On Theorizing

In the early public administration literature, theorizing as it is now understood in the social sciences—conditional causal reasoning—is difficult to distinguish from political philosophy, or normative reasoning. Certainly that was true of pre-World War II writers, for whom creating a modern, competent administrative system at municipal, state, and federal levels was a compelling preoccupation. Many of those drawn to the field were, and still are, concerned as much with promoting effective democracy as with exploring, in a scientific sense, how our institutions actually work and with what consequences.

A recognizable intellectual curiosity was nonetheless evident from the beginning. How does our system of government work? Why does it work that way? What does the answer imply for the emerging practice of public administration? That curiosity was seldom dispassionate or “academic.” It reflected an increasing sense of urgency, driven by the rapid transformation of American society, about the importance of the systematic study of administrative institutions and practices toward the goal of making them effective.

Much early study was descriptive, albeit often insightfully so. Scholarship was also becoming increasingly analytical, however, employing both inductive and deductive logic. Inductive reasoning drew on observation and reflection as well as on the rapidly growing body of evidence being accumulated by the bureaus of municipal research and other early research organizations. Deductive reasoning reflected public administration's roots in law and its founders' familiarity with European legalistic and cameralist traditions. The recent literature of political economics, whether formal or narrative, is, in contrast, primarily deductive, although not entirely so, and it is undoubtedly more inspired by academic concerns than by the need for institutional reform. Nevertheless, as Anthony Bertelli and I argue in our recent book, Madison's Managers (Bertelli and Lynn 2006), both traditional reasoning and the more formal, deductive reasoning of recent years converge to remarkably similar insights concerning the significance of the separation of powers for administrative practice and accountability and, as well, for the need for institutions and mechanisms that ensure responsible agency.

In public administration, the particular roles of theory, and of academic scholarship in general, are, as I see it, threefold. First is to reduce the unwieldy complexity of administrative practice to causal propositions concerning what goes on beneath the surface of the world as we observe it. Second is to create models that illuminate a significant number of important contexts. Third is to be useful for predicting, or for supporting informed conjectures about, the probable consequences of acting in particular ways. That is, if you do this under the circumstances, then that is a likely or plausible result, or, alternatively, that certain desired states tend to require certain preconditions that are at least partly under the control of policy makers and public managers.

The presumption of theory-based research in this sense is that people, organizations, and other social actors are conditionally alike in certain ways and that conceptualizing these ways will help us derive empirical knowledge that will enable us to be better policy makers, managers, and clinicians. In other words, political and administrative behavior are not so indeterminate, and public and private motives not so inscrutable and unstable, that instrumental knowledge of considerable value to administrative practice is impossible to obtain.

I do not claim universality for any particular reductive approach to theorizing and knowing. My own view is that all methods produce heuristics—aids to learning, discovery, and problem-solving. Theological arguments about what is “the truth” are less useful than consideration of how different heuristics, however produced, improve professional reasoning and provide insight concerning political and administrative phenomena. No matter the nature of one's evidence, what matters, beyond technical skill and imagination, is that the books have not been cooked and that arguments have been crafted with transparent intellectual integrity. I do claim that the kind of theorizing and theory-based work I am about to discuss is among our most valuable heuristics for comprehending and improving administrative practice.

Traditional Theorizing

The practice of theorizing in a positive sense began, as I said, during the founding days of the profession. The following examples are representative and cumulate to a coherent argument.

In the preface to Congressional Government: A Study In American Politics, Woodrow Wilson wrote: “My chief aim in these essays, has been … an adequate illustrative contrast [of cabinet and congressional organs of government] with a view to making as plain as possible the conditions of federal administration” (1885, quoted by Stillman 1973, 584). Wilson concluded, in a positive sense, that “For all practical purposes the national government is supreme over the state governments and Congress predominant over its so-called coordinate branches.” But this was no anodyne finding. Congressional dominance, in Wilson's view, “parceled out power and confused responsibility” and thus inhibited the emergence of effective administrative institutions.

