PA Assg#7–Please see attached instructions. Thank you.

https://www.jstor.org/stable/3525838?seq=1&cid=pdf- reference#references_tab_contents You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms and Oxford University Press are collaborating with JSTOR to digitize, preserve and extend access to Journal of Public Administration Research and Theory: J-PART This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms JPART 14:283-307 Institutional Isomorphism and Public Sector Organizations Peter Frumkin Harvard University Joseph Galaskiewicz University of Arizona ABSTRACT Although public sector organizations have long been seen as driving the institutionalization of business firms and nonprofit organizations, government agencies themselves have only occasionally been studied as subjects of institutional pressures. This research examines whether public sector organizations, when compared with organizations in the business and nonprofit sectors, are more or less as susceptible to mimetic, normative, and coercive pressures. Using data from the National Organizations Study, we discover that governmental organizations are in fact more vulnerable to all three types of institutional forces than other organizations, whereas the effect of institutional variables on for-profits and nonprofits is more sporadic. The susceptibility of public sector organizations to institutional pressures raises important questions for the field of public administration and has consequences for nonprofits and business firms, which are funded and regulated by government. Though it has not penetrated deeply into the public administration literature to date, the new institutionalism in organizational analysis has made significant contributions to our understanding of the dynamics of organizations. It has moved research away from overly rationalistic explanations of organizational behavior toward explanations that recognize that organizations are embedded in larger cultural and political contexts. It has shifted the focus of research from explaining why organizations are so heterogeneous to why organizations are so similar. Over the past decade, as its reach has extended across different topics, institutional theory has focused often on nonprofit organizations, such as art museums (DiMaggio 1991yf F R O O H J H V D Q G X Q L Y H U V L W L H V % U L Q W D Q G . D U D E H O . U D D W z 1998yf K X P D Q L W D U L D Q J U R X S V & K U L V W L D Q V H Q D Q G 0 R O L Q \f, associations (Halliday, Powell, and Granfors 1993yf D Q G G U X J D E X V H X Q L W V ' $ X Q Q R 6 X W W R Q D Q G 3 U L F H \f. A number of studies have demonstrated the relevance of the institutional approach to We would like to thank the team of researchers who conducted the National Organizations Study and allowed us to use their data. We are especially indebted to Peter V. Marsden, David Knoke, Ame L. Kalleberg, and Joe I. Spaeth. We also want to thank four anonymous JPART reviewers who gave us very useful suggestions for revising the manuscript, including one reviewer who went well above the call of duty to offer unusually detailed and helpful advice. DOI: 10.1093/jopart/muh028 Journal of Public Administration Research and Theory, Vol. 14, no. 3 ? 2004 Journal of Public Administration Research and Theory, Inc.; all rights reserved. This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms 284 Journal of Public Administration Research and Theory for-profit organizations, including large corporations (Davis and Greve 1997; Galaskiewicz and Wasserman 1989; Haunschild and Miner 1997; Holm 1995yf U D L O U R D G V ' R E E L Q \f, law firms (Tolbert 1988yf D Q G K L J K W H F K F R P S D Q L H V 3 R Z H O O 6 X F K P D Q \f. Some research has also looked at public sector organizations (Meyer et al. 1988; Meyer, Scott, and Strang 1987; Tolbert and Zucker 1983yf . On the whole, however, research in organizational theory has tended to view public sector agencies more as playing the role of catalyst and trigger to institutionalization in other organizations, rather than as the objects of institutional pressures. Through regulation, accrediting, oversight, and funding relations, public sector organizations have been de- scribed as forces pushing nonprofits and business firms toward greater levels ofhomogeneity. Our goal here is to reexamine how and why government organizations themselves might be susceptible to institutional pressures. Our exploration of this topic proceeds in four broad steps. First, we trace key developments in the evolution of neoinstitutional theory and the role public sector organizations have played in inaugurating the institutional transformation of other organizations. Second, drawing on the literature of public administration, we develop a theoretical basis for the claim that the impact of institutional forces on government organizations should actually be stronger than that in nonprofit or business organizations. Third, using a national study of organizations, we examine whether public sector organizations-compared with nonprofit and business organizations-are in fact more or less vulnerable to different kinds of institutional pressure. Finally, we explore whether any particular type of institutional pressure-that is, being subject to regulations, licensing, and accreditation (coercive isomorphismyf E H O R Q J L Q J W R D Q D V V R F L D W L R Q R I S H H U R U J D Q L ] D W L R Q s (normative isomorphismyf D Q G O R R N L Q J D W W K H S H U I R U P D Q F H R I R W K H U R U J D Q L ] D W L R Q V P L P H W L c isomorphismyf S O D \ V D G R P L Q D Q W U R O H L Q V K D S L Q J S X E O L F V H F W R U R U J D Q L ] D W L R Q V . THE PUBLIC SECTOR IN NEOINSTITUTIONAL THEORY In its present form, the new institutionalism in organizational analysis accommodates a broad range of theoretical, methodological, and substantive interests (Scott 1991, 1995yf . Although much work has been done to sort out and classify the many strands of institutional research, the theory remains more of an orientation than a scientific theory. Differences among institutional arguments are considerable, but a few central issues and themes do unite the approach. Early statements of the theory emphasized the symbolic and ceremonial transformation that organizations undergo, changes that reflect myths in the institutional environment rather than a detached calculus of costs and benefits (Meyer and Rowan [1977] 1991yf 2 W K H U H D U O \ Z R U N I R F X V H G R Q S U R F H V V H V V X F K D V L V R P R U S K L c transformation (Tolbert and Zucker 1983yf F R Q W U D G L F W L R Q V ) U L H G O D Q G D Q G $ O I R U G \f, persistence (Zucker 1988, [1977] 1991yf G L I I X V L R Q 7 R O E H U W D Q G = X F N H U \f, and institutionalization (Scott 1991, 1994a, 1994b, 1995yf , Q P D Q \ R I W K H V H W K H R U H W L F D l formulations and in the institutional case studies that have been conducted to date, government agencies play a critical role in triggering and inaugurating institutional change in nonprofit and for-profit organizations, sometimes by exerting pressure through their control of funding, sometimes by the exercise of their power to regulate. With its emphasis on legitimacy, satisfying behavior, structural decoupling, and symbols, the new institutionalism represents a major departure from rival theories such as transaction costs economics (Williamson 1981, 1985yf S R S X O D W L R Q H F R O R J \ + D Q Q D Q D Q d Freeman 1989yf D Q G U H V R X U F H G H S H Q G H Q F H W K H R U \ 3 I H I I H U D Q G 6 D O D Q F L N \f, which are This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms Frumkin and Galaskiewicz Institutional Isomorphism and Public Sector Organizations based to a greater or lesser extent on the model of rational action. Indeed, a common thread throughout neoinstitutional research is a commitment to seeing organizations as pursuing practices that may have little to do with maximizing efficiency and to understanding that structure may be decoupled from the organizational mission. Organizations do not always embrace strategies, structures, and processes that enhance their performance but, instead, react to and seek ways to accommodate pressures following external scrutiny and regulation. Drawing on three of the more prominent contributions to theory, we suggest that government action, or more generally state intervention, has consistently been conceived as playing a central function in initiating the structural transformation of other organizations. Although the new institutionalism is usually thought of as being primarily a cultural theory of organizations, emphasizing interorganizational diffusion of rituals and roles, the new institutionalism has a political core that points to public sector organizations as the drivers and triggers of institutionalization. Government licensing, inspection, and regulation are the levers that act on nonprofit and business organizations. The three seminal formulations of institutionalism can be seen as constituting together a slow progression toward an understanding of the centrality of public sector organizations in fueling the symbolic and isomorphic changes that these organizations undergo. Early work by Meyer and Rowan ([1977] 1991yf S R U W U D \ H G R U J D Q L ] D W L R Q D O V W U X F W X U H D s loosely coupled with work activity and saw roles, myths, and ceremonies as emerging and spreading within organizational fields. There are several sources of institutionalized myths, including dense relational networks and leadership of local organizations. These are not, however, the only factors that speed the spread of institutional myths. Political and legal constraints play a central role in this whole process: "The myths generated by particular organizational practices and diffused through relational networks have legitimacy based on the supposition that they are rationally effective. But many myths also have legitimacy based on legal mandates" (Meyer and Rowan [1977] 1991, 148yf 2 U J D Q L ] D W L R Q D O S U D F W L F H s may appear on the scene and gain acceptance because of external pressures from public sector organizations: "Legislative and judicial authorities create and interpret legal mandates; administrative agencies-such as state and federal government, port authorities, and school districts-establish rules of practice; and licenses and credentials become necessary in order to practice occupations" (Meyer and Rowan [1977] 1991, 148yf $ V O H J D l and regulatory pressures increase, nonprofit and business organizations respond with increasing levels of institutionalized rules and procedures. Organizational fields rich in myths and ceremonies are constructed when pressure is exerted on organizations by forces in the surrounding environment. Government agencies are some of the most potent and influential environmental actors, which come into contact at some level-be it through laws or regulations-with most organizations. Looking at some of these same issues, DiMaggio and Powell ([1983] 1991yf K D Y e identified three forces driving institutionalization: (1yf F R H U F L Y H L V R P R U S K L V P W K D