Week 1 - Assignment 1: Write a Memo About Basic Concepts and Theories Related to Public Management and Week 2 - Assignment: Analyze a Case Study on Management and Organizational Behavior

The_Public_Administration_Theory_Primer_----_3_-_Theories_of_Bureaucratic_Politics.pdf

Home >Business & Finance homework help >Accounting homework help >Week 1 - Assignment 1: Write a Memo About Basic Concepts and Theories Related to Public Management and Week 2 - Assignment: Analyze a Case Study on Management and Organizational Behavior

40 2: Th eories of Political Control of Bureaucracy

foundered, by common consent, but for various alleged reasons: It was em- pirically untrue to what happens and is impossible to operationalize; it was presumptuous if not impious, putting profane hands on a sacred scheme; it concealed ethical problems and encouraged illegal action. So, formally or os- tensibly, we put the dichotomy aside. But at the same time, it lingers, both as an idea and as a practice. And I don’t judge the lingering as simple inertia, a cultural lag. Th e twofold schema has too much going for it in logic and useful- ness simply to disappear. We do, commonsensically, decide and execute, set policy and administer. (1986, 153)

Th eories of political control of bureaucracy are, in sum, among the most em- pirically robust and theoretically elegant in public administration.

Frederickson, H. George, et al. The Public Administration Theory Primer, Taylor & Francis Group, 2015. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/ncent-ebooks/detail.action?docID=2039766. Created from ncent-ebooks on 2021-09-21 15:42:36.

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

3 Th eories of Bureaucratic Politics

Introduction: What Are Th eories of Bureaucratic Politics? Th eories of bureaucratic politics seek to explain the policymaking role of ad- ministration and bureaucracy. Such frameworks typically reject the politics- administration dichotomy underpinning theories of bureaucratic control, viewing this division as an analytical convenience that imposes too steep a cost on theoretical development. Specifi cally, the price of making theory more tracta- ble by separating administration from politics is held to be a willful ignorance of the central role of bureaucracy within the polity’s power structure.

Since bureaucracies and bureaucrats routinely engage in political behavior, the need to account theoretically for the bureaucracy’s political role is justifi ed. Politics is generically defi ned as the authoritative allocation of values, or the pro- cess of deciding “who gets what, when and how” (Easton 1965; Lasswell 1936). Numerous studies confi rm that bureaucracies and bureaucrats routinely allocate values and decide who gets what, that bureaucracies logically engage in “politics of the fi rst order” (Meier 1993, 7). Th eories of bureaucratic politics therefore be- gin by accepting what has long been empirically observed; that is, in practice, ad- ministration is not a technical and value-neutral activity separable from politics. Administration is politics (Waldo 1948).

Accordingly, theories of bureaucratic politics seek to breach the orthodox di- vide between administration and politics and attempt to drag the former into a systematic accounting with the latter. Th at traditional theoretical frameworks ac- count poorly for bureaucracy’s obvious and repeatedly observed political role has long been recognized. Even scholars traditionally credited with describing and supporting the politics-administration divide were well aware of the political role the bureaucracy plays, and the rigidity of the division accepted as their legacy has been described as a caricature of their arguments. Woodrow Wilson and Frank Goodnow, who both wrote at a time when public bureaucracies were ripe with patronage, incompetence, and even outright corruption, were well aware that

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

42 3: Th eories of Bureaucratic Politics

politics and administration represented a synthesis rather than two neatly separa- ble portions of the public policy enterprise (Lynn 2001). Other prominent public administration scholars argued during the fi rst half of the twentieth century that administrative theory had to account for politics, both in recognition of bureau- cracy’s real-world role and as a necessary element to building better explanatory frameworks within the discipline.

Among the most prominent of these was John Gaus (1931). He observed that federal agencies not only carried out clearly understood directives from Congress but also independently shaped those directives and exercised discretionary poli- cymaking authority while translating the vague intentions of statutes into specifi c government actions. Bureaucracy obviously wields political power. Th is being so, those who sought to understand public agencies could not simply carve off administration from politics and leave the complexities of the latter to political theorists. If bureaucracies were helping to determine the will of the state, they were inescapably political institutions, and Gaus argued that administrative the- ory ignored this fact at its peril. Most famously, in the fi nal sentence of an essay in Public Administration Review, he threw down an implied gauntlet to those who would fashion a theory of administration: “A theory of public administra- tion means in our time a theory of politics also” (1950, 168). Gaus thus succinctly summarized the purpose of theories of bureaucratic politics.

As the broader intellectual history of political theory makes clear, this is a dif- fi cult objective, and for more than half a century students of public administra- tion have had mixed success in meeting Gaus’s challenge. Th e issues raised here are more complex than those at the heart of the theories of bureaucratic control. Th e goal is not to locate the dividing line between politics and administration because no such line exists, nor is it to ascertain how bureaucracies can be made accountable to their democratic masters, although this is a question of some im- portance to theories of bureaucratic politics. Questions of political power are the central focus:

• To what extent do administrative processes, as opposed to democratic pro- cesses, determine public policy?

• Who controls or infl uences the exercise of bureaucratic power? • What is the role of bureaucracy in representing and advancing the goals of

particular clientele groups or organized interests? • To what extent do elective institutions and elected offi cials seek to shape and

control administration as a means to advance their own political interests? • What is the source of bureaucratic power? • How does the important political role of nonelected institutions based in

hierarchy and authority square with the fundamental values of democracy?

If anything has been learned by the eff orts expended on developing theories of bureaucratic politics, it is that such questions have no easy answers.

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

43Administrative Th eory as Political Th eory

Nevertheless, numerous studies have confi rmed the need for systematic frameworks accounting for the political role of the bureaucracy. Several re- sponses to this need have provided important insights into the political role of the bureaucracy, and in doing so, signifi cantly expanded our understanding of public administration.

Administrative Th eory as Political Th eory Th e seminal work that justifi ed the need for a theory of bureaucratic politics is Dwight Waldo’s Th e Administrative State (1948). Waldo did not construct a the- ory of bureaucratic politics in this book, but here and in later writings he made two critical contributions that have supported all subsequent eff orts to do so. First, he undertook a devastating critique of the extant research literature. He argued that public administration scholarship revolved around a core set of be- liefs that cumulatively served to constrain theoretical development. Key among these were the beliefs that effi ciency and democracy were compatible and that the work of government could be cleanly divided into separate realms of decision and execution. Th ese beliefs led public administration scholars to champion effi ciency as the central goal of public agencies, to develop a “science” of administration to maximize that effi ciency, and to ignore the political ramifi cations of these beliefs and the prescriptions they implied.

Second, and probably more important, Waldo argued that administrative scholarship was itself driven by a particular philosophy of politics. A good por- tion of Th e Administrative State is devoted to examining the scholarly public administration literature through the lens of fi ve key issues in political philos- ophy: (1) the nature of the Good Life, or a vision of what the “good society” should look like; (2) the criteria of action, or the procedures for determining how collective decisions should be made; (3) the question of who should rule; (4) the question of how the powers of the state should be divided and appor- tioned; and (5) the question of centralization versus decentralization, or the relative merits of a unitary state versus a federal system.

Waldo concluded that public administration scholarship was anchored by well-developed responses to all of these issues. Like theorists from Machiavelli to Marx, public administration scholars had a vision of what the “good society” looks like: It is industrial, urban, and centrally planned; it has no poverty, no cor- ruption, and no extremes of wealth. Science is its ideal, and waste and ineffi ciency are its enemy. Th ese same scholars also had a clear preference for the criteria of action: A scientifi c analysis of the facts should decide what should be done. Public administration orthodoxy espoused particularly fi rm beliefs about who should rule: “Th e assertion that there is a fi eld of expertise which has, or should have, a place in and claim upon the exercise of modern governmental functions—this is a fundamental postulate of the public administration movement” (1948, 89–90). Technocrats blessed with the requisite competence and expertise were public

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

44 3: Th eories of Bureaucratic Politics

administration’s equivalents of the Guardians in Plato’s Republic. On the par- ticularly American issues of the separation of powers and centralization versus decentralization, Waldo argued that the preferences of administration scholar- ship were equally clear: Administration scholars were hostile to the tripartite partition of power in the American system and sought to increase the power of the executive at the expense of the judiciary and the legislature. Th ey were also in favor of a centralized state. Th ey placed their faith in the competence of a professional administrator, who, given the requisite power and authority, could tackle the obstacles standing before the realization of the good life.