Frank Goodnow possessed a superb conceptual grasp of what he termed “the problem of administration” within the American constitutional scheme. Goodnow perfectly expressed the dilemma of reconciling the emerging need for administrative capacity with the constitutional imperative of political and judicial control:

Many years before his more celebrated works, in a law review article published as he was completing his Ph.D. at Harvard in 1924, John Gaus initiated, as far as I can tell, the popular practice of proclaiming “the new” in the world of public administration and management, in his case “the new administration” (Gaus 1923–1924, 220). By this term he meant a recognizable body of practice based on significant and increasing authority delegated to administrators to engage in discretionary actions that perforce posed a challenge to democratic control of the growing ranks of those unelected agents. Gaus also discussed judicial reaction against the growth of administrative power and the increasing control of administration by legislative and executive agents, including overhead agencies, thus sketching the relations among the branches with respect to the new administration.

A dozen years later, in the preface to The Frontiers of Public Administration, the authors (Gaus, Leonard White, and Marshall Dimock) wrote: “The student of public administration is now confronted with a literally overwhelming torrent of new administrative agencies and new problems to be assimilated and appraised” (1936, vii). That “[s]o many currents of experience from industry, from psychological research, from the study of comparative government, … to name but a few—are adding their contribution to the main stream require[s] both improvements in administrative technique and equally more accurate ideas concerning the nature of administration” (vii–viii). They wrote these words, it should be noted, when the field of public administration is said by some to have been baking to a crisp in the high noon of an inflexible and unreflective orthodoxy.

Just a few years later, White (1942) sounded some relatively new notes. “It is an arresting fact,” he observed, “that the great improvements in government since 1910⁴

have been designed to make the public service more effective in the management and direction of public affairs, not to keep it responsible or make it more amenable to control” (213). He continued: “Responsibility of officialdom to legislature and to courts, sometimes to the electorate, is a matter of law; … Responsiveness of officialdom to enacted policies, to broad public purposes, to sentiments and preferences, and to individual needs is not a matter of law but of the spirit. Both are important to the proper administration of laws in a democracy. Neither is achieved with full satisfaction by the formal relationship of dependency of administration on popular assemblies; the full responsiveness of officialdom to citizenry requires other and supplementary institutions of a more intimate nature.”

Of the textbooks of the 1930s by John Pfiffner (1935), Harvey Walker (1937), and White (1939), which codified the maturing field, Bertelli and I conclude in Madison's Managers that they reveal “a professional reasoning process that explores the interrelationships among the values of democracy; the dangers of an uncontrolled, politically corrupted, or irresponsible bureaucracy; the corruptibility of legislative processes; the imprecision of popular control of administration; and [the difficulties of designing] judicial and executive institutions that can balance capacity with control in a constitutionally appropriate manner” (Bertelli and Lynn 2006, 43).

Following World War II, public administration headed in divergent directions.⁵

Subsequent developments have produced both good news and bad news. The good news is that scholarship intended to “grasp the scheme of things entire” and to create a theory of practice that deepens our practical understanding of governance has been robust. The bad news—although just how bad depends on one's intellectual loyalties—is that much of this work is being done by scholars who do not identify themselves primarily with the field of public administration.

In his own Gaus lecture, James March (1997) criticized public administration's narrow concentration on administrative practice divorced from intellectual concerns. One reason for this narrowness seems clear. Following the attacks on traditional thinking by Dwight Waldo, Herbert Simon, Robert Dahl, Vincent Ostrom, and others, interest in structures, mandates, interests, bargaining, and the other institutional expressions of constitutional governance diminished in a field anxious to discover and describe new modes of practice and new ways of knowing about them, leaving an intellectual vacuum that others would soon fill.

The traditional mode of theorizing did not disappear entirely. Gaus's Reflections included observations that move the earlier tradition of theorizing forward. “[A]dministration is intermingled with the entire process of government” he argued, “and with the environment in which the people affected by the government exist” (125). He continues: “Government always presents this problem of balancing powers sufficient to enable a community to deal with its common problems on the one hand, with measures which will prevent or limit the evil effects of the abuse of power by the community's representatives” (134–5).

Gaus underscored the increasing resort to delegated authority to executive officials, with “the resources in knowledge and experience needed to apply the general principle to the localized situation.” “The lesson to us is … to concentrate our effort at political improvement at the point of the political agents … and the recruitment, education and training of the bureaucracy, and at procedures whereby their work and policies may be revised in the light of experience” (140). He continued: “The chief problem … is to get … agencies attached closely to the political leadership so that they participate naturally in the policy process as trusted and indispensable aids to that leadership and not as erratic particles following their own whims or as rival claimants to public attention” (142), in other words, as responsible agents.

In 1954, John Millett published what is the first public management book with a contemporary thrust: pragmatic and craft-oriented. At the same time, Millett regarded the political context as fundamental. “Are administrative agencies … to be regarded as a ‘fourth branch’ of government?” he asks. “I believe that they have no such exalted status. Rather, they are a kind of subordinate echelon of government subject in our scheme of things to the supervision of legislature, chief executive, and judiciary…. The administrator in the public service is concerned with all three, and ignores any one branch only at his peril. So it seems to me that the politics of public administration is concerned with how administrative agencies in our government are kept subject to popular direction and restraint in the interests of a free society, through the operation of three coordinate branches” (vii–viii).

The legacy of the early theorists is significant. They produced a coherent, synoptic perspective: given our constitutional scheme and its separation of powers, checks and balances, and federal structure, tensions between delegation and control are inevitable and not easily resolved. Ensuring the accountability and legitimacy of the administrative state is, therefore, both essential and problematic. Thus the founders of public administration laid hold of the theoretical problem at the core of public administration and management: depicting responsible administration within a constitutional scheme that forces administration to justify itself anew.

The New Political Economy

The vacuum in theorizing about administrative practice within our constitutional scheme that that I noted earlier has been filled by “the new political economy.” This style of thinking has won a considerable, and international, following. British political scientist Patrick Dunleavy (1992) lauds “the development and refinement of public choice into behaviourally realistic and theoretically diverse explanations of broad classes of political phenomena” (259). Canadian political scientist Peter Aucoin (1990) says that “[t]he public choice paradigm emphasizes the role to be played by political authorities as elected representatives in governance. It does not admit to a policy/administration dichotomy that would carve out spheres of responsibility for politicians on the one hand and bureaucrats on the other” (126–7). The public choice approach, says Aucoin, “sees politics as pervading management, that is, politics as being present in both the formulation and the implementation of policies.”

I want to note briefly some of the more synoptic contributions to the political economy of administrative practice. They illustrate how the administrative system is, as the founders understood, an artifact of political choice. More precise elaborations are to be found in specialist literatures.

To begin at the logical beginning, a synthesis of neoclassical and institutional economics (prominently including the work of Douglass North) is employed by Icelandic economist Thráinn Eggertsson (1990) to explain the purpose of the state, in a cameralist spirit, as creating a regime of property rights and transaction costs that enables it to maximize its technical economic potential. “[T]he willingness of individual owners to supply specific appropriable assets, essential for economic growth and full utilization of advanced technologies,” Eggertsson argues, “depends directly on the social rules structure, including the availability of relatively consistent and impartial dispute processing by a third party, which in most cases can be supplied only by the state” (320).

The logic of administrative practice then emerges from the creation of state constitutions. Dutch political scientist Jan-Erik Lane's Constitutions and Political Theory analyzes the existence, origins, and consequences of constitutions (1996). He argues that constitutions enhance national stability by recognizing unalterable rights, by creating veto players, and by creating highly qualified majorities to amend it. He notes, however, that these kinds of features of constitutions may tend to privilege the status quo at the expense of more participatory and responsive forms of democracy. Other students of endogenous institutions are elaborating on such entry-level ideas.

The administrative system then evolves, according to Norwegian political scientist Kaare Strøm (2000), from “a chain of delegation, in which those authorized to make political decisions conditionally designate others to make such decisions in their name and place” (266). Agents are accountable to their principals if they are obliged to act on the latter's behalf, and if the latter are empowered to reward or punish them for their performance in this capacity. The citizens' role is to control officials, but when we “consider the incumbents of political office as agents of the citizens, we have to acknowledge that they are constrained and frequently common agents, whose responsibilities may thus be manifold” (268). Scholars of the policy process, credible commitments, and related topics are further enriching the delegation literature.

American political economists, including Terry Moe, Mathew McCubbins, Barry Weingast, Elinor Ostrom, and many others, have explored the logic and implications of the delegation of sovereign authority to administrative entities, the assignment of tasks among agencies, and the principal-agent and control problems that result. In an important series of essays, Moe argues that “political rationality” results in public programs and organizations that are technically irrational. “As a result of regime restraints and the politics they authorize, the public manager may have to deal with inadequate resources, unreasonable or unrealistic workload or reporting requirements, inconsistent guidance, or missions defined so as to be virtually unachievable” (Lynn 2003, 21).

