The Problem of Bureaucratic Government_Peters 1981

Southern Political Science Association

The Problem of Bureaucratic GovernmentAuthor(s): B. Guy PetersSource: The Journal of Politics, Vol. 43, No. 1 (Feb., 1981), pp. 56-82Published by: Cambridge University Press on behalf of the Southern Political ScienceAssociationStable URL: http://www.jstor.org/stable/2130237Accessed: 19/02/2010 01:07

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available athttp://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unlessyou have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and youmay use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/action/showPublisher?publisherCode=cup.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.

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

Cambridge University Press and Southern Political Science Association are collaborating with JSTOR todigitize, preserve and extend access to The Journal of Politics.

http://www.jstor.org

The Problem of

Bureaucratic Government

B. GuY PErrms

IT IS BY NOW almost trite to say that bureaucracy and administra- tion are an increasingly significant - if not the most significant - feature of modern policy-making. Journalists and political candidates of almost all political persuasions find the public bureaucracy a convenient whipping boy to explain all varieties of social problems. Academic writers have been concerned with bureaucracy, although not unvaringly opposed to its increasing im- portance in policy-making. Their approaches have ranged from the theorists of post-industrial society who have welcomed the rational, technocratic decision-making processes of the bureaucracy as a means of saving them from a more fearsome ill - politics - to those who, like the journalists, go to great lengths to provide "proof" of the inadequacies of bureaucratic decision-making in government.' Somewhere between those two extreme views has been found the majority of students of public administration, gleeful over the in- creasing importance of public bureaucracy, but apprehensive over the weakness of their own models for explaining the policymaking role being assumed.2

Great imprecision has characterized both academic and popular attempts to analyze bureaucracy. One author has pointed to at

I For a discussion of bureaucracy in the context of post-industrial society see Timothy M. Hennessey and B. Guy Peters, "Political Paradoxes in Postindustrialism: A Political Economy Perspective," Policy Studies Journal, 3 (Spring, 1975), 233-239.

2 See, for example, Dwight Waldo, ed., Public Administration in a Time of Tur- bulence (San Francisco: Chandler, 1971); Frank Marini, ed., Toward a New Public Administration: The Minnowbrook Perspective (San Francisco: Chandler, 1971); Vin- cent Ostrom, The Intellectual Crisis in American Public Administration (University, Alabama: University of Alabama Press, 1973).

THE PROBLEM OF BUREAUCRATIC GOV ERNMENT 57

least seven different conceptions of bureaucracy which have been used in the literature.3 Likewise, there is a schizophrenia that has characterized most views of bureaucracy, and its inadequacies.4 On the one hand, bureaucracy has been seen as a Leviathan: an in- tegrated, monolithic institution unfettered by political checks and balances, and possessing an insatiable appetite for power. On the other hand, bureaucracy is also viewed as a Court Jester: a loose col- lection of agencies lacking ideas, coordination, and common sense, which at best muddles through, and at worst makes an absolute fool of itself.5 As Herbert Kaufman has pointed out,

. . . it is ironic that government workers who are often depicted as drab, 'faceless,' timid, and obscure should be called self-directing, dominant and sinister.... (They have also been described as incompetent, bungling, lazy and stupid, and at the same time as diabolically clever self-seeking conspirators.)6

There is likewise a distinction between talking about bureaucracy as a reified entity behaving as a monolith with a unified set of institu- tional values, and talking about bureaucracy, meaning more simply just those administrative agencies which do the majority of the ad- ministrative work for the political executive.

This paper is an attempt to look at the role of bureaucracies in modern government, especially in governments of advanced in- dustrial societies.7 If our analysis were to cover a wider range of na- tions, it would require different and expanded analytics. We will be developing a concept of "bureaucratic government" as an Ideal Type against which to compare the nature of real world bureaucratic systems. Thus, our basic question is the extent to which public bureaucracy has come to dominate or at least greatly influence policy-making. To some degree the answers to that ques- tion will depend upon the perspective of the individual answering it, but we will attempt to develop a useful set of criteria that can be used in arriving at the answer. However, before going on to look at this concept of "bureaucratic government," we should look at the ex-

3Martin Albrow, Bureaucracy (London: Pall Mall, 1970). 4Anthony Downs, Inside Bureaucracy (Boston: Little, Brown, 1967), 132-133. 5 Much of this literature, such as that by Parldnson, is humorous but yet it clearly

demonstrates the problems perceived in bureaucratic structures. 6 Herbert Kaufman, Fear of Bureaucracy: A Raging Pandemic, Edmund Janes

James Lecture, University of Illinois, April 12, 1978, 22. 7 Because of the interests of the author, the majority of examples will be drawn from

the United States, United Kingdom, Scandinavia and France. The same analytics would be applicable to other similar systems.

58 THE JOURNAL OF POLITICS, VOL. 43, 1981

isting literature on the role of bureaucracy in policy-making to see both what insights are provided, and what the significant gaps are.

BuREAucRAcY IN MODERN GOVERNMENT

There is an immense literature on public bureaucracy and public administration, and we will make no pretensions of thoroughly discussing that literature within the confines of this or any other single paper. However, several subsets of that literature do have special importance in understanding the place of bureaucracy in the governance of modern industrial societies.

The Positive Theorists

A number of scholars have undertaken to explain the behavior of bureaucracies and bureaucrats using the economic analysis of the "public choice" approach to politics.8 Although there are a number of differences among the writers using this approach, their basic orientation is that of bureaucracy as Leviathan, and as being com- posed of rational, maximizing actors. As Downs put it rather suc- cinctly: "The fundamental premise of the theory is that bureaucratic officials, like all other agents in the society, are significantly-though not solely-motivated by their own self- interests."9 Or Niskanen goes even further in stating, ... what, if anything, do bureaucrats maximize? An economist's initial response will be that the bureaucrat, like anyone else, maximizes his personal utility. By itself this is not very helpful, but it does suggest that a bureaucrat will engage in purposive behavior and that there are probably some elements in his utility other than the general welfare and the interests of the State.10

While these assumptions when stated so clearly do not appear revolutionary or even novel, when contrasted to the majority of the literature on public bureaucracy, they do represent an important departure from tradition. No longer is the bureaucrat assumed to act sine ire et studio, but is instead assumed to be operating in a political environment. Also, the bureaucrat is assumed to have in-

8 William A. Niskanen, Jr., Bureaucracy and Representative Government (Chicago: Aldine/Atherton, 1971); Downs, Inside Bureaucracy (Boston: Little, Brown, 1967); Albert Breton, The Economic Theory of Representative Government (Chicago: Aldine, 1974); Gordon Tullock, The Politics of Bureaucracy (Washington, D.C.: Public Affairs Press, 1965).

9 Inside Bureaucracy, 2. 10 Bureaucracy and Representative Government, 36.

THE PROBLEM OF BUREAUCRATIC GOVERNMENT 59

terests which he or she attempts to maximize through political ac- tion. 11 These basic assumptions further lead to some important ad- ditional assumptions and conclusions.

