Why Government Fails So Often: And How It Can Do Better (52 page)

Contracting out
. One of the most distinctive aspects of the federal bureaucracy is its extensive but opaque relationship with the private sector. Light points out that the government uses contracts, grants, and unfunded mandates to state and local governments not simply to deliver needed goods and services more effectively but also to make itself seem smaller than it actually is. A longstanding freeze on the number of federal employees (dating to 1951) has driven the growth of the “hidden government”; from 1990 to 2005 alone, Light estimates that as many as 2.5 million contracting jobs were created, while the civil service
shrank
by approximately 400,000. This has resulted in a prodigious number of such jobs: contract and grant-based employment by 2005 was, according to Light’s calculations, approximately 7,634,000 and 2,892,000, respectively, out of an estimated 13,165,000 people “employed” by the government (excluding military personnel). Such a workforce allows government to present itself as much smaller than it really is. Indeed, the above numerical estimates rest solely on Light’s speculation because the government does not keep—nor, it has been argued, would it even be able to keep—an exact count of the number of contractors and grantees it employs. This happens for two reasons: first, contractors themselves do not keep employee counts, Light argues, “in part because doing so would allow the federal government … to estimate actual labor costs and profit”; second, state and local agencies “make no effort whatsoever to quantify the number of hours their employees spend administering federal mandates.”
⁵⁵

Whether the tactical use of “hidden government,” “government by proxy,” and legal limits on agency staffing fools congressional and other critics of “bloated bureaucracy” is doubtful. Yet many profess to be shocked—
shocked!
—at the revelation in 2013 that core spying and data-mining functions of the National Security Agency are performed by private contractors.
Half a million of them have top-secret security clearances
, yet NSA control of them appears to be dangerously loose.
⁵⁶
Government-wide, no one seems to know how much of this outsourcing occurs, how much of it would be optimal in different agencies, how much discretion should be permitted given the risk of political favoritism in contracting, or many other aspects of how these practices work.
⁵⁷
As we have seen, the federal bureaucracy is highly differentiated and serves many political masters—Congress, the president, the public, its own workers, and powerful outside interest groups that lobby for contracts, as in the aerospace industry—each with different incentives to support outsourcing, and at different levels.

What seems clear, however, is that the vast amount of contracting reduces accountability and transparency, and raises serious questions about agencies’ ability to manage these contracts effectively and measure contractors’ performance.
⁵⁸
As contracting has become the fastest growing component of federal discretionary spending, these questions have taken on ever-greater urgency. In his 2008 book, Light noted that agency procurement offices are often too small to effectively oversee megacontracts that can involve thousands of subcontractors. “Contract management,” he observed, “has been a recurring item on the GAO ‘high risk’ list of roughly two dozen troubled federal programs and systems for the better part of two decades. GAO put Defense, Energy, and NASA contract management on its very first high-risk list in 1990, and has yet to remove them…. With contracts now running more than $500 billion a year and grants approaching the $100 billion mark, lack of accountability for results is particularly troubling.”
⁵⁹
By all accounts, Obamacare’s website contractors were egregiously mismanaged.

Isolation
. Writing of the public sector generally, but in terms that certainly apply to the federal service, Donahue describes the “segregation” that has occurred between the ethos and environment of public-and private-sector work in the last few decades. (He might have added that federal employees’ defined-benefit pensions afford them a measure of retirement security that most private sector workers can only envy.) In the private economy, “[p]eople endowed with native wit, education, ambition, and luck discovered ever-wider opportunities and ever-richer rewards.” But “by the early years of the twenty-first century the economic distances separating Americans were wider than they had been in living memory. Employment became a game with higher and higher stakes, and Americans learned to play the game hard….”

This didn’t happen in government. Today’s public sector mostly missed the transformation that swept over the rest of the working world. Government jobs, for good and for ill, tend to operate under the rules that defined the middle-class economy in the decades following World War II. Risk is dampened. So is opportunity. Rewards at the top are not all that different from those below. Nearly all workers, from janitors to governors, earn middle-class salaries. Unions thrive.
*
Change is gradual. Layoffs are rare. Promotions come slow. The role of money—as a motive and as a symbol—is circumscribed.
⁶⁰

Despite generations of blue-ribbon commissions and other reports urgently calling for much the same kinds of fundamental reforms of our civil service, it is increasingly a backwater.
⁶¹
The consequences of this increasingly demoralized, poorly equipped, marginalized, publicly scorned, and (literally) undisciplined workforce for government performance are deeply dismaying. These employees are, after all, the shock troops, the implementers, the ultimate instruments of our public policies. If they are not up to the job, then neither is our government. Alexander Hamilton, who ardently desired a strong, effective government, had no doubt on this score: “A feeble execution is but another phrase for a bad execution; and a government ill-executed, whatever it may be in theory, must be, in practice, a bad government.”
⁶²

*
Even the least rule-bound, most discretionary decisions made by a federal official—presidential pardons—are processed by a bureaucracy (here, the Office of the Pardon Attorney in the Department of Justice). See, generally, Rachel E. Barkow, “Prosecutorial Administration: Prosecutor Bias and the Department of Justice,”
University of Virginia Law Review
99 (2013): 271–342.

†
The exceptions are the successes to be discussed in
chapter 11
.

*
This total does not include the U.S. Postal Service, an independent agency within the executive branch, with 546,000 employees in 2012, the third largest civilian employer in the nation after the federal government and Walmart. Nor does it include the roughly thirteen million people who work indirectly for the federal government as employees of private firms and state or local agencies that are largely, if not entirely, supported by federal funds. Nearly three persons earn their living indirectly from the federal government for every one earning it directly.