If administration scholarship advanced such a distinct and defi nable political philosophy (some might say ideology), it raised an immediate and formidable intellectual obstacle to attempts at conceptually dividing politics and adminis- tration: How could students of administration claim that politics was largely ex- ternal to their interests when their intellectual history revealed such a systematic value-based philosophy of government? Waldo pointed out that administration is frequently claimed to be at the core of modern democratic government, and that this claim helps justify the entire discipline of public administration. If this claim has merit, it implies that democratic theory must deal with administration, and that administrative theory must deal with democratic politics. As a practical matter of explaining the operation and role of administration in government, not to mention as a point of intellectual honesty, students of administration cannot deal with the problems of politics by assuming them away.

Waldo argued that administrative scholarship’s failure to incorporate poli- tics explicitly into its theoretical development was a product of its early cultural and intellectual environment. While recognizing the impossibility of cleanly dating the beginning of public administration scholarship as a self-conscious body of thought, Waldo took as his starting point such writers as Woodrow Wilson, Frank Goodnow, and Frederick W. Taylor, namely, infl uential man- agement, administration, and organization theorists who wrote near the turn of the twentieth century. Th e work of these scholars refl ected not only the dom- inant cultural values of their time but also the contemporary problems in ad- ministration they sought to address. Cultural values led them to accept science as the surest path to knowledge and commerce as the central activity of society. Th e central problems they sought to address consisted of an unappetizing stew of ineffi ciency marinating in political cronyism and seasoned with graft .

One of the outcomes of these contextual forces was that, from the beginning, students of administration adopted effi ciency as their guiding principle. Th e term was vaguely defi ned, though “effi cient administration” clearly meant “good ad- ministration.” When administration scholars operationalized the concept, they mainly seemed to be talking about an input-output ratio, the most output for the least input being the implied objective (Waldo 1948, 201–202). A “good” decision or administrative act was thus one that maximized outputs for a given set of inputs. As Waldo pointed out, this is a concept fundamental to businesses

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

45Administrative Th eory as Political Th eory

operating in capitalistic markets, but it is not nearly so important to democratic government. Equity, consensus, or the satisfaction of particular interests is fre- quently the criterion for action in democratic processes, and none of these crite- ria is necessarily effi cient; indeed, these criteria are oft en inescapably ineffi cient.

Yet, as administration scholars accepted effi ciency as their central principle, they also accepted democracy—a notoriously ineffi cient basis of organization—as the central principle of the American political system. Th is presented a problem in developing administrative theory. Th e formative era of administrative schol- arship, with its focus on the scientifi c method, its guiding principle of effi ciency, and its position in the shadow of business, meant that it developed in a decidedly undemocratic context. Not only was democracy not synonymous with effi ciency and various other business and scientifi c practices incorporated into public ad- ministration orthodoxy, but also it was quite possibly hostile to them (Waldo 1952, 85). How could the principle central to the American political system be squared with the forces driving the theoretical development of public administra- tion as a discipline?

Waldo argued that the founders of public administration solved the conun- drum by accepting democracy as the guiding principle of the American po- litical system, but keeping it external to their professional interests through the politics-administration dichotomy. By separating the work of government into two distinct operations and limiting their attention to the “nonpolitical” element, administration scholars were free to push for centralized power in the executive branch, to prescribe hierarchical and authoritarian bureaucracies as the basis for organizing public agencies, and to call for passing greater responsibilities to the technocrat. As long as these reforms increased effi ciency in administration, and administration was kept separate from politics, theoretically the discipline did not have to square the contradictions these arguments presented to the egalitar- ian ideals of democracy.

As Waldo was careful to point out, the founders of public administration were not ideologically opposed to democracy. Th ey were progressive reform- ers who embraced the romantic ideal of democracy as the “best” or “proper” form of government. Th e reality they faced at the time, however, was a public administration characterized by disorganization, amateurism, and dishonesty. Nineteenth-century reforms springing from the presidency of Andrew Jackson had dispersed and factionalized the power of government. Elected offi cials multi- plied, the legislature took precedence over the executive, and government agencies were staff ed through the spoils system. If administration were the core of govern- ment, the net result of these reforms created a serious problem for democracy in the United States at the turn of the twentieth century: a public administration shot through with incompetence, ripe for corruption. For public administration to gain competency and effi ciency, it would have to cleanse itself of politics and learn the lessons of science and business. Good administration (and thus good government) could best be promoted by centralizing and concentrating power;

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

46 3: Th eories of Bureaucratic Politics

by running agencies according to sound, scientifi c management principles; by making technical competence the criterion for civil service employment; and by shielding these technical experts from whatever winds happened to be stirring the dust in the political arena.

Waldo thus viewed the political philosophy inherent in public administration scholarship not as an attempt to usurp democracy, but as a necessary corrective to save it. As Waldo put it, “Democracy if it were to survive, could not aff ord to ignore the lessons of centralization, hierarchy, and discipline. Put bluntly, it was the maxim ‘Autocracy during hours is the price of democracy aft er hours’” (1952, 87). Th eoretically, the undemocratic elements of administrative orthodoxy—its emphasis on effi ciency, hierarchy, and authority—could be seen in the service of democracy as long as the politics-administration orthodoxy held. An effi cient and expertly run administrative apparatus insulated from politics and under the authority of a powerful executive would increase accountability and promote ef- fective, competently run public programs and policies. If things did not work, ev- eryone would know whom to blame and why, and the representative institutions of democracy could act accordingly.

Th e problem, as numerous scholars have pointed out, was that the politics- administration dichotomy did not hold. As Waldo meticulously detailed in his literature review, there was ample evidence that bureaucracies pushed some val- ues over others, that bureaucracies acted as power brokers among competing special interests, and that lawmakers were increasingly reliant on and infl uenced by the expertise and opinions of administrators. Administrative theory simply could not ignore these realities and continue to usefully shape the direction of the discipline. At a minimum, Waldo argued, the concept of democracy and all its messy implications had to be brought back into administrative theory. Ad- ministrative scholars had to recognize that their central principle—effi ciency— was not value neutral, and that its uneasy relationship with democratic principles had to be recognized (Waldo 1952, 90).

Waldo suggested that continued attempts to create a science of adminis- tration would result in theoretical dead ends because “science” was, in eff ect, a code word for preserving the core principle of effi ciency, a signal for another attempt to inoculate administration against politics. In an essay in the Amer- ican Political Science Review, Waldo singled out Herbert Simon’s argument separating questions of administration into issues dealing with fact and issues dealing with values. Simon’s enormously infl uential Administrative Behavior (1947/1997) had essentially demolished the extant research seeking to defi ne and promulgate the “principles” or “laws” of a science of administration. Yet Simon sought to save the possibility of that science. He argued that it was con- ceivable if it limited its attention to decisions centered on facts (statements that can be tested to assess whether they are true or false) as opposed to values (statements that are validated by human fi at). Decisions of fact were central to the administrative realm, Simon argued, and could be scientifi cally guided

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

47Allison’s Paradigm of Bureaucratic Politics

toward the overall goal of effi ciency. Waldo said that Simon was simply recast- ing the problem by substituting a logical division of politics and administration for an institutional division, and was doing so to preserve the central principle of orthodox administrative theory, namely, effi ciency.