W V W H P s from political influence and the need for legitimacy, (2yf P L P H W L F L V R P R U S K L V P U H V X O W L Q g from standard responses to uncertainty, and (3yf Q R U P D W L Y H L V R P R U S K L V P D V V R F L D W H G Z L W h professionalization. These three mechanisms can overlap and intermingle, but they tend to derive from different conditions. At an analytic level, only coercive isomorphism is linked to the environment surrounding the organizational field. Mimetic and normative processes are internal to the field and help explain the spread of roles and structures. When organizations are subjected to outside coercive scrutiny, evaluation, and regulation, they tend to react defensively and gravitate toward isomorphic transformation. As the pressures 285 This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms 286 Journal of Public Administration Research and Theory from the outside grow, organizations are led to find ways to either diffuse or eliminate this pressure by changing their practices. One of the easiest ways to change is to adopt those routines and structures that are defined by law or government agencies as legitimate. To do so may ensure survival by minimizing conflict. The process of institutionalization is intimately tied with the idea of an organizational field. Rather than simply emerging as the product of the natural interaction of organizations, fields are constructed for a purpose. They are the ultimate product of coercive, mimetic, and normative isomorphism, and they reflect the slow homogenization and convergence of organizational forms. DiMaggio and Powell take a strong position on the emergence of fields, arguing that they are constructed and that they serve as information networks fueling standardization and professionaliza- tion. By spelling out and emphasizing the role of the state in forcing organizations in the direction of isomorphism and convergence, DiMaggio and Powell provide a significant addition to Meyer and Rowan's earlier formulation. Public sector organizations emerge as even more central actors propelling other organizations toward convergence in a subsequent institutional account. Fligstein's (1990, 1991yf F H Q W U D O W K H V L V L V W K D W W K H Y L D E L O L W \ R I $ P H U L F D Q L Q G X V W U L D O H Q W H U S U L V H L V U H O D W H G W R W K e long-term shifts in the conception of how the largest firms should operate to preserve their growth and profitability: "These shifts have occurred in response to a complex set of interactions between the largest firms, those who have risen to control those firms, and the government. They originated with managers and entrepreneurs who sought control over their internal and external environments. When one solution was blocked by the actions of the government, new solutions were created and diffused" (1990, 2yf 7 K L V O H D G V W R D W Z R - part institutional theory of change: (1yf R U J D Q L ] D W L R Q V D U H V X E M H F W W R J R Y H U Q P H Q W S U H V V X U e and embrace institutional myths, and (2yf V W U D W H J L H V D Q G V W U X F W X U H V D U H G L V V H P L Q D W H G D Q d a homogenized organizational field is created. Each time major new regulations and laws are put in place, the whole process begins anew. Thus, the regulatory state and the organizational field are locked in a mutually reinforcing game of moves and countermoves. In summary, three early and important statements in the development of institutional theory can be seen as expressing a consistent appreciation of the state's role in the production and dissemination of legitimate organizational forms. Meyer and Rowan discussed the function that government plays in institutionalization. DiMaggio and Powell recognized the centrality of state-driven coercive isomorphism but placed equal emphasis on the mimetic and normative isomorphism. In Fligstein's work, the public sector finally emerges as the central actor in the institutional plotline, with regulation driving corporations to homogenize and adopt distinctive conceptions of corporate control. Although public sector organizations occupy an important place in explanations of the drivers of institutionalization, these same organizations have not been adequately conceptualized as being subject to the same kinds of pressures that they help initiate. We argue that the essential features of public sector organizations make them not just capable of generating institutional pressures but also potentially very vulnerable to these same pressures. PUBLIC ADMINISTRATION THEORY AND PUBLIC SECTOR UNIQUENESS Institutional theory has never explicitly theorized that public sector organizations differ that greatly from all other organizations. If institutional pressures affected public sector organizations, then they did not affect these organizations any differently than they This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms Frumkin and Galaskiewicz Institutional Isomorphism and Public Sector Organizations affected nonprofits and for-profits. The same disinterest in public sector organizations sui generis is a feature of the broader organization theory literature. Indeed, when central figures in the development of organizational research like Herbert Simon (March and Simon 1958; Simon 1946, 1948, 1995; Simon, Smithburg, and Thompson [1950] 1991yf discussed sectoral issues, it was mainly to affirm that public agencies and private firms have more similarities than significant differences (Rainey 1997yf 7 K H S H U V S H F W L Y H R I W K H V e and other organizational theorists contributed over time to a broad conclusion that "organizations were simply organizations" and that "sector" was not a particularly useful distinction to make. The willingness of organization theory to blur public and private boundaries has been countered by a long tradition of research within public administration that holds that the sector of an organization needs to be an integral part of organizational research. Dahl and Lindblom's (1953yf D Q D O \ V L V R I P D U N H W V D Q G S R O \ D U F K L H V S R V L W H G D F R Q W L Q X X P R f organizations ranging from enterprises governed by private markets to agencies operated by government. Across the continuum of pure private to pure public organizations, a large group of hybrid and mixed organizations could be located. The incentive structures within enterprises and agencies were held to be different enough as to lead the private and public sectors to adopt substantially different systems to control costs and monitor performance. These early theoretical claims about sector differences have been informed by empirical work aimed at sorting out organizational characteristics that define the sectors. Over the past three decades or more, numerous empirical investigations were produced aimed at answering the question of whether public sector organizations are structured and behave differently than all other organizations. Some of these studies found higher levels of centralization and bureaucratization in government agencies than in other organizations, whereas other research concluded that the differences were in fact small (Rainey 1997yf . Although the empirical evidence is mixed, the public administration literature has coalesced around the view that organization theory provides an incomplete analysis of the internal structure and external environment of public organizations (Murray 1975; Perry and Kraemer 1983; Pitt and Smith 1981yf 6 H Y H U D O X V H I X O I U D P H Z R U N V K D Y H H P H U J H G W R K H O p further define the critical characteristics that define organizations in different sectors. From Wamsley and Zald's (1973yf F D W H J R U L ] D W L R Q D F F R U G L Q J W R S X E O L F R U S U L Y D W H I R U P V R I I X Q G L Q g and ownership to Bozeman's (1987yf G L V W L Q F W L R Q E H W Z H H Q S X E O L F D Q G S U L Y D W H I R U P V R f economic and political authority, the public administration literature has attempted to bring greater clarity and depth to the discussion of sectoral identity. Over time, the critique of organization theory by public administration has gotten quite broad and led to new research not just on organizational structure but also on management and leadership practices, decision-making processes, work-related behavior, and reward systems. Attempts to codify and organize the disparate research tracks have been limited by the sheer range and variety of empirical and theoretical rejoinders that the public administration literature has produced (Rainey 1989; Rainey, Backoff, and Levine 1976yf . When public administration theory has ventured into organizational sociology and taken up the issue of institutional pressures, it has usually equated this concept to the broader phenomenon of bureaucratization. Several important branches of public administration theory have advanced explanations for why public managers are often drawn to bureaucratic forms and processes. Early theories of bureaucratic politics, for example, emphasized that the budgetary process can lead to perverse decisions and make rational planning impossible (Pressman and Wildavsky 1973; Wildavsky 1964yf 7 K e 287 This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms 288 Journal of Public Administration Research and Theory inability of government agencies to free themselves from political and budgetary pressure can lead to the triumph of incrementalism (Lindblom 1965yf D Q G G L P L Q L V K H G F D S D F L W \ I R r change. Other work has focused on the phenomenon of bureaucratic inertia in which a vicious circle emerges that pushes public agencies toward ever greater levels of impersonalism. In the absence of strong stakeholder accountability mechanisms connected to service quality, bureaucratic forms emerge that often require less personal interaction among workers. Once personal connections are severed, rules and procedures are put in place to stop individuals from seeking to act alone to solve problems in the few remaining areas where rules to do not apply. Bureaucratization can also result when public sector organizations are suddenly subject to oversight and scrutiny by higher levels of government. Formalization and centralization can be simple responses to uncertainty and external pressure (Wilson 1989yf . Our empirical analysis here aims to bring the new institutionalism into fuller and deeper contact with public administration research. We revisit the question of public sector uniqueness by testing for differences in levels of institutional exposure among organizations from all three societal sectors. Before looking for differences in organizational response to the three main forms of institutional pressure, we need to ask why one might expect there to be differences among sectors in terms of vulnerability to institutional pressures. The public administration literature has focused on comparisons between government and business firms and has had relatively little to say about difference between government and nonprofit organizations. To answer the comparative questions related to all three sectors, it is useful to draw on both public