The Size and Power of the Bureaucracy

A bureaucracy is a large organization composed of appointed officials in which authority is divided among several managers. Bureaucracy is an obvious feature of all modern societies, but American governmental bureaucracy is distinctive in three ways. First, political authority over the bureaucracy is shared among several institutions. Second, most federal agencies share their functions with agencies of state and local government. Finally, America's adversary culture means that the actions of bureaucrats are often fought in court.

The Constitution makes little mention of the bureaucracy, other than to give the president power to appoint various sorts of officials. In 1789 Congress gave the president power to remove officials without congressional assent, but the question of who (if anyone) would actually control the bureaucracy has been hotly contested throughout American history.

Throughout most of American history, patronage was the chief means of determining who would hold federal jobs. Congress was the dominant institution, the president usually accommodated congressional preferences in appointments, and thus appointments were made to reward local supporters of Congress members or to build up local party organizations. By the middle of the nineteenth century there were a lot of federal jobs: from 1816 to 1861 the number of federal employees increased eightfold, with the Post Office accounting for most of this increase. The Civil War and postwar period saw the creation of many additional bureaus. A strong commitment to laissezfaire meant that these agencies did not for the most part regulate, but rather served specialized constituencies such as farmers or veterans. The bureaucracy as we know it today is the product of the New Deal (whose programs gave broad but vaguely defined powers to agencies) and of World War 11 (during which the government made use of the vastly increased revenues the income tax allowed).

The Supreme Court has interceded to restrict political patronage on constitutional grounds. The first step was taken in Elrod v. Burns (1976) in which the Court noted that important First Amendment interests in the protection of free speech must be taken into consideration in patronage firings. According to the majority, the public's interest in the effective implementation of policy "can be fully satisfied by limiting patronage dismissals to policy-making positions." Four years later, in Branti v. Finkel, the Supreme Court elaborated by explaining that "the question is whether the hiring authority can demonstrate that party affiliation is an appropriate requirement for the effective performance of the office." As such, the mere fact that a bureaucrat occupied a policy-making position no longer constituted the ultimate factor in a patronage firing. This line of cases was brought to conclusion with Rutan v. Republican Party of Illinois (1990), when the Court extended the Branti standard to "promotion, transfer, recall, and hiring decisions based on party affiliation and support." Thus patronage has reached the point of nearing political extinction.

People often think of big government in terms of the size of the bureaucracy, but the number of civilian federal employees has not been growing since World War 11. What has increased is the number of indirect federal employees-those working for state or local governments or private firms funded by federal programs. However, the power of the bureaucracy is a function not of its size but of the degree to which appointed officials have discretionary authority: the ability to choose courses of action and to make policies not spelled out in advance by laws. The vast increase in expenditures channeled through the bureaucracy, as well as the vast expansion in the number of regulations issued during the past thirty years, shows that the bureaucracy has indeed become very powerful.

Control of the Bureaucracy

Federal bureaucrats exercise a great deal of power, especially when operating under discretionary authority. It is therefore important to understand what influences bureaucratic conduct. In general, four factors explain the behavior of governmental officials:

1. Recruitment and reward. Once hired, a federal bureaucrat normally serves a oneyear trial period before being granted tenure. A tenured bureaucrat is extremely difficult to fire, with the average termination process (including appeals) lasting about two years. Thus, in practice, almost no one is ever fired and executives develop informal strategies for dealing with incompetent employees.

The Senior Executive Service (SES) was created in 1978 to provide presidents with a core group of neutral, professional managers in the upper grades of the bureaucracy. To ensure competence, members of the SES-who join on a voluntary basis-are subject to easier transfer and firing procedures as well as to pay increases determined by performance. The SES has not worked out as intended; almost no member of the group has been fired, and salary raises have been fairly automatic.

In spite of the merit system, hiring in federal agencies remains political, especially at the middle and upper levels. An agency can hire a particular individual on a name-request basis, giving rise to the buddy system. This practice allows the maintenance of issue networks based on shared policy views; bureaucrats in consumer-protection agencies, for example, may hire people from Naderite groups. The end-product of the recruitment and reward structure is that most bureaucrats become quite comfortable in their position and defensive about their agency, adopting an agency point of view.

2. Personal attributes. Bureaucrats at the middle and upper levels of government are not representative of the American public. They tend to be highly educated, middle-aged white males. But none of these factors explains much about the attitudes bureaucrats hold. Surveys have found top-level bureaucrats to be slightly more liberal than the average voter but not as liberal as members of the media. Yet even this generalization is a bit misleading. Attitudes tend to vary depending on the agency for which a bureaucrat works. Those employed by activist agencies (FTC, EPA) are much more liberal than those who work in traditional agencies (Commerce).

While attitudes differ, they do not necessarily influence bureaucratic behavior because much of bureaucratic work is governed by standardized rules and procedures. It is only where roles are loosely structured that a civil servant's attitudes come into play.

3. The nature of the job. Some agencies have a sense of n-tission, a clear doctrine that is shared by its members. Such agencies (the Forest Service, the FBI, and the Public Health Service) are easy to manage and have high morale but are hard to change and are resistant to political direction. To be sure, a sense of mission probably infiltrates most agencies to some degree; a survey by Kenneth Meier and Lloyd Nigro revealed that federal bureaucrats believe in the importance of their agency's work. Thus the mission of the agency may become synonymous with the public interest in the minds of many bureaucrats. An agency's mission, however, must be accomplished within an array of laws, rules, and regulations-dealing with hiring and firing, freedom of information, accounting for money spent, affirmative action, environmental impact, and administrative procedures. Agencies also have overlapping and even conflicting missions. These characteristics make controlling the bureaucracy difficult, no matter which party occupies the White House.

4. External forces. All government bureaus must cope with seven external forces: executive branch superiors, the president's staff, congressional committees, interest groups, the media, the courts, and other government agencies. All federal agencies are nominally subordinate to the president. In practice, agencies that distribute benefits among significant, discrete groups, regions, or localities within the United States (such as HUD, Agriculture, and Interior) tend to be closely overseen by Congress. Others (such as State, Treasury, or justice) are more under the control of the president. Bureaucrats, like people generally, desire autonomy-to be left alone, free of bureaucratic rivals and close political supervision. They may obtain autonomy through the skillful use of publicity to build public support, as did the FBI and NASA. A less risky strategy is to develop strong allies in the private sector that will provide political support in Congress. However, this limits the freedom of the agency; it must serve the interests of its clients. Thus the Maritime Administration supports high subsidies for the shipping industry, and the Department of Labor could never recommend a decrease in the minimum wage.

External forces influence agency decisions in the form of the so-called iron triangle-the informal policy network involving an agency, an interest group, and a congressional committee. Often, though, an agency will be faced with conflicting interest group demands. The National Farmers Union favors high subsidies to farmers, whereas the American Farm Bureau Federation takes a free-market position. Organized labor favors strict enforcement by the Occupational Safety and Health Adn-dnistration, whereas business is opposed. In these instances, issue networks emerge. These are an array of groups and individuals, often contentious, and split along ideological, partisan, and economic lines.

Congress has a forn-ddable array of powers to deal with the bureaucracy. First, congressional statutes establish the existence of an agency and occasionally specify in detail how agencies should behave. Lately, however, Congress has given broad discretion to agencies. Second, money must be authorized and then appropriated by Congress. The agency is thus beholden to the legislative committee that authorizes funds and to the Appropriations Committee of the House.

For many decades, Congress made increasing use of the legislative veto to control bureaucratic or presidential actions by vetoing a particular decision within a thirty- to ninety-day period. However, in June 1983, the Supreme Court declared the legislative veto unconstitutional (the Chadha case). This decision's exact effect on congressional oversight of the bureaucracy is still uncertain. Finally, congressional investigations are the most visible and dramatic form of oversight.

Bureaucratic Pathologies

There are five major problems with bureaucracies: red tape, conflict, duplication, imperialism, and waste.

1. Red tape is the existence of complex rules and procedures that must be followed to get something done. Any large organization must have some way of ensuring that one part of the organization does not operate out of step with another.

2. Conflict exists when some agencies work at cross-purposes with other agencies. The Agricultural Research Service tells farmers how to grow crops more efficiently, while the Agricultural Stabilization and Conservation Service pays farmers to grow fewer crops. Because Congress has 535 members and little strong leadership, it is not surprising that it passes laws that promote inconsistent or even contradictory goals.

3. Duplication occurs when two government agencies seem to be doing the same thing, such as when the Customs Service and the Drug Enforcement Administration both attempt to intercept illegally smuggled drugs.

4. Imperialism refers to the tendency of agencies to grow without regard to the benefits their programs confer or the costs they entail. Because government agencies seek vague goals and have vague mandates from Congress, it is not surprising that they often take the broadest possible view of their powers. If they do not, interest groups and judges may prod them into doing so.

5. Waste occurs when an agency spends more than is necessary to buy some product or service. An example would be the much-publicized purchase of $300 hammers by the military.

It should be clear that bureaucratic problems are hard to correct. Congress cannot make the hard policy choices and set the clear priorities necessary to eliminate conflict and duplication. Government exists partly to achieve the kind of vague goals that resist clear cost-benefit analysis; eliminating red tape might make coordination more difficult. Although Americans dislike "the bureaucracy" in general, studies show that they like the appointed officials with whom they deal.