Effi ciency could not remain the discipline’s talisman against politics, Waldo argued, because administration is political. In Waldo’s perspective, effi ciency it- self is a political claim. For example, how does one assess the effi ciency of, say, a library, or the Department of Defense? If effi ciency is defi ned as an input-output ratio, one has a choice of inputs and outputs to assess effi ciency in both instances, although none is the unassailably objective “factual” option. As choosing among these options unavoidably involves values not just facts, effi ciency can hardly be value neutral (Stone 2002, 65). If public administration insisted that its orthodox principles were politically neutral, Waldo argued, it would never be rid of the theoretical straitjacket it used to restrain itself from the world of politics. Wal- do’s argument bought a tart response from Simon (1952b), but even as Simon went on the off ensive, there were signs that Waldo’s point had sunk deep into the discipline. Published concurrently with Simon’s essay was another by an equally prominent scholar—Peter Drucker (1952)—who wholeheartedly agreed with Waldo’s assessment of the fundamentally political character of large-scale organi- zations, and suggested that, if anything, Waldo had not pushed his arguments far enough (see Simon 1952a for the complete essay on this point).

Waldo argued that at the heart of the problem with administrative theory is a version of the problem James Madison struggled with in Federalist No. 10: How do you preserve individual liberty without destroying the freedoms that make it possible? For Madison, it was the dilemma of constructing a government strong enough to protect individual liberty without making it vulnerable to the forces that would crush the liberties of others for their own selfi sh interests. For Waldo, “Th e central problem of democratic administrative theory, as of all democratic theory, is how to reconcile democracy . . . with the demands of authority” (1952, 102). How do we construct a theory that accommodates the hierarchical and authoritarian nature of the bureaucracy, the foundation of the modern adminis- trative state and a seemingly necessary component of contemporary government, with the seemingly contradictory egalitarian, ineffi cient ideals of democracy? Waldo bestowed this grand and sweeping question upon the discipline rather than provide its answer, but the question is surely enough to justify the need to meld administrative theory with political theory, to motivate the search for a theory of bureaucratic politics.

Allison’s Paradigm of Bureaucratic Politics In the two decades following the publication of Th e Administrative State (Waldo 1948), an embryonic theory of bureaucratic politics began to emerge from a series of studies examining decisionmaking in the executive branch. Th e

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

48 3: Th eories of Bureaucratic Politics

signifi cant claim generated by these studies was that government decisions were products of bargaining and negotiation among interested political actors. As these studies focused on the executive branch, the central player in these bar- gaining frameworks was the president. Th e president, however, was argued to have little unilateral decisionmaking power; he had to accommodate the inter- ests of the various institutional factions in the executive branch. Bureaucracies and bureaucrats, in short, played high-level politics, and usually played the game very well.

Th ese studies were discursive rather than explicitly theoretical, but the par- allels between them and the contemporary work on game theory—a highly for- malized and mathematical approach to explaining behavior—are unmistakable. Th e loose bargaining framework adopted by this research quickly proved a use- ful way to organize empirical research and produced many of the raw materials for a more comprehensive theory. Th e best-known studies of this early bureau- cratic politics literature include Samuel Huntington’s Th e Common Defense (1961), Warner Schilling’s 1962 essay on the politics of national defense, and, most famously, Richard Neustadt’s Presidential Power (1960). Bureaucracies and executive branch offi cials were not portrayed here as neutral agents of imple- mentation, but as active participants in determining the will of the state. Th ese studies steadily built a case for a general theory of bureaucratic politics centered on bargaining games in the executive branch.

Th e fi rst serious comprehensive attempt to produce such a framework was undertaken by Graham Allison in his book Essence of Decision (1971), and fur- ther refi ned by Allison and Morton Halperin (1972). Allison’s immediate focus in Essence of Decision was explaining why the governments of the United States and the Soviet Union did what they did during the Cuban missile crisis. With a nuclear exchange at stake, these were policies of particular importance, but Alli- son was aiming well beyond the confi nes of one case study. Essentially, he posed a broad question that cut to the heart of bureaucratic politics: Why do govern- ments do what they do? In other words, how is policy made, and who determines or infl uences it? To provide general answers to these questions, Allison articu- lated three theoretical models.

Th e fi rst was the rational actor model (what Allison termed “Model I,” or the classical model). Model I proposes that government decisions can be understood by viewing them as the product of a single actor in strategic pursuit of his own self-interest. Th e second model is the organizational process paradigm, or Model II, which argues that numerous actors are involved in decisionmaking, and de- cisionmaking processes are highly structured through standard operating pro- cedures (SOPs). When a problem occurs, Model I assumes that the government will identify the potential responses to that problem, assesses the consequences of those actions, and choose the action that maximizes benefi ts and minimizes costs. In contrast, Model II assumes that the government will rely on organizational routines instead of a rational cost-benefi t calculus to make that decision. Rather

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

49Allison’s Paradigm of Bureaucratic Politics

than search for all potential responses, the various components of government will act according to SOPs that, in eff ect, say, “When X happens, do Y.”

Th e seminal contribution of Allison (1971) and Allison and Halperin (1972) to the bureaucratic politics literature came in the articulation of an alternative to Models I and II. Model III, or the bureaucratic politics paradigm, explains government actions as the product of bargaining and compromise among the various organizational elements of the executive branch. Allison’s model of bu- reaucratic politics is constructed from four basic propositions. (1) Th e exec- utive branch is composed of numerous organizations and individuals having divergent objectives and agendas. Any given issue will attract the attention and involvement of a set of these actors, who bring to that issue their divergent in- terests and motivations. (2) No preponderant individual or organization exists; in other words, no one actor in the executive branch is able to act unilaterally. Th e president might be the most powerful actor on a given issue, but he will not be the only actor, and his infl uence will be limited. (3) Th e fi nal decision is a “political resultant”; in other words, what the government decides to do is the outcome of bargaining and compromise, the product of a political process. (4) Th ere is a diff erence between making policy and carrying it out. Once an action is decided upon, the task of implementing that decision is handed over to others who must also make decisions about the specifi c actions to take. Th ose decisions are in turn shaped by the operating procedures and interests of the implementers (Rosati 1981).

With these as a starting point, a policy analyst’s attention is immediately fo- cused on power and politics within and among executive branch bureaucracies. Within the confi nes of the executive branch, Allison’s model combines and makes little distinction between politics and administration, and in doing so seems to answer the challenge laid down by Gaus. In studying bureaucracy, as Allison put it, “the name of the game is politics: bargaining along regularized circuits among players positioned hierarchically within the government. Government behavior can thus be understood . . . as a result of these bargaining games” (1971, 144). Model III sees the components of the executive branch as semiautonomous orga- nizations that do not act in unison on a series of single strategic issues, but act on a variety of issues according to their own conceptions of national, organizational, and individual goals. Instead of making policy and implementation decisions ac- cording to rational self-interest, or according to the dictates of SOPs, government actors decide on the basis of the “pulling and hauling” that is politics.

From its general premises, Model III systematically goes about explaining spe- cifi c policies by seeking the answers to a few basic questions. (1) Who plays? In other words, what agencies or individuals have an important stake in a given is- sue, and whose behavior can have an important eff ect on government decisions and actions concerning that issue? (2) What determines each player’s stand? Th is question stems from Allison’s proposition that “where you stand depends on where you sit.” Diff erent agencies and individuals will have diff erent perceptions

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

50 3: Th eories of Bureaucratic Politics

of an issue and divergent preferences on what should be done based on their ob- jectives, values, and sense of mission. (3) How are players’ stands aggregated to yield governmental decisions and actions? Once it has been determined who are involved in a given issue and what their interests and objectives are, the challenge is to assess how these actors bargain to protect and advance their preferences. Th is means determining the relative infl uence of the players. Model III assumes that bargaining is highly structured and that “action channels” or “rules of the game” shape the process of decisionmaking and distribute power among players (Allison and Halperin 1972).