The first corollary is that bureaucrats will seek to maximize their own security, primarily through the inflation of the budgets and staffs of their agencies."2 A second aspect of bureaucratic behavior predicted by the positive theorists is that bureaucracies will, where possible, seek to provide public goods and to avoid any pricing or ra- tioning of the goods produced."3 Market mechanisms and marketable products make control of the bureaucracy easier for other political organizations which would seek to exert that control. The positive theorists have also placed substantial emphasis on the weapons available to bureaucrats in pursuing their goals. Among these are policies themselves, information, and the support of in- terest groups in the environment.'4

These assumptions, and deductions from the assumptions, con- cerning the nature of bureaucracy are important for an understand- ing of public bureaucracy, and also make a good beginning toward a conceptualization of bureaucratic government. The important first step of accepting an active role for the bureaucracy in government has been made, and we begin to understand that the motivations of bureaucrats are not necessarily "public" service and the "public" in- terest. However, as interesting and important as this work has been, we are left with a one-sided view of bureaucracy. We see it as a set of integrated agencies all striving to develop strategies for budget maximization. We do not see the professional and almost ideological behavior of many bureaucratic actors, who do at least apparently have ideas about policy which they seek to have im- plemented without regard to the budgetary consequences.'5 Fur-

11 See, for example, Mattei Dogan, "The Political Power of Western Mandarin," in Dogan, ed., The Mandarins of Western Europe (New York: Halstad, 1975), 3-24.

12 Downs, Inside Bureaucracy, 16-17; Niskanen, Bureaucracy and Representative Government, 36-40.

13 David Bartlett, Economic Foundations of Political Power (New York: The Free Press, 1973), 66-67.

14 For an interesting discussion of the strategies and weapons of bureaucracy in ac- tion see J.C.H. Jones, "The Bureaucracy and Public Policy: Canadian Foreign Policy and the Combines Branch, 1960-1971," Canadian Public Administration 18 (Summer, 1975), 269-296.

15 In general, however, agency ideology and budgetary interests tend to conform closely. The major threat is that the agency will lose credibility by advocacy of un- popular or unfeasible programs.

60 THE JOURNAL OF POLITICS, VOL. 43, 1981

ther, as Wade has pointed out, the principal decision-makers in public bureaucracy have relatively little to gain personally by hav- ing their agencies grow."' They typically are at the top of the salary schedule already, and can only acquire headaches, not more money. Thus, while this conclusion does constitute a good beginning, it re- mains only that, and we must press further to understand more fully the governmental role of bureaucracy.

The Descriptive Theorists

There have been two significant bodies of literature developed at- tempting to describe more completely the attitudes and behaviors of bureaucrats and bureaucracies. The first, generally associated with the work of Robert Putnam and his colleagues, has attempted to ex- plain the extent of bureaucratic involvement in policy-making by the attitudes of civil servants and the responsiveness of bureaucratic actors to political pressures.'7

This work has contrasted two polar - and perhaps Ideal - types of bureaucratic actors. The first is the Weberian, or "classical" bureaucrat. This individual

. . . operates with the monistic conception of the public interest-the 'national' interest or the 'interest of the State.' He believes that public issues can be resolved in terms of some objective standards of justice, or legality, or of technical practicality . . . the classical bureaucrat distrusts and rejects the institutions of politics.... 8

This "classical" bureaucrat may be contrasted with the "political" bureaucrat who accepts the rough-and-tumble of political life, and gladly joins in as a participant.

This work has been important in developing an understanding of the nature of contemporary administrative behavior, but it does not aid greatly in answering the question we have set forth: what is the nature and extent of bureaucratic involvement in policy-making?

16 L. L. Wade, "Public Administration, Public Choice and the Pathos of Reform," unpublished paper, University of California, Davis, Davis, California.

17 Robert Putnam, "The Political Attitudes of Senior Civil Servants in Britain, Ger- many and Italy," British Journal of Political Science 3 (July, 1973), 257-290; Thomas J. Anton, Claes Linde and Anders Mellbourn, "Bureaucrats in Politics: A Profile of the Swedish Administrative Elite," Canadian Public Administration 16 (Winter, 1973), 626-651; Samuel Eldersveld, Sonia Hubee-Boonzaaijer, and Jan Kooiman, "Elite Perceptions of the Political Process in the Netherlands," in Dogan, The Mandarins of Western Europe, 129-161.

18 Putnam, "The Political Attitudes...," 259.

THE PROBLEM OF BUREAUCRATIC GOVERNMENT 61

We learn a great deal about the attitudes of bureaucrats toward other policy-makers, but know very little about their interest, or ad- vocacy of, policy.

A second set of descriptive studies of bureaucracy and bureaucratic policymaking comes from analyses of foreign policy, and the role of bureaucracies in making defense and foreign policy. This is the literature characterized by the "bureaucratic politics" paradigm.'9 Beginning with the original Allison article, this ap- proach has argued that foreign policy decisions could be explained as well or better from a bureaucratic politics perspective than either from the perspective of the rational actor or the organizational proc- ess.20 In many ways, this approach has been a less rigorous version of the positive theory approach to-bureaucratic behavior, but with the intention to be more descriptive, and seeking to compare its results with traditional models of international politics and foreign policy as opposed to those of Wilsonian public administration.

The bureaucratic politics approach has been rather severely criticized by scholars of international relations, particularly related to its continuing state-centric assumptions in a world of in- terdependence.2' However, from our point of view, we must add that it apparently does not advance our understanding of the behavior of bureaucratic agencies beyond the more elegant work of the positive theorists. We are provided with more descriptive evidence than is generally true for the positive approach, but still lack an understanding of the politics of bureaucracy as a potential governing body-either in general or over foreign policy.

Theorists of Institutional Weakness

The positive theorists and the descriptive theorists assume a positive political role for the bureaucracy, either as policy en-

'9 Graham Allison, "Conceptual Models and the Cuban Missile Crisis," American Political Science Review, 63 (September, 1969), 689-718; Essence of Decision (Boston: Little, Brown, 1971); and Morton Halperin, "Bureaucratic Politics: A Paradigm and Some Implications," in R. Kanter and R. Allman, (eds.), Theory and Policy in Inter- national Politics, supplement to World Politics, 24 (1972), 40-79; Morton Halperin, Bureaucratic Politics and Foreign Policy (Washington, D.C.: The Brookings Institu- tion, 1974).

20 Allison, "Conceptual Models. 21 D. Krasner, "Are Bureaucracies Important? or Allison Wonderland," Foreign

Policy, 7 (Summer, 1972), 159-179; Robert Art, "Bureaucratic Politics and American Foreign Policy," Policy Sciences, 4 (1973), 467-490.

62 THE JOURNAL OF POLITICS, VOL. 43, 1981

trepreneur or as participant in the political process. A third body of literature stresses the weakness of the "political" institutions of government. As democratic governments have appeared headed down the slippery slope of "overload" and crisis, this literature has blossomed.