Problems of agency deepen as the chain of delegation reaches into government organizations themselves. Patrick Dunleavy (1992) analyzes the behavior of people inside such organizations. In his model, “bureaucrats' preferences … are endogenously determined within the budget-setting and bureau-shaping processes which underlie agencies' activity” (254). Bureau-shaping refers to activities directed at influencing the nature of the work and the work environment, with associated consequences for administrative practice and its outputs. Senior officials do not act like a unitary actor because they choose differently between individual or collective modes for improving their welfare. Because bureau-shaping presents fewer collective action problems than, say, budget maximizing, it is a superior way of pursuing personal goals.

This line of theorizing is hardly restricted to sterile propositions concerning what may seem implausibly rational actors. In a more sociological view of what Dunleavy calls “bureau-shaping,” John Brehm and Scott Gates (1999) argue that organizational cultures form when subordinates look to each other for appropriate behaviors. Cultures are common stocks of information. “Subordinate performance depends,” they argue, “first on functional preferences, second on solidary preferences, and lastly on the efforts of the supervisor” (195).⁶

They continue: “The chain of command looks to be a weak mode for principals—ultimately, democratic publics—to influence the performance of bureaucratic agents” (196). Further, “the strongest and most positive role that our [empirical] research identifies for the potential of supervisory control over bureaucracy is in the process of recruitment” (202), an argument developed at length in Madison's Managers.

In the face of such agency problems, Gary Miller (1990) offers a theory of leadership that is analytical rather than normative. Miller argues, first, that no system of incentives can entirely eliminate the non-cooperation that inhibits organizations from achieving their goals or priorities. Applying repeated game theory, he notes that organizational behavior yields indeterminate outcomes in the form of multiple equilibria. The role of the manager is, in effect, to select one of these equilibria and induce people to cooperate in its achievement. This process constitutes leadership, which consists in creating mutually reinforcing expectations concerning the benefits of cooperative behavior and teamwork on behalf of shared objectives.

Against the temptation to overstate the importance of bureau-shaping, cultures, and leadership, German political scientist Arthur Benz reminds us that,

In a similar vein, Dutch political scientists Peter Bogason and Theo Toonen (1998) are concerned that “[o]rganizations, law, constitutions, power, courts, governments, formal mandates and legal competencies” are disappearing from the study of public administration (208).

Finally, in these comments on recent literature, I want to note efforts by my colleagues and I to use this line of theorizing to gain sights from the very large, often unappreciated empirical literature concerned with public administration and management. Carolyn Heinrich, Carolyn Hill, and I (Lynn, Heinrich, and Hill 2001) conceptualize governance as a system of hierarchically-ordered institutions, created along the chain of delegation, which has a significant causal effect on administrative system outputs. Melissa Forbes and I have incorporated concepts of organizational effectiveness into this logic of governance (Forbes and Lynn 2007). Doing this highlights the fact that policy structures and processes both constrain and enable managerial actors, who are, then, both agents and principals in the administrative system, playing a mediating role between policy-making and service delivery. We have used this logic to reveal what the many research communities concerned in one way or another with the workings of our administrative system are learning and how they are learning it (Hill and Lynn 2005; Forbes and Lynn 2005; 2007; Forbes, Hill, and Lynn 2006; 2008).

Critical analysis of the empirical literature within a logic of governance into which organizational effectiveness has been incorporated suggests some hypotheses about the theory of administrative practice: (1) public agency direction, the function of public management, is governed primarily by the structures incorporated in public policies, but not to the point of extinguishing managerial initiative; (2) the outputs of service delivery processes are influenced primarily by public policies as mediated by public management (specifically, by managerial choices of structures and processes that shape the links between policy and service delivery), although outputs may also be determined directly by public policies if policies are designed so as to be, in effect, self-executing; and (3) the outcomes of public policies are determined by the outputs of service delivery processes, so that the study of outputs, not outcomes, is central to a theory of administrative practice.

The literature of political economics, then, 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 subfield itself cannot account for all of the particular factors that might contribute to government performance at all times and places. The determinants of that performance are, as my colleagues and I have shown elsewhere, multifarious and influenced by context. But Kenneth Arrow (1985) once argued on behalf of the principal agent model, that, as a heuristic, it accommodates a wide variety of insights into hierarchical relationships. In a similar vein, the new political economy can accommodate and put into broad institutional perspective a wide variety of contributions from many fields and disciplines while recognizing the most fundamental aspects of how political institutions frame and contribute to administrative practice, and how administrative practice becomes influential in its own right within that framework.