial meltdown as normal accident: The case of the American savings and loan industry. Accounting, Organizations, and Society, 19(2), 181 192. Mills, C.W. (1940). Situated actions and vocabularies of motive. American Journal of Sociology, 5(6), 904 913. Moerth, U. (2006). Soft regulation and global democracy. In M.L. Djelic & K. SahlinAndersson (Eds.), Transnational governance: Institutional dynamics of regulation (pp. 119 137). Cambridge, UK: Cambridge University Press. Monahan, S. (1999). Role ambiguity among protestant clergy: Consequences of the activated laity. Review of Religious Research, 41(1), 79 94. Murphy, J.A., & Hallinger, P. (1984). Policy analysis at the local level: A framework for expanded investigation. Educational Evaluation and Policy Analysis, 6, 5 13. Oliver, C. (1991). Strategic responses to institutional processes. Academy of Management Review, 16, 145 179. Orton, J.D., & Weick, K.E. (1990). Loosely coupled systems: A reconceptualization. Academy of Management Review, 15(2), 203 223. Pache, A., & Santos, F. (2010). When worlds collide: The internal dynamics of organizational responses to conicting institutional demands. Academy of Management Review, 35(3), 455 476. Pager, D., & Quillian, L. (2005). Walking the talk? What employers say versus what they do. American Sociological Review, 70(3), 355 380. Perrow, C. (1981). Normal accident at three mile Island. Society, 18(5), 17 26. Perrow, C. (1999). Organizing to reduce the vulnerabilities of complexity. Journal of Contingencies and Crisis Management, 7(3), 150 155. Pfeffer, J., & Salancik, G.R. (1978). The external control of organizations. New York: Harper & Row. Powell, W.W., & Colyvas, J.A. (2008). Microfoundations of institutional theory. In R. Greenwood, C. Oliver, K. Sahlin, & R. Suddaby (Eds.), The SAGE handbook of organizational institutionalism (pp. 276 298). Thousand Oaks, CA: Sage Publications. Power, M. (1994). The audit explosion. London: Demos. Power, M. (1997). The audit society. Oxford, UK: Oxford University Press. Pratt, M., & Foreman, P. (2000). Classifying managerial responses to multiple organizational identities. Academy of Management Review, 25(1), 18 42. Pressman, J.L., & Wildavsky, A. (1973). Implementation: How great expectations in Washington are Dashed in Oakland. Berkeley, CA: University of California Press. Provan, K.G. (1983). The federation as an interorganizational linkage network. Academy of Management Review, 8, 79 89. Putnam, R., & Campbell, R. (2010). American grace: How religion divides and unites us. New York: Simon and Schuster.

Downloaded by [Academy of Management ] at 19:27 21 June 2012

46

The Academy of Management Annals

Rao, H. (1998). Caveat emptor: The construction of nonprot consumer watchdog organizations. American Journal of Sociology, 103(4), 912 961. Risse, T., Ropp, S., & Sikkink, K. (1999). The power of human rights: International norms and domestic change. Cambridge: Cambridge University Press. Roberts, J. (2004). The modern rm: Organizational design for performance and growth. Oxford, UK: Oxford University Press. Room to Read website 2011. Envisioning our future: A roadmap for learning. Retrieved December 8, 2011, from [Link] Rowan, B. (2006). The new institutionalism and the study of educational organizations: Changing ideas for changing times. In H.D. Meyer & B. Rowan (Eds.), The new institutionalism in education (pp. 15 32). Albany, NY: State University of New York Press. Sahlin, K., & Wedlin, I. (2008). Circulating ideas: Imitation, translation and editing. In R. Greenwood, C. Oliver, K. Sahlin, & R. Suddaby (Eds.), The Sage handbook of organizational institutionalism (pp. 218 242). London: Sage. Sauder, M., & Espeland, W. (2009). The discipline of rankings: Tight coupling and organizational change. American Sociological Review, 74, 63 82. Schall, M.S. (1983). A communication-rules approach to organizational culture. Administrative Science Quarterly, 28, 557 581. Schneiberg, M., & Soule, S.A. (2005). Institutionalization as a contested, multilevel process: The case of rate regulation in American re insurance. In G. Davis, D. McAdam, W.R. Scott, & M. Zald (Eds.), Social movements and organizational theory (pp. 122 160). Cambridge, UK: Cambridge University Press. Schofer, E., & Hironaka, A. (2005). The effects of world society on environmental protection outcomes. Social Forces, 84(1), 25 47. Scott, W.R. (2008). Institutions and organizations: Ideas and interests (3rd ed.). Thousand Oaks, CA: Sage Publications, Inc. Scott, W.R., Ruef, M., Mendel, P.J., & Caronna, C.A. (2000). Institutional change and healthcare organizations: From professional dominance to managed care. Chicago, IL: University of Chicago Press. Selznick, P. (1949). TVA and the grass roots. Berkeley, CA: University of California Press. Selznick, P. (1957). Leadership in administration: A sociological interpretation. New York: Harper & Row. Seo, M., & Creed, W. (2002). Institutional contradictions, praxis, and institutional change: A dialectical perspective. Academy of Management Review, 27(2), 222 247. Short, J.L., & Toffel, M.W. (2010). Making self-regulation more than merely symbolic: The critical role of the legal environment. Administrative Science Quarterly, 55(3), 361 396. Snellman, K. (2011). Battles in boardrooms: The diffusion of shareholder value rhetoric and practice in Finland, 1990-2005 (Unpublished doctoral dissertation). Department of Sociology, Stanford University, Stanford, CA. Soltow, L. (1981). The rise of literacy and the common school in the United States: A socioeconomic analysis to 1870. Chicago, IL: University of Chicago Press. Spillane, J., Parise, L., & Sherer, J. (2011). Organizational routines as coupling mechanisms: Policy, school administration, and the technical core. American Educational Research Journal, 48(3), 586 619.

