Why Government Fails So Often: And How It Can Do Better (49 page)

Officials wishing to minimize litigation risk will operate their programs accordingly. They are more likely to reject innovations that might be challenged,
⁵⁴
substitute broad discretion for clear rules, make concessions to “squeaky wheels,” add more elaborate processes, alter internal information flows in order to stymie whistle-blowers, increase the ratio of lawyers to program staff, and restrain the disciplining of disgruntled or incompetent employees. Such responses to litigation risk, whatever their bureaucratic or political advantages, are bound to affect programs’ effectiveness. There is much to the claim that many complex projects that drove our progress in the not-so-distant past, such as the Hoover Dam and other infrastructure, could not be undertaken today due to the costs and delays created by strategic litigation—or the threat of it.

INERTIA

I explained in
chapter 6
(in the section titled “Inflexibility”) that once public law is on the books, it is very difficult to amend or repeal. (This is less true of common law rules where a more flexible principle of
stare decisis
operates.) Obviously, this inertia can be overcome—after all, statutes are sometimes repealed and policies changed—but a law’s staying power, absent some sunset provision, is striking. This inertial force reflects at least two factors. First, it is built into the constitutional system discussed in
chapter 3
, with its numerous veto points, separation of powers, and other obstacles to assembling an effective majority coalition. Second, legal inertia reflects the value accorded to vested interests and to the settled expectations that develop around them. Entitlement reform is said to be the “third rail” of American politics, making it extremely difficult to modify pensions, health care benefits, and other features of the status quo. For present purposes, the striking
point is that this inertia persists pretty much regardless of the law’s substantive content, and like the other limits of law discussed in this chapter, it applies to all public policies.

CROWDING-OUT EFFECTS

Law’s relationship to social norms is a complex and little-understood one. Law may sponsor an existing or new norm and may manage to transfer to the norm some of the law’s own prestige and power. This “halo effect,” of course, can cut both ways; advocates for same-sex marriage want law’s legitimating sanction, whereas it is precisely this legitimation that opponents fear.

But there are darker, more paradoxical possibilities than halos. Law’s embrace may cast a shadow over norms; indeed, it may kill them with its kindnesses. Its adoption or encouragement of a norm can actually discourage the desired act by altering the social meanings that people convey when they comply with or violate it.
⁵⁵
When law compels an act—charity or patriotism, for example—observers cannot readily distinguish spontaneous, voluntary behavior from an obligatory performance of the act, thus blurring what the act signals to others. Even when law merely rewards an act—subsidizing organ donations, for example—it may change the values that people ascribe to the act and thus alter their behavioral responses to it. Law may in effect cut off the norm’s oxygen supply, smothering the authenticity and autonomy of will that vitalized the norm, thereby endangering precisely what the law hoped to promote. The late William Stuntz, a criminal law professor, noted a similar paradox: when a community perceives enforcement as unfair or overzealous, criminals suffer less social stigma, in effect lowering the price of their crimes.
⁵⁶
The fact that the state has adopted a norm may actually encourage some people to defy and violate it, especially if it is only weakly enforced, as with laws against teenage smoking or drinking.
⁵⁷
Finally, law—however well-intended—can weaken or divide local communities in which “bonding” and “bridging” social capital are crucial to their effective functioning.
⁵⁸

*
Strenuous attempts to codify private law (contracts, torts, property, family law, and the like) and to make it more uniform across the states have met with some success. But simplification efforts in public law are few and far between; it is hard to think of a major example of success in codifying it.

CHAPTER 1 0

The Bureaucracy

E
very public policy is run through a bureaucracy. In America, bureaucracy is often used as an epithet, evoking ubiquitous red tape, rigidity, soullessness, waste, unreasonableness, impenetrability, and Kafkaesque cruelty and arbitrariness. Other than systematic scholars of the subject like Max Weber,
¹
James Q. Wilson,
²
and Donald Kettl,
³
few have a kind word to say about it. Nevertheless, bureaucracy—the exercise of legal authority by a large, complex, more or less permanent body of officials required to act according to rules, procedures, precedents, institutional and documented memory, hierarchical accountability, and a culture of legality—is a necessary (though insufficient) feature of a contemporary liberal democracy and the rule of law.
*

Chapters 1
and
2
explained my reasons for focusing on policy failures,
†
but there is an additional reason to do so here. The federal bureaucracy is not simply a neutral organizational instrument for carrying out public policies whose effectiveness is determined elsewhere. Far from being epiphenomenal, it independently shapes those policies by virtue of its own structural, endemic features. Some of these features (limited technical competence and mediocre education) may contribute to policy failures, while others (procedural
regularity and public spiritedness) can promote success. Before proceeding, some description of the bureaucracy is in order.