There is a growing body of literature emphasizing the crisis of government in many Western countries at this time. This work has been largely European, but some American problems and prospects are also included. The explanations and remedies of the authors writing vary widely. Crozier, for example, attributes much of the problem to economic crisis and institutional overload.22 Rose, Bell and others have emphasized the effects of attributed increasing en- titlements of populations for government benefits.23 Brittan has stressed the bidding-up of budgets by competitive political elites.24 King, LaPorte and Scharpf have stressed the multiple dependency relationships among political actors and the public, and the breakdowns of institutional coordination.'5 Scheuch has gone somewhat further and argued that the problems of governmental management are indicative of a more general rejection of large organization in society.26 Although varied in their approaches, these scholars all recognize the fundamental problems of current society and government, but some do see the bureaucracy as one alternative government, at least in the short-run, to provide some essential stability to the governing process.27

A second body of literature is more confined to the United Kingdom, and stresses the weaknesses of traditional party govern-

22 Michael Crozier, "Western Europe," in Crozier. Samuel P. Huntington and Joji Watanuki, The Crisis of Democracy (New York: Trilateral Commission, 1975).

23 Richard Rose, "Overloaded Governments," European Studies Newsletter, 3 (1975); Daniel Bell, "The Public Household-on 'Fiscal Sociology' and the Liberal Society," The Public Interest 37 (1974), 29-68.

24 Samuel Brittan, "The Economic Contradictions of Democracy," British Journal of Political Science, 5 (April, 1975), 129-159.

25 Anthony King, "Overloaded Governments," Political Studies, 23 (une/September, 1975), 284-296; Todd LaPorte, Organized Social Complexity (Princeton: Princeton University Press, 1975); Fritz W. Scharpf, "Public Organization and the Waning of the Welfare State," European Journal of Political Research, 5 (December, 1977), 339-362.

26 Erwin Scheuch, Wird die Bundesrepublik Uregeirbar? (Koln: Arbeitgeberver- band der Metallindustrie, 1976).

27 Bureaucracies are seen by many of the authors as at least a part of the cause of "overload," but some also regard it as a short-term alternative to political leadership which may be lacking.

THE PROBLEM OF BUREAUCRATIC GOVERNMENT 63

ment models in that political system, although there are certainly broader implications for the work. The most significant and com- prehensive work of this particular genre is Rose's The Problem of Party Government.28 Rose argues that although political parties continue to fulfill some aspects of their assigned tasks by providing competition for office and alternative sets of elites, they fail substan- tially in attempting to translate their electoral mandates into governmental actions. These specific problems and evidence cited by Rose come from the United Kingdom, but the problems also have more general applicability.29 The complexity and sheer size of modern governments makes the tasks of any set of political elites at best difficult and at worst impossible.

These weaknesses within the political decision-making institutions of government place the bureaucracy in a very strong position to govern, at least indirectly.- However, the deficiencies of the bureaucracy may be as great as those of the political institutions. The major disability of bureaucracy is commonly seen to be the lack of direction for policy. An overstated version of the argument of in- stitutional weaknesses is that if politicians have ideas, they will be prevented from implementing them, and civil servants do not and will not have ideas to implement. In sum, government is paralyzed and ineffective.

These theorists of institutional weakness add somewhat to our understanding of the role of bureaucracy in government. If nothing else, we do begin to see the void which the bureaucracy may fill, if we are not always sure that they will seek to fill it. Further, we find that some of the traditional bureaucratic virtues of organization and a concern for implementation actually provide them with significant weapons in a governmental process in which those virtues are scarce commodities. However, we must go fur- ther, and inquire how we may judge the extent to which public bureaucracy is, or may be, the major guiding institution of society.

Consensus, Dissensus and Bureaucratic Government

The three approaches to the involvement of bureaucracy in governmental leadership present very different pictures of that in-

28 Richard Rose, The Problem of Party Government (London: Macmillan, 1974). 29 Bruce Headey, British Cabinet Ministers (London: George Allen and Unwin,

1975); Edward Marples, "A Dog's Life in the Ministry," in Richard Rose (ed.), Policy- making in Britain (London: Macmillan, 1969), 128-132.

64 THE JOURNAL OF POLITICS, VOL. 43,1981

volvement. The positive theorists and the descriptive theorists both argue that bureaucracy is a positive force, although not necessarily for the commonweal or for technically elegant policies as is sometimes assumed. The theorists of institutional weakness accept to some degree the Weberian and Wilsonian conceptions of the bureaucracy, in that they argue that public bureaucrats tend not to have policy ideas of their own and that they are masters only of routine and not of policy.

There is some evidence that supports both conceptions of the governmental role and capabilities of the bureaucracy.30 We could stack and weigh evidence on both sides of the argument, but the question would remain unresolved. Two things seem necessary to provide a more complete answer to the questions raised in this argu- ment. The first is analytic and conceptual, involving the develop- ment of a more precise picture of what a bureaucratic government would look like. The second task is then more fully to analyze the existing literature with this conception of a bureaucratic govern- ment firmly in mind. Again, we would not expect an abdication by political institutions of their rights to bureaucrats, nor do we expect a declaration of bureaucratic government to emanate from the depths of some office building in Foggy Bottom, Whitehall, or Karlavagen. Rather, we are interested in the degree to which, given the lack of leadership alleged to be besetting traditional in- stitutions of government and the difficulties which even skilled leaders have in managing government departments, the bureaucracy is capable of providing needed direction and leader- ship. This has been largely assumed by theorists of post-industrial society, and we are now intending to provide some direction in con- ceptualization, measurement, and analysis.31

30 Alfred A. Diamant, "Tradition and Innovation in French Administration," Com- parative Political Studies, 1 (July, 1968), 251-274; Ezra Suleiman, Politics, Power and Bureaucracy in France (Princeton: Princeton University Press, 1974); Barbara Castle, "Mandarin Power," Sunday Times, June 10, 1973; Richard Crossman, The Diaries of a Cabinet Minister, Vols. I-III (London: Hamish Hamilton and Johnathan Cape, 1975-1979); Harold Seidman, Politics, Position and Power, 3rd. ed. (New York: Ox- ford University Press, 1980); Renate Mayntz and Fritz W. Scharpf, Policy-Making in the German Federal Bureaucracy (Amsterdam: Elsevier, 1975); P. Grottian, "Zum Plannungsbewusstein der Bonner Ministerialburokratie - Vorlaufige Ergebnisse einer emperische Studie," Politische Vierteiliahresschrift, Sonderheft 4 (1972); Daniel Tarschys, Petita (Stockholm: Liber, 1975); Bjorn Molin et al., Byrakrati och Politik (Stockholm: Almqvist and Wiksell, 1973).

31 Samuel P. Huntington, "Postindustrial Politics: How Benign Will It Be?," Com- parative Politics, 6 (anuary, 1974), 163-192.

THE PROBLEM OF BUREAUCRATIC GOVERNMENT 65

BUREAUCRATIC GOVERNMENT

What would a "bureaucratic government" look like, other than perhaps being uniformly gray? In other words, what criteria must be met before we would be satisfied that indeed the bureaucracy was capable of providing a viable government-meaning both policy direction and routine management - for a society. There are any number of criteria which might be applied, but those advanced by Rose in assessing the ability of political parties to provide govern- ment may fruitfully be adopted to serve our purpose here.32 Those criteria (as modified) are as follows:

(1) They must formulate policy intentions for enactment in of- fice.

(2) These intentions must be supported by statements of "not un- workable" means to the ends.

(3) There should be some competition over the allocation of resources.

(4) They should be in sufficient numerical strength in the most important positions in the regime.

(5) Those given office must have the skills necessary to running a large bureaucratic organization.

(6) High priority must be given to the implementation of goals.

