A State of Agents? Sharpening the Debate and Evidence over the Extent and Impact of the Transformation of Governance

INTRODUCTION

In Beyond Privatization, Salamon (1989) asserted that “the enterprise of government is under unusually severe attack in the United States” (3). Although acknowledging a steady, longer term transformation away from direct provision of goods and services through traditional structures of public administration toward more devolved authority and decentralized public services delivery, Salamon's language was remarkably alarmist. In describing how the basic “tools” of government action were changing, Salamon argued that “they put government in the position of operating by remote control, relying on other entities to deliver the services that government has authorized” (8–9). His concern was particularly strong for the US welfare state, which he declared “is not run by the state at all, but by a host of nongovernmental ‘third parties’” (xv).

Nearly two decades later, research on the growth of “third-party” government (also termed “indirect” or “quasi-” government) and new forms or “tools” of governance, including networks and other “hybrid” or collaborative arrangements, has suffused the literature on both the study and practice of public administration, albeit with far less alarmism. In this burgeoning body of work, scholars have sought to expound and clarify definitions of elemental concepts and terms such as “governance” and “third parties” and to formulate arguments and empirical tests of the implications of changing governance configurations against various criteria, including equity in access to public goods and services, responsiveness to “customers” (formerly “clients” or “patients”), accountability to elected officials and organized stakeholders, efficiency in service provision, and effectiveness in producing outcomes and results.

Scholars are also coming to grips with deeper questions about the magnitude and consequences of these changes. Some are concerned that “extra-formal democracy”1 usurps traditional forms of democratic accountability and imperils the legitimacy of government action (Frederickson and Frederickson 2006; Kettl 2002a; Skelcher 2007; Stoker 1998). Others argue that the de-bureaucratization of authority over publicly financed services enhances democratic legitimacy through increased transparency, responsiveness, and citizen participation in setting priorities and evaluating results (Bingham, Nabatchi, and O'Leary 2005; deLeon and deLeon 2002). Traditional hierarchical governance and the newer consociational structures and practices are viewed by still others as complementary, even arguing that the former is necessary for the latter to be effective (Frederickson and Smith 2003; Skelcher 2010).

It is our goal, and that of the “state of agents” symposium we have assembled, to engage some of the focal questions about the extent and impact of the increasing use of third parties and alternative structures and processes of governance to implement “the public will” (to use an older term) and, as well, to sharpen the discussion by bringing empirical evidence to bear on key arguments and assertions in the debate. What is the evidence, for example, that government control and democratic responsiveness are loosening or slipping away as a result of these changes? Is the public sector evolving (or drifting) into a “state of agents,” where government authority is widely dispersed and diluted and accountability to the public is weak? Or is the nature of government today simply a new configuration (with more players) of the same checks-and-balances and federalism “games” that have endured in one form or another since the founding of the American state? Perhaps the truth is that what we are witnessing is Theodore Lowi's “interest group liberalism”2 on steroids (Lowi 1979). The contribution of Christopher Skelcher enables the consideration of the European experience with both evolving supranational and subnational forms of governance and the new challenges they are presenting for institutional design and democratic control, representation, and accountability.

Although this initiative places greater emphasis on empirical scrutiny of these important issues and questions, a complementary goal of our collective effort is to consider theoretical research in this area. For example, are there recent advances in theories or models of what might be termed “socialized choice” (Hill and Lynn 2003), such as social network theory, that are helping us to better understand the nature of these changing governance mechanisms and their implications for government performance? In what ways have rational choice approaches such as positive political economy—through which many researchers have investigated the delegation, coordination, and control of public authority—evolved to encompass increasingly complex governance regimes and relationships? In exploring new forms of governance evolving in the European Union, Skelcher argues in this symposium that not only is the agent function of policy implementation devolving but so is the principal function of policy-making itself.

In this introduction, we review basic arguments and themes that have emerged in the academic literature and policy discussions about the growth and implications of the “new governance” in public affairs. In doing so, we clarify definitions of essential concepts and identify important, unsettled questions that may be amenable to empirical investigation. We conclude with an overview of the theoretical advances and empirical contributions of the research presented in this symposium and further consider the challenges of the transformation of governance for both the study and practice of public administration.