Allison’s model of bureaucratic politics has had a signifi cant impact on how bureaucracies are studied. It was not just a series of propositions formulated to explain one study, but rather a workable theory for understanding the policymak- ing role of bureaucracy. As such, the contribution of Model III to the theoretical development of public administration scholarship is hard to underestimate. Yet, although Allison undoubtedly reduced the paucity of theoretical substance in the fi eld, Model III has had mixed success as a general framework for the study of bureaucratic politics. Model III is predicated on a series of intuitively appealing assumptions: Government actions are the product of bargaining among the or- ganizational components of the executive branch, these actors have their own parochial interests, and their ability to translate those interests into policy is de- termined by their role in decisionmaking. Th ese assumptions logically lead to testable propositions: Policy outcomes will refl ect the parochial interests involved in the bargaining game, they also will refl ect the relative power of the players involved in the game, and the power of the players will be determined by the “action channels,” or regularized processes, used to structure decisionmaking. Unfortunately, subsequent scholarship has raised doubts about the empirical va- lidity of these hypotheses and the conceptual structure that supports them (Rosati 1981; Rhodes 1994; Bendor and Hammond 1992).

Politics, Power, and Organization Ironically, one of the implications of other work on bureaucratic politics is that Allison’s Model III was, if anything, too limited in scope rather than too am- bitious. In particular, Allison’s framework left important organizational issues underdeveloped, and, like the majority of the studies the framework sought to synthesize, it was almost exclusively focused on the executive branch. As some scholars were to make clear, bureaucratic politics is not confi ned to bargaining games within the executive branch; it is a fundamental component of a broad power structure that includes Congress, the courts, organized interest groups, in- tergovernmental relationships, and the public at large. Th e nature and context of this power structure, and the role and relative infl uence of bureaucracies within it, are heavily dependent upon organizational issues. How bureaucracies are or- ganized has been persuasively argued to play an important role in determining

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

51Politics, Power, and Organization

how power is distributed among various actors within the political system and in explaining how bureaucracies infl uence policymaking.

Th ere are two key organizational dimensions to bureaucratic politics theory. Th e fi rst deals with behavior. Th e primary goal here is to explain why bureaucrats and bureaucracies do what they do. Th e general presumption is that bureaucra- cies pursue important public missions and make numerous policy decisions, yet have only vague guidance from statutes. If legislatures, the institutions formally responsible for the goals of public agencies, only partially account for what bu- reaucracies do and why they do it, what explains the rest? Th e second deals with institutional structure and the distribution of power. Th e primary goal here is to understand how a bureaucracy’s formal lines of authority, its relationship to other institutions, and the programs and policies placed within its jurisdiction all combine to determine the relative political infl uence of a broad range of political actors.

Explanations for the political behavior of bureaucracy and bureaucrats have deep roots in the organization theory literature. For example, Robert K. Merton (1957) argued that institutions structured as classic bureaucracies shape the per- sonalities of the people who work for them. A bureaucratic environment, Merton argued, pressures people to conform to expected patterns of behavior—to follow rules, to be methodical and detailed. Given these pressures, bureaucracies will oft en substitute rules for ends, and they will adhere to SOPs even when those pro- cedures clearly interfere with the organization’s main mission. William Whyte Jr. echoed a similar theme in his work Th e Organization Man (1956). Whyte’s research detailed the willingness of managers in US corporations to adopt the goals of the organizations they worked for as their own, to subsume their per- sonalities into the larger organizational environment of their employment. Sim- ilar arguments about the pathologies of bureaucratic behavior have resurfaced more recently in such infl uential works as David Osborne and Ted Gaebler’s Reinventing Government (1992). Here the argument is that conscious eff ort is needed to break away from the predilection to see rules as ends, and instead shift to an outcomes-oriented entrepreneurial bureaucracy.

If organizational structure shapes the behavior of particular institutions and the individuals within them, this has broad implications for those seeking to ex- plain the policymaking role of bureaucracy. If bureaucrats make decisions that authoritatively allocate values, and organizational environment helps determine how those decisions are made, then organizational theory holds the potential to explain a good deal of how and why bureaucracy fulfi lls its political role.

One of the key contributions of organizational behavior scholarship to bureau- cratic politics theory is James Q. Wilson’s classic, Bureaucracy: What Government Agencies Do and Why Th ey Do It (1989). Wilson posed a similar question to Alli- son, though it was more focused toward administrative matters. Instead of asking why governments do what they do, Wilson asked why bureaucracies do what they do. Wilson argued that bureaucrats have discretion in their decisionmaking, and

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

52 3: Th eories of Bureaucratic Politics

that a complex set of factors determine how that discretion is exercised: “When bureaucrats are free to choose a course of action their choices will refl ect the full array of incentives operating on them: some will refl ect the need to manage a workload; others will refl ect the expectations of workplace peers and professional colleagues elsewhere; still others may refl ect their own convictions. And some will refl ect the needs of clients” (1989, 88). Before Wilson’s contribution, numerous scholars had argued that discretion in decisionmaking, in eff ect, made bureau- crats into policymakers, and bureaucracies into political actors. Wilson’s work provided a richly detailed study of how and why this discretion was exercised to produce government action.

Wilson took a disparate set of examples to develop his argument (he started with the German army of World War II, prison systems in Michigan and Texas, and a public school in Atlanta). Some of the agencies were successful, some of them were not, and the performance of some bureaucracies went from good to bad, or vice versa. Wilson sought to explain what separated the successful agencies from the not so successful, and to understand variation in bureau- cratic performance. In pursuit of this objective, he covered so much intellectual territory that it is diffi cult to provide a meaningful synopsis of the entire work. Key elements in his analysis, however, should convey some sense of the main arguments.

Wilson began with the presumption that the behavior of bureaucrats and bureaucracies was purposive; that is, it was motivated by some goal or objec- tive. He rejected the argument that the goals driving bureaucratic behavior were wholly, or even largely, determined by legislatures. Wilson noted that bureau- cratic missions encapsulated in law tend to be vague (the goal of the Department of State, for example, is to “promote the long-range security and well-being of the nation”). Fuzzy exhortations to “do the right thing” are politically appealing, but they provide no hint of the specifi c actions a bureaucracy is expected to un- dertake. In Wilson’s terms, these goals do not defi ne “operator tasks,” meaning they do not tell the frontline workers of a bureaucracy what they should be do- ing. Th ese workers, whom Wilson termed the “operators,” are those whose work actually justifi es the existence of a given organization—for example, teachers in a school, patrol offi cers in a law enforcement agency, or soldiers in an army (1989, 33–34).

Because goals are vague (or even contradictory), bureaucracies cannot sim- ply deploy their expertise to determine the best way of achieving the ends of policy. Something other than the product of the “politics” end of the politics- administration dichotomy must drive the behavior of bureaucrats and bureau- cracies. What is it? What determines the behavior of the cop on the beat, the teacher in the classroom, the private on the front lines? Wilson proposed several potential answers: situational imperatives (the day-to-day events operators must respond to), peer expectations, professional values, and ideology. He also argued that rules could substitute for goals. When goals are vague, following established

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

53Politics, Power, and Organization

procedures and “going by the book” provide operators with a guide to low-risk behavior. Wilson additionally argued that most large organizations, and cer- tainly many public agencies, have their own particular personality. Th ey have persistent, patterned ways of thinking about the purposes of the organization and the best means to achieve those purposes. Combined, these patterns constitute organizational culture, and they serve to socialize organizational novitiates into the “way things are done around here” (1989, 91–93). Recent work by Foldy and Buckley (2009) suggests that we only partially understand how street-level prac- tice might change. Th ey do fi nd, however, that the “way things are done around here” oft en endures, unless new goals are issued in a clear, coherent fashion, and that teams of workers are stable.