As noted, these criteria are somewhat modified and condensed ver- sions of those developed for political parties, but the damage done to the original intentions does not appear mortal. The basic idea that to govern it is necessary for individuals with ideas about policy to be able to implement those ideas through the existing structures of government comes through even in this modified version of the model. Let us now begin to examine these several criteria sepa- rately to assess the importance of each for the role of bureaucracy in governance.

Policy Intentions: the Agency Ideology

The first criterion is one which might ordinarily be regarded as the crucial shortcoming of public bureaucracy as a workable alter- native to other forms of governance. The bureaucracy has tradi- tionally been regarded in most societies as lacking ideas about what to do with the machinery of government they appear to control. However, we find that bureaucratic agencies frequently have well-

32 Richard Rose, The Problem of Party Government.

66 THE JOURNAL OF POLITICS, VOL. 43, 1981

developed ideas about what government should do. These ideas are not general statements, but rather are confined to the narrow area of the expertise of the agency. To understand better these ".agency ideologies" we should first differentiate them into two different categories, here labelled the "soft" and "hard" conceptions of bureaucratic ideologies.

The "soft" version of the agency ideology is that the existing pro- gram itself is a set of ideas which are favored by the bureaucracy, out of familiarity if for no other reason. Stated more positively, we may regard the ongoing program of an agency to constitute something of an agency ideology. Political executives coming into nominal positions of power over bureaucratic structures have almost invariably reported overt or covert resistance by their civil servants, and the existence of a "departmental view" about policy which limits the effectiveness of any political leader. For example, the British Foreign Office has commonly been regarded as being pro- Arab, and the Department of Education as being in favor of com- prehensivization of schools, so that any Minister coming into office with different policy views would have to overcome these pre- existing biases of his or her "servants."33 There are few commen- tators on bureaucracy or executive leadership in industrialized societies who have not commented on the existence of this "soft" ver- sion of a bureaucratic ideology, so that if we can accept this as a minimalist version of the existence of ideas about policy in a bureaucracy, then clearly such ideas do exist.

The "hard" version of the policy intention criterion is that the bureaucracy must not only be interested in administering the ex- isting policies of their agencies, but they must also seek to impose a new set of policies. Given that, on average, bureaucrats persist longer than do politicians, we might expect them to be able to alter policies over time to suit themselves. However, there are a number of other sources for policy change which may affect the civil servant's perception of what should be done.

First, bureaucracies and bureaucrats are increasingly intercon- nected via organizational and professional memberships, so that what bureaucrats want out of the policy process may tend to change over time to correspond to the "best practices" of their profession. Some organizations to which bureaucrats belong may be strictly

33See Joe Haines, The Politics of Power (London: Coronet, 1977); Maurice Kogan, The Politics of Education (London: Penguin, 1971).

THE PROBLEM OF BUREAUCRATIC GOVERNMENT 67

"bureaucratic," i.e., concerned entirely with public sector manage- ment, while others may be groups of subject-matter specialists for health, education, sanitary engineering, or whatever. In either case the bureaucracy may, through its professional contacts, generate challenges to the existing policies based upon new ideas or the diffusion of policy innovations.

In Europe, the existence of a number of transnational organiza- tions facilitates such diffusion, and bureaucratically developed policy agendas, such as minimum standards for social services, il- lustrate the beginning of an even more significant diffusion of stan- dards and practices.34 This in turn may indicate an even greater role for bureaucracy as a source of policy ideas.

Even without diffusion, however, bureaucrats do have policy ideas. These typically come from the increasing professional qualifications and training of members of the public bureaucracy. Mosher dates the rise of the "professional state" from the mid-1950s.35 This form of state organization is characterized by the dominance of specialized professional knowledge concentrated in bureaucratic agencies.36 The professionals in the agencies become the source of new policies within their sphere of competence, having both expert knowledge and some interest in the expansion of their agencies. Those bureaucrats interested in changing policies may have to wait a number of years before implementing their ideas, until sufficient popular and political support is developed. The movement for Medicare, and the development of community mental health programs are examples of policy changes generated within the bureaucracy and which typically required a very long time from formulation to implementation.37

34 Specifically, the "Best Practice" doctrine of the European Communities social policy seeks to bring standards in each of the Nine up to that currently prevailing in the country with the most generous provisions.

35 Frederick C. Mosher, Democracy and the Public Service (New York: Oxford University Press, 1968), 105ff.; "Professions in the Public Service," Public Administra- tion Review, 38 (March/April, 1978), 144-150.

36 Ibid.; Samuel H. Beer, "The Adoption of General Revenue Sharing: A Case Study in Public Sector Politics," Public Policy, 24 (Spring, 1976), 157-160; Francis E. Rourke, "Bureaucratic Autonomy and the Public Interest," American Behavioral Scientist, 22 (May/June, 1979), 537-546.

37 Theodore R. Marmor, The Politics of Medicare (Chicago: Aldine/Atherton, 1973); Henry A. Foley, Community Mental Health Programs: The Formative Process (Lexington, Mass.: Lexington Books, 1975); Martha Derthick, Policymaking for Social Security (Washington, D.C.: The Brookings Institution, 1979), 17-37.

68 THE JOURNAL OF POLITICS, VOL. 43, 1981

Taking either conception of the bureaucratic role in the genera- tion of policy ideas, we would expect significant cross-national dif- ferences in the role of bureaucracy. One source of these differences would be the relative independence of agencies from centralized political control. So, in the United States or Sweden where agencies (or styrelsen) have substantial independence, and where they must compete directly for funds, we would expect greater policy ad- vocacy than in political systems with more centralized ad- ministrative systems, e.g., those under Treasury control in the United Kingdom.38 Likewise-although Diamant's arguments would appear to refute the contention-the absence of effective political leadership would appear to allow greater bureaucratic discretion and policy advocacy than would a more stable and effec- tive political executive.39 So, Philip Williams writes that in France

.... long-range policies had been the work of officials rather than politicians in the Third Republic as well as the Fourth. This situa- tion was a by-product of ministerial instability; however undesirable in theory, it was preferable to no long-range policies at all."40 Much the same situation is alleged to have obtained during the latter days of the Nixon administration. By way of contrast, the doctrinal emphasis on ministerial responsibility in the United Kingdom makes even ineffective political leaders powerful in theory if not always in practice.

Thirdly, we would expect bureaucratic personnel systems which allow individuals to remain within a single or limited number of agencies during a career would experience greater bureaucratic policy advocacy than would administrative systems requiring more diverse career patterns. Thus, the Scandinavian countries in which civil servants are hired by individual agencies or ministries rather than through centralized personnel services, or the United States where careers tend to be confined to a single department, would be more likely to have stronger policy advocacy by bureaucratic agen- cies than would the United Kingdom or France where the senior

38 Tarschys, Petita: Hugh Heclo and Aaron Wildavsky, The Private Government of Public Money (Berkeley: University of California Press, 1974).

31 Diamant, "Tradition and Innovation..,"; See also Suleiman, Politics, Power and Bureaucracy in France, 160-170.

40 Philip Williams, Crisis and Compromise: Politics in the Fourth Republic (New York: Doubleday Anchor, 1966), 365-366; See also Lawrence Scheinmann, Atomic Energy Policy in France Under the Fourth Republic (Princeton: Princeton University Press, 1965).