KEY ARGUMENTS AND THEMES, ESSENTIAL AND OPERATIONAL DEFINITIONS, AND OPEN QUESTIONS

One would be hard-pressed to find meaningful debate in the literature as to whether there have been significant changes in how government manages its responsibilities and accomplishes its goals. The Shadow Government (Guttman and Willner 1976), a vociferous treatise focusing on the “invisible bureaucracy” of private, for-profit and nonprofit firms contracted with the federal government, was one of the first to call for more serious attention to changes in government's approach to carrying out its responsibilities. In introducing the book, Ralph Nader decried the “almost exclusive influence” of the invisible bureaucracy to “suggest, shape, and even implement much governmental policy in both its narrowest and broadest sense” (xii). Although debate as to how much influence nongovernmental agents have in the design, management, and execution of the government's work continues, there is basic agreement that third-party entities are playing increasingly influential roles, and how the government contracts or otherwise arranges for these agents to perform their roles has critical implications for what is accomplished and how. At the same time, important differences in the conceptualization of what constitutes “third-party” involvement that also vary with the use of the terms “government” and “governance” in the literature warrant clarification.3

In his earlier research on changing relationships between federal government and nonfederal government entities and new administrative “tools” (e.g., grants-in-aid made by the federal government to state and local governments), Salamon (1989, 9) defined “third-party government” as the “sharing of the exercise of discretion over the use of public authority and the spending of public funds” with “one or another nonfederal, often nonpublic, third-party implementer[s].” This definition is operationally distinct, however, from his more recent description of the “third sector” as private, nonprofit, “self-governing” organizations that “occupy the social space between the market and the state” (Salamon and Sokolowski 2003, 3). The use of the term “hybrid” or quasi-government organization (or “quango”) is likewise so widely varying, according to Koppell (2003), that it hinders any shared understanding of the phenomenon. For example, Koppell (1–2) characterizes a quango as an entity that “combines characteristics of public- and private-sector organizations” and may be publicly or privately owned (but typically “operates in the shadows”). In contrast, Bertelli (2006, 240–1), citing Van Thiel (2001), defines the quango as a quasi-autonomous, nongovernmental organization that is publicly funded but operates “at arm's length” of the central government. Clearly, there are conceptually important distinctions in researchers’ use of similar terminology—particularly, as to including federal and/or nonfederal government entities versus referring exclusively to private, nongovernmental agents—that warrant careful attention by those attempting to advance our understanding of the transformation of governance and its implications.

In addition, the distinction between “government” and “governance” has also been addressed by numerous scholars, and a move toward consensus or convergence in definitions has come primarily through generalization. Most broadly, governance is the exercise of authority, public or private, “concerned with creating the conditions for ordered rule and collective action” (Milward and Provan 2000, 360). In contrast, government, according to Milward and Provan, refers to the formal executive, legislative, and judicial institutions of the state and their legitimate exercise of coercive power; it is subsumed under governance.

It is in the nuances of the interpretation of these definitions (and various combinations of them, i.e., third-party government versus third-party governance), where conceptual slippage is more likely and yet also where clarity is essential for credible empirical research and synthesis. For example, is the “sharing” of public authority conceptually different from its devolution (or the “renting” of it) to an agent of the state? If a government grants authority to a nongovernmental organization to make policy decisions and spend public funds, is it necessarily relinquishing its own authority, as Stoker (1998) implies, that is, allowing autonomous, self-governing networks of actors to take control, or does it inevitably retain the ability to reclaim full authority as the statutory agent of constitutional authority? A more elaborate definition of public sector governance offered by Lynn, Heinrich, and Hill (2001),4 gives more explicit emphasis to the constitutional linkages embedded in these multilayered relationships and also to the responsibility of the state in serving the public interest. Ultimately, though, as Milward and Provan (2000, 360) suggest, the degree to which governance mechanisms “operate autonomously or are steered by the state” is an empirical question. Indeed, this is a focal question taken up in the research included in this symposium. Have we moved beyond the hollow state (Milward and Provan 2000), to where the state no longer has the ability to arrange contractual networks or otherwise carry out its functions without the assistance of agents? This is the metaphor we have adopted for this symposium—a state of agents—and a central question we have posed to our authors.

In considering the extent to which formal government institutions—with legitimacy and support from (and responsibility to) the electorate and to the courts for their actions—are able to retain control and uphold obligations to citizens in complex governance arrangements involving multiple, third-party actors, a number of unanswered (or unsettled) questions emerge. For example, in what circumstances, or with what type of third-party actor involvement, and at what level of government do we observe a loosening (or loss) of government control to be particularly problematic? Is it the weakening of federal control in the face of lengthening chains of delegation and increasing distance (political and physical) between the origins of government activity and the point of service delivery, as Lynn (2006) observes, that is most troublesome? Or is it the devolution of federal government responsibilities in combination with new vulnerabilities at the state and local levels, where discretion and control are also more widely dispersed and management capacity is more likely to be “hollowed-out” by extensive contracting or the deferral of responsibility for service provision to networks of nongovernmental providers, that has given cause for greater alarm? In discussing two aspects of what they call “the hollow state,” Milward and Provan (2000) ask whether the devolution of authority to third parties erodes the government's capability both to continue to manage and to maintain the capacity to take back tasks that have devolved. And how many links can be added in a chain of funding and authority (between the source and the use of funds) before public purpose and accountability are seriously eroded or attenuated? In their work on mental health, for example, Milward and Provan find that it is not at all unusual for as many as four funding links to exist between the source and use of public funds and for several of those links to be nongovernmental.