THE BUREAUCRACY: BASIC FACTS

Federal civilian employment—approximately two million workers—has remained fairly constant in size for more than fifty years,
*
and constitutes an ever-smaller share of the population.
⁴
(At the same time, federal contractors and consultants have mushroomed, along with state and local government employment.) From 1970 to 2010, almost all civilian executive branch agencies reduced their workforces. The major exception was the Department of Justice, which is largely explained by the Bureau of Prisons, whose staff grew by 90 percent between 1990 and 2009 (even as its prisoner population increased by 355 percent). About 85 percent of federal employees work outside the Washington, D.C., metropolitan area.
⁵

The bureaucracy’s power, however, is best measured not by the number of employees but by its discretionary authority—that is, the ability to choose courses of action and to make policies not spelled out in advance by laws. By this test, the power of the federal bureaucracy has grown enormously, particularly through congressional delegations of authority in three areas: (1) subsidies to particular groups and organizations; (2) conditional fiscal transfers to state and local governments; and (3) regulation of various sectors of society and the economy. As political scientists James Q. Wilson and John DiIulio put it, “appointed officials can decide, within rather broad limits, who shall own a television station, what safety features automobiles shall have, what kinds of scientific research shall be specially encouraged, what drugs shall appear on the market, which dissident groups shall be investigated, what fumes an industrial smokestack may emit, which
corporate mergers shall be allowed, what use shall be made of national forests, and what prices crop and dairy farmers shall receive for their products.”
⁶

The federal bureaucracy consists of three groups of federal employees—the competitive service, the Senior Executive Service (SES), and the excepted service—and a fourth group consisting of federal contractors, which are private for-profit or nonprofit organizations that federal agencies pay to perform specified tasks.

The competitive service
. Many officials are appointed only after they have passed a written examination administered by the Office of Personnel Management (OPM) or met certain selection criteria (such as training, education, or experience) devised by the hiring agency and approved by the OPM. When competitive candidates can be ranked, the agency must usually appoint one of the three top-ranking candidates. In recent years, the competitive service has become decentralized so that each agency now hires its own people without an OPM referral, and examinations have become less common for three reasons: dissatisfaction with the cumbersome OPM system; agencies’ need for more professionals who could not be ranked on the basis of a standard exam; and affirmative action. Agencies can also hire on a “name-request” basis in which they inform OPM that they want to hire a particular person.

The Senior Executive Service
. In 1978, Congress created the SES, about eight thousand top federal managers who would perform an integrating function between short-term political appointees and the career service. They could (in theory) be hired, fired, and transferred more easily than ordinary civil servants—although anyone removed from the SES was guaranteed a job elsewhere in the government. SES members would be paid more and be eligible for substantial cash bonuses if they performed their duties well. The disappointing experience with the SES reform is discussed later in this chapter.

The excepted service
. Almost half of federal employees are not hired by OPM but instead by agencies that have their own selection procedures. About 3 percent of the excepted employees are appointed on grounds other than, or in addition to, merit in order to
allow the president to select people for policy making and politically sensitive posts who agree with his policy views. These include (1) presidential appointments authorized by statute, many of which require Senate confirmation; (2) “Schedule C” appointments to jobs having a “confidential or policy-determining character” below the level of cabinet or subcabinet posts; and (3) “noncareer executive assignments” given to high-ranking members of the regular competitive service or to persons brought into the civil service at these high levels, who are deeply involved in the advocacy of presidential programs or participate in policy making. These three groups of excepted appointees constitute the patronage available to the president and his administration. The number of political appointees in the executive branch has increased dramatically over time, reaching more than three thousand, of which one thousand are agency heads (and ambassadors). They enable a new president to reach deep inside every corner of the government and put loyalists in charge. His campaign uses these positions to reward key supporters and to dump others.
⁷

When president John F. Kennedy took office in 1961, he had 451 political jobs to fill. When president Barack Obama took office in 2009, he had more than four times that number, including nearly four times the number of top cabinet posts. (In the nineteenth century, practically every federal job was a patronage job. For example, when Grover Cleveland, a Democrat, became president in 1885, he replaced some 40,000 Republican postal employees with Democrats.) A president can often “blanket in” patronage appointees already holding office, making it difficult or impossible for the next administration to fire them.

THE DISTINCTIVELY PROBLEMATIC NATURE OF THE FEDERAL BUREAUCRACY

In titling this section, I describe the federal bureaucracy not merely as problematic, but as
distinctively
so. By this, I mean to make a comparative claim that the rest of the chapter will support: although all bureaucracies in liberal democratic states raise fundamental questions
of public support and functional effectiveness, these questions are raised more acutely in the United States. Here our federal bureaucracy arouses not ambivalence but actual hostility and anxiety that are deeply rooted in American individualism and in other elements of our political institutions and culture (see
chapters 3
and
4
) and have no real counterpart in other modern democratic states. Private-sector bureaucracies, especially in large corporations, may exhibit some of these same problems,
⁸
but there the opportunities for reform, and the penalties for failure to do so, are far greater than in the public sector.

Paul Light’s recent book,
A Government Ill-Executed
,
*
warns that even optimally designed policies would be undermined or distorted by a poorly functioning bureaucracy. His message is crystal clear and profoundly dismaying: the federal career service is in serious crisis, experiencing a long-term decline in quality, accountability, vision, energy, and professional commitment.
⁹
Drawing on Light’s data and other sources, this chapter will analyze the main causes of these conditions. I organize the analysis around the following ten themes: (1) congressional influence over administration; (2) legalism; (3) leadership; (4) layering; (5) compensation, status, performance, and morale; (6) discipline; (7) senior executive service; (8) lower-level compliance; (9) contracting out; and (10) isolation.

Congressional influence over administration
. Virtually all advanced liberal democracies in the world are organized as Westminster-type systems in which the executive and legislative branches are under the unified control of the party that has won a majority of seats in the legislature (parliament). Among other things, this means that little if any separation of powers exists between the legislative and executive branches; together, they constitute the “government” under the leadership of the prime minister, who is also the leader of the controlling party (or coalition) in parliament.