*
The phrase is from Alexander Hamilton, perhaps our most fervent strong-government advocate.

*
The process took about two months to get president John F. Kennedy’s appointees confirmed by the Senate, and more than nine months for president George W. Bush’s. See Paul C. Light,
A Government Ill-Executed: The Decline of the Federal Service and How to Reverse It
(2008), 87–88.

*
Reflecting the exceptional growth in federal employment in the 1960s and 1970s, an estimated 60 percent of its white-collar workforce and 90 percent of the Senior Executive Service will be eligible to retire by 2016. Light,
A Government Ill-Executed
, 135.

*
This example now seems quaint in light of the revelations in 2013 about a rogue IRS unit’s targeting of conservative nonprofit groups for discriminatory treatment.

*
The federal government, the least unionized part of the public sector, is still more unionized than the most-unionized private industry (utilities). John Donahue,
The Warping of American Government Work
(2008), 77.

CHAPTER 11

Policy Successes

T
he preceding chapters have focused on policy failures, for three main reasons. First, there are so many significant failures. Taken individually, and especially in the aggregate, they exact a heavy toll on social welfare—under any plausible definition of that concept. Although some readers may view my criteria of success (see
chapter 2
) as too demanding, the criteria are really quite minimal. The very least that citizens can reasonably demand of a program is that it be socially beneficial on net, be cost-effective, meet its stated performance and cost goals, serve those who need it most, and use the most appropriate policy tools available. Second, and most obvious, failed policies are the ones that most need to be reformed or abandoned and that can therefore benefit most from the book’s diagnoses (
part 2
) and remedies (
chapter 12
). Third, triumphalist accounts of our system, sometimes referred to as American exceptionalism,
¹
are always in need of tempering. We do well to remember that our history is replete with failures.
*

But successes there have been, and they are worth revisiting here so that we might learn from them about why and to what extent they improved American society, and what light they can shed on today’s policy making. Although these programs are almost universally seen
as successes in retrospect, we shall see that the historical evidence tends to be more mixed, revealing some significant flaws that were criticized at the time or are apparent in retrospect.
²
For example, as noted below, while the Second Morrill Act wisely extended the land-grant system to more states, it also institutionalized segregated and inferior institutions for blacks, and the immense success of the Voting Rights Act is marred by its facilitation of perverse racial gerrymandering. More relevant to the book’s larger inquiry, some of these flaws roughly correspond to reasons for the endemic policy failures analyzed in earlier chapters.

After reviewing some historical successes, I shall devote the bulk of this chapter to contemporary ones, focusing on nine of them: (1) Title II of the Social Security Act (old age and survivors insurance); (2) the interstate highway system; (3) the food stamp program (now known as the Supplemental Nutrition Assistance Program); (4) the Voting Rights Act of 1965; (5) the Immigration Act of 1965; (6) the Earned Income Tax Credit of 1975; (7) the Airline Deregulation Act of 1978; (8) the Welfare Reform Act of 1996; and (9) the National Institutes of Health. Like earlier successes, these programs exhibit some important flaws, which are easier to identify (and in some cases, to quantify and perhaps rectify) than in the historical cases because we now have far better data, analytic techniques, and monitoring for policy assessment purposes. Some of these flaws are more readily rectified than others, but they serve to remind us of a fundamental point stressed in
chapter 2
: the term
policy success
is always a qualified, relative one; it is a matter of degree. The same is true for many
policy failures
, of course, although some of them—the War on Drugs, the ethanol program, and many policies directed at Native Americans, for example—appear to be unmitigated disasters (except to the narrow interests that benefit from them).

Finally, I am by no means suggesting that the policy successes discussed in this chapter are the only ones out there. Far from it. The National Institute of Standards and Technology facilitates commerce. Our national park system is peerless. Federal support for AIDS research
has led to great advances in curative techniques.
*
Despite occasional public health fiascos like the swine flu inoculation program,
³
the government can claim many important public health successes—for example, campaigns against lead poisoning, auto accidents, smoking,
†
and, most recently, the human papillomavirus.
⁴
The Centers for Disease Control and Prevention have gained great scientific credibility.
⁵

Some other policies are widely thought to have been successful, yet their governmental provenance is disputed—for example, the Internet.
⁶
Also, the government’s actual causal contributions to the improved social conditions may be questionable. The Civil Rights Act of 1964 is perhaps the most important example. The decline of discrimination on the basis of race, religion, gender, and sexual orientation in almost all areas of American life is stunning, albeit still incomplete.
⁷
Yet the complexity of social forces and the limits of law generally (see
chapter 8
) make it hard to know just how much the 1964 act produced this progress. Economist Thomas Sowell notes that the trajectory of black progress was the same before and after the act.
⁸
Some of the conditions that prompted its enactment were pushing independently in the same direction.
⁹
The act (against the expressed intent of its main sponsor, Hubert Humphrey) has spawned affirmative action, whose net benefits to equal opportunity are doubtful.
¹⁰
Economists June and Dave O’Neill recently reached a similar conclusion: that the civil rights laws—except for racial minorities in the South directly following the 1964 act—had little to do with the vastly improved job opportunities, wages, and working conditions for women and minorities, which overwhelmingly reflected other factors such as their greater education and skills.
¹¹
The same is true of the large presence
of women in the pharmacy industry.
¹²
While antidiscrimination laws surely played
some
positive role, they seem to have been but one factor among many. In contrast, the causal efficacy of the Voting Rights Act of 1965, discussed below, is incontrovertible.