Salamon (1989, 11) expressed serious concern about the “imperfect” control that federal managers have over third-party government and the “blurring of sector lines” that has occurred as devolution has progressed. Yet as Heinrich, Hill, and Lynn (2004) note, there is little evidence that there ever has been a tight link between government authority, activity, and outcomes or anything close to perfect control, an assertion supported by Carpenter's (2001) research on the development of bureaus in 19th and early 20th century America. It is more likely that in addition to the growing use of alternative governance arrangements, increasing attention to accountability for results over the last two decades has exposed weak linkages that were overlooked when a primary focus of management and oversight responsibilities (and public concern) was that bureaucracies comply with authoritative mandates.

Nonetheless, irrefutable consequences are associated with governments’ sharing or transfer of their duly authorized power and discretion to agents of the state, and an expanding body of work documents the potential for both beneficial and harmful results. For example, both researchers and policymakers have pointed to new opportunities to promote experimentation and innovations in program design and implementation through the devolution of responsibility to lower government levels and a more diverse array of hybrid and nongovernmental providers. Kamarck (2002) describes “networked government” as the leading form for social services and research and development, best suited to solving difficult problems through the freedom it accords employees from “rule-driven processes” and the opportunities for sharing best-practices among a diversity of organizations. A familiar related argument is that state or local governments and their nongovernmental partners are better able to discern and accommodate the diverse preferences, values, and needs among citizens and to also facilitate new channels for citizen participation (Rivlin 1992). In fact, tracing their origins back to the Federalist papers, Nathan (2006, 504) describes such virtues as the “essential political values of federalism”—diversity among regions, flexibility in problem solving, experimentation and innovation, and expanded opportunity for participation in political processes—and extols their “fundamental checking and balancing” role in American government. He also acknowledges, however, that “federalism is untidy, hard to manage, and uneven in its effects” (508), which are also some of the core concerns of critics of the brisk growth of new governance forms and arrangements.5

One of the tradeoffs most commonly discussed in the literature is that between the flexibility, innovation, and diversity of evolving governance forms and the constitutional values of transparency and accountability that are steadfastly associated with traditional bureaucratic structures and their central control mechanisms. As Kennedy (2006, 68) put it, “political efforts to keep government responsible and accountable—politically, fiscally, and constitutionally—depend on the ability to identify government and to recognize when the state has acted.” “‘Governance’,” she argued, “may be robbing citizens of the ability to make that crucial threshold identification.” If this occurs, then from the perspective of citizens and clients, we may indeed have devolved to a state of agents. In elaborating on this point, Kennedy focuses on the new challenges for legal accountability and draws on recent court cases to illustrate how the freedom from rule-driven processes lauded by some, or the “lack of comprehensible rules defining the actions we may legally attribute to the state” (70), blurs the boundaries of government's legal responsibilities and undercuts the role and efficacy of the courts in protecting and supporting citizens’ constitutional rights and privileges. And an even graver concern may be what she shows to be an intentional use of alternative governance mechanisms to evade “due process scrutiny” and state responsibility for policy actions. Given the heavy Anglo-American reliance on the legal authority of the courts for checking public and private power relationships (Bertelli 2005), it is not surprising that the struggle to uphold the legitimacy of government through devolved and dispersed networks of third-party actors and complex contracting relationships has become a central issue among leading US and European scholars studying governance (Kettl 2002a; Lynn 2006; Milward and Provan 2000; O'Toole 1997; Skelcher 2007).

In light of the growing distance between constitutional authority and the point of service delivery and suggestive evidence that “constitutionally authorized” forms of accountability are being evaded or compromised, Lynn (2006, 137, 154) asks how we can sustain “the legitimacy of government in the eyes of citizens,” and if it will require “wholly new structures, practices and institutions of democratic governance,” evolving from the institutional status quo. Both Lynn (2006) and Kettl (2002a) consider alternative mechanisms of accountability and their origins of authority, including executive tools (e.g., professionalism and representative bureaucracy), legislative channels, legal authority and popular (political) control through elections and participatory democracy, and contemplate whether a new equilibrium or change in reliance on one mechanism versus the others would be sufficient to better manage the new accountability challenges. Similarly, Romzek (1998, 23) submits that accountability relationships have long been complex—dense, multilayered, and overlapping webs of relationships that mix “active and reactive postures of legitimate authorities vis-à-vis the accountability of civil servants”—and suggests that multiple accountability types (hierarchical, legal, professional, and political) may be simultaneously present for a single entity, with relative priorities and demands among them changing over time. Indeed, these scholars generally concur that there has been a decided shift away from emphasis on hierarchical (tight, rule-based supervisor-subordinate relationships) and legal (compliance-oriented, external oversight) approaches to accountability and toward looser professional and political mechanisms, with greater deference to experts guided by professional norms and working within broader parameters that reflect expectations for responsiveness to stakeholders and for performance outcomes.