THE PROBLEM OF BUREAUCRATIC GOVERNMENT 69

civil service will have experienced a number of different types of posts, albeit within the framework of a grand corps in France.41

Finally, there are definite attitudinal configurations which ap- pear related to policy advocacy by bureaucracy. Putnam and his associates have uncovered such attitudinal configurations by quan- titative methods, and a number of more descriptive studies have in- dicated that such cultural and attitudinal patterns do exist.42 Thus, as with any other concept - or Ideal Type - there will be variation in the degree to which real-world cases correspond to the posited characteristics. However, there is some reason to believe that the bureaucracy can provide some policy leadership and that they are not only the masters of routine and paper-shuffling, but rather are frequently very important sources of ideas.

The Availability of "Not Unworkable" Means

If politicians are generally considered the masters of ideas, then certainly the bureaucracy is considered the master of routine and techniques. Thus, there should be little question about the bureaucracy presenting feasible means to carry out programs, whether those programs are generated internally or imposed upon them by politicians. In fact, the opposite may be the case: that which is feasible is often translated into policy. Thus, as with Lind- blom's reconstructed preferences, bureaucrats are frequently capable of molding not only techniques but also policies by the definition of what is feasible.43

The ability to reconstruct preferences makes the bureaucratic agency as much a victim of its own procedures as the master. The bureaucracy may be innovative, but is frequently limited by the reliance on accepted procedures for a definition of what can be done. Feasibility may be defined in terms of the ability of the pro- gram to be administered through the standard operating procedures of the agency, as with Allison's concept of the organizational process

4' These differences in career patterns are thoroughly described in Brian Chapman, The Profession of Government (London: Unwin University Books, 1959). Although dated, the information presented remains useful. See also B. Guy Peters, The Politics of Bureaucracy (New York: Longmans, 1978).

42 Putnam, "The Political Attitudes. . .;" Anton, Linde and Mellbourn, "Bureaucrats in Politics. . .;" Eldersveld, Hubee-Boonzaaijer and Kooiman, "Elite Perceptions. . ."

43 Charles E. Lindblom, The Policy-Making Process (Englewood Cliffs, N.J.: Prentice-Hall, 1968), 101-108.

70 THE JOURNAL OF POLITICS, VOL. 43, 1981

model of policy-making.44 Thus, while agencies may indeed develop feasible means, these means may in turn blind both bureaucrats and politicians to the range of available policy alter- natives.

Bureaucracies may clearly have procedures to implement any pro- gram they may wish to but, rather than being an undivided asset, this may at times be a liability. Agencies may be able to implement a weak conception of program advocacy mentioned above, but may be impeded in making any substantial changes in program because of accepted procedures and methodologies. Their agenda may be defined by how they are accustomed to doing business, rather than by what they would like to do. There is a tension, therefore, be- tween the role of the bureaucracy as advocates of innovation in policies and their role as conservers of procedures.45

The role of bureaucracies as conservers of procedures is a variable, just as is their role as policy advocate. It varies in part as a function of tradition and culture, but is also related to more specific political and structural considerations. One of the principal factors related to an emphasis on procedure would appear to be external pressures for control that would make administrators wary of ac- tions not justifiable as normal procedure. In all democratic -and most non-democratic -countries there have been increased pressures for better control and supervision of bureaucratic discre- tion.46 With such pressures, there is a natural tendency for bureaucrats to retreat to procedures for protection, with a conse- quent loss of innovation and flexibility.

Competition among Agencies One criterion for governance generally associated with

democratic and partisan government is competition among con- tenders for office. Bureaucrats already have office, and are un- likely to lose it. What they do not have is money. Thus, while the currency of partisan competition is votes, the currency of bureaucratic competition is currency. The competition for budgets among agencies may provide many of the same benefits at an

44 Allison, The Essence of Decision. 45 This tension is reflected in the "garbage can" model of decision-making in which

means seek ends, rather than vice versa. See Michael D. Cohen, James G. March and Johann P. Olsen, "The Garbage Can Model of Organizational Choice," Ad- ministrative Science Quarterly 17 (March, 1972), 1-25.

46 Peters, The Politics of Bureaucracy, Chapter 8.

THE PROBLEM OF BUREAUCRATIC GOVERNMENT 71

organizational level that partisan competition is assumed to provide in democratic politics. Just as partisan competition allows a voter to select among alternative governments, which in turn are sup- posed to be related to alternative policies, bureaucratic competition allows political and administrative personnel to choose more di- rectly among alternative policies.47 This competition is frequently conducted without direct political intervention, as with many deci- sions on spending within the British Treasury, or within Ministries of Finance in many countries.48

There is substantial disagreement among analysts of bureaucracy as to both the nature and efficacy of this competition among agen- cies. Some argue that the conflict is intense and pervasive, with the principal intention being to maximize the agency's budget.49 Others have argued that the competition is less frequent and more re- strained, seeking to preserve a "fair share" for the agency and even seeking cooperation in dividing the available budget pie.50 Some would argue, in fact, that agencies will frequently avoid conflict and agency growth if that growth may threaten their basic purpose and perhaps expose weaknesses in their existing programs.5' In ad- dition, Downs among others has argued that competition among bureaucracies, just as with industries in the model of the free market economy, is a positive force encouraging policy innovation and also serving as a check on bureaucratic autonomy.52 In any of the above conceptions, however, competition among agencies does have a place as a means of allocating resources among competing policies and thus allowing some to flourish, and some to languish or, less fre- quently, to die.53

No matter what the stakes of bureaucratic competition may be, it

47Breton also notes that competition is moved from a public arena to a private arena. The Economic Theory of Representative Government, 162-163.

48 Ibid.; Heclo and Wildavsky, The Private Government. . ., 76-128. 41 Niskanen argues that such competition should be expanded to provide a quasi-

market for services and (hopefully) improve the performance of public bureaucracies. William A. Niskanen, "Competition among Government Agencies," American Behavioral Scientist, 22 (May/June, 1979), 517-524.

50 Robert E. Goodin, "The Logic of Bureaucratic Backscratching," Public Choice 21 (1975), 53-68.

51 Matthew Holden, "Imperialism and Bureaucracy," American Political Science Review, 60 (December, 1966), 943-951.

52 Downs, Inside Bureaucracy, 198-199. 53 Few do die, as shown in Herbert Kaufman, Are Government Organizations Im-

mortal? (Washington, D.C.: The Brookings Institution, 1975).

72 THE JOURNAL OF POLITICS, VOL. 43, 1981

will occur in different degrees in different bureaucratic systems. The structure of some systems, e.g., the United States and Sweden, allows more latitude for bureaucratic competition and bargaining than is true in more centrally managed administrative systems. The existence of agencies with low levels of coordination, other than through the budgetary process, and the ability of agencies to argue for their own appropriations (at least indirectly), and to mobilize political support, makes competition a more important part of the lives of the agencies.54 It also means that they will be more capable of providing an alternative form of government than will agencies more constrained by central political and administrative control. Thus, bureaucratic competition seems to go hand-in-hand with the "hard" conception of policy advocacy, if for no other reason than policy and one means through which the competition is conducted.

The nature of bureaucratic competition has two principal effects on politics and government. First, it may in part account for some of the massive growth of the size of government - as reflected in public spending - over the past several decades.55 Old programs become institutionalized as commitments of governments - and en- titlements for citizens - and the need to compete for increased fund- ing produces new programs and new policies from the agencies.56 Some authors have argued exactly the opposite - that in fact com- petition among agencies would decrease the size of government, but that analysis seems severely to underestimate both the persistence of agencies and their ability to limit the scope of competition to areas outside their "heartlands."57 So long as there is competition for in- creasing expenditures, there will apparently be an increasingly big government, regardless of the preferences of citizens. This growth may, however, be favored by politicians because it provides them more benefits to distribute among constituents, and thus create the image of someone who can deliver for the folks back home.58

54 Tarschys, Petita; Seidman, Politics, Power and Position, 162-165. This may be important even in more centralized systems. See J. J. Richardson and A. G. Jordon, Governing Under Pressure (London: Martin Robertson, 1979), 53-59.

55 Richard Rose and Guy Peters, Can Government Go Bankrupt? (New York: Basic Books, 1978).

56 Daniel Bell, "The Public Household...," 39. 57 Downs, Inside Bureaucracy, 211-216. 58 Morris Fiorina, Congress: Keystone of the Washington Establishment (New

Haven: Yale University Press, 1978); See also Douglas Arnold, Congress and the Bureaucracy (New Haven: Yale University Press, 1979).

THE PROBLEM OF BUREAUCRATIC GOVERNMENT 73

The second major effect of bureaucratic competition is that it limits the internal consistency or "coherence" of government.59 The bureaucracy does not act as an integrated instrument to serve the public interest, but rather acts as a set of subgovernments each serv- ing a clientele group crucial in the political game of survival. These "whirlpools," "subgovernments" or "iron triangles" are a dominant characteristic of government.60 In other words, with highly com- petitive agencies, there may be bureaucratic governments, but no bureaucratic government. Or as Natchez and Bupp put it, "Priority setting in the Federal bureaucracy resembles nineteenth century capitalism: priorities are established by aggressive entrepreneurs at the operating levels of government."61

The Incumbency of Positions

Another necessary criterion for the ability of bureaucracies to pro- vide an alternative source of government is that they must occupy the most important positions in policy-making, and further, they must be in sufficient numbers to make their decisions effective. The bureaucracy clearly satisfies the quantitative aspect of this criterion, even though, as we will point out later, we can ensure that those in the lower echelons of the bureaucracy will always comply with the directives of their superiors. However, the bureaucracy may not appear to satisfy the qualitative aspect of the criterion. Politicians are traditionally thought to be in the most important positions in policy-making, and the bureaucrats only in a position to implement their decisions.

There are two points, however, which qualify the traditional

5' Samuel H. Beer, "Political Overload and Federalism," in Victoria Schuck and Josephine Milburn, (eds.), New England Politics (Cambridge, Mass.: Schenkman, forthcoming).

? While usually conceived of in the American context, e.g., J. Leiper Freeman, The Political Process (New York: Random House, 1955), the growth of corporatism and cooptation in most smaller European countries may be conceived of as a similar phenomenon. See Martin 0. Heisler and Ole P. Kristensen, "The Mixed Polity in the Welfare State: Corporate Pluralist Politics in Scandinavia in Comparative Perspective," paper presented to 1979 Annual Meeting of the Southern Political Science Association, Gatlinburg, Tennessee, November, 1979; Ole P. Kristensen, "The Logic of Political -Bureaucratic Decision-Making as a Cause of Governmental Growth," paper presented to the Sandberg Danish Political Science Association, March, 1979.

"1 Peter B. Natchez and Irving C. Bupp, "Policy and Priority in the Budgetary Proc- ess," American Political Science Review, 67 (September, 1973), 963.

74 THE JOURNAL OF POLITICS, VOL. 43, 1981

assumption. First, the contact of the bureaucracy with the environ- ment of the organization, as well as the concentration of technical expertise in the lower echelons of organizations, tends to give bureaucracies a substantial control over information and expertise which are crucial for policy-making. Thompson's analysis of the separation of expertise and authority in modern organizations is most important here, and the ability to control information is a major influence over policy in the hands of the bureaucrat.62 Fur- ther, to the extent to which information is passed through the bureaucratic hierarchy, it is selectively distorted. Thus, although there may be enough people in the bureaucracy - and perhaps a few extra -there may still be an imbalance between those making deci- sions at the top and those with the information for making the deci- sions at the bottom. 63

Political institutions have been attempting to break the monopoly on information which the bureaucracy appears to hold by creating their own independent sources of information. These "counter bureaucracies" are most numerous in the United States -for exam- ple, the agencies of the Executive Office of the President, the Con- gressional Budget Office, and the growing committee staffs of Con- gress -but also exist in a number of other political systems.64 Some have sought to provide this information through ministerial cabinets, while others have established research offices such as the Central Policy Review Staff in Britain.65 Still others have tried un- successfully to use their political parties as instruments for policy research.66 Despite these efforts, the bureaucracy retains a central role in the development and dissemination of policy relevant infor- mation, and thereby retains a powerful position in policy-making.

62 Victor Thompson, Modern Organizations (New York: Knopf, 1961). 63 Gordon Tullock, The Politics of Bureaucracy (Washington, D.C.: Public Affairs

Press, 1965), 137-141. " Thomas E. Cronin, "The Swelling of the Presidency," Saturday Review of So-

ciety, 20 Uanuary, 1973), 30-36; Allen Schick, "The Battle of the Budget," in Henry C. Mansfield, Jr., Congress Against the President (New York: Praeger, 1975), 64-69; Michael J. Malbin and M. A. Scalley, "Our Unelected Representatives," The Public In- terest, 47 (Spring, 1977), 16-48.

65 Suleiman, Politics, Power and Bureaucracy. . ., 201-238; Hugo Van Hassel, "Belgian Ministerial Cabinets," Res Publica 15 (1973), 357-369; "Belgian Civil Ser- vants and Political Decision Making," in Dogan, The Mandarins..., 187-195; Heclo and Wildavsky, The Private Government. . ., 304-339.

66 See the speech by Shadow Industry Minister John Silkin, The Guardian, 5 (December, 1979), 4.

THE PROBLEM OF BUREAUCRATIC GOVERNMENT 75

A second factor in assessing the relative importance of bureaucratic and political positions in policy-making concerns the importance of implementation in defining policy. It can easily be argued that "policy" is what happens rather than what it says in the legislation. Many public programs allow a substantial degree of latitude for the implementors of policy, e.g., in police work or in defining eligibility for social programs, and thus the lower echelons of the bureaucracy may be as important as those in "policy-making positions" in defining the realities of policy.67

Finally, the bureaucracy retains one principal advantage in a struggle for power and policy - it is simply so numerous. The sheer immensity of the task of controlling a large, growing and complex public bureaucracy possessing substantial expertise may defeat all but the hardiest politician. Even in the United States, which has a much larger than average number of political appointees who at- tempt to exercise this control, the size of the bureaucracy and its re- lationship to important political forces makes control difficult. In sum, bureaucrats may occupy the most important positions in gov- ernment simply because they occupy more positions than anyone else. They almost certainly will not agree on all aspects of the agency program, but they will agree that the program -and their jobs-should continue. As with competition among agencies, the effects of this characteristic of bureaucratic politics is to divide gov- ernment and give each agency a part of it.