administration theory (Bozeman 1987; Wamsley and Zald 1973yf D Q G W K e literature on nonprofit organizations (Hansmann 1980,1996; Weisbrod 1988yf 7 K H J U R Z L Q g research on nonprofit organizations has sought like public administration theory to draw lines and sectoral differences. Unlike the public administration literature, however, nonprofit research has focused on all three societal sectors in framing a theory of sectoral differences. Across these two literatures, we locate two bases for differentiating organizations from all the three sectors: (1yf G L I I H U H Q F H V L Q W K H D E L O L W \ W R D F F X U D W H O \ P H D V X U e organizational outputs or performance and (2yf G L I I H U H Q F H V L Q W K H F R Q W U R O R Y H U W K H I O R Z R f resources or inputs into the organization. Given that institutional pressures are likely to thrive where measurement and control are weak or imprecise, that is, where accountability is low, these distinctions allow us to generate two research hypotheses about how sector modifies the effect of institutional exposure on organizational structures. Differences in Measurement of Organizational Outputs A first way to look at sectoral differences is to focus on the way outputs are measured. In the business world, owners seek to maximize their return on investment by creating firms that will generate profits. Money generated by business activity is routinely distributed to those who have ownership status in the firm and who have taken on risk in investing. Performance is measured by profits, share price, earnings, or market share or through other metrics that allow owners including shareholders to measure how well management is furthering their interests. Managers who perform poorly are routinely dismissed, and shareholders are increasingly aggressive in asserting oversight and control, especially when performance lags (Useem 1993yf $ O W K R X J K W K H U H L V V R P H D U J X P H Q W R Y H U Z K L F K P H W U L c is the best predictor of firm success, there is little doubt that common quantitative measures of firm performance are fairly clear and easily interpreted. Regulatory agencies like the This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms Frumkin and Galaskiewicz Institutional Isomorphism and Public Sector Organizations Securities and Exchange Commission ensure that data on firm performance are consistently reported to all. Given these arrangements, owners and investors are in a very strong position indeed when it comes to oversight and measurement of organizational outputs. The situation of nonprofits and government agencies is more complicated because there are neither owners who have a material interest in the organization nor easily interpretable output measures. Both public agencies and nonprofits operate under what has been called the "nondistribution constraint" (Hansmann 1980yf : K H U H D V E X V L Q H V V I L U P s are able to distribute earnings to shareholders, government agencies and nonprofit organizations are not permitted to make such distributions but, rather, must retain any earnings. This means that there are no actors who have a direct financial interest in the organization, and thus no actor has a material stake in monitoring its operations. Furthermore, outputs are often more difficult to measure. The public management literature has consistently acknowledged that performance measurement is rendered more difficult given the nature of the goods and services produced (Moore 1995yf 2 I W H Q W K e outputs are indivisible; rarely is funding available to serious efforts at performance measurement; there are no large, mobilized groups pushing for a greater level of program assessment in government; sometimes benefits cannot be measured until far into the future; and considerable information asymmetries may be present. All of this makes performance measurement extremely difficult (Gormley and Weimer 1999yf . The presence of difficult-to-measure outputs can also be traced to the nondistribution constraint that binds public and nonprofit organizations. Because no private actor has claims to residual earnings, governments and nonprofits are thought to be in a better position to provide trust goods as well as collective goods that are often undersupplied in markets (Hansmann 1980yf 3 X E O L F D Q G Q R Q S U R I L W R U J D Q L ] D W L R Q V D U H D E O H L Q S U L Q F L S O H W o reassure citizens and clients that the public service goals of these organizations take precedence over the financial remuneration of any interested parties (Weisbrod 1988yf . Because there are certain services, such as child care and health care, whose benefits are difficult to measure or evaluate, governments and nonprofits are able to step in and meet this demand by promising that no investors will benefit by the cutting of corers or the delivery of unnecessary services. Furthermore, the nondistribution constraint helps to ensure wary donors and taxpayers that funds given to the organization for collective goods will be used as promised (Hansmann 1996yf $ O W K R X J K W K L V R S H Q V X S D Q G D O P R V W J X D U D Q W H H s certain markets for nonprofits and government establishments, it also means that many outputs of these organizations are difficult to measure. Compared with business organizations whose purpose is to earn profits for owners and which are accountable quarter by quarter, the performance of government and nonprofit organizations is less accurately measured, and monitoring is more sporadic (Rainey 1997, 75yf / D F N L Q J D V L Q J O H V W D N H K R O G H U J U R X S W R P R Q L W R U W K H R U J D Q L ] D W L R Q D Q G U L J R U R X V F U L W H U L a for evaluation, government and nonprofit organizations are more likely to embrace external referents of accountability to legitimate their operations. For the same reason, government and nonprofit organizations should be more susceptible to institutional pressures and more likely to be swayed by exposure to environmental pressures that promise an organization greater legitimacy. Because shareholders of business organizations constitute a dominant coalition that can more easily monitor the performance of managers, we hypothesize that these organizations will be less responsive to institutional influences than nonprofit and government organizations. Or, putting it another way, we expect that institutional forces 289 This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms 290 Journal of Public Administration Research and Theory will influence the behaviors of organizations whose outputs are less easily evaluated. Thus, our first hypothesis is: H1 The effects of institutional forces on organizational structures will be weaker in bus- iness firms than in government or nonprofit organizations. Differences in the Funding of Organizational Inputs Distinguishing organizations from different sectors in terms of oversight over their outputs is a first step. Answering the question of what distinguishes nonprofits from government organizations requires a second step, however, one that focuses attention on the control over organizational inputs or resources. By looking at how stakeholders make funding and allocation decisions about inputs into organizations, it is possible to sort out a core distinction between public and nonprofit organizations. Although both nonprofit and government organizations operate under the non- distribution constraint that bears on the use of profits or outputs, only nonprofits are exposed to a sea of stakeholders who are in a direct position to turn on or off the spigot of financial resources that flow into these organizations. Because many nonprofits offer services that are not provided by for-profits or that are too expensive to service targeted populations, finding patrons willing to subsidize charitable work is often critical. Many nonprofits receive contributions simultaneously from individuals, corporations, foundations, and government agencies at the city, state, and federal levels. Weaving these funding streams together is one of the most important challenges in nonprofit management, one that requires a combination of marketing, negotiation, and sales skills (Gronbjerg 1989yf $ W W L P H V W K H L Q W H U H V W V D Q d desires of funders can be in conflict with one another, and nonprofits must learn to satisfy the various demands of those who provide the charitable dollars they need to operate. Many donors make demands for recognition and put conditions on the use of their funds or limit support to a specified number of years. As the demands of donors on nonprofits have risen, many organizations have turned to earned income to supplement the contributions they receive. This stream of revenue is most often made up of fees for services that are charged to clients (Weisbrod 1998yf 8 V X D O O \ W K H V H I H H V D U H G L V F R X Q W H G E H O R Z P D U N H W U D W H D Q G L Q P D Q y cases they are based on the ability of the client to pay. From community health clinics that operate with a sliding scale of fees based on the client's income to sporting clubs that charge their members annual dues for the privilege of using their facilities, nonprofits that depend on earned income end up varying greatly in their missions and target clientele. This often creates substantial choice for clients and consumers-and pressure on nonprofits to satisfy the demands of those paying the organization's bills. The dependence of nonprofits on contributed and earned income links them to a number of powerful stakeholders. Donors and users do not have direct financial interests in nonprofit organizations, and neither group can be clearly identified per se as a key ownership group (Hansmann 1996yf + R Z H Y H U E H F D X V H R I W K H L U F O R V H F R Q W U R O R I W K H I L Q D Q F L D O U H V R X U F H V R r inputs needed by nonprofits to survive, both donors and users are positioned to influence nonprofit mission and activity. In any particular nonprofit, the relative strength of these two influence groups will depend on how an organization is funded. Donors often hold a close grip on nonprofit organizations that depend heavily on charitable contributions: when a small group of foundations or a single public sector agency is holding a nonprofit's purse strings, there is a significant though often unspoken power imbalance separating the two sides. In This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms Frumkin and Galaskiewicz Institutional Isomorphism and Public Sector Organizations nonprofit organizations that are largely driven by the service fees charged to clients, users, and members, the situation is somewhat different, owing to the fact that the number of stakeholders is likely to be quite large and the size of the stakes, relatively small. Still, if survival is to be ensured, nonprofits that depend on earned income must be responsive to those who use and pay for their services. Of course, there are many nonprofits that are simultaneously beholden to a group of influential contributors and also exposed to pressures from clients and members seeking improved services