At the same time, research to date suggests that adapting accountability mechanisms to new structural and professional contexts and redefining or realigning accountability relationships presents substantial challenges for public managers. An example is the expanding body of work on the design and implementation of performance management and performance-based contracting systems, which shift accountability relationships from directing and controlling based on hierarchical authority (as with direct service provision) to devising incentives, negotiating standards or expectations, and assigning “residual rights of control” through contract provisions for renegotiation, renewal and/or bonuses or sanctions based on performance (Hart, Schleifer, and Vishny 1997; Heinrich 2007; Heinrich and Choi 2007). As Kettl (2002b) observes, control over the flow of funds endures as a primary tool of accountability in these relationships, although tracking and monitoring financial flows and encouraging or measuring/verifying their use in ways that promote or support program goals and core public values (e.g., justice, equity in access, quality, efficiency, and responsiveness) is considerably more difficult to accomplish. Indeed, measuring the performance of government agencies that rely largely on third parties to carry out their core functions and responsibilities is one of the central challenges in public management today, as Frederickson and Frederickson (2006) demonstrate in their case studies of federal health care agencies that make heavy use of a diverse set of policy tools and third parties to perform their primary work.

Posner (2002) and Smith and Ingram (2002) point out that third-party actors may bring different norms to these relationships and may draw on independent bases of political power and information to determine priorities and guide their actions. A possible long-term result, suggest Smith and Ingraham, is a “restructuring” of the relationship between government and its citizens with consequences for democratic participation. Skelcher (2007) likewise grapples with this issue and calls for more empirical research to address the larger question: Do these changing governance configurations imply an unwelcome shift toward “extra-formal democracy” (i.e., outside electoral processes) or an opportunity to strengthen democratic foundations by opening up new channels for citizen participation?

THE QUESTION OF EVIDENCE

As noted above, there is little meaningful debate over whether the changes in governance are in fact occurring and to an extent justifying claims that a “new governance” is either transforming public administration or has already done so. Unfortunately, the profession is much better at proclaiming “the new” than at making evidence-based arguments for such claims. “Compelling” cases of “the new” are staples of such proclamations. Too seldom are such cases weighed against matched cases that are inconclusive or contradictory, let alone against equally compelling illustrations of the continued importance and relevance of whatever is viewed as “the old.” Assertions based on examples are more common than arguments based on evidence and warranted by theory.

Such tendencies do not go unnoticed. Peters (1998, 408) has argued, for example, that “[i]t is difficult to take issue with the importance of networks and other forms of state and society in contemporary government. This pattern of state-society interaction is certainly evident in a number of policy areas and in most if not all industrialized democracies. The identification of the existence of networks does not, however, address several other important issues. One is whether those networks or something like them have not always been present, and what has changed is the theory rather than the reality. Have any observed changes been more matters of degree than of type, given that there have always been some interactions of government and groups in society?” In a similar spirit, George Frederickson (2005, 293) says that “[g]overnance theorists must be ready to explain not only what governance is, but also what it is not [and] be up-front about the biases in the concept and the implications of those biases.” Likewise, Smith (2003, 624) has argued that “[m]oving from description to analysis in this subject area first means sharpening the focus upon change in the power relationships observed in and between subnational, national and supranational arenas.” He urges drawing careful distinctions between the relative importance of different kinds of changes, engaging in longitudinal studies in order to distinguish superficial from genuine change, and, beginning at the local level, recognizing the influence of politics; that is, distributions of power and especially processes of institutionalization and legitimization.

This is not to suggest that empirical evidence concerning administrative structures and processes is altogether lacking. The extent of third-party government—of contracting out, outsourcing, procurement, and nonprofit sector participation—has been extensively documented. Light (2008) has revealed trends in “the true size of government,” that is, in the amount of employment generated by federal contracts and grants to public and private sector agents. World Bank researchers have developed a Worldwide Governance Indicators database covering 212 countries and territories and measuring six dimensions of governance between 1996 and 2006. These include: voice and accountability, political stability and absence of violence, government effectiveness, regulatory quality, rule of law, and control of corruption (Kaufmann, Kraay, and Mastruzzi 2007).

Unclear from these sources, however, is how the distribution of authority over the availability, cost, and quality of public service delivery might be changing. Literally, hundreds of studies published in recent years have reported causal relationships among the institutions and levels of governance. Following an analysis of over 800 empirical research studies published between 1990 and 2001, Hill and Lynn found that “the vast majority of studies adopt a top-down perspective on governance…. Influence is modeled as flowing downward from legislation and management toward treatments and consequences, and this general pattern is evident for virtually every level of governance being modeled….” (2005, 179).