The Possession of Managerial Skills

Political leaders frequently have been shown to be lacking in the skills necessary to manage large complex organizations such as the public bureaucracy.68 It is assumed that the bureaucrats who oc- cupy those organizations will have the skills, if for no other reason than they do seem to manage the organizations on a day-to-day basis. So, just as with their comments on attempting to change policies, politicians coming into office may find that the organiza- tions of which they are nominally in charge tend to run on their own with little direction from above.

67 Kenneth Culp Davis, et al., Discretionary Justice in Europe and America (Ur- bana, Ill.: University of Illinois Press, 1976.)

68 Richard Rose, The Problem of Party Government; Bruce Headey, British Cabinet Ministers (London: Allen and Unwin, 1975); B. Chenot, Etre Ministere (Paris: Plon, 1967); Hugh Heclo, A Government of Strangers.

76 THE JOURNAL OF POLITICS, VOL. 43, 1981

When compared to some absolute scale, rather than to the abilities of the politicians, the skills of the bureaucrats may not ap- pear so overwhelming. In fact, many of the standard complaints against bureaucracy, and more specifically public bureaucracy, concern their internal managerial dysfunctions. Discussions of rigidity, "red-tape," displacement of goals, and general inefficiency have filled the literature on bureaucracy. We therefore must be seriously concerned if these internal problems within bureaucracy are not sufficiently great to limit the ability of bureaucracy to pro- vide effective governance when conventional political institutions also prove themselves ineffective.

The "publicness" of the public bureaucracy, and the lack of any readily measurable outputs, both contribute to these difficulties. Be- ing public, these organizations must be more concerned about the adherence to norms, procedures, etc. than are private organiza- tions. They are responsible for public money and act in the name of the people and must therefore be accountable to the public. Ac- countability, in turn, may force the bureaucrat to protect himself against possible complaints, and this protection comes through adherence to rules and procedures.89 This protection is as impor- tant when dealing with superiors as it is when dealing with clients, so that policy leadership from the top may be thwarted by bureaucratic rigidities and procedures within the organization. The best prepared policy innovations will fail if the administrators im- plementing these innovations are more concerned about their own protection than about implementation.

These general problems of control are exaggerated by the lack of any measurable output in public bureaucracies.70 The major means of evaluating the success of public bureaucracies is consumption and not production. Therefore, lacking a measurement such as profit to assess whether or not the organization is functioning effectively, managers are forced to use rules, regulations and hierarchical con- trol more than would be the case in other types of organizations. What is attained, however, is as often non-functioning than smooth functioning.

The degree to which internal managerial dysfunctions beset

'9 Michael Crozier, The Bureaucratic Phenomenon, 213-220. 70 The measurement of outputs of the public sector has become something of a

search for the philosopher's stone. For a classic discussion see Ludwig von Mises, Bureaucracy (New Haven: Yale University Press, 1972).

THE PROBLEM OF BUREAUCRATIC GOVERNMENT 77

public bureaucracy is a function of many factors. Some are purely organizational, while others appear related to cross-national dif- ferences in conceptions of authority and hierarchical control.7' While many of the dysfunctions are universal, they are particularly apparent when there is a resistance to impersonal authority, where the costs of sanctions are high to the individual, and where there is little peer protection against authority. These relationships are now being examined more systematically, but the general finding is that the internal dynamics of organizations may be a limiting factor in the ability of bureaucracies to provide direction to society.

Finally, one of the truisms about modern government is that there is increasing interdependence of the public and private sectors.72 Not only does policy have to come down from above, it must be cleared below. This fact puts bureaucrats in a strategic position as the linkage between the public and the private sectors, but also makes their tasks that much more difficult. Not only must there be compliance within the organization, but compliance in society is now required, with many more built-in "clearance points."73 Thus while bureaucrats occupy key positions, government and manage- ment are no longer as simple as they once were, or as it appears on organization charts.

A High Priority Given to the Implementation of Policy within the Bureaucracy

We have already been discussing the problems of internal management within public organizations. There are a number of problems with communications which block the smooth flow of in- formation upward, and problems of internal rigidities blocking the smooth flow of authority downward. This sixth criterion is more directly concerned with the translation of decisions made at the top of the organization into effective policy. To a great extent the "real" policy of government is that policy which is implemented, rather than that policy which is adopted by the legislature of the

71 See, for example, Michael Crozier, The Bureaucratic Phenomenon (Chicago: University of Chicago Press, 1964), 210-264; On ne change pas la societe par decret (Paris: Grasset, 1979), 81-110.

72 B. Guy Peters and Martin 0. Heisler, "The Growth of Government: What is Growing, Why, How and How Do We Know?," unpublished paper, Tulane Univer- sity, January, 1980.

73 King, "Overloaded Governments," 293.

78 THE JOURNAL OF POLITICS, VOL. 43, 1981

upper echelons of the bureaucracy. Administrators in day-to-day contact with citizens who actually dispense services and decide upon eligibility-social case workers, clerks, teachers, policemen-may be as important in molding policy as those who are commonly thought of as being in policy-making positions. These client- contact level employees often have pressures on them - from clients, supervisors, and perhaps most importantly from peers -that inhibit their ability to conform to the stated policies of the organization. A number of studies have documented the existence of these pressures, and ensuring compliance and effective implementation remains a central administrative problem.74 It should be noted that this perspective of the discretion available to lower echelon workers to some degree conflicts with our earlier discussion of domination by procedure and "red tape." However, what is often found is com- pliance with the form of policies and regulations but avoidance of the substance of those policies.

A significant portion of this failure of implementation can be ex- plained by political factors rather than organizational factors. As an administrator finds himself or herself farther and farther from the center of organizational power, there is a loss of political support and policy reinforcement. The administrator becomes more subject to political pressures from outside their organization, if for no other reason than that these pressures are more relevant and more im- mediate than those from the home office. Kaufman's now classic study of the U.S. Forest Service is a case in point of local pressures on a field officer.75 Pressman and Wildavsky's study of implementa- tion illustrates this problem in an intergovernmental context, as do the problems of ensuring compliance in the decentralized ad- ministrative structure of West Germany.76 Even the haughty French prefet must negotiate and attempt to co-opt local political

74 See Dietrich Garlichs and Chris Hull, "Central Control and Information Dependence: Highway Planning in the Federal Republic of Germany," in Kenneth Hanf and Fritz W. Scharpf, Interorganizational Policy Making (London: Sage, 1978), 143-166.

75 Herbert Kaufman, The Forest Ranger (Baltimore: Johns Hopkins University Press, 1960), 75-80.

78 Jeffrey L. Pressman and Aaron Wildavsky, Implementation (Berkeley: University of California Press, 1973); Garlichs and Hull, "Central Control"; Kenneth Hanf, "Ad- ministrative Developments in East and West Germany: Stirrings of Reform," Political Studies, 21 (1973), 35-44; Neville Johnson, Federalism and Decentralization in the Federal Republic of Germany (London: HMSO, 1973).