Wilson was not just interested in identifying the behavioral motivations of operators; he also identifi ed two other kinds of bureaucrats: managers (peo- ple who coordinate the work of operators to achieve organizational goals) and executives (people responsible for maintaining their organizations). He also identifi ed systematic elements to the behavior of each level and how they inter- act with each other. For example, he argued that managers of public agencies have a diff erent set of constraints upon their behavior than managers in private fi rms. At a most basic level, managers must have a clear sense of an organiza- tion’s mission if they are to coordinate the work of operators toward that end, and executives must be capable of defending that mission and of supporting their organization’s pursuit of a given objective. In the public sector, even these most basic elements of administration are complex because public organiza- tions cannot control their broadly defi ned missions, nor can they unilaterally control other critical elements of management—revenues, personnel, and the means of production.

Wilson concluded that successful bureaucracies are those in which executives have created a clear sense of mission, identifi ed the tasks that must be achieved to fulfi ll that mission, distributed authority within the organization according to those tasks, and provided subordinates (particularly operators) with enough autonomy to achieve the task at hand (1989, 365)—a fairly tall order given the complex environment of public agencies. Wilson’s argument suggests that agen- cies given clear objectives and high levels of autonomy are more likely to be suc- cessful in achieving those objectives. Yet, as Wilson acknowledged, it is diffi cult to see how clear objectives can be routinely manufactured as an end product of a democratic process. If the administrative arm of government is given greater levels of autonomy, and if clearer goals are not forthcoming from the democratic institutions of government, the likely result is the transfer of increasing amounts of policymaking power to the bureaucracy. In its call for clearer missions and less centralization in public bureaucracy, Wilson’s argument is ultimately prescrip- tive. Reorganization through deregulation, however, turns out to be an extension of bureaucratic politics rather than a way to channel it toward universally desired ends. Organization helps determine not only how bureaucracies and bureaucrats

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

54 3: Th eories of Bureaucratic Politics

behave but also how power and infl uence are distributed among the various ac- tors in the political system.

Government organization, or, more accurately, reorganization, is a subject near and dear to the discipline of public administration and a perennial feature of American politics. For virtually the length of the twentieth century, and continu- ing into the twenty-fi rst, critics have argued that the central problem of govern- ment is poor management. In other words, the basic problem with government is administrative: It is ineff ectively organized and ineffi ciently run. Th e orthodox response of public administration scholarship to this problem is to impose the politics-administration dichotomy, and on the administrative side, to organize government agencies by functional responsibility, put them into a logical hierar- chy with one another, and clearly assign authority and responsibility within these hierarchies. In various guises through various administrations, such eff orts were repeatedly made long aft er Waldo, Gaus, and others had pointed out that the con- ceptual foundation that supported such eff orts was untenable. All these eff orts at reorganization largely failed to meet their objectives when they ran into political diffi culties.

In recent decades, the orthodox solutions have increasingly been abandoned for a “new” organizational paradigm that seeks to bring economy and effi ciency to government by adopting market-oriented management practices. Th e “reinven- tion” movement of the Clinton administration, for example, sought to eliminate hierarchy, to put “customers” fi rst, and to prize performance over accountabil- ity. Yet the reinvention movement also ran into political obstacles. Regardless of whether it is an orthodox call for centralization and reliance on the competence of the technocrat or a less traditional appeal for decentralization and reliance on market-based processes, the purported objective is the same: to improve the ef- fectiveness and effi ciency of government through reorganization.

Students of the organizational connections to bureaucratic politics argue that the reason government reorganization is never far from the public agenda, and the reason it never achieves its supposed goals, is because the organization of the government’s administrative arm has little to do with economy or effi ciency. Organization of the government’s administrative arm is about power and politics. One of the most astute proponents of this argument is Harold Seidman, whose Politics, Position, and Power: Th e Dynamics of Federal Organization was immedi- ately recognized as a landmark in the study of bureaucratic politics when it was fi rst published in 1970 (it has since gone through several editions). Seidman’s central argument was this: Th e institutional location and environment of a policy or program and the organizational structure, process, and procedures that govern it help determine the distribution of power and infl uence within the polity. Th is includes the distribution of power among executive branch bureaucracies, but also encompasses the balance of power among the three branches of the federal government, between the federal government and state and local governments, and between the government and organized interest groups. As Allison, Richard

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

55Politics, Power, and Organization

Neustadt, and others have demonstrated, bureaucratic politics within the execu- tive branch almost certainly aff ected policy. Such scholars as Seidman suggested that bureaucracies—their organization, staff s, authority, and responsibilities— were involved in and oft en the focus of much broader and more intense political games.

Seidman supported his argument by examining the well-known organiza- tional eccentricities of the executive branch through a political rather than an ad- ministrative lens. From the perspective of public administration orthodoxy, many elements of the executive branch are perversely designed. Th ere are overlapping jurisdictions, unclear lines of authority, programs assigned to agencies with little regard to the functional priorities of the organization, and agencies built on a variety of organizational blueprints using a bewildering variety of organizational processes and procedures. To a public administration analyst steeped in the in- violability of the politics-administration dichotomy and prizing effi ciency as a guiding principle, this makes little sense.

But it makes perfect sense from a political point of view. For example, fi ve federal agencies regulate banks, savings and loans, and credit unions. Why the duplication? Why put up with the consumption of extra resources, the inevitable turf wars, and the confusion over regulatory authority? Administrative orthodoxy would call for consolidating regulation of depository agencies under one federal agency. Yet the banking industry has successfully resisted all eff orts to achieve such administrative concentration. Why? Seidman argued that the duplication allows commercial banks to pick their regulators according to the activity they engage in. Duplication, in short, shift s power from the regulators to the regu- lated, and the banking industry has had enough infl uence in Congress to keep the “eccentric” administration of banking regulations. It is not particularly effi cient or eff ective, but it is a politically desirable (or at least acceptable) way to regulate depository agencies (Seidman 1998, 14).

Looking at federal agencies through a political lens also off ers numerous other insights into why programs and policies succeed or fail. A key determinant of a program’s success or failure is where it gets assigned. Programs and policies will be neglected if they are assigned to an agency that considers them peripheral to its primary mission (a phenomenon also observed by Wilson). Th e National Oce- anic and Atmospheric Administration (NOAA), for example, considers science its primary mission. When resources become scarce, the nonscience programs the NOAA administers (nautical and aeronautical charting, for example) are the fi rst to suff er (Seidman 1998, 16). Such assignment and organizational issues de- termine not only the success or failure of the program but also the balance of political power. A program assigned to an executive department will be subject to diff erent lines of authority and accountability than a program assigned to an independent agency, a government corporation, or any one of the other bewil- dering variety of organizational arrangements in the federal bureaucracy. Institu- tional type thus helps determine how power and infl uence over a given program

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

56 3: Th eories of Bureaucratic Politics

are distributed among the executive, the legislature, various organizational com- ponents within each branch, and organized interest groups.

Accordingly, we should expect Congress to be intensely interested in the or- ganizational makeup of the executive branch and deeply involved in questions of administration. Indeed, Seidman argued that one of the central reasons for the “eccentric” organizational makeup of the executive branch is the jockeying for political power among the various elements of the legislative branch. Congressio- nal committees have historically operated as highly autonomous minilegislatures that routinely struggle for jurisdictional supremacy over policies or programs. In 1966, for example, two agencies were created to administer highway safety: the National Highway Safety Agency and the National Traffi c Agency. Both agencies were headed by one appointee. In other words, structurally built into the admin- istration of highway safety programs and policies are duplication, confused lines of authority, and many other management and organization issues that would tend to promote ineffi ciency. Th is makes sense if administration is viewed as in- exorably intertwined with politics rather than separable from it. Th e reason for two agencies rather than one was a simple matter of intrachamber politics: Two Senate committees wanted to confi rm the agency head, and the creation of two agencies achieved this purely political goal. Th e organizational and administra- tive “problems” of the executive branch are thus oft en “nothing but mirror im- ages of jurisdictional confl icts within the Congress” (Seidman 1998, 27).