How, Hill and Lynn ask, can this finding be reconciled with the strong current toward redefined governance? Although methodological shortcomings may have some influence, the robustness of the evidence suggests to them that: “The seemingly ‘paradigmatic’ shift away from hierarchical government and toward horizontal governing … is less fundamental than it is tactical: new tools or administrative technologies are being added that facilitate public governance within a hierarchical system. Shedding the strictures of hierarchy may seem refreshing (in a normative, positive, or symbolic sense), but constitutional authority (manifested in hierarchy) and the “fiscal spine” ‘of appropriated funds remain the structures within which relational and networked forms are enabled to flourish (2005, 189).”

The matter is hardly settled, however, and an important question remains: What kinds of evidence would support claims that new forms of governance are emerging, for whatever reason? How might changes in the boundary between the state and civil society be documented? We might expect to see, for example:

• significant reductions in the proportion of human services resources flowing through governments and thereby subject to legislative authorization and appropriation processes;

• evidence of both devolution and deregulation of human services program administration, including noninsurance transfers, a laying off of federal and state hands (e.g., a reduction in the number of pages of regulations governing human services administration or instances of language changes that reallocate “steering authority” to nongovernmental or hybrid entities);

• evidence of changes in agency enforcement strategies, not necessarily lax enforcement but increased reliance on noncoercive, voluntary, and negotiated forms of enforcement across significant swaths of public responsibility;

• a growing chorus of voices from the field, both from public program administrators and from their private sector agents and other civil society institutions, that government has become less intrusive, that human service professionals are feeling more empowered and less burdened by official rules, and that local justice determinations are being given greater weight in street-level practice;

• a strengthening of interest group and citizen involvement and influence at all levels of policy-making and bargaining over program features;

• shifts of interest group activity from higher to lower levels of government and from political action to broader forms of social mobilization;

• an increasing number and coverage (of individuals and households) by alternative, nontraditional service delivery arrangements, evidence, that is, that more aspects of well being for more people are in nongovernmental hands;

• some reduction in the extent of co-optation of the charitable nonprofit sector by government as public resources and guidance are replaced by civil society resources; or

• civil society institutions themselves—for example, foundations, United Way-type organizations, and large nonprofits—relinquishing their own authority over priorities and methods of implementation in favor of community and grass-roots entities.

In the absence of a preponderance of evidence one way or another, it is arguable that global, regional, national, and subnational changes are in fact strengthening the role and authority of governments. The recent global financial crisis, brought about by the triumph of market fundamentalist ideas, may well serve as the impetus for a return of the state in areas where its writ receded during the last 30 years. Although this is most apparent in calls for more effective regulation of financial markets and the government becoming an equity partner in banks and financial institutions, the Obama administration has indicated its intent to shrink the number of contractors and increase the number of government employees who oversee contractors because they believe that contractors have too much power vis-à-vis the federal government (Hedgpeth 2009, A01). A corollary to such an argument, however, is that governments under pressure to perform will necessarily need to rely on a wide array of structures and processes, involving, as appropriate, both hierarchical command and control and devolution and consociation, if they are to be effective in grappling with complex problems requiring coordinated action.

THE QUESTION OF THEORY

If there is little meaningful debate about whether and to what extent changes in governance are occurring, there is even less debate over why such changes may be happening.

The tendency within the profession to proclaim “the new”—a “new public management,” a “new governance”—evident in America since the early 1920s, enjoys worldwide popularity. Claims concerning “the new” have various foundations. They are often based, as noted above, on a perceived pattern of change or a perceived transformation of practice supported by examples or a selective assortment of evidence. Such inductive claims also occur when someone calls attention to certain facts of how governments and administrators appear to be approaching their work and in the nature of administrative work itself.

Among the reasons typically cited for transformative changes are contemporaneous developments in the external environment: globalization, technological innovation, the emergence of new policy, and programmatic challenges. Explanations seldom make reference to theories of institutional transformation within the American constitutional scheme or within democratic societies, although potentially relevant theories and analytic narratives abound in the social sciences. The new is seldom viewed as a consequence of a historical or path-dependent logic of change, and the old and the new are hardly ever conceived as on the same pathway of processual institutional evolution or adaptation.

The rarity of theoretical support for claims of administrative change seems puzzling in the light of the popularity in recent decades of “the new institutionalism” and other synoptic accounts of administrative system creation and evolution (for an overview, see Lynn 2001, 2006). Indeed, explanations of administrative transformation based on institutional theories or political economics are more likely to be invoked by those who are skeptical of such claims because correlates in the political and legal environments of public administration cannot readily be identified. Some of these skeptics tend to see change, but from a different perspective: general patterns of stability and path dependence that perhaps are occasionally interrupted by “punctuations” that shift states and their governments onto new trajectories of evolution and adaptation. Others tend to see in the data on administrative change an evolution in the variety of instruments, administrative technologies, and practice values that modify rather than replace existing institutions.

Much positive theorizing that does occur reflects microlevel analysis based on causal accounts or models in which organizations or interorganizational consociations are the unit of analysis and organizational theory is the intellectual rubric. Investigators may then generalize their findings to the administrative state as a whole. The problem with such reasoning is that entire levels within a political-administrative system are typically left out of the analysis. In other words, work of this kind fails to relate the parts to a coherent “whole” account of why and how the administrative state is organized as it is and functions as it does, what Terry Moe has termed “the politics of bureaucratic structure.”

Elsewhere, Lynn and colleagues sketch one approach to accomplishing such intellectual integration within a “logic of governance” (Lynn, Heinrich, and Hill 2001). “The literature of political economics,” Lynn (2008, 7) argues, “elaborates on how institutions, organizations, and individuals together design and shape the outputs of the administrative system. Within a rigorous, relatively parsimonious theoretical paradigm, a rich, subtle, and variegated depiction of administrative practice emerges without losing sight of the role of constitutional and political institutions in shaping that practice. The intellectual controversies thereby generated are the right ones for a field that aspires to understand administration fundamentally, in its political context.”

This particular reasoning is offered as an example, not as exemplary. The question of theory is, in general, a question of explaining administrative change, such as the emergence and form of a state of agents, within the context of a political-administrative system that incorporates both multilevel and intertemporal dimensions.

CONTRIBUTIONS OF THIS SYMPOSIUM

To this point, we have brought attention to only some of the many important questions that researchers are pursuing or for which further empirical investigation is essential to enlighten debates about the consequences of expanding third-party, quasi-governmental, and nongovernmental involvement in the design, management, and implementation of public policy. In this section, we describe the specific contributions of this symposium to these efforts to improve our understanding of new governance arrangements and its implications. The articles cover a broad range of governance and policy topics, including contracting, networks and other management tools and structures, collaboration and democratic governance, with applications as diverse as national security, education, and social welfare services delivery, the latter of which has seen some of the greatest growth in third-party involvement and for which there is also profound concern about the impacts of governance changes on vulnerable constituencies.

We view the studies presented in this symposium as exemplars of the cutting-edge of governance research in terms of both the theoretical and empirical contributions they are making. Although the primary objective of most of the papers in this symposium is to bring empirical evidence to bear on some of the major arguments asserted in the study of third-party governance, a central focus of one of the papers, “Relational Contracting and Network Management” by Anthony Bertelli and Craig Smith, is to advance theoretical perspectives in relational contracting and their application in the study of how modern governance arrangements and networks emerge and perform. Bertelli and Smith draw from economic theories of relational contracting and other interdisciplinary paradigms to explore and enrich our understanding of how economic, social, and political institutions—including rules, norms, and laws—influence contract design and contractual exchanges, both formal and informal. They are particularly interested in long-term contractual relationships that involve services for which it is difficult to develop complete contracts, measure performance, and hold providers accountable through traditional “command-and-control” approaches.

Bertelli and Smith elucidate key concepts in the study of relationship contracting, such as credibility, incentive power, integrated and nonintegrated transactions, optimal contracts and the brokering role of public managers, in describing how differing forms of relational contracting and network structures emerge and are sustained or terminated. They explain how both vertical and horizontal forms of control can simultaneously exist and be effectively managed, including in the face of unexpected contingencies or shocks to these relationships. They suggest that although the growth in third-party governance may “stretch” traditional lines of authority and accountability, bureaucratic control firmly continues, in part through the use and effective management of relational contracts in the public sector.

In their article on “Contracting for Complex Products” in the Coast Guard's Deepwater project, Trevor Brown, Matthew Potoski, and David Van Slyke also advance contracting theory in the context of complex and incomplete contracting arrangements. They specifically study the case in which the government seeks to acquire a product that is produced by a nongovernmental firm, in circumstances that are characterized by a high degree of uncertainty, including limited information about the costs, quality, and value of inputs and outputs, and unknown or unforeseen economic and political contingencies. Using the prisoners’ dilemma lens, they set forth a framework for analyzing principal-agent problems that arise in these contexts, such as contract lock-in, perfunctory behavior that emphasizes unilateral versus mutual gains, and the challenges of establishing credible commitments and cooperative relationships that promote more favorable contract outcomes.