Downloaded by [Academy of Management ] at 19:27 21 June 2012

From Smoke and Mirrors to Walking the Talk

47

Strathern, M. (2000). Audit cultures: Anthropological studies in accountability, ethics and the Academy. New York: Routledge. Thomas, D.C. (2001). The Helsinki effect: International norms, human rights, and the demise of communism. Princeton, NJ: Princeton University Press. Thomas, J. (1983). Justice as interaction: Loose coupling and mediations in the adversary process. Symbolic Interaction, 6, 243 260. Thomas, J. (1984). Some aspects of negotiated order, loose coupling, and mesostructure in maximum security prisons. Symbolic Interaction, 7, 213 231. Thompson, J.D. (1967). Organizations in action. New York: McGraw Hill. Thornton, P.H. (2004). Markets from culture: Institutional logics and organizational decisions in higher education publishing. Stanford, CA: Stanford University Press. Thorton, P.H., & Ocasio, W. (1999). Institutional logics and the historical contingency of power in organizations: Executive succession in the higher education publishing industry, 1958-1990. American Journal of Sociology, 105, 801 843. Tilcsik, A. (2010). From ritual to reality: Demography, ideology, and decoupling in a post-communist government agency. Academy of Management Journal, 53(6), 1474 1498. Tyack, D., & Cuban, L. (1995). Tinkering toward utopia: A century of public school reform. Cambridge, MA: Harvard University Press. Vaughan, D. (1982). Toward understanding unlawful organizational behavior. Michigan Law Review, 80, 1377 1402. Vaughan, D. (1996). The challenger launch decision. Chicago, IL: University of Chicago Press. Weaver, G., Trevino, L., & Cochran, L. (1999). Corporate ethics practices in the mid1990s: An empirical study of the Fortune 1000. Journal of Business Ethics, 18, 283 294. Weber, M. ([1921] 1968). G. Roth & C. Wittech (Eds.), Economy and society. New York: Bedminster. Weber, M. (1930 [1992]). The Protestant ethic and the spirit of capitalism. (T. Parsons, Trans.). London: Routledge. Weber, K., Davis, G., & Lounsbury, M. (2009). Policy as myth and ceremony: The global spread of stock exchanges, 1980 2005. Academy of Management Journal, 52(6), 1319 1347. Weick, K.E. (1976). Educational organizations as loosely coupled systems. Administrative Science Quarterly, 21, 1 19. Weick, K.E. (1982). Administering education in loosely coupled schools. Phi Delta Kappan, 63(10), 673 676. Westphal, J., Gulati, R., & Shortell, S. (1997). Customization or conformity? An institutional and network perspective on the content and consequences of TQM adoption. Administrative Science Quarterly, 42, 366 394. Westphal, J., & Zajac, E. (1994). Substance and symbolism in CEOs long-term incentive plans. Administrative Science Quarterly, 39, 367 390. Westphal, J., & Zajac, E. (1998). The symbolic management of stockholders: Corporate governance reforms and shareholder reactions. Administrative Science Quarterly, 43(1), 127 153. Westphal, J., & Zajac, E. (2001). Decoupling policy from practice: The case of stock repurchase programs. Administrative Science Quarterly, 46(2), 202 228.

Downloaded by [Academy of Management ] at 19:27 21 June 2012

48

The Academy of Management Annals

Zajac, E., & Westphal, J. (2004). The social construction of market value: Institutionalization and learning perspectives on stock market reactions. American Sociological Review, 69, 433 457. Zott, C., & Nguyen Huy, Q. (2007). How entrepreneurs use symbolic management to acquire resources. Administrative Science Quarterly, 52, 70 105. Zucker, L.G. (1987). Institutional theories of organization. Annual Review of Sociology, 13, 443 464.

Downloaded by [Academy of Management ] at 19:27 21 June 2012