THE PROBLEM OF BUREAUCRATIC GOVENMENT 79

forces in order to govern his territory successfully.77 The necessity of mobilizing local political support for policy, and its attendant need to bargain away some policy intentions of the central govern- ment, may be a fundamental part of the policy process. However, as Pressman and Wildavsky point out, it is perhaps the one area of the policy process about which we have the least reliable informa- tion. 78

Given this paucity of information, we are somewhat lacking in the ability to develop reliable hypotheses concerning sources of variation. One very plausible hypothesis is that in administrative systems which are highly decentralized, with West Germany as the most obvious example, there will be greater difficulties in ensuring proper implementation throughout the hierarchy. Likewise, we would hypothesize that the degree of sectionalism and local autonomy would be related to failures of implementation, as would the ability of political leaders with strong sectional power bases to place pressures on bureaucracies for special considerations.79 Also, the degree of vertical separation of the client-contact level from the center of the organization may also make it more likely-and necessary - that lower echelon workers bargain more with local and client interests. Finally, the lack of political support for an organization, as with the independent regulatory commissions in the United States, may make it crucial for organizations to develop operative policies very different from the intended policies.

In addition to the pull of clients and geographical interests, there are other factors within public bureaucracies which limit their effec- tive implementation of policies. The organizational factors affect- ing implementation have been well documented in a number of studies.80 More importantly, as we outlined in the discussion of agency ideologies, organizations may have goals of their own, and consequently may not accept the goals of their nominal political

77 Jean-Pierre Worms, "Le Prefet et ses notables," Sociologie du Travail, 3 (1966), 149-175.

78 Implementation, 166ff. 79 See Basil Chubb, "Going Around Persecuting Bureaucrats: The Role of the Irish

Parliamentary Representative," Political Studies, 11 (1963), pp. 272-286; Allan Korn- berg and William Mishler, Influence in Parliament: Canada (Durham, N.C.: Duke University Press, 1976), 191ff.

80 See especially Chris Hood, The Limits of Administration (New York: Wiley, 1976).

80 THE JOURNAL OF POLITICS, VOL. 43, 1981

masters. Opposition to the ideas and policies of politicians is rarely overt, as this might violate the fundamental relationships between elective and permanent officials in government. More commonly, bureaucrats defeat politicians by obfuscation, delay and the use of rules, regulations and procedures.8' Politicians, being short-term occupants of their positions, rarely understand either the procedural mechanisms or the substance of policy as well as their nominal ser- vants, and consequently are frequently at the mercy of civil ser- vants. They are particularly at the mercy of their civil servants when the policy in question falls among several departments, so that the policy is the result of the ". . . groups of officials in the thousands of interdepartmental meetings, luncheons, and telephone calls that take place every day."82 An interorganizational network exists in government, both among departments and among levels of govern- ment, and an individual needs substantial length of service in order to learn the network and how to get what he or she wants out of it.83 Civil servants have that longevity, while politicians rarely do.

We should not count politicians out too readily, however, and politicians have developed a number of mechanisms to attempt to restore their control over the structures and policies of government, increasingly considered to be dominated by bureaucracy. We have already mentioned the use of ministerial cabinets in France and Belgium, as well as the role of the Central Policy Review Staff in the United Kingdom. Also, there is an increasing use of political ap- pointees in positions in which they were previously infrequent. Mrs. Thatcher has placed several appointees into the Treasury, while in Sweden appointees have become more common in the central ministries.84 In West Germany there has been a long tradition of patronage appointments at the local government level to ensure the responsiveness of bureaucracies to elected officials, and the develop- ment of "matrix organizations" in the Bund ministries has been, in

81 Leslie Chapman, Your Disobedient Servant (Harmondsworth, England: Penguin, 1978).

82 Dudley Sears, "The Structure of Power," in Hugh Thomas, (ed.), Crisis in the Civil Service (London: Anthony Blond, 1968).

83 See the work in Hanf and Scharpf, Interorganizational Policy-Making. 84 As a parody on practice in the Soviet Union, these officials are referred to as

politruker. See Neil Elder, "The Functions of the Modern State," in Jack Hayward and R. N. Berki, State and Society in Contemporary Europe (New York: St. Martin's, 1979), 66.

THE PROBLEM OF BUREAUCRATIC GOVERNMENT 81

part, an attempt to improve control over these organizations since it was so difficult to exercise control over Land administrations.85

Implementation remains perhaps the central problem in contem- porary democracies. Breakdowns of implementation represent a fundamental failure of those systems to transform political ideas into effective action. Bureaucracies are a central cause of this failure, although usually not from malice but more from the rigidities built into their structures, or from their sincere belief in the policies they are already pursuing.

SUMMARY AND CONCLUSIONS

We have been exploring the question of the ability of bureaucratic institutions to provide government -a set of coherent policy inten- tions and the implementation of those intentions - for modern societies. Our findings, or rather analysis, is that although bureaucracy may be able to go some distance in providing such leadership, it is also thwarted by many of the same problems which appear to hinder politicians seeking to exercise governance. Those whose primary concern is democratic politics and popular control of government may welcome this analysis initially, but upon reflection may be chastened. What this analysis indicates is that there are dif- ficulties in public management and government in industrial societies which are more basic than the short-term political and economic forces usually cited as the causes of current problems.

The problems - termed "overload" in much of the European literature - appear to be more fundamental and deep-seated. They have to do with the loss of confidence of citizens, the decline of obe- dience and quiescence, the exhaustion of budgetary appeals to citizens by politicians, and, last but not least, the machinery of government itself. As we have been discussing throughout, the sheer bulk and inertia of bureaucracy, combined with its need for external political support from clientele groups, tends to fragment control and divert attention from problems of governance to prob- lems of organizational survival. The political life and, to some ex- tent, the values of bureaucratic agencies are tied up in questions of

85 Renate Mayntz and Fritz W. Scharpf, Policy-Making in the German Federal Bureaucracy (Amsterdam: Elsevier, 1975), 63-78. George Otte, "The Political Role of the German Municipal Bureaucracy," paper presented at 1979 Annual Meeting of the American Political Science Association, Washington, D.C., September, 1979.

82 THE JOURNAL OF POLITICS, VOL. 43, 1981

organizational survival. Therefore, they are almost inherently in- capable of considering broad allocative and governance questions for the society. Therefore, to the extent that bureaucracies appear to be gaining in influence over policy and government, a nation will have many governments, but no government.

Thus, government by bureaucracy is a problem. Bureaucracy may be capable of supplying government, but unlike political par- ties which supply government by "directionless consensus," govern- ment supplied by bureaucracy may be government by "non- consensual directions." The government supplied will not go in any single direction, but in many dependent upon the agency and its relationship to its clientele. For the same reason it will be non- consensual and incoherent government. There would be no in- tegrating ideology or philosophy, only a set of specific ideologies about specific policy problems, These ideologies, rather than in- tegrating the activities of government tend to fragment government and render it a set of competing, or at least not co-operating, fief- doms.

Bureaucratic government is a threat to those who see the central position of bureaucracy in modern policy-making as a threat to traditional democratic values. It is also a threat to those who desire an effective government as, if not a requisite for democracy, at least a confederate of stable democracy. We have in this one paper, merely explored the dimensions of the problems without attempting to provide any definitive solution to them. The search for these solutions should be a high priority for students of politics and bureaucracy, as well as for those who practice the arts of govern- ment.