Brown, Potoski, and Van Slyke describe three contractual layers that specified the terms of exchange and the process by which the Coast Guard and the contracting firm would collaborate in the design and production of the Deepwater products (i.e., small and large boats, planes, and helicopters linked through communication and integration technologies). They evaluate the strategies pursued by both the Coast Guard and its contractor to navigate the contract complexities and work toward a set of products that would satisfy expectations and the Coast Guard's needs. They conclude, however, that both parties were unhappy with the contract outcomes and that blame was cast on both sides for leadership failures, cost overruns, quality deficiencies, and schedule delays. Brown, Potoski, and Van Slyke analyze possible explanations for these outcomes.

President Bush signed the US No Child Left Behind (NCLB) Act into law in 2002 “to close the achievement gap [in public education] with accountability, flexibility, and choice.” In “Third-Party Governance under No Child Left Behind: Accountability and Performance Management Challenges,” Carolyn Heinrich studies a key NCLB feature—the required provision of supplemental educational services for children in public schools that are performing below minimum standards—that embodies these elements in its use of market-like mechanisms (i.e., choice and competition) to improve educational opportunities and increase student academic achievement. NCLB directs state and local educational agencies to identify and contract with third-party organizations that provide supplemental educational services (e.g., tutoring) consistent with state and local instructional programs, and states are required to withdraw approval from providers that fail to increase student academic achievement. Heinrich brings empirical and case study evidence to bear in investigating whether the use of market-like mechanisms and devolution of primary responsibility for public services delivery to third parties is able to achieve results where the public sector was identified as failing.

Using data from Milwaukee public schools, Heinrich analyzes the performance of supplemental educational service providers and the relationship between provider effectiveness and their market (student) share over time. Her study findings, showing that provider market shares were not highly correlated with estimates of provider performance or other relevant vendor characteristics, suggest that relying on parents and students alone to check the market through their choice of providers is unlikely to bring about the intended improvements in student academic achievement. Local educational agencies were not satisfied with their level of control over service quality and educational outcomes, and despite ambitious efforts to measure provider performance and hold them accountable for outcomes, in practice, control over the flow of funds was their primary lever for managing these third-party relationships.

In “The Role of Private Agents in Affordable Housing Policy,” Elizabeth Graddy and Raphael Bostic describe how high-profile failures in federally funded affordable housing construction brought an end to subsidized production programs in the 1970s and opened the door for the use of multiple policy instruments to encourage private developers (businesses and nonprofits) to produce affordable housing. This transformation in affordable housing policy, combined with a devolution of authority for implementation from the federal government to lower levels of government, has been carried out primarily via three policy instruments: the formal transfer of housing policy control from the federal government to the state and local governments through block grants; the increasing use of housing vouchers that allow recipients to access the private housing market; and the provision for tax credits that promote the production of low-income rental housing. The objective of the research by Graddy and Bostic research is to explore the consequences of these policy changes for public authority, responsibility, accountability, and effectiveness in affordable housing policy in the United States.

Graddy and Bostic conduct a comparative case study of two states, Massachusetts and New Jersey, that have adopted different governance processes and structures in the production of affordable housing. They also empirically explore how differences in incentives and governance processes, such as the extent of state government support for affordable housing development, lead to different affordable housing outcomes. Graddy and Bostic conclude that coherent statewide plans, whether generated by the judiciary or the legislature, can effectively overcome local resistance to affordable housing development, and both states’ plans likely fostered improvements in meeting the goal of affordable housing for all households. In many states, however, there is no mechanism for enforcement by regional or state-level authorities to ensure the local production of affordable units and no tracking of the production of these units at the state level. They suggest that this decentralized approach to affordable housing policy oversight not only generates little information with which to assess compliance but also likely perpetuates under-provision of affordable housing.

Since the passage of the Adoption and Safe Families Act in 1997 and performance-focused state laws, child welfare systems have been pressured to identify innovative service models that move foster children out of state custody efficiently, without compromising their safety or well being. In “Governance in Motion: Service Provision and Child Welfare Outcomes in a Performance-Based, Managed Care Contracting Environment,” Bowen McBeath and William Meezan describe how policymakers and public managers have sought to reorganize child welfare programs to improve performance, in reaction to claims that they are unnecessarily inefficient, inflexible, and unresponsive to child and family needs. Performance in child welfare systems has been traditionally measured by tracking the proportion of foster youth who are successfully placed and retained in permanent placements, such as reunification with their biological families or adoption.

McBeath and Meezan identify three mechanisms through which public or nonprofit organizations might affect child welfare outcomes in market-based environments: the technical and task environment surrounding frontline service provision (including staff structures and interorganizational relationships); the actual services provided by frontline caseworkers to children and families over their foster care stay; and service intensity. In their empirical analysis of foster children and families served by six nonprofit child welfare agencies in Wayne County (Detroit), Michigan, from 2001 to 2004, they examine the relationship between each of these three mechanisms and child welfare outcomes. They find significant differences in the intensity with which nearly all services were provided, with greater service-related resources (e.g., the development of parent-child visitation schedules and increased caseworker-child-parent interactions) expended to reunify children with biological parents than to achieve other outcomes. They also conclude, however, that agencies under performance pressures (via the performance-based, managed care contracting regime) choose service technologies and pressure caseworkers to serve clients in ways that negatively affect client outcomes and, over time, organizational and system performance.