or lower fees. In most cases, the complex and contingent funding of nonprofits applies pressure on these organizations to be responsive and effective in their work. In public sector organizations, these funding pressures are substantially different. At the federal, state, and local levels, government collects revenue through both the charging of fees and the levying of taxes. Although government does have the power to coerce and punish those who violate the law, the ability of government to act is constrained by both elections and interest groups. The need to manage the political environment and to marshal support for spending programs tempers the power of government and ensures that real constraints exist to sudden shifts in the financial flows in government. A significant collective action problem also buffers government from sudden and decisive action by stakeholders with regard to financial inputs. This system works because citizens entrust government to provide certain collective goods and services in the name of the larger collectivity-and because citizens are willing to yield substantial control to the state in large measure because of the legitimacy of government. Although devolution and privatization have pushed many functions down to lower levels of government and out to nonprofit and private firms, the progressive hollowing of the state (Milward and Provan 1993yf K D V Q R W V R O Y H G D O O W K H S U R E O H P V W K D W S X E O L F V H F W R r agencies face. Research has shown that the environment around public organizations is fluid and capable of exerting influence over the action of public managers (Moore 1995yf . Although the flow of financial and human resources into government may be less contingent than that in nonprofit organizations, many public managers still devote important resources to controlling and shaping the political environment around their organizations. External supervision by government oversight institutions and authorities may also affect the ability of lower levels of government to carry out significant changes in policy and procedure. Ultimately, the ability of government to maintain a fair measure of control over the flow of inputs is a function of its ability not just to collect taxes and assess fees but also to win public approval so that decisions about the allocation of resources end up favoring one purpose or agency over another. The way nonprofit organizations differ from government organizations is thus a func- tion of the degree of control over inputs. Because investors and grant makers exercise control over firms and nonprofits, both types of organizations have to be especially responsive to those controlling financial resources. In government, the flow of financial inputs is harder for stakeholders to change and demands a much higher level of coordination and collective action than in nonprofits or business. The sources of government funding-taxpayers- exercise less direct control over government organizations, which creates accountability problems. Indeed, this is often one rationale that public administration theory offers for why public sector organizations are more bureaucratized. Thus, our second hypothesis is: H2 The effects of institutional forces on organizational structures and practices will be stronger in government organizations than in nonprofit organizations or business firms. 291 This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms 292 Journal of Public Administration Research and Theory Summary Attempts to define sectoral differences are not without controversy and contention, and some argue that the differences among organizations across the three sectors are disappearing, as nonprofits become more commercial and business firms take on social responsibilities and as government seeks closer community ties (Frumkin and Andre- Clark 2000; Weisbrod 1998yf 1 H Y H U W K H O H V V V W U X F W X U D O G L I I H U H Q F H V H P H U J H W K D W O H D G X V W o expect that (1yf L Q V W L W X W L R Q D O S U H V V X U H V Z L O O O H D V W L Q I O X H Q F H I R U S U R I L W V E H F D X V H R Z Q H U V D U e able to measure performance and control the flow of financial resources into business firms, (2yf J R Y H U Q P H Q W R U J D Q L ] D W L R Q V Z L O O E H P R V W L Q I O X H Q F H G E \ L Q V W L W X W L R Q D O S U H V V X U H s because outputs are very difficult to evaluate and the flow of resources is more shielded from sudden interruption by collective action problems, and (3yf Q R Q S U R I L W V Z L O O I D O l somewhere in the middle because their outputs are also difficult to evaluate but they must be responsive to their funders in order to survive financially. Building on distinctions in Wamsley and Zald (1973yf D Q G % R ] H P D Q \f frameworks, the relative positions of public, nonprofit, and business organizations can be roughly depicted across two axes (figure 1yf . In the analysis that follows, we test our two hypotheses. We are also interested in exploring whether organizations in public, nonprofit, and business sectors are particularly sensitive not just to institutional pressure broadly conceived but to specific kinds of institutional pressure. Here we examine three types: coercive, normative, and mimetic (DiMaggio and Powell [1983] 1991yf , W L V L P S R U W D Q W W R U H F R J Q L ] H W K D W W K H U H L V F R Q V L G H U D E O e variation in the possible forms that convergence or isomorphism can take. On the one hand, some institutional forces might make organizations more like their peers. For example, belonging to associations of similar organizations or paying attention to the practices of other organizations may make government agencies more like other government establishments, or nonprofits more like other nonprofits (DiMaggio and Powell [1983] 1991yf 2 Q W K H R W K H U K D Q G H [ S R V X U H W R R W K H U L Q V W L W X W L R Q D O S U H V V X U H V P L J K W P D N e organizations more similar to a more abstract ideal type of a bureaucratic organization. That is, exposure leads for-profits, nonprofits, and government establishments to embrace a generic bureaucratic form. In the former case, institutional pressures homogenize organizations within different categories. In the latter case, institutional pressures lead all organizations to look more like a single organizational form. A third alternative is that institutional forces may work to break down the boundaries across sectors and result in all organizations looking very similar to the statistical average. It remains an empirical question as to how these effects might play out. Because there is little theory to guide us, our analysis is somewhat exploratory when it comes to the differential effects of these forces on different kinds of organizations. Accordingly, after detailing our methods for our study, we proceed both to test our hypotheses and to search in a more exploratory mode for differential institutional effects across sectors. DATA AND MEASUREMENT OF VARIABLES We used the data collected in the 1991 National Organizations Study (NOSyf 7 K H S U L Q F L S D l advantage of the NOS data was that it had information on for-profit, nonprofit, and governmental establishments as well as data on institutional pressures on organizations. As described by Knoke and Kalleberg: This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms Frumkin and Galaskiewicz Institutional Isomorphism and Public Sector Organizations Figure 1 Measurement and Control in Public, Nonprofit, and Business Organizations Direct Business External control over Nonprofit inputs Indirect Government Imprecise Precise Ability to measure outputs [The NOS is] a multiplicity survey of establishments (single-site workplaces, including businesses, governmental, and nonprofit organizationsyf F R Q G X F W H G E \ W K H 6 X U Y H \ 5 H V H D U F h Laboratory (SRLyf D W W K H 8 Q L Y H U V L W \ R I , O O L Q R L V 5 H V S R Q G H Q W V L Q W K H * H Q H U D O 6 R F L D O 6 X U Y H y (GSSyf L G H Q W L I L H G W K H L U S O D F H V R I H P S O R \ P H Q W D Q G W K R V H R I W K H L U H P S O R \ H G V S R X V H V , Q D G H T X D W e information was obtained from 300 of the 1,427 GSS respondents who reported occupations. The SRL completed interviews with informants (typically personnel directorsyf I U R P R I W K e remaining 1,127 establishments, yielding a response rate of 64.5 percent. Because the 1991 GSS used area-cluster sampling, 39 organizations were duplicates: This left 688 unique establish- ments. (1994, 540; see also Kalleberg et al. 1996yf Knoke and Kalleberg (1994yf F D X W L R Q W K D W E H F D X V H H V W D E O L V K P H Q W V Z H U H L G H Q W L I L H G W K U R X J h a sample of individuals, this gives establishments of larger size a greater probability of being selected. To correct for this, all analyses were done weighting the NOS establishment sample by number of employees.1 Thus, all inferences are made to the size-weighted sample of establishments and not to a population of equally weighted establishments. l As noted by Kalleberg et al., "If the NOS sample is not weighted, statistics will describe the typical work settings to which employees in the U.S. labor force are exposed: each worker is given an equal weight.... In contrast, if observations in the NOS sample are weighted inversely proportional to the number of employees in an establishment, statistics will represent the population of U.S. establishments, as if each establishment has an equal probability of inclusion" (1996, 46yf 7 K H Z H L J K W L Q J S U R F H G X U H L V U H O D W L Y H O \ V W U D L J K W I R U Z D U G % H F D X V H R I W K H V D P S O L Q J G H V L J Q Z K L F h resulted in larger establishments being overrepresented in the sample, the observations in the NOS sample were weighted inversely proportional to the number of employees in an establishment. 293 This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms 294 Journal of Public Administration Research and Theory Dependent Variables This study focused on ways that organizations structured their activities. Hypotheses 1 and 2 were stated in very general terms. The focus was on the interaction of sector (public, for-profit, nonprofityf H [ S R V X U H W R L Q V W L W X W L R Q D O L Q I O X H Q F H V D Q G R U J D Q L ] D W L R Q D l structures. There were clearly many different organizational structures that we could have studied. The NOS collected particularly good data on organizational structures and practices related to human resource management. Of the various functions that we could have focused on, human resource management seems to be the one most generic to all three types of organizations. Because our focus was more on environmental influences and sector effects, we decided to identify three dependent variables and then test to see if the effects of institutional factors varied by sector across these three dependent variables. Because we were particularly interested in bureaucratic controls, we selected three variables that measure the extent of bureaucratization within the work establishment. We looked at whether hiring decisions were made at high or low levels within the organization, the extent to which rules governing hiring were written down or formalized, and the extent to which various functions in the organization were organized as departments. The NOS team measured the degree to which personnel decisions were centralized or decentralized. The SRL asked respondents: "We are interested in who is responsible for making different kinds of decisions in your workplace. I am going to read a list of decision areas and ask you to tell me who actually makes the final decision in each area. Is it the head of (establishmentyf V R P H R Q H E H O R Z W K D W > R U V R P H R Q H D W O D U J H U R U J D Q L ] D W L R Q L f establishment is part of a larger organizationyf @ " 7 K H G H F H Q W U D O L ] D W L R Q V F D O H Z D V W K H P H D n level at which the eight decisions were made in the establishment (decent [Kalleberg et al. 1996, 77]yf 3 The NOS team also constructed an index measuring the degree to which estab- lishments' hiring practices were formalized. To create an index of formalization, the SRL asked respondents, "Do each of the following documents exist at (establishment nameyf ? What about... ?"4 The formalization scale was the proportion of the seven documents present in an establishment (formal [Kalleberg et al. 1996, 75]yf . They also computed a measure of departmentalization. They asked respondents, "Is there a separate department or section responsible for (functionyf " 7 K H G H S D U W P H Q W D O - ization scale was the proportion of the eight departments present in an establishment (depart [Kalleberg et al. 1996, 74]yf ) R U H D F K R I W K H V H L W H P V W K H 1 2 6 W H D P L P S X W H G Y D O X H s where there were missing data. Descriptive statistics are in table 1. 2 The eight issues were performance evaluation, scheduling/overtime, which employees to hire, promotions, use of subcontractors/temporaries, discharges/layoffs, wage/salary levels, and number of employees. Respondents were able to respond: someone at larger organization (1yf H V W D E O L V K P H Q W K H D G D Q G V R P H R Q H D W O D U J H U R U J D Q L ] D W L R Q \f, establishment head (3yf H V W D E O L V K P H Q W K H D G D Q G V R P H R Q H E H O R Z \f, or someone below establishment head (5yf . 3 The variable names are taken from the NOS 1 data file and codebook. They are included only to inform readers who are familiar with the data file as to the exact variables we used. 4 The items were rules and procedures manual, documents on fringe benefits, written job descriptions, documents on safety and hygiene, written performance records, documents on hiring/firing procedures, and documents on personnel evaluation. 5 The functions were finance, personnel/labor relations, accounting, health and safety, public relations, research and development, long-range planning, and marketing or sales. This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms Frumkin and Galaskiewicz Institutional Isomorphism and Public Sector Organizations Table 1 Variable Means, Standard Deviations, Minimum and Maximum Values, and Number of Cases (Weighted by Number of Employees in the Establishment [N = 688]yf Variable Mean SD Minimum Maximum n Control Variables Number of employees in establishment 1.170 0.867 0 10.7 683 Branch of an organization 0.198 0.399 0 1 685 Union pressure scale 1.060 0.230 1 3 683 Products 0.073 0.261 0 1 685 Services 0.853 0.354 0 1 685 Competitive product/service markets 2.310 0.779 0 4 577 Competitive product/service markets dummy for missing cases 0.085 0.280 0 1 685 Sector Variables For-profit 0.893 0.310 0 1 685 Nonprofit 0.037 0.189 0 1 685 Government 0.070 0.256 0 1 685 Institutional Variables Subject to reviews/accreditation/licensing 0.343 0.475 0 1 683 Belongs to association of similar organizations 0.436 0.496 0 1 683 Pays attention to practices of other organizations 0.678 0.468 0 1 682 Organizational Structure Variables Decentralized decision-making scale 2.940 0.650 1 5 661 Proportion of (7yf G R F X P H Q W V I R U P D O L ] H G 5 Proportion of (8yf I X Q F W L R Q V G H S D U W P H Q W D O L ] H G 7 Control Variables Many factors could affect which organizational structures were adopted, and Kalleberg et al. (1996yf D U H Y H U \ X V H I X O L Q L G H Q W L I \ L Q J W K H V H I D F W R U V : H U H Y L H Z H G W K H L U I L Q G L Q J V D Q d identified seven variables that were consistently important in explaining variance across these three dependent variables. In all our analyses we controlled for the log of the number of employees (Insize [Kalleberg et al. 1996, 49]yf Z K H W K H U R U Q R W W K H H V W D E O L V K P H Q W Z D V D n independent company (i.e., the parentyf R U D E U D Q F K L H W K H F K L O G Y > . D O O H E H U J H W D O . 1996, 65]yf X Q L R Q S U H V V X U H X Q L R Q S > . D O O H E H U J H W D O @ \f, the level of competition in their main market or service area and a dummy for missing values on this variable (v328yf Z K H W K H U R U Q R W W K H H V W D E O L V K P H Q W R Q O \ S U R G X F H G D S U R G X F W S U R G X F W > . D O O H E H U J H W D O . 1996, 17]yf D Q G Z K H W K H U R U Q R W L W R Q O \ S U R Y L G H G V H U Y L F H V V H U Y L F H V > . D O O H E H U J H W D O , 48]yf 7 K H V H O D V W W Z R G X P P L H V Z H U H X V H G L Q F R Q W U D V W W R H V W D E O L V K P H Q W V W K D W S U R Y L G H G E R W h products and services. Sector The SRL asked respondents to identify the sector of their establishment. For-profit estab- lishments constituted 65.7 percent of the unweighted sample, public sector establishments 6 Nineteen percent of the cases had missing data for the level of competition in the main market or service area. To address this we used mean substitution (values ranged from 1 = none to 4 = a great dealyf D Q G F U H D W H G D G X P P y variable for those cases that had missing data, which we included in all our analyses. 295 This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms 296 Journal of Public Administration Research and Theory were 26.9 percent of the total, and nonprofit establishments were only 7.4 percent of the total. Weighting by number of employees, the percentages changed: 89.3 percent of the establishments were for-profit, 7.0 percent were public (pubnonpyf D Q G S H U F H Q W Z H U e nonprofit (privnonpyf 7 K L V L V E H F D X V H R I W K H I D F W W K D W W K H I R U S U R I L W I L U P V W H Q G H G R Q D Y H U D J e to be smaller than public or nonprofit organizations (Kalleberg et al. 1996, 47yf . Institutional Exposure The SRL asked respondents: "Is (establishmentyf V X E M H F W W R D S H U L R G L F U H Y L H Z E \ D Q R X W V L G e accreditation or licensing organization [iv315]? Does (establishmentyf E H O R Q J W R D n association of organizations like it [iv316]? In evaluating (establishment'syf S H U I R U P D Q F H , to what extent do you pay attention to the practices of other organizations like (establishmentyf > L Y @ " 7 K H I L U V W W Z R L W H P V Z H U H G L F K R W R P R X V D Q G Z H U H U H F R G H G Q R = 0, yes = 1yf D Q G W K H W K L U G K D G W K U H H U H V S R Q V H V W K D W Z H U H F R G H G Y H U \ O L W W O H V R P H Z K D W = 0, very much = 1yf 7 K H I L U V W L W H P Z D V R X U P H D V X U H R I F R H U F L Y H S U H V V X U H W K H V H F R Q G Z D V R X r measure of normative pressure, and the third was our measure of mimetic pressures. For cases that had missing data, the NOS team imputed values if the respondent answered the other items (Kalleberg et al. 1996, 61yf ' H V F U L S W L Y H V W D W L V W L F V I R U W K H V H Y D U L D E O H V E D V H G R n the weighted sample are in table 1. Our analysis builds on and extends that of Kalleberg et al. (1996yf L Q W Z R L P S R U W D Q t ways. We first break apart their measure of institutionalization and treat coercive, normative, and mimetic pressures separately. This allows us to examine whether different isomorphic pressures shape particular kinds of human resource practices. Beyond adding some clarity to the explanatory model, this decision to disassemble the earlier measure of institutionalization seems justified given the only modest level of internal consistency in the broader measure of institutionalization (Cronbach alpha = 0.56yf 7 K H V H F R Q G Z D \ R X r research is different from Kalleberg and associates' is that we are able to test research hypotheses about the interaction effect of sector on the link between isomorphic pressures and human resource practices. Societal sector was included in Kalleberg et al.'s study only as a control, and little theorizing about the effect of sector was attempted. We put sector at the forefront of the analysis. MODELS We now make specific predictions about which sectors should be more vulnerable to institutional effects and the direction that the effects should take. Our dependent variables are the three structural variables, and our model tries to explain differences in their values across organizations. As noted earlier, Kalleberg et al. (1996yf K D Y H P D G H F R Q V L G H U D E O e headway in explaining these differences. Our goal is now to see if exposure to institutional pressures can explain additional variance in these three dependent variables. Our hypothesis is that among for-profits, the three institutional variables should have little effect; among government organizations, the effects should be significantly stronger; and 7 In computing the weighted scores we used the weight variable in the NOS data file (which is the inverse of the number of establishment employeesyf D Q G W K H D Q D O \ W L F Z H L J K W V R S W L R Q L Q 6 7 $ 7 $ . 