****For C. Owens Only****

American Federalism and the Search for Models of Management 671

Robert Agranoff Indiana University–Bloomington

Michael McGuire University of North Texas

American Federalism and the Search for Models of Management

Changes in the United States federal system mean that managers must operate by taking into account multiple interacting governments and nongovernmental organizations; dealing with nu- merous programs emanating from Washington and state capitols; and engaging in multiple inter- governmental transactions with an expanding number of intergovernmental instruments. Four models of management within this changing system are identified. The top-down model emphasizes ex- ecutive-branch control and is embedded in enforcement and exchange related to the laws, regu- lations, funding rules, program standards, and guidelines associated with federal/state grant, procurement, and regulation programs. The donor-recipient model emphasizes mutual depen- dence or shared program administration, where two-party bargaining or reciprocal interactions among government officials is the norm. The jurisdiction-based model is defined by the initiated actions of local officials and managers who seek out program adjustments and other actors and resources to serve the strategic aims of their governments. The network model highlights the ac- tions of multiple interdependent government and nongovernmental organizations pursuing joint action and intergovernmental adjustment. Although the first two models are long-standing and the latter two are emergent, all appear to be alive and well on the intergovernmental scene, posing complex challenges for public managers.

Public administration and the processes of federalism have merged to a nearly indistinguishable point. Since the “cooperative federalism” of the 1930s, managing within the federal system has become an increasingly more im- portant activity. However, federalism is not static. As policy responsibilities between the national and subnational gov- ernments have evolved and devolved, governing authority has overlapped across levels to a point where all actors are involved simultaneously to varying degrees (Wright 1988). Attention must be given to operations in such a system.

Managing across governments and across organizations within the complex and continuously changing processes of federalism deserves the attention of both scholars and practitioners. Such activity has become the very heart of public administration and management. The continuing growth of federal grants and new regulatory programs, in- creased federal–state programming, the continuation of some federal–local programs, federal initiatives to nongov-

ernmental organizations, and expanded roles for state gov- ernment have changed the context of public administra- tion from single-organization operations to boundary-span- ning operations (Agranoff and McGuire 1998a). Not only do local public managers now operate within their home agency and jurisdiction, they also perform numerous iden-

Robert Agranoff is an associate dean and professor in the School of Public and Environmental Affairs at Indiana University–Bloomington, where he spe- cializes in federalism, intergovernmental relations and management, and community and economic development. His writings include Dimensions of Human Services Integration, Intergovernmental Management: Human Ser- vices Problem Solving in Six Metropolitan Areas, and New Governance for Rural America: Creating Intergovernmental Partnerships, as well as many journal contributions. Since 1990, he has been affiliated with the Instituto Universitario Ortega y Gasset in Madrid. Email: [email protected].

Michael McGuire is an associate professor of public administration at the University of North Texas. His research on interorganizational networks, col- laborative public management, and rural development has been published in many journals, including Public Administration Review, Journal of Public Administration Research and Theory, State and Local Government Review, Publius: The Journal of Federalism, Policy Studies Review, and Economic Development Quarterly. Email: [email protected].

672 Public Administration Review • November/December 2001, Vol. 61, No. 6

tifiable activities within the vertical realm, which includes the state and federal governments, and also horizontal ac- tivities, which involve other local governments and many nongovernmental organizations (Agranoff and McGuire 1998b; Jennings and Krane 1994; Mandell 1990; Wright and Krane 1998). These forces have put a premium on collaborative actions and transactions across governmen- tal boundaries.

While the resilience of federalism as a form of gover- nance is undeniable—its shape and operation has caused and been caused by changing social, economic, and po- litical trends (Watts 1996)—the search for appropriate management models within the changing processes of fed- eralism remains a difficult task. To help focus that search, this article explores models of management. Two vener- able models, top-down and donor-recipient, and two emer- gent models, jurisdiction-based and network, are pre- sented. Each model’s prevalence and applicability to twenty-first-century federalism are examined. While they are adaptable to explanation in a number of policy are- nas, the emerging models are confirmed by our empirical study of 237 city governments and the intergovernmen- tal and collaborative activity of government officials pro- moting economic development. The primary concern of this paper is to contribute to the understanding of public- management approaches by demonstrating how emergent models exist alongside more traditional models as a re- sult of shifts in federalism.

The term “model” employed here follows Kaplan’s (1964, 266–7) usage: It is a “scientific metaphor” that di- rects attention to certain resemblances between theoreti- cal entities and the real subject-matter; one type of system can be shown to be a consistent interpretation of another. Our search here is not for characteristic metaphors of fed- eralism itself (Wright 1988), but how policy making and management can vary within federalism across time and policy realms. Thus, each model is not only described, its temporal and policy-specific relevance is also analyzed.

Concern for management models is hardly new. Elazar (1964, 248) suggests that, from the founding period, inter- governmental cooperation was necessary, and methods of providing for collaboration among the various parts of the federal system were sought out continually. Similarly, Grodzins (1966) equates administrative practices in feder- alism with shared functions, and Leach (1970) identifies the management of grants programs as involving joint ac- tion and manpower from all levels of government. Even during Nixon’s New Federalism attempt to streamline the intergovernmental system, Walker (1974, 30) claimed that managing within federalism was still in “a state of consid- erable confusion.” Since the expansion of federal program- ming in the 1960s and the consequent administrative com- plexity, the term intergovernmental management (IGM)

entered the public management lexicon somewhat ambigu- ously as a way to describe the activities of management across governments and as a means for “insiders” to cap- ture the essence of the routines that administrators must follow within federalism. IGM has been characterized as a term of “recent vintage, specialized usage, limited visibil- ity and uncertain maturity” (Wright and Krane 1998) be- cause it includes so many disparate actions: seeking funds from other governments, coping with an expanding regu- latory burden, and developing interlocal cooperative agree- ments, as well as many interlocal collaborative activities. We concur with this observation, and, although our em- pirical studies of economic development directly involve many degrees of intergovernmental management, we em- ploy a more general, more careful, and potentially more acceptable approach by merely discussing models of man- aging within federalism, only some of which occur exclu- sively between governments.