Th e bureaucracy is politically important not only to the president and to Congress but also to a broad range of organized interests. Seidman pointed out that the public bureaucracy has a parallel private bureaucracy—businesses that perform contract work for the government—heavily invested in the status quo. Contracting with a private fi rm to perform various public functions has its advan- tages. Private companies, for example, are subject to lower levels of oversight and accountability, which gives them an operational fl exibility that public agencies frequently lack. Using private companies also helps reduce the number of civil servants on the public payroll, an important consideration for presidents dealing with the size of the public bureaucracy, always a politically sensitive issue. Th e downside to these arrangements is the loss of accountability and the high resis- tance of private fi rms to changes in the public bureaucracy because their liveli- hoods are dependent upon preserving the status quo (Seidman 1998, 15).

Networks and Bureaucratic Politics Th is fact that bureaucratic politics extends beyond the bureaucracy itself was highlighted by Laurence O’Toole (1997b) in his admonition to take networks seriously. For public administration, networks can be thought of as a set of orga- nizations that are interdependent; that is, they share goals, interests, resources, or values. Th ese interdependencies tie together not just public bureaucracies within, between, and among diff ering political jurisdictions, but private and

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

57Networks and Bureaucratic Politics

nonprofi t agencies as well, in the process creating new forms of organizational and management practices that are employed to achieve collective or public ends. O’Toole (1997a) argued that networked administration is not only com- mon, but also increasingly important, for fi ve main reasons. First, “wicked” policy problems require the mobilization of a variety of actors, both inside and outside government. Such problems are the result of multiple causes, and typi- cally span more than one jurisdiction. A single agency will not be able to address these problems without help, from actors both inside and outside of government and from across levels of government. Second, political demands for limited government, but without reductions in demands for action, give rise to networks that include nonstate actors through contracting. As shown in Chapter 5, the implication of contracting is something we are only beginning to understand. Th ird, the need for bureaucracy to be responsive to the public naturally leads to the inclusion of citizen and industry groups in decisionmaking. Networks may indeed increase accountability to the public, but, as will be discussed, it is unclear if they always produce the democratic eff ects we expect. Fourth, as sophisticated program evaluations have revealed indirect or second-order eff ects of policies, implementation networks have been established to refl ect those rela- tionships. Fift h, O’Toole (1997b) noted that many mandates have multiple layers that essentially require program management to become networked. Here, he used the example of transportation program managers needing to account for the rights of disabled people.

O’Toole’s (1997a) networked bureaucratic world raises important questions regarding our understanding of bureaucratic politics, governance, and account- ability. Do these networks threaten democracy or enhance it? Should expecta- tions of accountability and oversight be changed as the result of acknowledging that so many actors are involved in bureaucratic decisions? How do agencies wield political power in these networks? Some recent research has identifi ed a “dark side” of networks—that network managers respond to elements of the net- work that are more politically infl uential, and thus the result is that networks may actually exacerbate already present inequality (O’Toole and Meier 2004). Th is raises important and disturbing questions about the nature and implications of the political power in networked administration. Other concerns, as addressed in Chapter 5, involve the “hollowing out” of the state, where the state’s main task is to arrange networks rather than manage hierarchies, which is the traditional task of government.

Th e need to understand a networked bureaucracy is obvious, but it is unclear whether we have made much theoretical headway since the mid-1990s. Most literature has focused on how to manage networked systems, rather than on im- plications for politics and governance (O’Toole and Meier 2004). O’Toole and Meier argue that networks should be treated as political institutions, since their establishment is oft en for political reasons (to perhaps avoid having to deal di- rectly with a controversial issue) and always has political implications. Decisions

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

58 3: Th eories of Bureaucratic Politics

to contract or privatize functions of government are inherently political, as they involve decisions to shift the locus of state power, and certainly represent choices to move public resources to other network members, including private compa- nies or not-for-profi t organizations. Yet beyond the call by O’Toole and Meier to focus on these political implications, the fi eld has not yet produced the necessary work. Scott Robinson (2006), for example, argues we lack the conceptual tools to understand the governance implications of diff erent types of networks and how political context shapes their creation, membership, goals, and outcomes. Given the explosive growth of networked administration and its poorly understood implications for public policy and eff ect on democratic values, there can hardly be a better example of the practical and critical need for theory development, not just in the realm of bureaucratic politics, but also in the general fi eld of public administration.

Given the highly political nature of bureaucracy that Seidman, O’Toole, and others have described, eff orts to make the administrative arm of government more eff ective and effi cient persistently fail because the real objectives of bureau- cracy have nothing to do with effi ciency and better management practices. Power is really at stake in reorganization, and this is the reason the president, Congress, and other political actors take such an intense interest in administration. Reorga- nization has become such a perennial part of politics that it is increasingly pur- sued for its own sake—a political objective with no underlying administrative strategy whatsoever. During the 1980s, for example, House Republicans proposed abolishing the Departments of Education, Energy, Commerce, and Housing and Urban Development. Th e 1996 Republican presidential nominee, Bob Dole, also campaigned on a promise of eliminating the Internal Revenue Service. Th ese proposals were largely calculated to make political profi ts from popular negative stereotypes of the bureaucracy and made no real sense from an administrative point of view. No one made serious proposals for the wholesale elimination of programs, no one had a strategic plan to reassign these programs, and no one made any real argument that the end result would be more eff ective and effi cient government. Th e whole point seemed to be to attack the administrative infra- structure in the belief that smaller government was better government. Yet if there were to be no wholesale elimination of public programs, government would not get smaller, just more confused, and, in all likelihood, increasingly privatized (Seidman 1998, 110).

Such political games with the bureaucracy are not the sole province of Re- publicans. Th e Clinton/Gore administration played a particularly cynical game in its reinvention eff orts, repeatedly publicizing the shrinking federal payroll. Th e quarter-million federal positions eliminated by the reinvention movement were mostly supervisors, personnel specialists, budget analysts, accountants, auditors, and the like. Th ese people primarily oversaw third-party operations, namely, the private contractors the government increasingly uses to carry out public programs and policies. Contract employees who indirectly do the public’s

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

59Representative Bureaucracy

business vastly outnumber employees in the federal civil service, and reinvention shrank these numbers not at all. If anything, the cuts in the federal payroll made it much more diffi cult to hold third-party contractors accountable (Seidman 1998, 112–113).

Neither Wilson’s nor Seidman’s arguments constitute fully developed theo- retical frameworks, and Wilson (1989, xi) explicitly raised doubts about whether a comprehensive theory of organizational behavior was even possible. Yet Wil- son and Seidman both provide a series of empirically testable propositions that are characteristic of theoretical frameworks. From Wilson comes a rich set of hypotheses, which can be confi rmed by observing bureaucratic behavior, about everything from professional norms to the substitution of rules for goals. Seid- man’s work points analysts toward the high political stakes surrounding organi- zation and administration, and, in doing so, makes sense of the “eccentricities” that defi ed the expectations of traditional theoretical frameworks. Combined, both make it easier to understand why bureaucracies are the way they are, and why they do the things they do.

Although Seidman’s work and Wilson’s work are discursive rather than the- oretical, more explicitly theoretical eff orts from organization literature seek to explain at least some elements of the political behavior bureaucracies indulge in. John Kingdon’s Agendas, Alternatives, and Public Policies (1995), for example, at- tempts to explain why government addresses some problems while ignoring oth- ers. Kingdon’s analysis shows that government agencies have an important role in shaping the public agenda, not so much in determining agenda priorities but in acting as key members of “policy communities.” Th ese communities consist of actors who, through their specialized interests in particular policies and the den- sity of their interconnections and common interests, can decide the fate of policy proposals. A fragmented community (for example, one in which agencies have confl icting goals on a particular issue) dissipates support for a policy proposal and severely limits its potential for success (Kingdon, 1995, 116–144). Although organizational context is shown here to play an important role in shaping the po- litical role of bureaucracy, that role is not the primary focus of the theory.