8 The fourth institutional variable, how much operations are regulated by government agencies, was not used because all government organizations are subject to such controls. This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms Frumkin and Galaskiewicz Institutional Isomorphism and Public Sector Organizations among nonprofits, the effects should be stronger than they are among for-profits but not as strong as they are among government organizations. We also want to examine the direction of the effects, but we do this in an exploratory analysis. As noted earlier, institutional forces might make organizations more like their peers, more similar to the ideal bureaucratic type, or more like a statistical average. For now it remains an empirical question as to how these effects might play out. All analyses included multiplicative interaction terms and took the following general form: Yj = aj + bjXl + bjX2 + bjX3 + bjZI + bjZ2 + bj(XI*ZIyf E M ; , = \f + bj(X2*ZIyf + bj(X2*Z2yf E M ; = O \f + bj(X3*Z2yf E M L : M L H j XI, X2, andX3 were dummy variables that measured the extent to which organizations were exposed to coercive, normative, and mimetic influences. Zl and Z2 were dummy variables indicating if the organization was government or nonprofit (for-profit organizations was the reference categoryyf : L L Q F O X G H G W K H F R Q W U R O Y D U L D E O H V W D N H Q I U R P W K H . D O O H E H U J H W D O \f analyses: the number of employees (loggedyf Z K H W K H U R U Q R W W K H H V W D E O L V K P H Q W Z D V D E U D Q F h of a parent organization, union pressure, the level of competition in markets, a dummy for missing data on this last item, and dummies for product-only and service-only establishments. We centered all our regressors, except for the dummies, so as to make our intercepts more intelligible. Yj indicated three dependent variables including decentralized decision making, formalization, and departmentalization. In all our models we tested for interaction effects. All the analyses used weighted least squares. To test H1 and H2, we will examine the absolute size of the regression coefficients associated with XI, X2, and X3. Thus, we hypothesize: (HIyf E O ) R U S U R I L W O > E O 1 3 2 , O E O * R Y W l (H'yf O E ) R U S U R I L W O @ E 1 3 2 , E L * R Y W l (Hiyf / E ) R U S U R I L W O E 1 3 2 E O L * R Y W l (H2yf O E ) R U S U R I L W O , E O 1 3 2 , , E * R Y W l (H2yf , E ) R U S U R I L W E O 1 3 2 , O E * R Y W l (H2yf O E ) R U S U R I L W O E O 1 3 2 , O E * R Y W O 9 This simply formalizes our hypotheses that the effects of coercive, normative, and mimetic influences (XI, X2, and X3yf Z L O O E H V W U R Q J H V W I R U J R Y H U Q P H Q W V D Q G Z H D N H V W I R U I R U S U R I L t business establishments and that nonprofits will be somewhere in-between. RESULTS We begin our discussion by reviewing the human resource practices of for-profit, nonprofit, and government establishments. In table 2 we present three one-way analyses of variance. We see that for all three of the dependent variables, differences across sectors are statistically significant at the 0.001 level. It is important to notice that in almost every case the mean value for the variable is comparable to the mean value for business establishments. This is not surprising because 89 percent of the weighted sample was 9 Note, bl,for-profit = for-profit businesses; bl,NPO = nonprofit organizations; and bl,Govt = governments. 297 This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms 298 Journal of Public Administration Research and Theory Table 2 One-Way Analysis of Variance in Human Resource Practices by Sector (N = 688yf Decentralized Formalization of Departmentalization of Decision Making Documents Functions Sector Mean SD Mean SD Mean SD For-profit 2.98 0.607 0.204 0.305 0.022 0.080 Nonprofit 3.07 0.453 0.473 0.257 0.131 0.222 Government 2.60 0.950 0.818 0.272 0.149 0.294 Total 2.94 0.650 0.257 0.342 0.035 0.122 F (p valueyf \f 99.3 (0.000yf \f for-profit. We also see that government establishments had the most extreme values. They tended to be more centralized, to be more formalized, and to have more departments than the other two types of establishments. These findings are not unexpected because re- searchers often associate greater bureaucracy with government organizations. Turning to table 3, we see that the control variables were exceptionally strong predictors in our models. In the three models, the control variables themselves (log size, branch facility, union pressure, products, services, market competition, and the market competition dummy for missing casesyf D F F R X Q W H G I R U S H U F H Q W S H U F H Q W D Q G 3 percent of the variance in our four models. Thus, the difference in R2 between the models with only the control variables and those with the substantively interesting variables and interactions ranged from 4.9 to 10.4 percent. Two control variables had strong effects in almost every model. Establishments that had more employees were more decentralized, were more formalized, and had more departments. Establishments that were branches of organizations headquartered elsewhere were more centralized, were more formalized, and had fewer departments. Establishments with a strong union presence were more centralized, and establishments that only produced a product or only provided a service (in contrast to establishments that did bothyf Z H U H P R U H F H Q W U D O L ] H G . The purpose of this article is to see if the sector of an organization is important in modifying the effect of the institutional environment on organizational behavior. We expected that, in general, government would be most vulnerable to institutional pressures; business firms, least vulnerable; and nonprofits to fall somewhere in between. We also were interested in the direction that the influence took. When institutional forces had an effect, establishments could either become more like their peers, more like the Weberian bureaucracy typified by the government establishment, or more like the statistical average. The results in table 3 are difficult to interpret, and so we provide table 4 as an aid. Table 4 has the same information as table 3, but we did the arithmetic to compute the values of "bl,For-profit," "bl,NPO," "bl,Govt," "b2,For-profit," "b2,NPO," "b2,Govt," "b3,For-profit," "b3,NPO," and "b3,Govt," and we reran the analyses to derive the proper standard error for each effect. The values in the parentheses are the differences between the regression coefficients for different combinations of organizational sectors, and the asterisks indicate the statistical significance. Our first research question was: are for-profits less vulnerable to institutional effects than nonprofits (H1yf " : H F R P S D U H G U R Z V D Q G L Q W D E O H ) R U W K L V Z H I L U V W O R R N H G D W H D F h of our three institutional variables separately. First, in two analyses (formalized documents and departmentalizationyf W K H H I I H F W V R I F R H U F L Y H L Q I O X H Q F H V Z H U H V W U R Q J H U I R U Q R Q S U R I L W s than for business establishments at the 0.05 level and at the 0.01 level; for decentralized This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms Table 3 Weighted Least Squares Predicting Three Human Resource Practices Decentralized Decision Making Formalized Documents Departmentalized Functions Dependent Variable (1yf \f (3yf \f (5yf \f (7yf \f (9yf Model b b b b b b Number of employees (lnyf . Branch -0.887**** -0.878**** -0.911** 0.404**** 0.404**** 0.418**** -0.038*** -0.033*** -0.036*** Union pressures -0.236*** -0.217*** -0.205** -0.006 -0.016 -0.020 0.004 -0.001 -0.005 Products -0.389**** -0.413**** -0.410**** -0.069* -0.058 -0.063 0.001 0.001 0.003 Services -0.194*** -0.212*** -0.195** -0.011 -0.000 -0.007 0.000 -0.000 0.000 Market competition -0.043 -0.040 -0.040 0.020* 0.019* 0.019* 0.010* 0.008 0.009 Market competition dummy for missing cases 0.041 -0.012 0.010 -0.104** -0.097* -0.106** 0.014 0.002 0.016 Government organization (Govtyf 0 Nonprofit organization (NPOyf 4 Subject to reviews/accreditation/ licensing -0.142*** -0.091** -0.109** 0.016 -0.009 -0.001 -0.009 -0.017 -0.015 Govt*Subject to reviews/ accreditation/licensing 0.604**** -0.376**** -0.187**** NPO*Subject to reviews/ accreditation/licensing -0.225 0.219** 0.147*** Association membership 0.117*** 0.063 0.109** 0.001 0.023 0.005 0.033*** 0.017* 0.031 Govt*Association membership 0.503**** -0.242**** 0.119*** NPO*Association membership 0.090 -0.029 0.250**** Pays attention to others 0.143*** 0.127*** 0.167**** 0.060*** 0.067**** 0.058*** 0.009 0.011 0.003 Govt*Pays attention to others -0.694**** 0.249*** 0.095** NPO*Pays attention to others -0.051 0.036 0.193** Constant 3.96**** 4.01** 3.98**** 0.543** 0.512**** 0.515** 0.154**** 0.142**** 0.145** R2 0.473** 0.464**** 0.468**** 0.654** 0.638** 0.636**** 0.267**** 0.267**** 0.229**** n 654 654 654 678 678 678 672 672 672 *p < 0.10; ** < 0.05; *** < 0.01; ****p < 0.001. w to 'a UJ sDThis content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms ?A Table 4 Summary of Interaction Effects in Table 3 (Change in Value of Dependent Variable Given Exposure to Institutional Environmentyf Decentralized Decision Making Formalized Documents Departmentalized Functions Subject to Subject to Subject to reviews/ Pays reviews/ Pays reviews/ Pays accreditation/ Association attention accreditation/ Association attention accreditation/ Association attention licensing membership to others licensing membership to others licensing membership to others Sector b b b b b b b b b For-profit (ForPyf 3 Nonprofit (NPOyf * (NPO-ForPyf \f (0.090yf \f (0.219**yf \f (0.036yf \f (0.250 ***yf \f Government (Govtyf * (Govt-ForPyf \f (0.503****yf \f (-0.376****yf \f (0.249***yf \f (0.119***yf \f (Govt-NPOyf \f (0.413*yf \f (-0.595****yf \f (0.213*yf \f (-0.131**yf \f < 0.10; ** < 0.05; *p < 0.01; ****p < 0.O1. This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms Frumkin and Galaskiewicz Institutional Isomorphism and Public Sector Organizations decision making, the difference was not significant. Second, in only one analysis (departmentalizationyf Z D V W K H U H D V W D W L V W L F D O O \ V L J Q L I L F D Q W G L I I H U H Q F H L Q W K H H I I H F W R f associational membership on organizational structure (p < 0.001yf 7 K L U G L Q R Q O \ R Q e analysis (departmentalizationyf Z D V W K H H I I H F W R I P L P H W L F L Q I O X H Q F H V V W U R Q J H U I R U Q R Q S U R I L W s than for business establishments (p < 0.001yf 7 K X V Z H I R X Q G V W D W L V W L F D O O \ V L J Q L I L F D Q t differences between nonprofits and for-profits in only four of the nine tests, therefore finding little support for the first half of H1. Our second research question was: are government establishments more vulnerable to institutional effects than for-profits (H1, H2yf " : H F R P S D U H G U R Z V D Q G L Q W D E O H 7 K e results are simpler to summarize. For all three dependent variables, the effects of coercive, normative, and mimetic influences were stronger for government establishments than for business establishments. Governmental agencies were more vulnerable to institutional influences than business establishments, thus amply supporting the second half of H1 and the first half of H2. Our third research question was: are government establishments more vulnerable to institutional effects than nonprofits (H2yf " : H F R P S D U H G U R Z V D Q G L Q W D E O H $ J D L Q Z e looked at each of our three institutional variables separately. The situation is not easy to summarize. First, we see that the effects of being subject to reviews/accreditation/licensing were in a different direction for nonprofits and governments. Thus, the effects in some absolute sense were not that different (the absolute values of the effects were similaryf E X t the effects were not similar in nature. Second, the pattern for the association membership variable was mixed. For all three, the signs were in the same direction, but the institutional effects in two out of the three analyses were not significantly different at the 0.05 level, and in the third (departmentalizationyf W K H H I I H F W Z D V V W U R Q J H U I R U Q R Q S U R I L W V W K D Q I R r government establishments. Third, we see a complicated pattern when looking at the mimicry effect as well. Only in one case was the effect significantly different at the 0.05 level and in the hypothesized direction (decentralizationyf L Q W K H R W K H U W Z R F D V H V W K H H I I H F W s were significantly different only at the 0.10 