In “Governance and Collaboration: An Evolutionary Study of Two Mental Health Networks,” Brinton Milward, Keith Provan, Amy Fish, Kimberly Isett, and Kun Huang conduct a comparative analysis of the evolution of two community mental health networks that have contracts with the State of Arizona. Although both operate under the same set of rules, one is governed by a for-profit firm that both produces services directly and buys them from a network of nonprofit agencies, and the other is managed by a community-based nonprofit that contracts with four separate nonprofit networks to offer services. They show how money flows from federal, state, and county sources into these systems.

In treating individuals with serious mental illness, the authors note that a continuum of care—in which patients are served in various domains, from medication and therapy to vocational rehabilitation and housing—is considered to be more effectively provided through a wide variety of agencies that collaborate through an integrated network of care providers. A primary question they address is: Does whether a network is governed by a for-profit or a nonprofit entity affect decision making in the provision of these services? Using social network analysis, they examine these two county systems at two points in time, comparing the more mature of the two networks to the newer one and the for-profit–governed system to the not-for-profit–managed system. They find that although there was a distinct difference in contracting behavior between the two networks, the informal relationships in each resembled one other, and both networks matured over time. They conclude that isomorphic pressures likely lead nonsectarian, nonprofit organizations (competing in a market economy with for-profit organizations) to behave like for-profits and that the sector of the governing organization plays a mostly insignificant role in the provision of quality services.

Based on interviews conducted with a variety of federal agencies that rely on third parties for the delivery of services or the production of knowledge, Dubnick and Frederickson argue that accountability for achieving the goals of a program is a function of the goodness of fit, or articulation, between parts of a program. Articulation, in turn, is a function of the quality and nature of the connections by which separate actors in the network are coupled and combined. This leads to their finding that program accountability is enhanced by relational contracting and network management. The authors also point out the irony of federal performance and accountability mechanisms that refuse to acknowledge that an agency can, by law, have conflicting goals or that performance is dependent on the actions of third parties operating under contract.

Finally, in “Fishing in Muddy Waters: Principals, Agents, and Democratic Governance in Europe,” Chris Skelcher describes the evolution of new forms of public and third-party governance in Europe—including supranational institutions and governance networks and partnerships with a high degree of autonomy in their specific jurisdictions—that are creating new challenges for democratic policy-making, accountability and control. He discusses how the theory of representative democracy, which assumes a principal-agent chain linking citizens, elected representatives, public managers, and third-party providers, comes up short as a framework for explaining how governance operates in Europe. An objective of his research is to explore alternatives to principal-agent theory and develop methodologies for democratic analysis that do not presume representative government.

In conducting this research, Skelcher draws data from a study of 27 governance partnerships in two local (UK) government areas, covering a wide range of policy issues such as economic development, crime reduction, health, and education. In addition, he also uses a criteria-based Governance Assessment Tool that was developed to assess the formal institutional features of each partnership. The data he collects are supplemented with interview data that help to illuminate the processes of institutional design and transformation and informal practices that emerge alongside the formal rules. Skelcher concludes that public administration scholars need both new theoretical frameworks and analytical tools for “fishing” in the “muddy water” of complex delegations, accountabilities, and authority in governance collaborations, partnerships, and networks, in order to better understand how third-party governments respond to the challenges of democracy (rather than taking representative democracy as a given.)

CONCLUDING THOUGHTS

In the research presented in this symposium, it is clear that government exercises very limited oversight of “agents of the state”—whether for-profit or nonprofit providers—that conduct the primary work of public service provision. Yet the organizational structures and incentives that governments establish to promote service quality, efficiency, and effectiveness appear to weigh heavily on service outcomes. In other words, even if they are not engaging directly in service provision, governments play a far more vital role than just funding the services. And government managers cannot expect the competitive market and its third-party entities to just work their “magic” in improving service outcomes without government input.

As public pressures for performance accountability have increased, public managers have introduced performance- and cost-related criteria into purchase of service contracts and fee-for-service agreements, sometimes (as in the case of child welfare) with unintended and harmful consequences. The research described in this symposium suggests that governments also cannot simply rely on assumed benevolent, quality-of-service motives of the nonprofit sector to ensure that citizen or client interests are being promoted and protected. Government mechanisms for enforcement and oversight of the public interest by regional, state-, and local-level authorities have often been weak or too far removed from the point of service delivery, but as in the case of affordable housing, they may still make a difference in promoting basic public values such as equity in access to public goods and services and accountability and responsiveness to citizens for service outcomes.

References