Th e bottom line is that organization theory has provided an important lens for such works as Wilson’s and Seidman’s, signaled the importance of the growing phenomenon of networked administration, and in doing so has given ample jus- tifi cation for pursuing comprehensive explanations of the political role of bureau- cracy. As yet, however, organization theory has not provided that comprehensive explanation.

Representative Bureaucracy Th e theory of representative bureaucracy is perhaps the most explicit attempt to address the central problem of democratic administrative theory raised by Waldo (1952, 102): How can a theory that embraces the hierarchical and authoritarian

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

60 3: Th eories of Bureaucratic Politics

nature of bureaucracy be reconciled with the seemingly contradictory egalitar- ian and ultimately ineffi cient values of democracy? Th e work of such scholars as Waldo, Allison, Wilson, and Seidman strongly suggests that bureaucracies are political policymaking institutions. Yet if bureaucracies are powerful policy ac- tors engaged in “politics of the fi rst order,” they are also largely insulated from the ballot box and only partially held accountable to elected offi cials (Meier 1993, 7; Mosher 1982). Th is contradiction between bureaucracies making policy and basic democratic values raises one of the most important challenges for public admin- istration theory: “How does one square a permanent [and, we would add, power- ful] civil service—which neither the people by their vote nor their representatives by their appointments can readily replace—with the principle of government ‘by the people’?” (Mosher 1982, 7). Any democratic theory of administration, Waldo suggested, must be capable of answering this question.

Th e theory of representative bureaucracy focuses on fi nding a way to legitimate the bureaucracy’s political power in the context of democratic values. Th e central tenet of the theory is that a bureaucracy refl ecting the diversity of the community it serves is more likely to respond to the interests of all groups in making policy de- cisions (Krislov 1974; Selden 1997). If bureaucracies are sensitive to such a diver- sity of interests, and these interests are represented in bureaucratic decisions and behavior, the argument is that bureaucracy itself can be considered a representa- tive institution. If bureaucracy is a representative institution, its long-recognized political role can be accommodated with such basic democratic values as majority rule, minority rights, and equal representation. Th is again refl ects the relation- ship between politics and administration refl ected in fi gure 2.3d, which shows that agencies are inherently legitimate policy actors and have an ethical responsibility to act in the interests of citizens and protect the underrepresented.

Th e notion of legitimating bureaucratic power by treating bureaucracy as a representative institution was formally introduced by J. Donald Kingsley in Representative Bureaucracy (1944). Kingsley’s work, a study of the British public service, advanced the argument that the civil service should refl ect the character- istics of the ruling social class. To carry out its role eff ectively within the polity, Kingsley argued, the civil service has to be sympathetic to the concerns and val- ues of the dominant political group. Th ese shared values connect the exercise of discretionary authority on the part of the bureaucrat to the will of the democratic state. Although Kingsley coined the term “representative bureaucracy,” the basic idea he articulates is quite old. In the United States, the spoils system instituted during the nineteenth century resulted in a civil service that was dominated by major party loyalists (Meier 1975). Such a bureaucracy can be viewed as an exten- sion of the majority party, and therefore of the preferences expressed at the ballot box. Such patronage systems, of course, also invite just the sort of problems that prompted such scholars as Goodnow and Wilson to seek some division between politics and administration: technical incompetence, favoritism in administrative decisionmaking, and outright corruption.

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

61Representative Bureaucracy

More contemporary advocates of representative bureaucracy reject patronage or spoils systems as an appropriate model for a representative bureaucracy for exactly these reasons. Instead, most accept the need for organizational arrange- ments as prescribed by administrative orthodoxy, namely, public agencies based on the Weberian rational-legal bureaucracy (Selden 1997). In contrast to the spoils system, the latter is seen as conferring various benefi ts, among them effi ciency, making merit the basis of public-sector employment, and strengthening the role of technical expertise in decisionmaking (Meier 1993). Although this means ac- cepting arguments from orthodox administrative theory, advocates of represen- tative bureaucracy reject the notion of a politics-administration dichotomy. Th e theoretical and empirical lessons from the likes of Gaus, Waldo, Allison, Seidman, Wilson, and numerous others simply make it impossible to ignore or assume away the political role of the bureaucracy.

Th e theory of representative bureaucracy thus begins with the assumption that there are good reasons for public agencies to be organized the way they are (i.e., undemocratically) and that these undemocratic agencies exercise consid- erable political power. As Kenneth Meier puts it, “Th e theory of representative bureaucracy begins by recognizing the realities of politics. In a complex polity, such as the United States, not all aspects of policy decisions are resolved in the ‘political’ branches of government” (1975, 527). Th e basis of bureaucratic power is assumed to derive from the discretionary decisionmaking authority that, as a practical matter, has to be granted to them because not all implementation and enforcement scenarios can be conceived of and accounted for in statutes. Elected offi cials may have numerous tools at their disposal to restrict bureaucratic power, but strong forces place practical limits on the use of these tools. Public support of programs or agency objectives, the information advantage bureaucrats oft en hold over elected offi cials because of their technical expertise, and simple political ex- pediency all work to limit the constraints placed on bureaucratic power.

Perhaps the best-known argument that individual bureaucrats have an un- avoidable policymaking role is Michael Lipsky’s Street-Level Bureaucracy: Di- lemmas of the Individual in Public Services (1980). Lipsky’s central premise is that street-level bureaucrats—policemen, teachers, and the like—routinely have to make decisions that are not dictated by the mission of the organizations they work for, or the rules they are supposed to enforce. Street-level bureaucrats thus make policy as a result of their behavior. For example, no matter what the law says the speed limit is, in practice it is determined by the individual traffi c cop. Th e discretion to make such on-the-spot decisions, which in eff ect are policy de- cisions, is going to be considerable, even for bureaucrats working within a dense tangle of rules designed to guide their behavior. It is simply a fact of political life that nonelected individuals, protected by civil service mechanisms and working for hierarchical (even authoritarian) bureaucracies, wield signifi cant policymaking power in democratic polities. Given this, a key challenge for administrative theory is to account for this fact in the context of democratic values (Selden 1997, 13–26).

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

62 3: Th eories of Bureaucratic Politics

In meeting this challenge, those who advocate the theory of representative bureaucracy begin by seeking an answer to the same question posed by Wilson: Why do bureaucrats do what they do? Specifi cally, the focus is on explaining the behavior of bureaucrats when they exercise discretionary authority. Generally, it is assumed that bureaucrats are rational actors in the sense that they pursue self-interested goals when faced with discretionary choices. Proponents of rep- resentative bureaucracy argue that the goals driving behavior are supplied by the individual values of the decisionmaker. Th us, “if the administrative apparatus makes political decisions, and if bureaucracy as a whole has the same values as the American people as a whole, then the decisions made by the bureaucracy will be similar to the decisions made if the entire American public passed on the is- sues. . . . If values are similar, rational decisions made so as to maximize these values will also be similar” (Meier 1975, 528). Th is suggests that bureaucratic power can be harnessed to diverse and representative social interests even though the orthodox organizational arrangements of public administration are insulated from the basic processes and values of democracy. If the ranks of the civil service refl ect the diverse interests and values of society, bureaucracy becomes a repre- sentative “fourth branch of government” with a legitimate basis for exercising power in a democratic system.

Th e fi rst scholars to formulate and apply the basic arguments of representative bureaucracy in the United States were David Levitan (1946) and Norton Long (1952). Long adopted the most extreme stance, arguing that the national legisla- ture, which was heavily tilted toward the upper strata of society, did not represent a variety of important national interests. Instead, “these interests receive more eff ective and more responsible representation through administrative channels” (1952, 808). Long’s claim was that the bureaucracy had more of a democratic character than the legislature because the ranks of the federal civil service were much more refl ective of the American public. Th at diversity was refl ected in ad- ministrative decisions, even as narrower interests dominated the decisionmaking of Congress. Th e normative claim was that the bureaucracy actually made up for the representative defi ciencies of the legislature.