level. Either the effects of institutional variables among nonprofits were different than those among governmental agencies, but not in the hypothesized direction, or we found no significant differences between the two. Thus, the second half of H2 received little support. Further Issues Institutional forces have long been thought to lead to conformity. But "conformity to what"? Because institutional theory has been vague about the precise nature of iso- morphism, we took the opportunity to scrutinize this central question. The issue is whether institutional effects resulted in organizations looking like organizations similar to them- selves, to some ideal type (e.g., a Weberian bureaucracyyf R U W R D Q D Y H U D J H W K D W F X W V D F U R V s all types of establishments. Drawing on our analysis, we focus on the different ways institutional effects impacted for-profits, nonprofits, and government organizations. We can address these questions by again looking at table 4. Looking first at coercion, we see that being subject to reviews/accreditation/licensing seemed to make for-profits and nonprofits more bureaucratic, that is, more like government organizations. For-profits and nonprofits that were subject to these effects were more centralized at the 0.01 and 0.10 levels, respectively. Nonprofits that were subject to coercive influences also were more formalized and had more departments. In contrast, government organizations, which 301 This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms 302 Journal of Public Administration Research and Theory were subject to coercive influences, were more decentralized, less formalized, and less departmentalized than those that were not subject to such pressures. That is, they looked more like for-profit and nonprofit organizations. Thus, it appears that the coercive influences studied here did not make organizations more like their peers; nor did they lead them to become more bureaucratized. Rather, coercion made them look more the same, converging on a middle ground. Looking at normative pressures, we again see some convergence toward a middle ground, but the effect is not as dramatic. For-profits and nonprofits that were members of associations of similar organizations had more departments to handle their human service needs. The effects were significant at the 0.10 and 0.05 levels, respectively. However, neither nonprofits nor for-profits that belonged to associations tended to be more centralized or more formalized. In contrast, being exposed to normative institutional influences resulted in government agencies being more decentralized and less formalized. However, for some reason, these organizations also had more departments, which is a more bureaucratic pattern. Finally, we looked at mimetic effects. Here we found something different. Nonprofits that paid attention to other nonprofits were more departmentalized, a pattern converging on the bureaucratic model. For-profits that paid attention to others like themselves were more formalized, but they were also more decentralized. Finally, government establishments that paid attention to organizations similar to themselves-which we presume were other government establishments-were more centralized, more formalized, and more de- partmentalized. Mimesis resulted in government agencies becoming more like the prototypical bureaucratic governmental organization. Thus, in five of the six significant effects, paying close attention to similar organizations led to greater bureaucratic controls. CONCLUSION The results presented here have implications for both theory and practice. The significance of our findings for organization and public administration theory is threefold. First, as predicted, institutional pressures do not affect all organizations the same. Hypothesis 1 argued that institutional pressure would be weaker for for-profits than for nonprofits and government establishments because the former have owners who monitor performance and assert claims over residual earnings whereas the latter do not. In contrast, the latter would be vulnerable to institutional or legitimizing influences because of their lack of accountability for performance indicators such as sales and profits. This is the reason cited by many in the public administration literature for why government establishments tended more to bureaucratic structures. However, if exposed to institutional forces, these organizations seemed to adapt and change, seeking legitimacy by conforming to insti- tutional pressures. We found that governmental organizations were much more vulnerable to institutional forces than for-profits. However, the differences between nonprofits and for-profits were not as great. Hypothesis 2 argued that for-profits, nonprofits, and government establishments have different levels of external control over inputs. Nonprofits, like for-profits, have funders that are in a position to cut off funding either by denying grants or disinvesting. Government agencies are administered by managers who have considerably more protection from input providers or taxpayers, but with this separation come accountability problems. Contrary to what we expected, we did not find a clear pattern of differences This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms Frumkin and Galaskiewicz Institutional Isomorphism and Public Sector Organizations between nonprofits and government establishments. Either the differences were only marginally different (i.e., several effects were significant at the 0.10 levelyf W K H L Q V W L W X W L R Q D l effect for nonprofits was stronger, or the effects were in the opposite direction. These two sets of findings highlight that government and for-profits are quite different in their vulnerability to institutional pressures but that the differences between nonprofits and the other two sectors are not that clear-cut. Second, our results suggest that institutional theorists rethink what they mean by "conformity" and the "norm." In DiMaggio and Powell's ([1983] 1991yf F O D V V L F D U W L F O H W K e "norm" was the bureaucratic organization of Max Weber, which epitomized the rationality agenda. Yet even in their discussion it soon becomes apparent that the bureaucratic form is not the only form that could spread, via institutional processes, across actors in an organizational field. Any number of experiments, innovations, technologies, and man- agement styles could be driven by exposure to institutional influences (see, e.g., Davis and Greve 1997; Galaskiewicz and Wasserman 1989; Haunschild and Beckman 1998; Haunschild and Miner 1997yf . In our analysis, it appears that there was no clear norm and that different institutional pressures were having different effects. Institutional forces-particularly coercive and, to a lesser degree, normative forces-seemed to be working to break down the differences among for-profit, nonprofit, and governmental agencies, resulting in the adoption of hybrid structural patterns of control. Exposure to these types of institutional influences homog- enized organizational structures across the entire organizational population. In contrast, paying attention to others like oneself (mimetic influencesyf V H H P H G W R P D N H R U J D Q L ] D W L R Q s look more like the classic bureaucratic model-more centralized and more formalized. These findings were illustrated nicely in the case of government establishments. Looking at the simple descriptive statistics in table 2, we learned that government agencies were more bureaucratized than for-profits and nonprofits. Decision-making structures were more centralized, rules were formalized, and there were more functional departments. Yet our analysis discovered that exposure to coercive as well as normative influences made government organizations less like traditional bureaucracies and more like for-profits and nonprofits. In contrast, exposure to mimetic pressures made government establishments more like traditional bureaucracies. One way to understand these finding is that when public managers are subject to external oversight or when they are embedded in professional networks, they are liberated to some extent from the internal pressures they face on a daily basis to bureaucratize. Our results suggest that public sector managers respond to external oversight and professional norms by finding ways to capitalize on the presence of these external pressures to move their organizations away from hierarchy and formalization and toward the mean of all organizations regardless of sector. We therefore believe that one possible interpretation of these results is that the presence of external influences on public sector organizations may rival internal controls within the agencies and open the door for more liberating structural transformations. This is an interesting contrast to the prediction that bureaucratic theory would usually make. Finally, we believe that these results have practical implications. The vulnerability of public sector organizations to institutional pressures is significant for two reasons. The first relates to government's role as the principal funder of the nonprofit human services. With increasing amounts of nonprofit agency finance flowing from public sector agencies of all kinds, the vulnerability of government agencies to institutional pressure may have clear 303 This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms 304 Journal of Public Administration Research and Theory spillover effects on nonprofits that could shape the scope and character of contracting relations. One effect might well be a narrowing of the range of programmatic innovations that government agencies are willing to consider under conditions of uncertainty. This could in the long run have serious implications for the diversity of nonprofit initiatives and might well put a damper on the drive for innovation that nonprofits often bring to their work. The second reason that our findings are significant is connected to government's traditional role as a regulator of business through a myriad of laws and statutes in areas as diverse as antitrust, environmental protection, and consumer rights. Because government takes an active role in regulating business activity, the vulnerability of government agencies to institutional pressures is significant in that it may well impose externalities on business firms, especially if regulations rest more on institutional considerations than on rational requirements for increased oversight. However, to the extent that institutional pressures lead toward broad convergence toward a universal model, some of these issues may be minimized. In thinking about the effect of institutional pressures, it is useful to recall that many of the recent efforts to "reinvent government" have aimed to make government organizations more responsive to client groups. How to achieve this shift has been the subject of considerable debate. It is tempting to view both the coercive and the normative influences discovered in this article as evidence that exposure to some institutional pressures may further the cause of "reinvention." Belonging to professional associations and being subject to periodic reviews, accreditation, and licensing appear to make government agencies less rigid and more flexible. Whether or not this translates into being more responsive to the general public remains a far different and more complex question that public administration research, not institutional theory, will need to resolve. REFERENCES Bozeman, Barry. 1987. All organizations are public: Bridging public and private organizational theories. San Francisco, CA: Jossey-Bass. Brint, Steven, and Jerome Karabel. 1991. Institutional origins and transformations: The case of American community colleges. In The new institutionalism in organizational analysis, ed. Walter W. Powell and Paul J. DiMaggio, 337-60. Chicago: University of Chicago Press. Christensen, Sorren, and Jan Molin. 1995. Origin and transformation of organizations: Institutional analysis of the Danish Red Cross. In The institutional construction of organizations: International and longitudinal studies, ed. W. Richard Scott and Sorren Christensen, 67-90. Thousand Oaks, CA: Sage. Dahl, Robert A., and Charles E. Lindblom. 1953. Politics, economics, and welfare. New York: Harper Collins. D'Aunno, T., R. I. Sutton, and R. H. Price. 1991. Isomorphism and external support in conflicting institutional environments: A study of drug abuse treatment units. Academy of Management Journal 14:636-61. Davis, Gerald F., and Henrich R. Greve. 1997. Corporate elite networks and governance changes in the 1980s. American Journal of Sociology 103:1-37. DiMaggio, Paul J. 1991. Constructing an organizational field as a professional project: US art museums, 1920-1940. In The new institutionalism in organizational analysis, ed. Walter W. Powell and Paul J. DiMaggio, 267-92. Chicago: University of Chicago Press. DiMaggio, Paul J., and Walter W. Powell. [1983] 1991. The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. In The new institutionalism in organizational analysis, ed. Walter W. Powell and Paul J. DiMaggio, 63-82. Chicago: University of Chicago Press. This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms Frumkin and Galaskiewicz Institutional Isomorphism and Public Sector Organizations Dobbin, Frank. 1995. The origins of economic principles: Railway entrepreneurs and public policy in 19th century America. In The institutional construction of organizations: International and longitudinal studies, ed. W. Richard Scott and Sorren Christensen, 277-302. Thousand Oaks, CA: Sage. Fligstein, Neil. 1990. The transformation of corporate control. Cambridge, MA: Harvard University Press. . 1991. The structural transformation of American industry: An institutional account of the causes of diversification in the largest firms, 1919-1979. In The new institutionalism in organizational analysis, ed. Walter W. Powell and Paul J. DiMaggio, 311-36. Chicago: University of Chicago Press. Friedland, Roger, and Robert R. Alford. 1991. Bringing society back in: Symbols, practices and institutional contradictions. In The new institutionalism in organizational analysis, ed. Walter W. Powell and Paul J. DiMaggio, 232-63. Chicago: University of Chicago Press. Frumkin, Peter, and Alice Andre-Clark. 2000. When missions, markets, and politics collide: Values and strategy in the nonprofit human services. Nonprofit and Voluntary Sector Quarterly 39 (1yf . Galaskiewicz, Joseph, and Stanley Wasserman. 1989. Mimetic processes within an interorganizational field: An empirical test. Administrative Science Quarterly 39:654-78. Gormley, William T., and David L. Weimer. 1999. Organizational report cards. Cambridge, MA: Harvard University Press. Gronbjerg, Kirsten. 1989. Understanding nonprofit funding. San Francisco, CA: Jossey-Bass. Halliday, Terence C., Michael J. Powell, and Mark G. Granfors. 1993. After minimalism: Transformation of state bar associations from market dependence to state reliance, 1918 to 1950. American Sociological Review 58:515-35. Hannan, Michael T., and John Freeman. 1989. Organizational ecology. Cambridge, MA: Harvard University Press. Hansmann, Henry. 1980. The role of nonprofit enterprise. Yale Law Journal 89:835-901. . 1996. The ownership of enterprise. Cambridge, MA: Harvard University Press. Haunschild, Pamela R., and Christine M. Beckman. 1998. When do interlocks matter?: Alternate sources of information and interlock influence. Administrative Science Quarterly 43:815-76. Haunschild, Pamela R., and Anne S. Miner. 1997. Modes of interorganizational imitation: The effects of outcome salience and uncertainty. Administrative Science Quarterly 42:472-500. Holm, Peter. 1995. The dynamics of institutionalization: Transformation processes in Norwegian fisheries. Administrative Sciences Quarterly 40 (3yf . Kalleberg, Are L., David Knoke, Peter V. Marsden, and Joe I. Spaeth. 1996. Organizations in America. Thousand Oaks, CA: Sage. Knoke, David, and Ame L. Kalleberg. 1994. Job training in the U.S. organizations. American Sociological Review 59:537-46. Kraatz, Matthew S. 1998. Learning by association? Interorganizational networks and adaptions to environmental change. Academy of Management Journal 41 (6yf . Lindblom, Charles E. 1965. The intelligence of democracy. New York: Free Press. March, J. G., and Herbert A. Simon. 1958. Organizations. New York: Wiley. Meyer, John W., and Brian Rowan. [1977] 1991. Institutionalized organizations: Formal structure as myth and ceremony. In The new institutionalism in organizational analysis, ed. Walter W. Powell and Paul J. DiMaggio, 143-64. Chicago: University of Chicago Press. Meyer, John W., W. Richard Scott, and David Strang. 1987. Centralization, fragmentation, and school district complexity. Administrative Sciences Quarterly 32:186-201. Meyer, John W., W. Richard Scott, David Strang, and Andrew Creighton. 1988. Bureaucratization without centralization: Changes in the organizational system of American public education: 1940-1980. In Institutionalpatterns and organizations, ed. Lynne G. Zucker, 139-67. Cambridge, MA: Ballinger. Milward, H. Brinton, and Keith G. Provan. 1993. The hollow state: Private provision of public services. In Public policy for democracy, ed. Helen Ingram and Steven Rathgeb Smith, 222-40. Washington, DC: Brookings Institution. Moore, Mark H. 1995. Creating public value. Cambridge, MA: Harvard University Press. Murray, H. A. 1975. Comparing public and private management: An exploratory essay. Public Administration Review 35:364-71. 305 This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms 306 Journal of Public Administration Research and Theory Perry, James L., and K. L. Kraemer. 1983. Public management. Mountain View, CA: Mayfield. Pfeffer, Jeffrey, and Gerald Salancik. 1978. The external control of organizations. New York: Harper and Row. Phalen, Marilyn E. 1985. Nonprofit enterprises: Law and taxation, vol. 1. Wilmette, IL: Callaghan and Co. Pitt, D. C., and B. C. Smith. 1981. Government departments: An organizational perspective. London: Routledge. Powell, Walter W. 1998. Learning from collaboration: Knowledge and networks in the biotechnology and pharmaceutical industries. California Management Review 40 (3yf . Pressman, Jeffrey L., and Aaron Wildavsky. 1973. Implementation. Berkeley: University of California Press. Rainey, Hal G. 1989. Public management: Recent research on the political context and managerial roles, structures, and behaviors. Yearly Review ofManagement of the Journal of Management 15:229-50. .1997. Understanding and managing public organizations. San Francisco, CA: Jossey-Bass. Rainey, Hal G., R. W. Backoff, and C. L. Levine. 1976. Comparing public and private organizations. Public Administration Review 36:233-46. Scott, W. Richard. 1991. Unpacking institutional arguments. In The new institutionalism in organizational analysis, ed. Walter W. Powell and Paul J. DiMaggio, 164-82. Chicago: University of Chicago Press. . 1994a. Institutional analysis: Variance and process theory approaches. In Institutional environments and organizations: Structural complexity and individualism, W. Richard Scott, John W. Meyer, et al., 81-99. Thousand Oaks, CA: Sage. .1994b. Institutions and organizations: Toward a theoretical synthesis. In Institutional environments and organizations: Structural complexity and individualism, W. Richard Scott, John W. Meyer, et al., 55-80. Thousand Oaks, CA: Sage. .1995. Introduction: Institutional theory and organizations. In The institutional construction of organizations: International and longitudinal studies, ed. W. Richard Scott and Sorren Christensen, xi-xxiii. Thousand Oaks, CA: Sage. Simon, Herbert A. 1946. The proverbs of administration. Public Administration Review 6:53-67. 1948. Administrative behavior. New York: Free Press. 1995. Organizations and markets. Journal of Public Administration Research and Theory 5:273- 94. Simon, Herbert A., Donald W. Smithburg, and Victor A. Thompson. [1950] 1991. Public administration. New Brunswick, NJ: Transaction Publishers. Suchman, Mark. 1995. Localism and globalism in institutional analysis: The emergence of contractual norms in venture finance. In The institutional construction of organizations: International and longitudinal studies, ed. W. Richard Scott and Sorren Christensen, 39-66. Thousand Oaks, CA: Sage. Tolbert, Pamela S. 1988. Institutional sources of organizational culture in major law firms. In Institutional patterns and organizations, Lynne G. Zucker, 101-13. Cambridge, MA: Ballinger. Tolbert, Pamela S., and Lynne G. Zucker. 1983. Institutional sources of change in the formal structure of organizations: The diffusion of civil service reform, 1880-1935. Administrative Science Quarterly 28:22-39. Useem, Michael. 1993. Executive defense: Shareholder power and corporate reorganization. Cambridge, MA: Harvard University Press. Wamsley, Gary L., and M. N. Zald. 1973. The political economy ofpublic organizations. Lexington, MA: Heath. Weisbrod, Burton A. 1988. The nonprofit economy. Cambridge, MA: Harvard University Press. . 1998. To profit or not-to-profit: The commercial transformation of the nonprofit sector. New York: Cambridge University Press. Wildavsky, Aaron. 1964. The politics of the budgetary process. Boston: Little, Brown. Williamson, Oliver E. 1981. The economics of organizations: The transaction cost approach. American Journal of Sociology 87:548-77. . 1985. The economic institutions of capitalism. New York: Free Press. Wilson, James Q. 1989. Bureaucracy. New York: Free Press. This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms Frumkin and Galaskiewicz Institutional Isomorphism and Public Sector Organizations 307 Zucker, Lynne G. 1988. Where do institutional patters come from? In Institutional patterns and organizations, ed. Lynee G. Zucker, 3-21. Cambridge, MA: Ballinger. . [1977] 1991. The role of institutionalization in cultural persistence: Postscript: Micro-foundations of institutional thought. In The new institutionalism in organizational analysis, ed. Walter W. Powell and Paul J. DiMaggio, 83-107. Chicago: University of Chicago Press. This content downloaded from 128.6.45.205 on Thu, 26 Aug 2021 00:30:58 UTC All use subject to https://about.jstor.org/terms