Four Models of Managing within Federalism

Top-Down Model The top-down model is predicated on the growth of na-

tional programming and tipping the balance within the fed- eral system toward executive control, with the federal gov- ernment somehow “managing” its programs through state and local governments’ managers. For example, as a part of administrating federal job training, economic develop- ment, and environmental programs, many large city gov- ernments have a mayor/manager’s level of compliance/ oversight function. Their role is to monitor intergovern- mental programs, ensure that federal money is spent prop- erly, and see to it that rules, regulations, and standards are followed. Intellectually, this model goes back to Leonard D. White’s (1939, 144) classic textbook on public admin- istration. He refers to the importance of central adminis- tration in national–state relations because of the “gradual acceptance of the idea that in certain fields a national mini- mum standard is in the public interest, and by the inter- vention and application of nation-wide procedures, largely administrative in nature, to secure such standards.” The expansion of national programming came as a result of congressional frustration with state governments’ inaction in solving domestic problems of national scope after World War II. Along with national legislation and funding came rules and standards intended to direct state (and local) ac- tion (Walker 2000).

The governmental heritage of top-down management flows from attempts by the executive office of the Presi- dent and the Brownlow Committee in the 1930s and the Kestnbaum Commission on Intergovernmental Relations

American Federalism and the Search for Models of Management 673

of the 1950s to forge a federally led “executive-centered intergovernmental management.” Stever (1993, 76–7) char- acterizes this approach as operating the federal system from the perspective of the president, field staff loyal to the presi- dent, and use of the Bureau of the Budget (later the Office of Management and Budget) capacities to manage inter- governmental programs. The approach is based on two normative premises: that the federal system must be con- sidered as a single system, and that de facto interdepen- dence of the federal system mandates the application of executive-centered logic (Sundquist and Davis 1969). The model represents the bureaucratic resolution to the basic dilemma of “how to achieve goals and objectives that are established by the national government, through the ac- tions of other governments, state and local, that are legally independent and politically may even be hostile” (Sundquist and Davis 1969, 12). When intergovernmental programs appear to lack an overall intergovernmental strat- egy—evidenced by confusing legislative enactments, over- lapping program authority, uncertain responsibility, and duplication of effort—state and local officials may com- plain about these conditions, but the “easiest” solution is for the federal government to manage by enforcing or re- quiring coordination. Toward that end, mandated inter- agency linkages, comprehensive plans, joint boards and councils, special planning areas, and quasi-governmental planning units are common requirements associated with grant legislation.

One aspect of top-down management is local compli- ance. Coordinating managerial activity, with higher gov- ernments or with interlocal partners, is difficult and often not desirable. Fierce competition for limited resources among localities results in a kind of impulsive grantsman- ship: managers look to the federal government for assis- tance with little regard for how, or even if, such funds should be spent locally. More demonstrative of top-down manag- ing, however, is the impact of regulatory programs or “gov- ernment by remote control.” The pace of administrative rule making and of new regulatory and pre-emptive enact- ments has increased over the past few decades (Conlan 1991). Regulation led Joseph Zimmerman (1992, xi) to conclude that federalism has shifted from cooperative models to what he calls “the accretion of political power in the United States at the national level.” Moreover, this increase in control has not become a “zero-sum” game, with a corresponding loss by the state governments; states also have increased their latent powers and increased their control over local governments. This situation has led state and local officials (managers in particular) into a compli- ance-management mode, usually in situations where no stream of funding accompanies national mandates, with important internal budgetary and other resource implica- tions. In such a bureaucratic environment, the “local-man-

ager-as-agent” perspective emphasizes the risks of discre- tion and the importance of compliance with external regu- lations and grant requirements.

Thus, the top-down model has a great deal of temporal relevance. While the momentum of program growth may have subsided since the 1980s, new programs continue to emerge as a reflection of national intent, and manag- ers remain in a position where they must deal with na- tional and state aims. For public managers, the top-down model is a clear manifestation of the bureaucratic tradi- tion of the twentieth century. Whether the manager fol- lows the federal or state legal template to the letter, pro- grams are guided in a normative framework that implies substantial (if not essential) compliance. Likewise, ac- tions are subject to reporting, review, evaluation, and au- dit in an administrative manner. From our field studies and many other studies (Church and Nakamura 1993; Kettl 1987; Williams 1981), it is clear that considerable managerial time is spent on transactions relating to meet- ing program intent and expectation—that is, trying to fit programs into standards, rules, and regulations. Local governments as creatures of their state governments are heavily supervised by the states (except in the case of substantial home-rule situations) and operate largely within an extensive top-down web of expectations. Most observers would agree that, along with conformity, fed- eral standards have contributed to a more professional and accountable public administration at the subnational level than before the onset of national programming. While it is clearly not the only way to manage within federalism, it is a time-honored activity that is expected to continue.

Donor-Recipient Model The major limitation of the top-down model of manag-

ing federalism is its false presupposition that some actor possesses the necessary information, the requisite exper- tise, and the political skill to singularly steer courses of action and to deliver (command) policy outputs that are consistent with the multiplicity of societal interests. An alternative management model is based on mutually de- pendent actors in the intergovernmental system rather than complete federal control and influence. For example, a growing number of cities negotiate the workload and fi- nancial burden that federal environmental regulations im- pose. In the case of drinking water, for instance, they often try to get the U.S. Environmental Protection Agency to remove requirements for costly testing of the federal agency’s entire list of chemicals, some of which are highly unlikely to affect the city or its outlying water source. Moreover, cities are likely to propose that they develop their own set of water-quality standards, in priority order, with the most important local risks at the top of the list.

674 Public Administration Review • November/December 2001, Vol. 61, No. 6

Partly as a result of governing problems in domestic policy, such as inadequate coordination and control, and partly as a result of a worldwide trend toward decentraliza- tion, many initiatives over the past decade have been de- signed to balance or realign autonomy and influence within intergovernmental relations (Kincaid 1998). The donor-re- cipient model reflects the administrative reality of such changes by describing a form of management that recog- nizes policy is not easily or even hierarchically executed. Actors at the other end of the intergovernmental chain are involved in programming as well (Elmore 1985), creating what is known as a “bottom-up” effect. If the critical com- ponent of policy success in the top-down model is that local actors carry out national goals, policies are considered suc- cessful in the donor-recipient model when they leave room for local decision making and provide local actors with suf- ficient policy discretion, resources, and autonomy to carry out national goals while fulfilling local needs.

The empirical basis of this model comes largely from policy-implementation studies that take a view from the “bottom.” Pressman and Wildavsky (1984) were among the first to understand policy implementation as a process of interaction or shared administration that occurs at the intersection of setting goals and carrying out actions geared to achieve them. The utility of viewing intergovernmental relations in terms of a donor (the actor providing resources) and a recipient was first identified in Jeffrey Pressman’s Federal Programs and City Politics (1975). Examining federal aid at the local level, Pressman concludes that any federal partner depends on the actions of others to achieve its own objectives; the recipient may need money, but do- nors need fundable applications and implementation abil- ity at the local level. The donor does not have the time, resources, inclination, and, in many cases, authority to in- tervene on a regular basis in the recipient’s locality. Al- though donors acknowledge their dependence on the ac- tions of the recipient, thus leading to cooperation, conflict is also a common aspect of managing donor-recipient re- lations. Pressman argues, “Donor and recipient need each other, but neither has the ability to control fully the actions of the other. Thus, the aid process takes the form of bar- gaining between partly cooperative, partly antagonistic, and mutually dependent sets of actors” (1975, 106–7). Other studies demonstrate that grant programs are not necessar- ily instruments of federal control, but opportunities to bar- gain (Ingram 1977). While federal agents would like to bind subnational jurisdictions to federal policy, subnational agencies seek the maximum possible leeway to pursue their own goals and objectives with federal help. Donor-recipi- ent managing can involve several different approaches: compromise, participation in a problem-solving discussion, unilateral initiative, or cooperation with a third party who is trying to solve the controversy (Liebschutz 1991).

According to the donor-recipient model, recipient man- agers navigate between top-down legal compliance or sub- stantial conformity and some desired change at the bot- tom, in a direction that is desirable to the jurisdiction. Some actions provide adequate resources and discretion for the jurisdiction, some actions do not, but, like the top-down model, all action must ultimately be consistent with the donor’s goals. Recipient managers might ignore national standards, either by inaction or open defiance, placing the jurisdiction at risk of penalties such as withdrawal of funds, large financial penalties, or even protracted litigation. Other alternative means of navigation include engaging the do- nor in a forum where adjustments can be made or using the recipient’s jurisdictional resources to counter the donor’s attempt at conformity.

The donor-recipient model is temporally relevant because its effects compound over time as new programs are en- acted. While controls are exerted and subnational govern- ments can be controlled, this is not always the case. Be- cause the federal system is not centralized, implementing governments have their political and day-to-day program control weapons to counteract law and money, should they choose to take action. There is also a learning-curve effect at work. When subnational officials achieve negotiating suc- cess in one program arena, the experience can be carried over into others. Moreover, federal officials need program successes, and those successes depend on state and local government action. In turn, state and local governments want federal program authority and federal assistance, but they want to direct it to jurisdictional needs and priorities. This sets up the possibility that human engagement over federal (or state) programs could move beyond compliance to a form of negotiated adjustment or settlement that maximizes solutions for the two parties. The donor-recipient model says that not everybody will comply and not everybody will defy. Some jurisdictions do find a way to manage a solu- tion within the overall policy framework.

Jurisdiction-Based Model The jurisdiction-based model of management is most

apparent in a highly complex administrative context. The authors’ research on economic development in cities has defined jurisdiction-based activity as the extent to which local officials (managers) seek out the normative adjust- ments, actors, and resources necessary to adequately serve the jurisdiction (Agranoff and McGuire 1998b). An ex- ample of this type of management would be when a city government receives the input of the major economic ac- tors—government and nongovernmental—and formulates a long-range development plan that proposes several in- tergovernmental adjustments. Examples of such locally initiated adjustments include a three-way land swap with county government and the school district, an appeal to

American Federalism and the Search for Models of Management 675

the transit and sewer districts for infrastructure improve- ments, unsolicited applications for state and federal grants, financial-support negotiations with the state to change eli- gibility regulations or to receive a waiver on the state pub- lic works improvement program, and extended discussions and agreements with the U.S. Department of Housing and Urban Development regarding a city-developed “creative- use initiative” of its community development block grant entitlement money.

The operational reality for many cities at the end of the twentieth century is a milieu of multiple incentives, orga- nizations, actors, and agencies (Judd and Swanstrom 1997). Unlike the donor-recipient model, the jurisdiction-based model reflects the fact that grants in aid are just one policy instrument among many used to govern, and the federal government is just one governing actor among many in the intergovernmental system (Agranoff and McGuire 1999a). In many policy areas, including economic devel- opment, education, human services, and natural resources, state government has become the most important intergov- ernmental actor, possessing the critical resources of fund- ing, information, and expertise. Such institutional fragmen- tation allows any and all actors to play an important, irreplaceable role in local governance, depending on the policy problem and the context. This enables the local manager to pursue resources from political actors that are closer to home, relying less on federal programs and more on the variety of resources available from all governmen- tal, private, and nonprofit organizations. Far from carrying out a specific function on the intergovernmental organiza- tional chart or adapting needs to fit federal or state pro- gramming, local managers practicing jurisdiction-based management strategically interact with actors to design and administer policies that meet the jurisdiction’s goals.

Jurisdiction-based activity emphasizes local managers tak- ing strategic action with multiple actors and agencies from various governments and sectors. This model portrays the local manager as driven by a strategic rationale that is, as the model’s name indicates, based in the jurisdiction. For such a manager, achieving local goals and completing particular tasks is paramount. Strategic behavior in the complex, inter- dependent world of the jurisdiction-based manager consists of seeking out and contacting actors who possess the resources (legal authority, funding, organization, expertise, informa- tion) that local managers need to achieve their goals. Achiev- ing goals in complex settings requires interaction and adjust- ment with critically positioned or endowed actors. As a result, jurisdiction-based managers may recognize numerous fed- eral programs, actors, and agencies—the many possible do- nors—that constitute the policy environment, but they may contact only those agencies that can provide targeted, place- oriented resources for the manager’s jurisdiction. By calcu- lating the costs and benefits of each available program, the

manager of the jurisdiction-based model proceeds more stra- tegically through the grant-seeking thicket than the recipient in the donor-recipient model.

Bargaining and negotiation are important instruments of jurisdiction-based management. Bargaining by local managers within programs of vertical (state or federal gov- ernment) or horizontal (metropolitan, regional, intersectoral) origin provide alternatives to unilateral con- cession, resulting in a “mutually beneficial solution” (Agranoff and Lindsay 1983). Adjustments are often initi- ated by localities requesting treatment that is not techni- cally or apparently within standards or regulations. Ad- justment-seeking activity on the part of the local jurisdiction attempts to redefine program goals from purely federal terms to local or at least intergovernmental terms that ben- efit all relevant levels of government. Thus, jurisdiction- based management emphasizes calculation and negotia- tion rather than accommodation and acclimation.

The types of interactions that characterize jurisdiction- based activity are more varied than other management models. Managers who are actively engaged in jurisdic- tion-based activity display a complex grid of managerial behaviors that are not merely a part of the manager’s job, they are the job itself. Such tasks are based not in a con- stant battle over the distribution of power, as assumed by top-down and donor-recipient models, but in a daily sort- ing out of issues within a particular policy area. The stra- tegic, interactive foundation of the jurisdiction-based model is consistent with a practical and theoretical move away from the classical perspective of management, toward an approach based in strategy and managing interdependen- cies (Bozeman 1993; Bryson and Roering 1996; Kettl 1996). The jurisdiction-based management model describes the type of locally initiated interaction that research indi- cates is alive and well in many American cities.

As more responsibilities are left to the local level or are devolved by state and federal governments and more policy decisions are decentralized, the temporal relevance of the jurisdiction-based management model should increase as well. If, over time, federal grant programs include elements of broad or discretionary categoricals and/or the block-grant structure, and if local governments become more proac- tive and innovative in addressing environmental and regu- latory concerns, then jurisdiction-based management will become more relevant, displacing top-down or coopera- tive models of managing within federalism. The jurisdic- tion-based model presupposes that local jurisdictions wish to assert their will in intergovernmental matters, that they are willing to bypass existing relationships and establish new relationships with bearers of resources for the pur- pose of achieving local goals. If such behavior is encour- aged, either politically or administratively, jurisdiction- based managers will become common.

676 Public Administration Review • November/December 2001, Vol. 61, No. 6

Network Model The final model of management describes public policy

making and governance that takes place in networks con- sisting “of various actors (individuals, coalitions, bureaus, organizations) none of which possesses the power to de- termine the strategies of other actors” (Kickert, Klijn, and Koppenjan 1997, 9). In a small city, for example, the city government, a public–private economic development cor- poration (EDC), the local chamber of commerce, county government, an investor-owned utility serving the area, three private investors, and a local bank work together as partners to expand an existing industrial park at the edge of town. With the staff support of the EDC, the actors net- work to make the project a reality by developing a plan that includes pledged business expansions, a private land donation, city government property tax abatements, a county tax contribution, a utility subsidy, and EDC prepa- ration of several business expansion and venture capital grants and loans. The mayor and the EDC director go to the state government and the Economic Development Ad- ministration of the U.S. Department of Commerce for in- tergovernmental assistance to underwrite the remaining costs of the project.

Whereas the jurisdiction-based model is premised on achieving the goals of the subnational jurisdiction (em- phasizing the jurisdiction’s (desired) superior position to other actors), and the top-down and donor-recipient mod- els focus on the achievement of national goals, the net- work model is based in interdependency. As in the juris- diction-based model, the network model extends the analytic focus to intersectoral and intergovernmental rela- tionships, but in this case leadership is collaborative. Ac- tors in a federal system depend on each other because they need the other’s resources to achieve their goals. Interde- pendence implies that all actors benefit in some way, re- sulting from their joint interest in a specific activity, and that a particular problem cannot be solved unless partici- pating actors pursue the solution strategically and collaboratively. From this perspective, involvement in a network is not an administrative choice, such as choosing to merge two departments or streamline a budgeting pro- cess. Multiorganizational arrangements, like the variety that characterize many policy-making domains, are arrange- ments for solving interorganizational problems that can- not be achieved by single organizations (Radin et al. 1996).

Networks consist of a variety of public and private ac- tors, each with their own goals and policy strategies. Such “structures of interdependence involving multiple organi- zations or parts thereof, where one unit is not merely the formal subordinate of the others in some larger hierarchi- cal arrangement” (O’Toole 1997, 45) include no central actor, no ruler-defined relationships, and no pre-established goals of a single actor that provide administrative guid-

ance and control. Although there are often strong incen- tives for operating in networked settings, policy processes in networks are unpredictable and complex. Actors’ pref- erences may change during the course of the interaction, and actors do not know in advance which outcomes are likely to occur and which targets they can meet in the pro- cess (Kickert, Klijn, and Koppenjan 1997, 32).

Networks are conceived in terms of a cluster of orga- nizations, a purposeful whole, rather than as many dif- ferent organizations performing unrelated tasks (Mandell 1988). Symbiotic relationships occur when interdepen- dent organizations, possessing different types and levels of technologies and resources needed for fulfilling a col- lective task, increase the frequency and intensity of com- munication among organizations, which in turn forces decisions to be made jointly and actions to be carried out collectively (Alter and Hage 1993). Such solutions are not merely compromises between parties’ “go-alone” strategies, they are adjustments designed to provide a solution that benefits all parties (Kickert and Koppenjan 1997). Robert Agranoff’s study of interlocal management of federal grant and regulatory programs reveals consid- erable mutual adjustment on the part of officials at all levels in “the middle-ground between strict compliance and open defiance,” involving gray areas of interpreta- tion, priority, direction, substantial conformance, spirit of the law, and mutually agreed upon actions that are deemed acceptable (1986, 193).

The parties involved in networks jointly reach decisions about direction and operation. This is perhaps what distin- guishes networks from other multiorganizational structures, such as markets. Markets display characteristics of inter- dependence, but market participants do not meet, agree on issues, and forge decisions. Through formal processes of grant applications or less formal decision meetings, the parties must always identify and refine the problem and agree about a course of action. The parties need to develop strategy, mechanisms for operation and implementation, and some means for determining whether the expected re- sults are being accomplished (Agranoff 1998).

The role of management in the network model is one of problem resolution by facilitating and furthering interac- tion, as well as creating and changing network arrange- ments for better coordination. The public manager acts within the network and will “use information to heighten the salience of preferred choices; convey knowledge about how cooperation can serve interests of the others; honor confidences; and focus participants’ perceptions on ele- ments crucial for success” (O’Toole 1997, 48). A central feature of furthering interaction is promoting an underly- ing commitment or program rationale throughout the net- work (Mandell 1990). Because the effectiveness of a net- work is as much a function of where and how the various

American Federalism and the Search for Models of Management 677

parties are positioned within the network as it is of the actions of the actors themselves (Burt 1992; Provan and Milward 1995), the manager of a network may modify the network when it proves impossible to solve problems within the existing structure (Kickert, Klijn, and Koppenjan 1997). An adept manager will determine the configuration of or- ganizations that is needed to perform a particular function.

Examples of networked settings exist in many different policy areas and for many different purposes. Networked settings are commonly found in human services because many organizations from the nonprofit sector take part in delivering such services. In this area, networks are both intergovernmental and intersectoral. Another prominent example of strategic activity within networked settings is the State Rural Development Councils that exist as part of the National Rural Development Partnership, in which rep- resentatives from all levels of government and from pri- vate and nonprofit agencies work collaboratively on rural development issues (Radin et al. 1996). Each council pur- sues problem solving strategically by including the right combination of actors, using the right blend of resources, and removing critical barriers to action. These intergov- ernmental councils demonstrate the efficacy of mutual ac- tion and the possibility of strategic efforts between levels of governments that are collaborative and not competitive.

In many policy areas, including economic development, the key network management issue is not simply the ex- tent to which networks are the primary organizational set- ting for designing and executing policy, but, more impor- tantly, the number and type of networks that exist within the policy-making realm of a single city. In economic de- velopment, local development managers attend to multiple strategic tasks within the context of designing and execut- ing economic policy. Like the hub of a multispoked wheel, the development manager is connected to all of the spokes, each representing a different strategic task, each consist- ing of networks of different composition, scope, and size, and each with its own set of management challenges and responsibilities (Agranoff and McGuire 1998c). Manag- ers in networks are continually faced with problems that can lead to instability (Milward 1996), and these concerns are compounded by the number of networks that consti- tute local economic policy making.

The temporal relevance of this model is clear. The “age of the network” has arrived, supplementing previous group, hierarchy, and bureaucratic eras (Lipnack and Stamps 1994). There is every reason to expect that managing in the federal system according to the network model will increase over time. Despite misunderstandings about the causes and applicability of network operations, “the emer- gence of networks in public management is not a passing fad” (O’Toole 1997, 47). Salamon argued nearly two de- cades ago that the increasing use of policy instruments in-

volving a host of third parties “has significantly altered the practice of public management …” (1981, 255). Milward echoes this sentiment by arguing that an increas- ingly networked public management is resulting in a “hol- low state,” a concept that is analogous to the “hollow cor- poration,” which operates through a lean headquarters and contracts out all production capability to other organiza- tions (Milward, Provan, and Else 1993). If public manage- ment in networked settings effectuates policy to a greater degree than public management in traditional, bureaucratic contexts, if it is more “consequential” (O’Toole 1996, 241), then the network model of managing may become as preva- lent in intergovernmental matters as the top-down and do- nor-recipient models have been in the past.

Federalism and Models of Managing Increasingly, public managers are concerned with how

to operate within a changing federal system that includes multiple governments, organizations, programs, and instru- ments. Greater proportions of public managers’ time are caught up in handling the interdependencies between their organizations and others (Kettl 1996), including horizon- tal and vertical actions. In the 237-city study, economic development managers reported they spend an average of 20 percent of their time on these activities. The emergence of the jurisdiction-based model is a result of federalism changes since the Reagan years, in which local govern- ments (and other federal aid recipients) have been encour- aged to wean themselves from dependence on federal aid, gain control over their own destinies, become attractive to business, compete with each other, and adapt to national economic trends. “City officials were asked to become more than experts at grantsmanship; they were asked to become entrepreneurs” (Morgan 1989, 296). Networking among local actors for exchange of local services (such as fire, water, and libraries) has occurred for some time. However, as federalism has become more challenging—in terms of the problems approached, nongovernmental organizations invoked, and expanded managerial vehicles (regulations, loans, tax credits, and the like)—more emphasis has been placed on horizontal and vertical actors working together. Administrators remain involved in more standard, top-down regulatory and funding compliance activities. As federal money has become more scarce, Congress has chosen to accelerate the regulatory burden on governments, ratcheting up the time spent by administrators on compliance activ- ity. Finally, increased programming of different types, along with experience in dealing with the federal system, offers more opportunities for bargaining and adjustment between donors and recipients. The sheer number and variety of federal and state programs places implementation control in the hands of more local government and nongovern-

678 Public Administration Review • November/December 2001, Vol. 61, No. 6

mental agents, who have more options in seeking program adjustments. Thus, the emergent models do not replace the longer-standing models.

To the manager, the federal system is a field of interorganizational complexity. While the term “intergov- ernmental” is often thought of as involving dyadic trans- actions between governments, in fact, a variety of links are engaged. The days when a city could simply fill out a federal grant application and operate a small program at- tached to a city department are becoming less frequent. The manager is more likely to propose a leveraged fund- ing arrangement with a state-funded program, along with a private investment, to match a federal government loan. This strategic move may be worked out among city gov- ernment, county government, a development association, and a public–private economic development corporation. Managing the interdependencies may also lead to a vari- ety of somewhat unpredictable connections. Our data in- dicate that, on average, cities experience around 60 such linkages over a two-year period, involving not only county government and development corporations, but also cham- bers of commerce and other government (township, nearby cities, substate regional) and nongovernmental organiza- tions (private industry council, developers, foundations, utilities, neighborhood associations). Moreover, this type of complex interorganizational managerialism is rarely lim- ited to a single plane. Most horizontal transactions are linked to a corresponding set of vertical connections around the same project effort. As we suggested earlier, locally initiated negotiations surrounding a project to build an in- dustrial park often triggers vertical contacts—a series of EDA and SBA loans, for example, plus work with the EPA and state environmental agency connections over compli- ance with water-supply regulations and licensing and per- mitting with other state agencies. Indeed, the managers’ job is more deeply involved in a variety of linkages that extend far beyond traditional grantsmanship.

The work identified here also involves a different kind of public management than the “classical” hierarchical work within a single organization. Managing linkages with other organizations is qualitatively different. This is not the place to explore the multifaceted dimensions of transboundary management (Agranoff and McGuire 1999b; O’Toole 1997), but each of the four models suggests a dif- ferent type of interorganizational management. The top- down approach suggests compliance-oriented, discerning donor intentions transmitted through “buzzwords,” rules, standards, and expectations. Donors and recipients engage in two-party bargaining across jurisdictions and try to reach mutually agreeable accommodative decisions. Jurisdiction- based activity invokes leadership by a unit in forging stra- tegic design and packaging of programs, wrapped around dozens of horizontal and vertical consultations, toward

programmatic adjustments that meet the recipient’s rather than the donor’s intent. To network intergovernmentally, a host of federal, state, and local government (plus nongov- ernmental) actors are involved in collective strategy build- ing and programming where no one is in charge (Bryson and Crosby 1992). These suggestions oversimplify a con- siderably more involved set of cross-agency managerial skills and processes (Bardach 1998), but they illustrate an emerging type of work that is substantially different from the core approaches in standard public administration texts.

A related implication of the emergent approaches is that managers in a host of agencies and organizations are be- coming the “policy makers” as they propose strategies and work through adjustments at the boundaries of their orga- nizations. In our research, whether the city is the lead ac- tor or a partner in a network, organizational representation and decisions are made by administrative officials. City managers, economic development directors, chamber of commerce executive directors, county planners, economic development corporation operating officers, utility man- agers, and foundation executives are the players who make the core decisions. Their decisions, collective or accom- modated, form the jurisdictional positions, and from their ranks the contacts and negotiations with state and federal governments ensue. Representatives of legislative bodies or governing boards are less often in the central loop of this type of intergovernmental decision making. As Wright (1983) maintains, collective strategy and policy issues in the intergovernmental scene place the greatest emphasis on appointed administrators.

The two emergent models underscore the importance of “governance” in the term intergovernmental, in the sense that Frederickson (1997, 84) uses it to include “the wide range of types of organizations and institutions that are linked together and engaged in public activities.” As it is described in urban-regime theory, arrangements among public bodies and private interests function together to make and carry out governing decisions (Stone 1989). The task is to “identify major stakeholders and activate them on behalf of a widely beneficial program of social improve- ment” (Stone et al. 1999, 357). As complexity increases, governmental actors link with nongovernmental actors, who become necessary components of the local delivery system (Stoker and Mossberger 1994). To a certain extent, local officials have worked together for some time on mat- ters of mutual service agreements, contracting for service, boundary adjustments, and other means of interlocal ad- aptation as a surrogate for regional government in metro- politan America. Governance in the sense underscored here takes managing within the intergovernmental system fur- ther, involving many entities, governmental and nongov- ernmental, in a host of strategy making, project-based, and resource-exchange activities through networks.

American Federalism and the Search for Models of Management 679

Is government, then, just another actor in this array of organizations? This question goes to the heart of power relationships within multiorganizational governance ar- rangements. Federal, state, and local governments appear to possess some legal and fiscal levers that can keep gov- ernmental units at the center of intergovernmental trans- actions and can be counterweights to any financial, infor- mational, operational, or other asymmetries that nongovernmental organizations might bring to the table. Sharpe (1986) maintains that assessing power distribution in intergovernmental relations is monstrously difficult, but to press interorganizational-relations theory beyond the lim- its of exchange relationships and to relegate government to the status of just another organization is inaccurate. Our analysis of economic development puts municipalities at the hub of horizontal and vertical connections as they en- gage in contacts and transactions with multiple federal and state agencies and interlocal players.

Along with other facets of public management, manag- ers who deal with the federal system have entered the in- formation age through expanded contacts and networks. While bureaucracy was the hallmark of the industrial age, interorganizational teamwork and networks are the hall-

mark of the information age. Managing across organiza- tions meets the need to deal with greater complexity, scope, speed, flexibility, and adaptability (Alter and Hage 1993). Lipnack and Stamps’ (1994) five “teamnet” principles are intended to represent management activity in this new age: shared commitment to goals and projects, independent but cooperative jurisdictions and organizations, voluntary link- ages involving extensive participation and crisscrossing relationships, multiple leaders, and people working at dif- ferent levels within and between organizations. These prin- ciples increasingly characterize managing within federal- ism, particularly the emergent jurisdiction-based and network models.

Acknowledgments

This research was sponsored by the Ameritech Foundation Research Program in Management and Organizational Studies, Institute for Developmental Strategies, Indiana University. An earlier version of this article was delivered at the annual meeting of the American Political Science Association, Atlanta, Geor- gia, September 2–5, 1999. The authors thank Dale Krane, Julianne Mahler, and anonymous reviewers for their helpful com- ments.

680 Public Administration Review • November/December 2001, Vol. 61, No. 6

References

Agranoff, Robert. 1986. Intergovernmental Management: Hu- man Services Problem-Solving in Six Metropolitan Areas. Albany, NY: State University of New York Press.

———. 1998. Partnerships in Public Management: Rural En- terprise Alliances. International Journal of Public Adminis- tration 21(11): 1533–75.

Agranoff, Robert, and Valerie A. Lindsay. 1983. Intergovern- mental Management: Perspectives from Human Services Problem Solving at the Local Level. Public Administration Review 43(3): 227–37.

Agranoff, Robert, and Michael McGuire. 1998a. The Intergov- ernmental Context of Local Economic Development. State and Local Government Review 30(3): 150–64.

———. 1998b. A Jurisdiction-Based Model of Intergovernmen- tal Management in U.S. Cities. Publius: The Journal of Fed- eralism 28(4): 1–20.

———. 1998c. Multinetwork Management: Collaboration and the Hollow State in Local Economic Policy. Journal of Pub- lic Administration Research and Theory 8(1): 67–91.

———. 1999a. Expanding Intergovernmental Management’s Hidden Dimensions. American Review of Public Administra- tion 29(4): 352–69.

———. 1999b. Managing in Network Settings. Policy Studies Review 16(1): 18–41.

Alter, Catherine, and Jerald Hage. 1993. Organizations Working Together. Newbury Park, CA: Sage Publications.

Bardach, Eugene. 1998. Getting Agencies to Work Together. Washington, DC: Brookings Institution.

Bozeman, Barry. 1993. Preface to Public Management: The State of the Art, edited by Barry Bozeman, xiii–xv. San Francisco, CA: Jossey-Bass.

Bryson, John M., and Barbara C. Crosby. 1992. Leadership for the Common Good. San Francisco, CA: Jossey-Bass.

Bryson, John M., and William D. Roering. 1996. Strategic Plan- ning Options for the Public Sector. In Handbook of Public Administration, 2nd ed., edited by James M. Perry, 479–98. San Francisco, CA: Jossey-Bass.

Burt, Ronald. 1992. Structural Holes: The Social Structure of Competition. Cambridge, MA: Harvard University Press.

Church, Thomas W., and Robert Nakamura. 1993. Cleaning Up the Mess: Implementation Strategies in Superfund. Washing- ton, DC: Brookings Institution.

Conlan, Timothy. 1991. And the Beat Goes On: Intergovernmen- tal Mandates and Preemption in an Era of Deregulation. Publius: The Journal of Federalism 21(3): 43–57.

Elazar, Daniel J. 1964. Federal-State Collaboration in the Nine- teenth-Century United States. Political Science Quarterly 79(2): 248–81.

Elmore, Richard F. 1985. Forward and Backward Mapping: Re- versible Logic in the Analysis of Public Policy. In Policy Implementation in Federal and Unitary Systems, edited by Kenneth Hanf and Theo A.J. Toonen, 33–70. Dordrecht, Hol- land: Martinus Nijhoff Publishers.

Frederickson, H. George. 1997. The Spirit of Public Adminis- tration. San Francisco, CA: Jossey-Bass.

Grodzins, Morton. 1966. The American System. Chicago: Rand McNally.

Ingram, Helen. 1977. Policy Implementation Through Bargain- ing: The Case of Federal Grants-in-Aid. Public Policy 25(4): 499–526.

Jennings, Edward T., and Dale Krane. 1994. Coordination and Welfare Reform: The Quest for the Philosopher’s Stone. Pub- lic Administration Review 54(4): 341–8.

Judd, Dennis, and Todd Swanstrom. 1997. City Politics: Private Power and Public Policy. 2nd ed. Reading, MA: Addison Wesley.

Kaplan, Abraham. 1964. The Conduct of Inquiry. San Francisco, CA: Chandler.

Kettl, Donald F. 1987. The Regulation of American Federalism. Baltimore, MD: Johns Hopkins University Press.

———. 1996. Governing at the Millennium. In Handbook of Public Administration, 2nd ed., edited by James L. Perry, 5– 18. San Francisco, CA: Jossey-Bass.

Kickert, Walter J.M., and Joop F.M. Koppenjan. 1997. Public Management and Network Management: An Overview. In Managing Complex Networks, edited by Walter J.M. Kickert, Erik-Hans Klijn, and Joop F.M. Koppenjan, 35–61. London: Sage Publications.

Kickert, Walter J.M., Erik-Hans Klijn, and Joop F.M. Koppenjan. 1997. Introduction: A Management Perspective on Policy Net- works. In Managing Complex Networks, edited by Walter J.M. Kickert, Erik-Hans Klijn, and Joop F.M. Koppenjan, 1–13. London: Sage Publications.

Kincaid, John. 1998. The Devolution Tortoise and the Decen- tralization Hare. New England Economic Review (May/June): 13–40.

Leach, Richard. 1970. American Federalism. New York: W.W. Norton.

Liebschutz, Sarah F. 1991. Bargaining Under Federalism: Con- temporary New York. Albany, NY: State University of New York Press.

Lipnack, Jessica, and Jeffrey Stamps. 1994. The Age of the Net- work. New York: John Wiley and Sons.

Mandell, Myrna P. 1988. Intergovernmental Management in Interorganizational Networks: A Revised Perspective. Inter- national Journal of Public Administration 11(4): 393–416.

———. 1990. Network Management: Strategic Behavior in the Public Sector. In Strategies for Managing Intergovernmental Policies and Networks, edited by Robert W. Gage and Myrna P. Mandell, 29–53. New York: Praeger.

Milward, H. Brinton. 1996. Symposium on the Hollow State: Capacity, Control, and Performance in Interorganizational Settings. Journal of Public Administration Research and Theory 6(2): 193–5.

American Federalism and the Search for Models of Management 681

Milward, H. Brinton, Keith G. Provan, and Barbara A. Else. 1993. What Does the Hollow State Look Like? In Public Manage- ment: The State of the Art, edited by Barry Bozeman, 309– 22. San Francisco, CA: Jossey-Bass.

Morgan, David R. 1989. Managing Urban America. Pacific Grove, CA: Brooks/Cole.

O’Toole, Laurence J. 1996. Hollowing the Infrastructure: Re- volving Loan Programs and Network Dynamics in the Ameri- can States. Journal of Public Administration Research and Theory 6(2): 225–42.

———. 1997. Treating Networks Seriously: Practical and Re- search-Based Agendas in Public Administration. Public Ad- ministration Review 57(1): 45–52.

Pressman, Jeffrey L. 1975. Federal Programs and City Politics: The Dynamics of the Aid Process in Oakland. Berkeley, CA: University of California Press.

Pressman, Jeffrey, and Aaron Wildavsky. 1984. Implementation. 3rd ed. Berkeley, CA: University of California Press.

Provan, Keith, and H. Brinton Milward. 1995. A Preliminary Theory of Interorganizational Effectiveness: A Comparative Study of Four Community Mental Health Systems. Adminis- trative Science Quarterly 40(1): 1–33.

Radin, Beryl A., Robert Agranoff, Ann O’M. Bowman, Gregory C. Buntz, Steven J. Ott, Barbara S. Romzek, and Robert H. Wilson. 1996. New Governance for Rural America: Creating Intergovernmental Partnerships. Lawrence, KS: University Press of Kansas.

Salamon, Lester. 1981. Rethinking Public Management: Third Party Government and the Changing Forms of Government Action. Public Policy 29(3): 255–75.

Sharpe, Laurence J. 1986. Intergovernmental Policy-Making: The Limits of Subnational Autonomy. In Guidance, Control, and Evaluation in the Public Sector, edited by Franz-Xaver Kaufmann, Giandomenico Majone, and Vincent Ostrom, 159– 81. New York: Walter de Gruyter.

Stever, James A. 1993. The Growth and Decline of Executive- Centered Intergovernmental Management. Publius: The Jour- nal of Federalism 23(1): 71–84.

Stoker, Gerry, and Karen Mossberger. 1994. Urban Regime Theory in Comparative Perspective. Environment and Plan- ning C: Government and Policy 12(2): 195–212.

Stone, Clarence. 1989. Regime Politics. Lawrence, KS: Univer- sity Press of Kansas.

Stone, Clarence, Kathryn Doherty, Cheryl Jones, and Timothy Ross. 1999. Schools and Disadvantaged Neighborhoods: The Community Development Challenge. In Urban Problems and Community Development, edited by Ronald F. Ferguson and William T. Dickens, 339–80. Washington, DC: Brookings In- stitution.

Sundquist, James L., and David W. Davis. 1969. Making Feder- alism Work: A Study of Program Coordination at the Com- munity Level. Washington, DC: Brookings Institution.

Walker, David B. 1974. How Fares Federalism in the Mid-Sev- enties. Intergovernmental Relations in America Today, An- nals of the American Academy of Political and Social Sci- ence. Philadelphia: Stanford University Press.

———. 2000. The Rebirth of Federalism. 2nd ed. New York: Chatham House.

Watts, Ronald L. 1996. Comparing Federal Systems in the 1990s. Kingston, Ontario: Institute of Intergovernmental Relations, Queen’s University.

White, Leonard D. 1939. Introduction to the Study of Public Administration. Revised ed. New York: Macmillan.

Williams, Walter W. 1981. Government by Agency: Administer- ing Grants-in-Aid Programs. New York: Academic Press.

Wright, Deil S. 1983. Managing the Intergovernmental Scene. In Handbook of Organizational Management, edited by Wil- liam B. Eddy, 417–54. New York: Marcel Dekker.

———. 1988. Understanding Intergovernmental Relations. 3rd ed. Belmont, CA: Wadsworth.

Wright, Deil S., and Dale Krane. 1998. Intergovernmental Man- agement (IGM). In International Encyclopedia of Public Policy and Administration, edited Jay M. Shafritz, 1162–8. New York: Henry Holt.

Zimmerman, Joseph F. 1992. Contemporary American Federal- ism: The Growth of National Power. New York: Praeger.