Although subsequent scholars have generally made less radical normative claims than Long, the two key questions driving Long’s work have remained the basic focus of work on representative bureaucracy: (1) Do public agencies broadly represent the interests and values of the American public? (2) Are these interests and values refl ected in the policy actions of bureaucracy? Th e fi rst of these ques- tions deals with the concept of “passive representation,” or the extent to which the bureaucracy refl ects the composition of society. Kingsley (1944) suggested that socioeconomic class should be the basic yardstick for comparing the com- position of the civil service with that of the public. Kingsley’s study, however, was focused on the British civil service. In the United States, Samuel Krislov (1974) argued that a more appropriate basis of comparison is race, ethnicity, and sex. Th ese factors are assumed to be a key source of socialization, and thus of values.

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

63Representative Bureaucracy

A large portion of empirical research on representative bureaucracy in the United States is thus devoted to examining the extent to which bureaucracy refl ects the basic demographic composition of society. Th e general fi nding of this research is that minorities and women are proportionally represented in bureaucracy as a whole but are underrepresented in the upper levels of bureaucratic hierarchies (Selden 1997, 45).

Th e second question deals with the concept of “active representation,” or the relationship between passive representation and policy outputs or out- comes. Again, it was Krislov who made the key contribution to shaping schol- arly thought on this issue. He argued that the demographic composition of the bureaucracy provides only indirect evidence of the representative nature of bureaucracy. Th e social profi le of any given bureaucrat—race, sex, education, and so forth—provides only a limited indication of that bureaucrat’s ability to advance the interests of these demographic groups. It is not enough, in other words, to fi nd that women and minorities are roughly proportionally repre- sented in the ranks of the civil service. Any serious claim that bureaucracy is a representative institution requires evidence that passive representation trans- lates into active representation, that the more women and minorities join the civil service, the more the policy outputs of bureaucracies represent the broad interests of women and minorities.

Considering its importance to the theory of representative bureaucracy, it is not surprising that there has been a growing body of empirical work on this lat- ter issue. Studies by Kenneth Meier and various colleagues (Meier, Stewart, and England 1989; Meier and Stewart 1992; Meier, Wrinkle, and Polinard 1999) have consistently found that minority representation in the civil service is related to policy outputs that favor the minority group. Th ese studies have exclusively fo- cused on education and the eff ects of minority representation on policy outputs (the research examined the impact of minority representation in teaching, on administrative and school board positions concerning school policies, and on outputs that aff ected minorities). Some research outside education has produced more mixed results (Hindera 1993a, 1993b; Selden 1997). However, much other research indicates that the conditions found by Meier and his colleagues do exist for other agency types as well as for representation for women (Keiser, Wilkins, Meier, and Holland 2002; Meier and Nicholson-Crotty 2006; Lim 2006; Wilkins and Keiser 2006). One more recent example is on the enforcement of immigra- tion law (Lewis et al. 2012), where municipal police forces, if granted discretion by the city council, enforce the law less intensively if the police chief is Hispanic/ Latino.

A further development in the literature is to incorporate the concept of sym- bolic representation, which, unlike active representation, works cognitively on the public. Th us, when bureaucrats share the identifi cation, experience, and characteristics of a portion of the public, that audience will perceive the actions of those bureaucrats as legitimate, even if the bureaucrats are not purposefully

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

64 3: Th eories of Bureaucratic Politics

representing that group. Nick Th eobald and Donald Haider-Markel (2009), by examining citizen attitudes about actions by police offi cers, show that actions by bureaucrats are more likely to be perceived as legitimate if citizens and bu- reaucrats share demographic characteristics. If this holds across agencies, it sug- gests that citizen attitudes about bureaucrats and policy implementation can be changed without actions on the part of bureaucrats that are expressly designed to represent certain groups (active representation). Furthermore, they argue that methods used by those studying representative bureaucracy have relied on ag- gregate data, which makes it diffi cult to know if their fi ndings demonstrate active or symbolic representation. Considering the implications for democratic gover- nance, it is important to clarify this question.

Th e key to representative bureaucracy’s attempt to build a bridge between orthodox public administration theory and democratic theory thus still rests to no small extent on the ability of future empirical studies to support the theo- ry’s central hypothesis that passive representation will lead to active representa- tion. Although the literature has expanded greatly since 2000, the issues raised by Th eobald and Haider-Markel imply the more individual-level empirical work is needed.

Conclusions It is probably fair to say that public administration scholarship has been more successful in demonstrating the need for theories of bureaucratic politics than in actually producing those frameworks. It has been more than half a century since such scholars as Waldo and Gaus exposed the rickety foundations of the politics- administration dichotomy and made a convincing brief that administrative theory had to share common ground with political theory. Since then, numerous studies have empirically confi rmed the political role of the bureaucracy. Some of these, including those of Wilson and Seidman, center on a series of empirically testable propositions. Even if the works themselves are explicitly discursive, they contain the basic materials for constructing theory. To date, however, that construction project remains incomplete.

Does the relative lack of success in producing widely applicable bureaucratic politics frameworks mean the eff ort to do so should be reassessed? Th e progeni- tors of the bureaucratic politics movement would surely answer no, for the simple reason that the most important characteristic of public administration is its po- litical nature, and we ignore this at our peril. Long once wrote that “there is no more forlorn spectacle in the administrative world than an agency and a program possessed of statutory life, armed with executive orders, sustained in the courts, yet stricken with paralysis and deprived of power. An object of contempt to its enemies and of despair to its friends” (1949, 257).

Long’s point was that the ability of a public agency to get things done was not dependent upon the responsibilities and authority granted to it by statute. Th e

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

65Summary

decentralized nature of the American system meant a program’s success or failure was tied to the political muscle of the bureaucracy it was entrusted to. As Long succinctly put it, “Th e lifeblood of administration is power. Its attainment, main- tenance, increase, dissipation, and loss are subjects the practitioner and student can ill aff ord to neglect” (1949, 257). Long argued that ignoring the political role of the bureaucracy robs administrative theory of a crucial connection to the real world and consigns any number of the prescriptive conclusions of scholarly work to failure.

Summary Such scholars as Long, Gaus, and Waldo argue that, like it or not, bureaucracy is a political institution and that any useful theoretical framework has to recognize and account for this simple fact of political life. Public administration theory, in other words, must also be political theory. Th eories of bureaucratic politics are designed with this objective in mind, and pursuit of this goal remains a profi table activity for students of public administration.

Allison’s Model III and the theory of representative bureaucracy represent two of the better-known and most widely employed bureaucratic politics frame- works. Although it is hard to underestimate Allison’s contribution, it clearly falls short of a generally applicable theoretical framework. Allison’s Model III is likely to continue to fi nd gainful employment in structuring administrative studies, but evidence has steadily mounted that it is underperforming in practice. Wilson’s work shows that Allison’s Model III was perhaps too limited in scope, as it leaves out organizational issues and ignores the fact that bureaucratic politics exist in a complex network, far beyond the confi nes of the executive branch.

Although having a considerably older lineage than Model III, the theory of representative bureaucracy in one sense remains curiously underemployed. Th e basic model is parsimonious, and its predictive hypotheses are intuitively easy to grasp. Simply stated, the theory argues that a civil service refl ecting the diverse interests and values of the community it serves will take those interests into ac- count when exercising its discretionary authority. Th e validity of the theory is tied to the hypothesis that passive representation will lead to active representation. Although there are diffi culties in operationalizing such tests, a growing number of published studies are aimed squarely at empirically assessing this claim. So far, these have produced mixed and sometimes contradictory results. Since the literature on representative bureaucracy is also increasing in sophistication, per- haps it represents the most promising theoretical development for understanding bureaucratic politics.

C o p yr

ig h t ©

2 0 1 5 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .